From e05b59fe7927bc648ac3af3d59dc64a7ee6b22e2 Mon Sep 17 00:00:00 2001 From: Corey Minyard Date: Wed, 19 Apr 2006 22:40:53 +0200 Subject: [WATCHDOG] Pre-Timeout flags Some watchdog timers support the concept of a "pretimeout" which occurs some time before the real timeout. The pretimeout can be delivered via an interrupt or NMI and can be used to panic the system when it occurs (so you get useful information instead of a blind reboot). Signed-off-by: Corey Minyard Signed-off-by: Wim Van Sebroeck Signed-off-by: Andrew Morton --- include/linux/watchdog.h | 3 +++ 1 file changed, 3 insertions(+) (limited to 'include') diff --git a/include/linux/watchdog.h b/include/linux/watchdog.h index 1192ed8f4fe..a99c937f665 100644 --- a/include/linux/watchdog.h +++ b/include/linux/watchdog.h @@ -28,6 +28,8 @@ struct watchdog_info { #define WDIOC_KEEPALIVE _IOR(WATCHDOG_IOCTL_BASE, 5, int) #define WDIOC_SETTIMEOUT _IOWR(WATCHDOG_IOCTL_BASE, 6, int) #define WDIOC_GETTIMEOUT _IOR(WATCHDOG_IOCTL_BASE, 7, int) +#define WDIOC_SETPRETIMEOUT _IOWR(WATCHDOG_IOCTL_BASE, 8, int) +#define WDIOC_GETPRETIMEOUT _IOR(WATCHDOG_IOCTL_BASE, 9, int) #define WDIOF_UNKNOWN -1 /* Unknown flag error */ #define WDIOS_UNKNOWN -1 /* Unknown status error */ @@ -41,6 +43,7 @@ struct watchdog_info { #define WDIOF_POWEROVER 0x0040 /* Power over voltage */ #define WDIOF_SETTIMEOUT 0x0080 /* Set timeout (in seconds) */ #define WDIOF_MAGICCLOSE 0x0100 /* Supports magic close char */ +#define WDIOF_PRETIMEOUT 0x0200 /* Pretimeout (in seconds), get/set */ #define WDIOF_KEEPALIVEPING 0x8000 /* Keep alive ping reply */ #define WDIOS_DISABLECARD 0x0001 /* Turn off the watchdog timer */ -- cgit v1.2.3-70-g09d2 From 58b519f3e5e491d5a3e320dc525f58ac439bdde4 Mon Sep 17 00:00:00 2001 From: Wim Van Sebroeck Date: Sun, 21 May 2006 12:48:44 +0200 Subject: [WATCHDOG] add WDIOC_GETTIMELEFT ioctl Some watchdog drivers have the ability to report the remaining time before the system will reboot. With the WDIOC_GETTIMELEFT ioctl you can now read the time left before the watchdog would reboot your system. The following drivers support this new IOCTL: i8xx_tco.c, pcwd_pci.c and pcwd_usb.c . Signed-off-by: Wim Van Sebroeck Signed-off-by: Andrew Morton --- Documentation/watchdog/watchdog-api.txt | 9 +++++++++ drivers/char/watchdog/i8xx_tco.c | 28 +++++++++++++++++++++++++++- drivers/char/watchdog/pcwd_pci.c | 30 +++++++++++++++++++++++++++++- drivers/char/watchdog/pcwd_usb.c | 23 +++++++++++++++++++++++ include/linux/watchdog.h | 7 ++++--- 5 files changed, 92 insertions(+), 5 deletions(-) (limited to 'include') diff --git a/Documentation/watchdog/watchdog-api.txt b/Documentation/watchdog/watchdog-api.txt index 7dc2c1c6f77..d738ec25eaa 100644 --- a/Documentation/watchdog/watchdog-api.txt +++ b/Documentation/watchdog/watchdog-api.txt @@ -134,6 +134,15 @@ There is also a get function for getting the pretimeout: Not all watchdog drivers will support a pretimeout. +Get the number of seconds before reboot: + +Some watchdog drivers have the ability to report the remaining time +before the system will reboot. The WDIOC_GETTIMELEFT is the ioctl +that returns the number of seconds before reboot. + + ioctl(fd, WDIOC_GETTIMELEFT, &timeleft); + printf("The timeout was is %d seconds\n", timeleft); + Environmental monitoring: All watchdog drivers are required return more information about the system, diff --git a/drivers/char/watchdog/i8xx_tco.c b/drivers/char/watchdog/i8xx_tco.c index fa2ba9ebe42..bfbdbbf3c2f 100644 --- a/drivers/char/watchdog/i8xx_tco.c +++ b/drivers/char/watchdog/i8xx_tco.c @@ -205,6 +205,23 @@ static int tco_timer_set_heartbeat (int t) return 0; } +static int tco_timer_get_timeleft (int *time_left) +{ + unsigned char val; + + spin_lock(&tco_lock); + + /* read the TCO Timer */ + val = inb (TCO1_RLD); + val &= 0x3f; + + spin_unlock(&tco_lock); + + *time_left = (int)((val * 6) / 10); + + return 0; +} + /* * /dev/watchdog handling */ @@ -272,6 +289,7 @@ static int i8xx_tco_ioctl (struct inode *inode, struct file *file, { int new_options, retval = -EINVAL; int new_heartbeat; + int time_left; void __user *argp = (void __user *)arg; int __user *p = argp; static struct watchdog_info ident = { @@ -320,7 +338,7 @@ static int i8xx_tco_ioctl (struct inode *inode, struct file *file, return -EFAULT; if (tco_timer_set_heartbeat(new_heartbeat)) - return -EINVAL; + return -EINVAL; tco_timer_keepalive (); /* Fall */ @@ -329,6 +347,14 @@ static int i8xx_tco_ioctl (struct inode *inode, struct file *file, case WDIOC_GETTIMEOUT: return put_user(heartbeat, p); + case WDIOC_GETTIMELEFT: + { + if (tco_timer_get_timeleft(&time_left)) + return -EINVAL; + + return put_user(time_left, p); + } + default: return -ENOIOCTLCMD; } diff --git a/drivers/char/watchdog/pcwd_pci.c b/drivers/char/watchdog/pcwd_pci.c index 2451edbefec..1f40ecefbf7 100644 --- a/drivers/char/watchdog/pcwd_pci.c +++ b/drivers/char/watchdog/pcwd_pci.c @@ -21,7 +21,7 @@ */ /* - * A bells and whistles driver is available from: + * A bells and whistles driver is available from: * http://www.kernel.org/pub/linux/kernel/people/wim/pcwd/pcwd_pci/ * * More info available at http://www.berkprod.com/ or http://www.pcwatchdog.com/ @@ -390,6 +390,24 @@ static int pcipcwd_get_temperature(int *temperature) return 0; } +static int pcipcwd_get_timeleft(int *time_left) +{ + int msb; + int lsb; + + /* Read the time that's left before rebooting */ + /* Note: if the board is not yet armed then we will read 0xFFFF */ + send_command(CMD_READ_WATCHDOG_TIMEOUT, &msb, &lsb); + + *time_left = (msb << 8) + lsb; + + if (debug >= VERBOSE) + printk(KERN_DEBUG PFX "Time left before next reboot: %d\n", + *time_left); + + return 0; +} + /* * /dev/watchdog handling */ @@ -512,6 +530,16 @@ static int pcipcwd_ioctl(struct inode *inode, struct file *file, case WDIOC_GETTIMEOUT: return put_user(heartbeat, p); + case WDIOC_GETTIMELEFT: + { + int time_left; + + if (pcipcwd_get_timeleft(&time_left)) + return -EFAULT; + + return put_user(time_left, p); + } + default: return -ENOIOCTLCMD; } diff --git a/drivers/char/watchdog/pcwd_usb.c b/drivers/char/watchdog/pcwd_usb.c index 3fdfda9324f..0d072bed501 100644 --- a/drivers/char/watchdog/pcwd_usb.c +++ b/drivers/char/watchdog/pcwd_usb.c @@ -317,6 +317,19 @@ static int usb_pcwd_get_temperature(struct usb_pcwd_private *usb_pcwd, int *temp return 0; } +static int usb_pcwd_get_timeleft(struct usb_pcwd_private *usb_pcwd, int *time_left) +{ + unsigned char msb, lsb; + + /* Read the time that's left before rebooting */ + /* Note: if the board is not yet armed then we will read 0xFFFF */ + usb_pcwd_send_command(usb_pcwd, CMD_READ_WATCHDOG_TIMEOUT, &msb, &lsb); + + *time_left = (msb << 8) + lsb; + + return 0; +} + /* * /dev/watchdog handling */ @@ -422,6 +435,16 @@ static int usb_pcwd_ioctl(struct inode *inode, struct file *file, case WDIOC_GETTIMEOUT: return put_user(heartbeat, p); + case WDIOC_GETTIMELEFT: + { + int time_left; + + if (usb_pcwd_get_timeleft(usb_pcwd_device, &time_left)) + return -EFAULT; + + return put_user(time_left, p); + } + default: return -ENOIOCTLCMD; } diff --git a/include/linux/watchdog.h b/include/linux/watchdog.h index a99c937f665..011bcfeb9f0 100644 --- a/include/linux/watchdog.h +++ b/include/linux/watchdog.h @@ -30,6 +30,7 @@ struct watchdog_info { #define WDIOC_GETTIMEOUT _IOR(WATCHDOG_IOCTL_BASE, 7, int) #define WDIOC_SETPRETIMEOUT _IOWR(WATCHDOG_IOCTL_BASE, 8, int) #define WDIOC_GETPRETIMEOUT _IOR(WATCHDOG_IOCTL_BASE, 9, int) +#define WDIOC_GETTIMELEFT _IOR(WATCHDOG_IOCTL_BASE, 10, int) #define WDIOF_UNKNOWN -1 /* Unknown flag error */ #define WDIOS_UNKNOWN -1 /* Unknown status error */ @@ -40,9 +41,9 @@ struct watchdog_info { #define WDIOF_EXTERN2 0x0008 /* External relay 2 */ #define WDIOF_POWERUNDER 0x0010 /* Power bad/power fault */ #define WDIOF_CARDRESET 0x0020 /* Card previously reset the CPU */ -#define WDIOF_POWEROVER 0x0040 /* Power over voltage */ -#define WDIOF_SETTIMEOUT 0x0080 /* Set timeout (in seconds) */ -#define WDIOF_MAGICCLOSE 0x0100 /* Supports magic close char */ +#define WDIOF_POWEROVER 0x0040 /* Power over voltage */ +#define WDIOF_SETTIMEOUT 0x0080 /* Set timeout (in seconds) */ +#define WDIOF_MAGICCLOSE 0x0100 /* Supports magic close char */ #define WDIOF_PRETIMEOUT 0x0200 /* Pretimeout (in seconds), get/set */ #define WDIOF_KEEPALIVEPING 0x8000 /* Keep alive ping reply */ -- cgit v1.2.3-70-g09d2 From d8deac5094988c7ad1127ee61f52c59a952fcabb Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Mon, 20 Jun 2005 21:15:16 -0700 Subject: [PATCH] devfs: Remove devfs from the kernel tree This is the first patch in a series of patches that removes devfs support from the kernel. This patch removes the core devfs code, and its private header file. Signed-off-by: Greg Kroah-Hartman --- fs/Makefile | 1 - fs/compat_ioctl.c | 1 - fs/devfs/Makefile | 8 - fs/devfs/base.c | 2836 --------------------------------------- fs/devfs/util.c | 97 -- include/linux/compat_ioctl.h | 5 - include/linux/devfs_fs.h | 41 - include/linux/devfs_fs_kernel.h | 18 - 8 files changed, 3007 deletions(-) delete mode 100644 fs/devfs/Makefile delete mode 100644 fs/devfs/base.c delete mode 100644 fs/devfs/util.c delete mode 100644 include/linux/devfs_fs.h (limited to 'include') diff --git a/fs/Makefile b/fs/Makefile index d0ea6bfccf2..89135428a53 100644 --- a/fs/Makefile +++ b/fs/Makefile @@ -66,7 +66,6 @@ obj-$(CONFIG_MSDOS_FS) += msdos/ obj-$(CONFIG_VFAT_FS) += vfat/ obj-$(CONFIG_BFS_FS) += bfs/ obj-$(CONFIG_ISO9660_FS) += isofs/ -obj-$(CONFIG_DEVFS_FS) += devfs/ obj-$(CONFIG_HFSPLUS_FS) += hfsplus/ # Before hfs to find wrapped HFS+ obj-$(CONFIG_HFS_FS) += hfs/ obj-$(CONFIG_VXFS_FS) += freevxfs/ diff --git a/fs/compat_ioctl.c b/fs/compat_ioctl.c index d8ecfedef18..d8d50a70c58 100644 --- a/fs/compat_ioctl.c +++ b/fs/compat_ioctl.c @@ -44,7 +44,6 @@ #include #include #include -#include #include #include #include diff --git a/fs/devfs/Makefile b/fs/devfs/Makefile deleted file mode 100644 index 6dd8d1245e2..00000000000 --- a/fs/devfs/Makefile +++ /dev/null @@ -1,8 +0,0 @@ -# -# Makefile for the linux devfs-filesystem routines. -# - -obj-$(CONFIG_DEVFS_FS) += devfs.o - -devfs-objs := base.o util.o - diff --git a/fs/devfs/base.c b/fs/devfs/base.c deleted file mode 100644 index 51a97f13274..00000000000 --- a/fs/devfs/base.c +++ /dev/null @@ -1,2836 +0,0 @@ -/* devfs (Device FileSystem) driver. - - Copyright (C) 1998-2002 Richard Gooch - - This library is free software; you can redistribute it and/or - modify it under the terms of the GNU Library General Public - License as published by the Free Software Foundation; either - version 2 of the License, or (at your option) any later version. - - This library is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - Library General Public License for more details. - - You should have received a copy of the GNU Library General Public - License along with this library; if not, write to the Free - Software Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. - - Richard Gooch may be reached by email at rgooch@atnf.csiro.au - The postal address is: - Richard Gooch, c/o ATNF, P. O. Box 76, Epping, N.S.W., 2121, Australia. - - ChangeLog - - 19980110 Richard Gooch - Original version. - v0.1 - 19980111 Richard Gooch - Created per-fs inode table rather than using inode->u.generic_ip - v0.2 - 19980111 Richard Gooch - Created .epoch inode which has a ctime of 0. - Fixed loss of named pipes when dentries lost. - Fixed loss of inode data when devfs_register() follows mknod(). - v0.3 - 19980111 Richard Gooch - Fix for when compiling with CONFIG_KERNELD. - 19980112 Richard Gooch - Fix for readdir() which sometimes didn't show entries. - Added <> option to . - v0.4 - 19980113 Richard Gooch - Created function. - v0.5 - 19980115 Richard Gooch - Added subdirectory support. Major restructuring. - 19980116 Richard Gooch - Fixed to not search major=0,minor=0. - Added symlink support. - v0.6 - 19980120 Richard Gooch - Created function and support directory unregister - 19980120 Richard Gooch - Auto-ownership uses real uid/gid rather than effective uid/gid. - v0.7 - 19980121 Richard Gooch - Supported creation of sockets. - v0.8 - 19980122 Richard Gooch - Added DEVFS_FL_HIDE_UNREG flag. - Interface change to . - Created to support symlink(2). - v0.9 - 19980123 Richard Gooch - Added check to to check inode is in devfs. - Added optional traversal of symlinks. - v0.10 - 19980124 Richard Gooch - Created and . - v0.11 - 19980125 C. Scott Ananian - Created . - 19980125 Richard Gooch - Allow removal of symlinks. - v0.12 - 19980125 Richard Gooch - Created . - 19980126 Richard Gooch - Moved DEVFS_SUPER_MAGIC into header file. - Added DEVFS_FL_HIDE flag. - Created . - Created . - Fixed minor bug in . - 19980127 Richard Gooch - Changed interface to , , - , and . - Fixed inode times when symlink created with symlink(2). - v0.13 - 19980129 C. Scott Ananian - Exported , - and . - 19980129 Richard Gooch - Created to support unlink(2). - v0.14 - 19980129 Richard Gooch - Fixed kerneld support for entries in devfs subdirectories. - 19980130 Richard Gooch - Bugfixes in . - v0.15 - 19980207 Richard Gooch - Call kerneld when looking up unregistered entries. - v0.16 - 19980326 Richard Gooch - Modified interface to for symlink traversal. - v0.17 - 19980331 Richard Gooch - Fixed persistence bug with device numbers for manually created - device files. - Fixed problem with recreating symlinks with different content. - v0.18 - 19980401 Richard Gooch - Changed to CONFIG_KMOD. - Hide entries which are manually unlinked. - Always invalidate devfs dentry cache when registering entries. - Created to support rmdir(2). - Ensure directories created by are visible. - v0.19 - 19980402 Richard Gooch - Invalidate devfs dentry cache when making directories. - Invalidate devfs dentry cache when removing entries. - Fixed persistence bug with fifos. - v0.20 - 19980421 Richard Gooch - Print process command when debugging kerneld/kmod. - Added debugging for register/unregister/change operations. - 19980422 Richard Gooch - Added "devfs=" boot options. - v0.21 - 19980426 Richard Gooch - No longer lock/unlock superblock in . - Drop negative dentries when they are released. - Manage dcache more efficiently. - v0.22 - 19980427 Richard Gooch - Added DEVFS_FL_AUTO_DEVNUM flag. - v0.23 - 19980430 Richard Gooch - No longer set unnecessary methods. - v0.24 - 19980504 Richard Gooch - Added PID display to debugging message. - Added "after" debugging message to . - 19980519 Richard Gooch - Added "diread" and "diwrite" boot options. - 19980520 Richard Gooch - Fixed persistence problem with permissions. - v0.25 - 19980602 Richard Gooch - Support legacy device nodes. - Fixed bug where recreated inodes were hidden. - v0.26 - 19980602 Richard Gooch - Improved debugging in . - 19980607 Richard Gooch - No longer free old dentries in . - Free all dentries for a given entry when deleting inodes. - v0.27 - 19980627 Richard Gooch - Limit auto-device numbering to majors 128 to 239. - v0.28 - 19980629 Richard Gooch - Fixed inode times persistence problem. - v0.29 - 19980704 Richard Gooch - Fixed spelling in debug. - Fixed bug in parsing "dilookup". - v0.30 - 19980705 Richard Gooch - Fixed devfs inode leak when manually recreating inodes. - Fixed permission persistence problem when recreating inodes. - v0.31 - 19980727 Richard Gooch - Removed harmless "unused variable" compiler warning. - Fixed modes for manually recreated device nodes. - v0.32 - 19980728 Richard Gooch - Added NULL devfs inode warning in . - Force all inode nlink values to 1. - v0.33 - 19980730 Richard Gooch - Added "dimknod" boot option. - Set inode nlink to 0 when freeing dentries. - Fixed modes for manually recreated symlinks. - v0.34 - 19980802 Richard Gooch - Fixed bugs in recreated directories and symlinks. - v0.35 - 19980806 Richard Gooch - Fixed bugs in recreated device nodes. - 19980807 Richard Gooch - Fixed bug in currently unused . - Defined new type. - Improved debugging when getting entries. - Fixed bug where directories could be emptied. - v0.36 - 19980809 Richard Gooch - Replaced dummy .epoch inode with .devfsd character device. - 19980810 Richard Gooch - Implemented devfsd protocol revision 0. - v0.37 - 19980819 Richard Gooch - Added soothing message to warning in . - v0.38 - 19980829 Richard Gooch - Use GCC extensions for structure initialisations. - Implemented async open notification. - Incremented devfsd protocol revision to 1. - v0.39 - 19980908 Richard Gooch - Moved async open notification to end of . - v0.40 - 19980910 Richard Gooch - Prepended "/dev/" to module load request. - Renamed to . - v0.41 - 19980910 Richard Gooch - Fixed typo "AYSNC" -> "ASYNC". - v0.42 - 19980910 Richard Gooch - Added open flag for files. - v0.43 - 19980927 Richard Gooch - Set i_blocks=0 and i_blksize=1024 in . - v0.44 - 19981005 Richard Gooch - Added test for empty <> in . - Renamed to and published. - v0.45 - 19981006 Richard Gooch - Created . - v0.46 - 19981007 Richard Gooch - Limit auto-device numbering to majors 144 to 239. - v0.47 - 19981010 Richard Gooch - Updated for VFS change in 2.1.125. - v0.48 - 19981022 Richard Gooch - Created DEVFS_ FL_COMPAT flag. - v0.49 - 19981023 Richard Gooch - Created "nocompat" boot option. - v0.50 - 19981025 Richard Gooch - Replaced "mount" boot option with "nomount". - v0.51 - 19981110 Richard Gooch - Created "only" boot option. - v0.52 - 19981112 Richard Gooch - Added DEVFS_FL_REMOVABLE flag. - v0.53 - 19981114 Richard Gooch - Only call on first call to - . - v0.54 - 19981205 Richard Gooch - Updated for VFS change in 2.1.131. - v0.55 - 19981218 Richard Gooch - Created . - 19981220 Richard Gooch - Check for partitions on removable media in . - v0.56 - 19990118 Richard Gooch - Added support for registering regular files. - Created . - Update devfs inodes from entries if not changed through FS. - v0.57 - 19990124 Richard Gooch - Fixed to only initialise temporary inodes. - Trap for NULL fops in . - Return -ENODEV in for non-driver inodes. - v0.58 - 19990126 Richard Gooch - Switched from PATH_MAX to DEVFS_PATHLEN. - v0.59 - 19990127 Richard Gooch - Created "nottycompat" boot option. - v0.60 - 19990318 Richard Gooch - Fixed to not overrun event buffer. - v0.61 - 19990329 Richard Gooch - Created . - v0.62 - 19990330 Richard Gooch - Don't return unregistred entries in . - Panic in if entry unregistered. - 19990401 Richard Gooch - Don't panic in for duplicates. - v0.63 - 19990402 Richard Gooch - Don't unregister already unregistered entries in . - v0.64 - 19990510 Richard Gooch - Disable warning messages when unable to read partition table for - removable media. - v0.65 - 19990512 Richard Gooch - Updated for VFS change in 2.3.1-pre1. - Created "oops-on-panic" boot option. - Improved debugging in and . - v0.66 - 19990519 Richard Gooch - Added documentation for some functions. - 19990525 Richard Gooch - Removed "oops-on-panic" boot option: now always Oops. - v0.67 - 19990531 Richard Gooch - Improved debugging in . - v0.68 - 19990604 Richard Gooch - Added "diunlink" and "nokmod" boot options. - Removed superfluous warning message in . - v0.69 - 19990611 Richard Gooch - Took account of change to . - v0.70 - 19990614 Richard Gooch - Created separate event queue for each mounted devfs. - Removed . - Created new ioctl()s. - Incremented devfsd protocol revision to 3. - Fixed bug when re-creating directories: contents were lost. - Block access to inodes until devfsd updates permissions. - 19990615 Richard Gooch - Support 2.2.x kernels. - v0.71 - 19990623 Richard Gooch - Switched to sending process uid/gid to devfsd. - Renamed to . - Added DEVFSD_NOTIFY_LOOKUP event. - 19990624 Richard Gooch - Added DEVFSD_NOTIFY_CHANGE event. - Incremented devfsd protocol revision to 4. - v0.72 - 19990713 Richard Gooch - Return EISDIR rather than EINVAL for read(2) on directories. - v0.73 - 19990809 Richard Gooch - Changed to new __init scheme. - v0.74 - 19990901 Richard Gooch - Changed remaining function declarations to new __init scheme. - v0.75 - 19991013 Richard Gooch - Created , , - and . - Added <> parameter to , , - and . - Work sponsored by SGI. - v0.76 - 19991017 Richard Gooch - Allow multiple unregistrations. - Work sponsored by SGI. - v0.77 - 19991026 Richard Gooch - Added major and minor number to devfsd protocol. - Incremented devfsd protocol revision to 5. - Work sponsored by SGI. - v0.78 - 19991030 Richard Gooch - Support info pointer for all devfs entry types. - Added <> parameter to and - . - Work sponsored by SGI. - v0.79 - 19991031 Richard Gooch - Support "../" when searching devfs namespace. - Work sponsored by SGI. - v0.80 - 19991101 Richard Gooch - Created . - Work sponsored by SGI. - v0.81 - 19991103 Richard Gooch - Exported . - Work sponsored by SGI. - v0.82 - 19991104 Richard Gooch - Removed unused . - 19991105 Richard Gooch - Do not hide entries from devfsd or children. - Removed DEVFS_ FL_TTY_COMPAT flag. - Removed "nottycompat" boot option. - Removed . - Work sponsored by SGI. - v0.83 - 19991107 Richard Gooch - Added DEVFS_FL_WAIT flag. - Work sponsored by SGI. - v0.84 - 19991107 Richard Gooch - Support new "disc" naming scheme in . - Allow NULL fops in . - Work sponsored by SGI. - v0.85 - 19991110 Richard Gooch - Fall back to major table if NULL fops given to . - Work sponsored by SGI. - v0.86 - 19991204 Richard Gooch - Support fifos when unregistering. - Work sponsored by SGI. - v0.87 - 19991209 Richard Gooch - Removed obsolete DEVFS_ FL_COMPAT and DEVFS_ FL_TOLERANT flags. - Work sponsored by SGI. - v0.88 - 19991214 Richard Gooch - Removed kmod support. - Work sponsored by SGI. - v0.89 - 19991216 Richard Gooch - Improved debugging in . - Ensure dentries created by devfsd will be cleaned up. - Work sponsored by SGI. - v0.90 - 19991223 Richard Gooch - Created . - Work sponsored by SGI. - v0.91 - 20000203 Richard Gooch - Ported to kernel 2.3.42. - Removed . - Work sponsored by SGI. - v0.92 - 20000306 Richard Gooch - Added DEVFS_ FL_NO_PERSISTENCE flag. - Removed unnecessary call to in - . - Work sponsored by SGI. - v0.93 - 20000413 Richard Gooch - Set inode->i_size to correct size for symlinks. - 20000414 Richard Gooch - Only give lookup() method to directories to comply with new VFS - assumptions. - Work sponsored by SGI. - 20000415 Richard Gooch - Remove unnecessary tests in symlink methods. - Don't kill existing block ops in . - Work sponsored by SGI. - v0.94 - 20000424 Richard Gooch - Don't create missing directories in . - Work sponsored by SGI. - v0.95 - 20000430 Richard Gooch - Added CONFIG_DEVFS_MOUNT. - Work sponsored by SGI. - v0.96 - 20000608 Richard Gooch - Disabled multi-mount capability (use VFS bindings instead). - Work sponsored by SGI. - v0.97 - 20000610 Richard Gooch - Switched to FS_SINGLE to disable multi-mounts. - 20000612 Richard Gooch - Removed module support. - Removed multi-mount code. - Removed compatibility macros: VFS has changed too much. - Work sponsored by SGI. - v0.98 - 20000614 Richard Gooch - Merged devfs inode into devfs entry. - Work sponsored by SGI. - v0.99 - 20000619 Richard Gooch - Removed dead code in which used to call - . - Work sponsored by SGI. - v0.100 - 20000621 Richard Gooch - Changed interface to . - Work sponsored by SGI. - v0.101 - 20000622 Richard Gooch - Simplified interface to and . - Simplified interface to . - Work sponsored by SGI. - v0.102 - 20010519 Richard Gooch - Ensure terminates string for root entry. - Exported to modules. - 20010520 Richard Gooch - Make send events to devfsd. - Cleaned up option processing in . - 20010521 Richard Gooch - Fixed bugs in handling symlinks: could leak or cause Oops. - 20010522 Richard Gooch - Cleaned up directory handling by separating fops. - v0.103 - 20010601 Richard Gooch - Fixed handling of inverted options in . - v0.104 - 20010604 Richard Gooch - Adjusted to account for fix. - v0.105 - 20010617 Richard Gooch - Answered question posed by Al Viro and removed his comments. - Moved setting of registered flag after other fields are changed. - Fixed race between and . - Global VFS changes added bogus BKL to : removed. - Widened locking in and . - Replaced stack usage with kmalloc. - Simplified locking in and fixed memory leak. - v0.106 - 20010709 Richard Gooch - Removed broken devnum allocation and use . - Fixed old devnum leak by calling new . - v0.107 - 20010712 Richard Gooch - Fixed bug in which could hang boot process. - v0.108 - 20010730 Richard Gooch - Added DEVFSD_NOTIFY_DELETE event. - 20010801 Richard Gooch - Removed #include . - v0.109 - 20010807 Richard Gooch - Fixed inode table races by removing it and using - inode->u.generic_ip instead. - Moved into . - Moved into . - v0.110 - 20010808 Richard Gooch - Fixed race in for uni-processor. - v0.111 - 20010818 Richard Gooch - Removed remnant of multi-mount support in . - Removed unused DEVFS_FL_SHOW_UNREG flag. - v0.112 - 20010820 Richard Gooch - Removed nlink field from struct devfs_inode. - v0.113 - 20010823 Richard Gooch - Replaced BKL with global rwsem to protect symlink data (quick - and dirty hack). - v0.114 - 20010827 Richard Gooch - Replaced global rwsem for symlink with per-link refcount. - v0.115 - 20010919 Richard Gooch - Set inode->i_mapping->a_ops for block nodes in . - v0.116 - 20011008 Richard Gooch - Fixed overrun in by removing function (not needed). - 20011009 Richard Gooch - Fixed buffer underrun in . - 20011029 Richard Gooch - Fixed race in when setting event mask. - 20011114 Richard Gooch - First release of new locking code. - v1.0 - 20011117 Richard Gooch - Discard temporary buffer, now use "%s" for dentry names. - 20011118 Richard Gooch - Don't generate path in : use fake entry instead. - Use "existing" directory in <_devfs_make_parent_for_leaf>. - 20011122 Richard Gooch - Use slab cache rather than fixed buffer for devfsd events. - v1.1 - 20011125 Richard Gooch - Send DEVFSD_NOTIFY_REGISTERED events in . - 20011127 Richard Gooch - Fixed locking bug in due to typo. - Do not send CREATE, CHANGE, ASYNC_OPEN or DELETE events from - devfsd or children. - v1.2 - 20011202 Richard Gooch - Fixed bug in : was dereferencing freed pointer. - v1.3 - 20011203 Richard Gooch - Fixed bug in : was dereferencing freed pointer. - Added process group check for devfsd privileges. - v1.4 - 20011204 Richard Gooch - Use SLAB_ATOMIC in from . - v1.5 - 20011211 Richard Gooch - Return old entry in for 2.4.x kernels. - 20011212 Richard Gooch - Increment refcount on module in . - 20011215 Richard Gooch - Created and exported . - Increment refcount on module in . - Created . - v1.6 - 20011216 Richard Gooch - Added poisoning to . - Improved debugging messages. - v1.7 - 20011221 Richard Gooch - Corrected (made useful) debugging message in . - Moved in to - 20011224 Richard Gooch - Added magic number to guard against scribbling drivers. - 20011226 Richard Gooch - Only return old entry in if a directory. - Defined macros for error and debug messages. - v1.8 - 20020113 Richard Gooch - Fixed (rare, old) race in . - v1.9 - 20020120 Richard Gooch - Fixed deadlock bug in . - Tag VFS deletable in if handle ignored. - v1.10 - 20020129 Richard Gooch - Added KERN_* to remaining messages. - Cleaned up declaration of . - v1.11 - 20020219 Richard Gooch - Changed to allow later additions if not yet empty. - v1.12 - 20020406 Richard Gooch - Removed silently introduced calls to lock_kernel() and - unlock_kernel() due to recent VFS locking changes. BKL isn't - required in devfs. - v1.13 - 20020428 Richard Gooch - Removed 2.4.x compatibility code. - v1.14 - 20020510 Richard Gooch - Added BKL to because drivers still need it. - v1.15 - 20020512 Richard Gooch - Protected and - from changing directory contents. - v1.16 - 20020514 Richard Gooch - Minor cleanup of . - v1.17 - 20020721 Richard Gooch - Switched to ISO C structure field initialisers. - Switch to set_current_state() and move before add_wait_queue(). - 20020722 Richard Gooch - Fixed devfs entry leak in when *readdir fails. - v1.18 - 20020725 Richard Gooch - Created . - v1.19 - 20020728 Richard Gooch - Removed deprecated . - v1.20 - 20020820 Richard Gooch - Fixed module unload race in . - v1.21 - 20021013 Richard Gooch - Removed DEVFS_ FL_AUTO_OWNER. - Switched lingering structure field initialiser to ISO C. - Added locking when updating FCB flags. - v1.22 -*/ -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include - -#include -#include -#include -#include -#include -#include - -#define DEVFS_VERSION "2004-01-31" - -#define DEVFS_NAME "devfs" - -#define FIRST_INODE 1 - -#define STRING_LENGTH 256 -#define FAKE_BLOCK_SIZE 1024 -#define POISON_PTR ( *(void **) poison_array ) -#define MAGIC_VALUE 0x327db823 - -#ifndef TRUE -# define TRUE 1 -# define FALSE 0 -#endif - -#define MODE_DIR (S_IFDIR | S_IWUSR | S_IRUGO | S_IXUGO) - -#define DEBUG_NONE 0x0000000 -#define DEBUG_MODULE_LOAD 0x0000001 -#define DEBUG_REGISTER 0x0000002 -#define DEBUG_UNREGISTER 0x0000004 -#define DEBUG_FREE 0x0000008 -#define DEBUG_SET_FLAGS 0x0000010 -#define DEBUG_S_READ 0x0000100 /* Break */ -#define DEBUG_I_LOOKUP 0x0001000 /* Break */ -#define DEBUG_I_CREATE 0x0002000 -#define DEBUG_I_GET 0x0004000 -#define DEBUG_I_CHANGE 0x0008000 -#define DEBUG_I_UNLINK 0x0010000 -#define DEBUG_I_RLINK 0x0020000 -#define DEBUG_I_FLINK 0x0040000 -#define DEBUG_I_MKNOD 0x0080000 -#define DEBUG_F_READDIR 0x0100000 /* Break */ -#define DEBUG_D_DELETE 0x1000000 /* Break */ -#define DEBUG_D_RELEASE 0x2000000 -#define DEBUG_D_IPUT 0x4000000 -#define DEBUG_ALL 0xfffffff -#define DEBUG_DISABLED DEBUG_NONE - -#define OPTION_NONE 0x00 -#define OPTION_MOUNT 0x01 - -#define PRINTK(format, args...) \ - {printk (KERN_ERR "%s" format, __FUNCTION__ , ## args);} - -#define OOPS(format, args...) \ - {printk (KERN_CRIT "%s" format, __FUNCTION__ , ## args); \ - printk ("Forcing Oops\n"); \ - BUG();} - -#ifdef CONFIG_DEVFS_DEBUG -# define VERIFY_ENTRY(de) \ - {if ((de) && (de)->magic_number != MAGIC_VALUE) \ - OOPS ("(%p): bad magic value: %x\n", (de), (de)->magic_number);} -# define WRITE_ENTRY_MAGIC(de,magic) (de)->magic_number = (magic) -# define DPRINTK(flag, format, args...) \ - {if (devfs_debug & flag) \ - printk (KERN_INFO "%s" format, __FUNCTION__ , ## args);} -#else -# define VERIFY_ENTRY(de) -# define WRITE_ENTRY_MAGIC(de,magic) -# define DPRINTK(flag, format, args...) -#endif - -typedef struct devfs_entry *devfs_handle_t; - -struct directory_type { - rwlock_t lock; /* Lock for searching(R)/updating(W) */ - struct devfs_entry *first; - struct devfs_entry *last; - unsigned char no_more_additions:1; -}; - -struct symlink_type { - unsigned int length; /* Not including the NULL-termimator */ - char *linkname; /* This is NULL-terminated */ -}; - -struct devfs_inode { /* This structure is for "persistent" inode storage */ - struct dentry *dentry; - struct timespec atime; - struct timespec mtime; - struct timespec ctime; - unsigned int ino; /* Inode number as seen in the VFS */ - uid_t uid; - gid_t gid; -}; - -struct devfs_entry { -#ifdef CONFIG_DEVFS_DEBUG - unsigned int magic_number; -#endif - void *info; - atomic_t refcount; /* When this drops to zero, it's unused */ - union { - struct directory_type dir; - dev_t dev; - struct symlink_type symlink; - const char *name; /* Only used for (mode == 0) */ - } u; - struct devfs_entry *prev; /* Previous entry in the parent directory */ - struct devfs_entry *next; /* Next entry in the parent directory */ - struct devfs_entry *parent; /* The parent directory */ - struct devfs_inode inode; - umode_t mode; - unsigned short namelen; /* I think 64k+ filenames are a way off... */ - unsigned char vfs:1; /* Whether the VFS may delete the entry */ - char name[1]; /* This is just a dummy: the allocated array - is bigger. This is NULL-terminated */ -}; - -/* The root of the device tree */ -static struct devfs_entry *root_entry; - -struct devfsd_buf_entry { - struct devfs_entry *de; /* The name is generated with this */ - unsigned short type; /* The type of event */ - umode_t mode; - uid_t uid; - gid_t gid; - struct devfsd_buf_entry *next; -}; - -struct fs_info { /* This structure is for the mounted devfs */ - struct super_block *sb; - spinlock_t devfsd_buffer_lock; /* Lock when inserting/deleting events */ - struct devfsd_buf_entry *devfsd_first_event; - struct devfsd_buf_entry *devfsd_last_event; - volatile int devfsd_sleeping; - volatile struct task_struct *devfsd_task; - volatile pid_t devfsd_pgrp; - volatile struct file *devfsd_file; - struct devfsd_notify_struct *devfsd_info; - volatile unsigned long devfsd_event_mask; - atomic_t devfsd_overrun_count; - wait_queue_head_t devfsd_wait_queue; /* Wake devfsd on input */ - wait_queue_head_t revalidate_wait_queue; /* Wake when devfsd sleeps */ -}; - -static struct fs_info fs_info = {.devfsd_buffer_lock = SPIN_LOCK_UNLOCKED }; -static kmem_cache_t *devfsd_buf_cache; -#ifdef CONFIG_DEVFS_DEBUG -static unsigned int devfs_debug_init __initdata = DEBUG_NONE; -static unsigned int devfs_debug = DEBUG_NONE; -static DEFINE_SPINLOCK(stat_lock); -static unsigned int stat_num_entries; -static unsigned int stat_num_bytes; -#endif -static unsigned char poison_array[8] = - { 0x5a, 0x5a, 0x5a, 0x5a, 0x5a, 0x5a, 0x5a, 0x5a }; - -#ifdef CONFIG_DEVFS_MOUNT -static unsigned int boot_options = OPTION_MOUNT; -#else -static unsigned int boot_options = OPTION_NONE; -#endif - -/* Forward function declarations */ -static devfs_handle_t _devfs_walk_path(struct devfs_entry *dir, - const char *name, int namelen, - int traverse_symlink); -static ssize_t devfsd_read(struct file *file, char __user *buf, size_t len, - loff_t * ppos); -static int devfsd_ioctl(struct inode *inode, struct file *file, - unsigned int cmd, unsigned long arg); -static int devfsd_close(struct inode *inode, struct file *file); -#ifdef CONFIG_DEVFS_DEBUG -static ssize_t stat_read(struct file *file, char __user *buf, size_t len, - loff_t * ppos); -static const struct file_operations stat_fops = { - .open = nonseekable_open, - .read = stat_read, -}; -#endif - -/* Devfs daemon file operations */ -static const struct file_operations devfsd_fops = { - .open = nonseekable_open, - .read = devfsd_read, - .ioctl = devfsd_ioctl, - .release = devfsd_close, -}; - -/* Support functions follow */ - -/** - * devfs_get - Get a reference to a devfs entry. - * @de: The devfs entry. - */ - -static struct devfs_entry *devfs_get(struct devfs_entry *de) -{ - VERIFY_ENTRY(de); - if (de) - atomic_inc(&de->refcount); - return de; -} /* End Function devfs_get */ - -/** - * devfs_put - Put (release) a reference to a devfs entry. - * @de: The handle to the devfs entry. - */ - -static void devfs_put(devfs_handle_t de) -{ - if (!de) - return; - VERIFY_ENTRY(de); - if (de->info == POISON_PTR) - OOPS("(%p): poisoned pointer\n", de); - if (!atomic_dec_and_test(&de->refcount)) - return; - if (de == root_entry) - OOPS("(%p): root entry being freed\n", de); - DPRINTK(DEBUG_FREE, "(%s): de: %p, parent: %p \"%s\"\n", - de->name, de, de->parent, - de->parent ? de->parent->name : "no parent"); - if (S_ISLNK(de->mode)) - kfree(de->u.symlink.linkname); - WRITE_ENTRY_MAGIC(de, 0); -#ifdef CONFIG_DEVFS_DEBUG - spin_lock(&stat_lock); - --stat_num_entries; - stat_num_bytes -= sizeof *de + de->namelen; - if (S_ISLNK(de->mode)) - stat_num_bytes -= de->u.symlink.length + 1; - spin_unlock(&stat_lock); -#endif - de->info = POISON_PTR; - kfree(de); -} /* End Function devfs_put */ - -/** - * _devfs_search_dir - Search for a devfs entry in a directory. - * @dir: The directory to search. - * @name: The name of the entry to search for. - * @namelen: The number of characters in @name. - * - * Search for a devfs entry in a directory and returns a pointer to the entry - * on success, else %NULL. The directory must be locked already. - * An implicit devfs_get() is performed on the returned entry. - */ - -static struct devfs_entry *_devfs_search_dir(struct devfs_entry *dir, - const char *name, - unsigned int namelen) -{ - struct devfs_entry *curr; - - if (!S_ISDIR(dir->mode)) { - PRINTK("(%s): not a directory\n", dir->name); - return NULL; - } - for (curr = dir->u.dir.first; curr != NULL; curr = curr->next) { - if (curr->namelen != namelen) - continue; - if (memcmp(curr->name, name, namelen) == 0) - break; - /* Not found: try the next one */ - } - return devfs_get(curr); -} /* End Function _devfs_search_dir */ - -/** - * _devfs_alloc_entry - Allocate a devfs entry. - * @name: the name of the entry - * @namelen: the number of characters in @name - * @mode: the mode for the entry - * - * Allocate a devfs entry and returns a pointer to the entry on success, else - * %NULL. - */ - -static struct devfs_entry *_devfs_alloc_entry(const char *name, - unsigned int namelen, - umode_t mode) -{ - struct devfs_entry *new; - static unsigned long inode_counter = FIRST_INODE; - static DEFINE_SPINLOCK(counter_lock); - - if (name && (namelen < 1)) - namelen = strlen(name); - if ((new = kmalloc(sizeof *new + namelen, GFP_KERNEL)) == NULL) - return NULL; - memset(new, 0, sizeof *new + namelen); /* Will set '\0' on name */ - new->mode = mode; - if (S_ISDIR(mode)) - rwlock_init(&new->u.dir.lock); - atomic_set(&new->refcount, 1); - spin_lock(&counter_lock); - new->inode.ino = inode_counter++; - spin_unlock(&counter_lock); - if (name) - memcpy(new->name, name, namelen); - new->namelen = namelen; - WRITE_ENTRY_MAGIC(new, MAGIC_VALUE); -#ifdef CONFIG_DEVFS_DEBUG - spin_lock(&stat_lock); - ++stat_num_entries; - stat_num_bytes += sizeof *new + namelen; - spin_unlock(&stat_lock); -#endif - return new; -} /* End Function _devfs_alloc_entry */ - -/** - * _devfs_append_entry - Append a devfs entry to a directory's child list. - * @dir: The directory to add to. - * @de: The devfs entry to append. - * @old_de: If an existing entry exists, it will be written here. This may - * be %NULL. An implicit devfs_get() is performed on this entry. - * - * Append a devfs entry to a directory's list of children, checking first to - * see if an entry of the same name exists. The directory will be locked. - * The value 0 is returned on success, else a negative error code. - * On failure, an implicit devfs_put() is performed on %de. - */ - -static int _devfs_append_entry(devfs_handle_t dir, devfs_handle_t de, - devfs_handle_t * old_de) -{ - int retval; - - if (old_de) - *old_de = NULL; - if (!S_ISDIR(dir->mode)) { - PRINTK("(%s): dir: \"%s\" is not a directory\n", de->name, - dir->name); - devfs_put(de); - return -ENOTDIR; - } - write_lock(&dir->u.dir.lock); - if (dir->u.dir.no_more_additions) - retval = -ENOENT; - else { - struct devfs_entry *old; - - old = _devfs_search_dir(dir, de->name, de->namelen); - if (old_de) - *old_de = old; - else - devfs_put(old); - if (old == NULL) { - de->parent = dir; - de->prev = dir->u.dir.last; - /* Append to the directory's list of children */ - if (dir->u.dir.first == NULL) - dir->u.dir.first = de; - else - dir->u.dir.last->next = de; - dir->u.dir.last = de; - retval = 0; - } else - retval = -EEXIST; - } - write_unlock(&dir->u.dir.lock); - if (retval) - devfs_put(de); - return retval; -} /* End Function _devfs_append_entry */ - -/** - * _devfs_get_root_entry - Get the root devfs entry. - * - * Returns the root devfs entry on success, else %NULL. - * - * TODO it must be called asynchronously due to the fact - * that devfs is initialized relatively late. Proper way - * is to remove module_init from init_devfs_fs and manually - * call it early enough during system init - */ - -static struct devfs_entry *_devfs_get_root_entry(void) -{ - struct devfs_entry *new; - static DEFINE_SPINLOCK(root_lock); - - if (root_entry) - return root_entry; - - new = _devfs_alloc_entry(NULL, 0, MODE_DIR); - if (new == NULL) - return NULL; - - spin_lock(&root_lock); - if (root_entry) { - spin_unlock(&root_lock); - devfs_put(new); - return root_entry; - } - root_entry = new; - spin_unlock(&root_lock); - - return root_entry; -} /* End Function _devfs_get_root_entry */ - -/** - * _devfs_descend - Descend down a tree using the next component name. - * @dir: The directory to search. - * @name: The component name to search for. - * @namelen: The length of %name. - * @next_pos: The position of the next '/' or '\0' is written here. - * - * Descend into a directory, searching for a component. This function forms - * the core of a tree-walking algorithm. The directory will be locked. - * The devfs entry corresponding to the component is returned. If there is - * no matching entry, %NULL is returned. - * An implicit devfs_get() is performed on the returned entry. - */ - -static struct devfs_entry *_devfs_descend(struct devfs_entry *dir, - const char *name, int namelen, - int *next_pos) -{ - const char *stop, *ptr; - struct devfs_entry *entry; - - if ((namelen >= 3) && (strncmp(name, "../", 3) == 0)) { /* Special-case going to parent directory */ - *next_pos = 3; - return devfs_get(dir->parent); - } - stop = name + namelen; - /* Search for a possible '/' */ - for (ptr = name; (ptr < stop) && (*ptr != '/'); ++ptr) ; - *next_pos = ptr - name; - read_lock(&dir->u.dir.lock); - entry = _devfs_search_dir(dir, name, *next_pos); - read_unlock(&dir->u.dir.lock); - return entry; -} /* End Function _devfs_descend */ - -static devfs_handle_t _devfs_make_parent_for_leaf(struct devfs_entry *dir, - const char *name, - int namelen, int *leaf_pos) -{ - int next_pos = 0; - - if (dir == NULL) - dir = _devfs_get_root_entry(); - if (dir == NULL) - return NULL; - devfs_get(dir); - /* Search for possible trailing component and ignore it */ - for (--namelen; (namelen > 0) && (name[namelen] != '/'); --namelen) ; - *leaf_pos = (name[namelen] == '/') ? (namelen + 1) : 0; - for (; namelen > 0; name += next_pos, namelen -= next_pos) { - struct devfs_entry *de, *old = NULL; - - if ((de = - _devfs_descend(dir, name, namelen, &next_pos)) == NULL) { - de = _devfs_alloc_entry(name, next_pos, MODE_DIR); - devfs_get(de); - if (!de || _devfs_append_entry(dir, de, &old)) { - devfs_put(de); - if (!old || !S_ISDIR(old->mode)) { - devfs_put(old); - devfs_put(dir); - return NULL; - } - de = old; /* Use the existing directory */ - } - } - if (de == dir->parent) { - devfs_put(dir); - devfs_put(de); - return NULL; - } - devfs_put(dir); - dir = de; - if (name[next_pos] == '/') - ++next_pos; - } - return dir; -} /* End Function _devfs_make_parent_for_leaf */ - -static devfs_handle_t _devfs_prepare_leaf(devfs_handle_t * dir, - const char *name, umode_t mode) -{ - int namelen, leaf_pos; - struct devfs_entry *de; - - namelen = strlen(name); - if ((*dir = _devfs_make_parent_for_leaf(*dir, name, namelen, - &leaf_pos)) == NULL) { - PRINTK("(%s): could not create parent path\n", name); - return NULL; - } - if ((de = _devfs_alloc_entry(name + leaf_pos, namelen - leaf_pos, mode)) - == NULL) { - PRINTK("(%s): could not allocate entry\n", name); - devfs_put(*dir); - return NULL; - } - return de; -} /* End Function _devfs_prepare_leaf */ - -static devfs_handle_t _devfs_walk_path(struct devfs_entry *dir, - const char *name, int namelen, - int traverse_symlink) -{ - int next_pos = 0; - - if (dir == NULL) - dir = _devfs_get_root_entry(); - if (dir == NULL) - return NULL; - devfs_get(dir); - for (; namelen > 0; name += next_pos, namelen -= next_pos) { - struct devfs_entry *de, *link; - - if (!S_ISDIR(dir->mode)) { - devfs_put(dir); - return NULL; - } - - if ((de = - _devfs_descend(dir, name, namelen, &next_pos)) == NULL) { - devfs_put(dir); - return NULL; - } - if (S_ISLNK(de->mode) && traverse_symlink) { /* Need to follow the link: this is a stack chomper */ - /* FIXME what if it puts outside of mounted tree? */ - link = _devfs_walk_path(dir, de->u.symlink.linkname, - de->u.symlink.length, TRUE); - devfs_put(de); - if (!link) { - devfs_put(dir); - return NULL; - } - de = link; - } - devfs_put(dir); - dir = de; - if (name[next_pos] == '/') - ++next_pos; - } - return dir; -} /* End Function _devfs_walk_path */ - -/** - * _devfs_find_entry - Find a devfs entry. - * @dir: The handle to the parent devfs directory entry. If this is %NULL the - * name is relative to the root of the devfs. - * @name: The name of the entry. This may be %NULL. - * @traverse_symlink: If %TRUE then symbolic links are traversed. - * - * Returns the devfs_entry pointer on success, else %NULL. An implicit - * devfs_get() is performed. - */ - -static struct devfs_entry *_devfs_find_entry(devfs_handle_t dir, - const char *name, - int traverse_symlink) -{ - unsigned int namelen = strlen(name); - - if (name[0] == '/') { - /* Skip leading pathname component */ - if (namelen < 2) { - PRINTK("(%s): too short\n", name); - return NULL; - } - for (++name, --namelen; (*name != '/') && (namelen > 0); - ++name, --namelen) ; - if (namelen < 2) { - PRINTK("(%s): too short\n", name); - return NULL; - } - ++name; - --namelen; - } - return _devfs_walk_path(dir, name, namelen, traverse_symlink); -} /* End Function _devfs_find_entry */ - -static struct devfs_entry *get_devfs_entry_from_vfs_inode(struct inode *inode) -{ - if (inode == NULL) - return NULL; - VERIFY_ENTRY((struct devfs_entry *)inode->u.generic_ip); - return inode->u.generic_ip; -} /* End Function get_devfs_entry_from_vfs_inode */ - -/** - * free_dentry - Free the dentry for a device entry and invalidate inode. - * @de: The entry. - * - * This must only be called after the entry has been unhooked from its - * parent directory. - */ - -static void free_dentry(struct devfs_entry *de) -{ - struct dentry *dentry = de->inode.dentry; - - if (!dentry) - return; - spin_lock(&dcache_lock); - dget_locked(dentry); - spin_unlock(&dcache_lock); - /* Forcefully remove the inode */ - if (dentry->d_inode != NULL) - dentry->d_inode->i_nlink = 0; - d_drop(dentry); - dput(dentry); -} /* End Function free_dentry */ - -/** - * is_devfsd_or_child - Test if the current process is devfsd or one of its children. - * @fs_info: The filesystem information. - * - * Returns %TRUE if devfsd or child, else %FALSE. - */ - -static int is_devfsd_or_child(struct fs_info *fs_info) -{ - struct task_struct *p = current; - - if (p == fs_info->devfsd_task) - return (TRUE); - if (process_group(p) == fs_info->devfsd_pgrp) - return (TRUE); - read_lock(&tasklist_lock); - for (; p != &init_task; p = p->real_parent) { - if (p == fs_info->devfsd_task) { - read_unlock(&tasklist_lock); - return (TRUE); - } - } - read_unlock(&tasklist_lock); - return (FALSE); -} /* End Function is_devfsd_or_child */ - -/** - * devfsd_queue_empty - Test if devfsd has work pending in its event queue. - * @fs_info: The filesystem information. - * - * Returns %TRUE if the queue is empty, else %FALSE. - */ - -static inline int devfsd_queue_empty(struct fs_info *fs_info) -{ - return (fs_info->devfsd_last_event) ? FALSE : TRUE; -} /* End Function devfsd_queue_empty */ - -/** - * wait_for_devfsd_finished - Wait for devfsd to finish processing its event queue. - * @fs_info: The filesystem information. - * - * Returns %TRUE if no more waiting will be required, else %FALSE. - */ - -static int wait_for_devfsd_finished(struct fs_info *fs_info) -{ - DECLARE_WAITQUEUE(wait, current); - - if (fs_info->devfsd_task == NULL) - return (TRUE); - if (devfsd_queue_empty(fs_info) && fs_info->devfsd_sleeping) - return TRUE; - if (is_devfsd_or_child(fs_info)) - return (FALSE); - set_current_state(TASK_UNINTERRUPTIBLE); - add_wait_queue(&fs_info->revalidate_wait_queue, &wait); - if (!devfsd_queue_empty(fs_info) || !fs_info->devfsd_sleeping) - if (fs_info->devfsd_task) - schedule(); - remove_wait_queue(&fs_info->revalidate_wait_queue, &wait); - __set_current_state(TASK_RUNNING); - return (TRUE); -} /* End Function wait_for_devfsd_finished */ - -/** - * devfsd_notify_de - Notify the devfsd daemon of a change. - * @de: The devfs entry that has changed. This and all parent entries will - * have their reference counts incremented if the event was queued. - * @type: The type of change. - * @mode: The mode of the entry. - * @uid: The user ID. - * @gid: The group ID. - * @fs_info: The filesystem info. - * - * Returns %TRUE if an event was queued and devfsd woken up, else %FALSE. - */ - -static int devfsd_notify_de(struct devfs_entry *de, - unsigned short type, umode_t mode, - uid_t uid, gid_t gid, struct fs_info *fs_info) -{ - struct devfsd_buf_entry *entry; - struct devfs_entry *curr; - - if (!(fs_info->devfsd_event_mask & (1 << type))) - return (FALSE); - if ((entry = kmem_cache_alloc(devfsd_buf_cache, SLAB_KERNEL)) == NULL) { - atomic_inc(&fs_info->devfsd_overrun_count); - return (FALSE); - } - for (curr = de; curr != NULL; curr = curr->parent) - devfs_get(curr); - entry->de = de; - entry->type = type; - entry->mode = mode; - entry->uid = uid; - entry->gid = gid; - entry->next = NULL; - spin_lock(&fs_info->devfsd_buffer_lock); - if (!fs_info->devfsd_first_event) - fs_info->devfsd_first_event = entry; - if (fs_info->devfsd_last_event) - fs_info->devfsd_last_event->next = entry; - fs_info->devfsd_last_event = entry; - spin_unlock(&fs_info->devfsd_buffer_lock); - wake_up_interruptible(&fs_info->devfsd_wait_queue); - return (TRUE); -} /* End Function devfsd_notify_de */ - -/** - * devfsd_notify - Notify the devfsd daemon of a change. - * @de: The devfs entry that has changed. - * @type: The type of change event. - * @wait: If TRUE, the function waits for the daemon to finish processing - * the event. - */ - -static void devfsd_notify(struct devfs_entry *de, unsigned short type) -{ - devfsd_notify_de(de, type, de->mode, current->euid, - current->egid, &fs_info); -} - -static int devfs_mk_dev(dev_t dev, umode_t mode, const char *fmt, va_list args) -{ - struct devfs_entry *dir = NULL, *de; - char buf[64]; - int error, n; - - n = vsnprintf(buf, sizeof(buf), fmt, args); - if (n >= sizeof(buf) || !buf[0]) { - printk(KERN_WARNING "%s: invalid format string %s\n", - __FUNCTION__, fmt); - return -EINVAL; - } - - de = _devfs_prepare_leaf(&dir, buf, mode); - if (!de) { - printk(KERN_WARNING "%s: could not prepare leaf for %s\n", - __FUNCTION__, buf); - return -ENOMEM; /* could be more accurate... */ - } - - de->u.dev = dev; - - error = _devfs_append_entry(dir, de, NULL); - if (error) { - printk(KERN_WARNING "%s: could not append to parent for %s\n", - __FUNCTION__, buf); - goto out; - } - - devfsd_notify(de, DEVFSD_NOTIFY_REGISTERED); - out: - devfs_put(dir); - return error; -} - -int devfs_mk_bdev(dev_t dev, umode_t mode, const char *fmt, ...) -{ - va_list args; - - if (!S_ISBLK(mode)) { - printk(KERN_WARNING "%s: invalide mode (%u) for %s\n", - __FUNCTION__, mode, fmt); - return -EINVAL; - } - - va_start(args, fmt); - return devfs_mk_dev(dev, mode, fmt, args); -} - -EXPORT_SYMBOL(devfs_mk_bdev); - -int devfs_mk_cdev(dev_t dev, umode_t mode, const char *fmt, ...) -{ - va_list args; - - if (!S_ISCHR(mode)) { - printk(KERN_WARNING "%s: invalide mode (%u) for %s\n", - __FUNCTION__, mode, fmt); - return -EINVAL; - } - - va_start(args, fmt); - return devfs_mk_dev(dev, mode, fmt, args); -} - -EXPORT_SYMBOL(devfs_mk_cdev); - -/** - * _devfs_unhook - Unhook a device entry from its parents list - * @de: The entry to unhook. - * - * Returns %TRUE if the entry was unhooked, else %FALSE if it was - * previously unhooked. - * The caller must have a write lock on the parent directory. - */ - -static int _devfs_unhook(struct devfs_entry *de) -{ - struct devfs_entry *parent; - - if (!de || (de->prev == de)) - return FALSE; - parent = de->parent; - if (de->prev == NULL) - parent->u.dir.first = de->next; - else - de->prev->next = de->next; - if (de->next == NULL) - parent->u.dir.last = de->prev; - else - de->next->prev = de->prev; - de->prev = de; /* Indicate we're unhooked */ - de->next = NULL; /* Force early termination for */ - return TRUE; -} /* End Function _devfs_unhook */ - -/** - * _devfs_unregister - Unregister a device entry from its parent. - * @dir: The parent directory. - * @de: The entry to unregister. - * - * The caller must have a write lock on the parent directory, which is - * unlocked by this function. - */ - -static void _devfs_unregister(struct devfs_entry *dir, struct devfs_entry *de) -{ - int unhooked = _devfs_unhook(de); - - write_unlock(&dir->u.dir.lock); - if (!unhooked) - return; - devfs_get(dir); - devfsd_notify(de, DEVFSD_NOTIFY_UNREGISTERED); - free_dentry(de); - devfs_put(dir); - if (!S_ISDIR(de->mode)) - return; - while (TRUE) { /* Recursively unregister: this is a stack chomper */ - struct devfs_entry *child; - - write_lock(&de->u.dir.lock); - de->u.dir.no_more_additions = TRUE; - child = de->u.dir.first; - VERIFY_ENTRY(child); - _devfs_unregister(de, child); - if (!child) - break; - DPRINTK(DEBUG_UNREGISTER, "(%s): child: %p refcount: %d\n", - child->name, child, atomic_read(&child->refcount)); - devfs_put(child); - } -} /* End Function _devfs_unregister */ - -static int devfs_do_symlink(devfs_handle_t dir, const char *name, - const char *link, devfs_handle_t * handle) -{ - int err; - unsigned int linklength; - char *newlink; - struct devfs_entry *de; - - if (handle != NULL) - *handle = NULL; - if (name == NULL) { - PRINTK("(): NULL name pointer\n"); - return -EINVAL; - } - if (link == NULL) { - PRINTK("(%s): NULL link pointer\n", name); - return -EINVAL; - } - linklength = strlen(link); - if ((newlink = kmalloc(linklength + 1, GFP_KERNEL)) == NULL) - return -ENOMEM; - memcpy(newlink, link, linklength); - newlink[linklength] = '\0'; - if ((de = _devfs_prepare_leaf(&dir, name, S_IFLNK | S_IRUGO | S_IXUGO)) - == NULL) { - PRINTK("(%s): could not prepare leaf\n", name); - kfree(newlink); - return -ENOTDIR; - } - de->info = NULL; - de->u.symlink.linkname = newlink; - de->u.symlink.length = linklength; - if ((err = _devfs_append_entry(dir, de, NULL)) != 0) { - PRINTK("(%s): could not append to parent, err: %d\n", name, - err); - devfs_put(dir); - return err; - } - devfs_put(dir); -#ifdef CONFIG_DEVFS_DEBUG - spin_lock(&stat_lock); - stat_num_bytes += linklength + 1; - spin_unlock(&stat_lock); -#endif - if (handle != NULL) - *handle = de; - return 0; -} /* End Function devfs_do_symlink */ - -/** - * devfs_mk_symlink Create a symbolic link in the devfs namespace. - * @from: The name of the entry. - * @to: Name of the destination - * - * Returns 0 on success, else a negative error code is returned. - */ - -int devfs_mk_symlink(const char *from, const char *to) -{ - devfs_handle_t de; - int err; - - err = devfs_do_symlink(NULL, from, to, &de); - if (!err) { - de->vfs = TRUE; - devfsd_notify(de, DEVFSD_NOTIFY_REGISTERED); - } - - return err; -} - -/** - * devfs_mk_dir - Create a directory in the devfs namespace. - * new name is relative to the root of the devfs. - * @fmt: The name of the entry. - * - * Use of this function is optional. The devfs_register() function - * will automatically create intermediate directories as needed. This function - * is provided for efficiency reasons, as it provides a handle to a directory. - * On failure %NULL is returned. - */ - -int devfs_mk_dir(const char *fmt, ...) -{ - struct devfs_entry *dir = NULL, *de = NULL, *old; - char buf[64]; - va_list args; - int error, n; - - va_start(args, fmt); - n = vsnprintf(buf, 64, fmt, args); - if (n >= 64 || !buf[0]) { - printk(KERN_WARNING "%s: invalid argument.", __FUNCTION__); - return -EINVAL; - } - - de = _devfs_prepare_leaf(&dir, buf, MODE_DIR); - if (!de) { - PRINTK("(%s): could not prepare leaf\n", buf); - return -EINVAL; - } - - error = _devfs_append_entry(dir, de, &old); - if (error == -EEXIST && S_ISDIR(old->mode)) { - /* - * devfs_mk_dir() of an already-existing directory will - * return success. - */ - error = 0; - goto out_put; - } else if (error) { - PRINTK("(%s): could not append to dir: %p \"%s\"\n", - buf, dir, dir->name); - devfs_put(old); - goto out_put; - } - - devfsd_notify(de, DEVFSD_NOTIFY_REGISTERED); - - out_put: - devfs_put(dir); - return error; -} - -void devfs_remove(const char *fmt, ...) -{ - char buf[64]; - va_list args; - int n; - - va_start(args, fmt); - n = vsnprintf(buf, sizeof(buf), fmt, args); - if (n < sizeof(buf) && buf[0]) { - devfs_handle_t de = _devfs_find_entry(NULL, buf, 0); - - if (!de) { - printk(KERN_ERR "%s: %s not found, cannot remove\n", - __FUNCTION__, buf); - dump_stack(); - return; - } - - write_lock(&de->parent->u.dir.lock); - _devfs_unregister(de->parent, de); - devfs_put(de); - devfs_put(de); - } -} - -/** - * devfs_generate_path - Generate a pathname for an entry, relative to the devfs root. - * @de: The devfs entry. - * @path: The buffer to write the pathname to. The pathname and '\0' - * terminator will be written at the end of the buffer. - * @buflen: The length of the buffer. - * - * Returns the offset in the buffer where the pathname starts on success, - * else a negative error code. - */ - -static int devfs_generate_path(devfs_handle_t de, char *path, int buflen) -{ - int pos; -#define NAMEOF(de) ( (de)->mode ? (de)->name : (de)->u.name ) - - if (de == NULL) - return -EINVAL; - VERIFY_ENTRY(de); - if (de->namelen >= buflen) - return -ENAMETOOLONG; /* Must be first */ - path[buflen - 1] = '\0'; - if (de->parent == NULL) - return buflen - 1; /* Don't prepend root */ - pos = buflen - de->namelen - 1; - memcpy(path + pos, NAMEOF(de), de->namelen); - for (de = de->parent; de->parent != NULL; de = de->parent) { - if (pos - de->namelen - 1 < 0) - return -ENAMETOOLONG; - path[--pos] = '/'; - pos -= de->namelen; - memcpy(path + pos, NAMEOF(de), de->namelen); - } - return pos; -} /* End Function devfs_generate_path */ - -/** - * devfs_setup - Process kernel boot options. - * @str: The boot options after the "devfs=". - */ - -static int __init devfs_setup(char *str) -{ - static struct { - char *name; - unsigned int mask; - unsigned int *opt; - } devfs_options_tab[] __initdata = { -#ifdef CONFIG_DEVFS_DEBUG - { - "dall", DEBUG_ALL, &devfs_debug_init}, { - "dmod", DEBUG_MODULE_LOAD, &devfs_debug_init}, { - "dreg", DEBUG_REGISTER, &devfs_debug_init}, { - "dunreg", DEBUG_UNREGISTER, &devfs_debug_init}, { - "dfree", DEBUG_FREE, &devfs_debug_init}, { - "diget", DEBUG_I_GET, &devfs_debug_init}, { - "dchange", DEBUG_SET_FLAGS, &devfs_debug_init}, { - "dsread", DEBUG_S_READ, &devfs_debug_init}, { - "dichange", DEBUG_I_CHANGE, &devfs_debug_init}, { - "dimknod", DEBUG_I_MKNOD, &devfs_debug_init}, { - "dilookup", DEBUG_I_LOOKUP, &devfs_debug_init}, { - "diunlink", DEBUG_I_UNLINK, &devfs_debug_init}, -#endif /* CONFIG_DEVFS_DEBUG */ - { - "mount", OPTION_MOUNT, &boot_options}, { - NULL, 0, NULL} - }; - - while ((*str != '\0') && !isspace(*str)) { - int i, found = 0, invert = 0; - - if (strncmp(str, "no", 2) == 0) { - invert = 1; - str += 2; - } - for (i = 0; devfs_options_tab[i].name != NULL; i++) { - int len = strlen(devfs_options_tab[i].name); - - if (strncmp(str, devfs_options_tab[i].name, len) == 0) { - if (invert) - *devfs_options_tab[i].opt &= - ~devfs_options_tab[i].mask; - else - *devfs_options_tab[i].opt |= - devfs_options_tab[i].mask; - str += len; - found = 1; - break; - } - } - if (!found) - return 0; /* No match */ - if (*str != ',') - return 0; /* No more options */ - ++str; - } - return 1; -} /* End Function devfs_setup */ - -__setup("devfs=", devfs_setup); - -EXPORT_SYMBOL(devfs_mk_dir); -EXPORT_SYMBOL(devfs_remove); - -/** - * try_modload - Notify devfsd of an inode lookup by a non-devfsd process. - * @parent: The parent devfs entry. - * @fs_info: The filesystem info. - * @name: The device name. - * @namelen: The number of characters in @name. - * @buf: A working area that will be used. This must not go out of scope - * until devfsd is idle again. - * - * Returns 0 on success (event was queued), else a negative error code. - */ - -static int try_modload(struct devfs_entry *parent, struct fs_info *fs_info, - const char *name, unsigned namelen, - struct devfs_entry *buf) -{ - if (!(fs_info->devfsd_event_mask & (1 << DEVFSD_NOTIFY_LOOKUP))) - return -ENOENT; - if (is_devfsd_or_child(fs_info)) - return -ENOENT; - memset(buf, 0, sizeof *buf); - atomic_set(&buf->refcount, 1); - buf->parent = parent; - buf->namelen = namelen; - buf->u.name = name; - WRITE_ENTRY_MAGIC(buf, MAGIC_VALUE); - if (!devfsd_notify_de(buf, DEVFSD_NOTIFY_LOOKUP, 0, - current->euid, current->egid, fs_info)) - return -ENOENT; - /* Possible success: event has been queued */ - return 0; -} /* End Function try_modload */ - -/* Superblock operations follow */ - -static struct inode_operations devfs_iops; -static struct inode_operations devfs_dir_iops; -static const struct file_operations devfs_fops; -static const struct file_operations devfs_dir_fops; -static struct inode_operations devfs_symlink_iops; - -static int devfs_notify_change(struct dentry *dentry, struct iattr *iattr) -{ - int retval; - struct devfs_entry *de; - struct inode *inode = dentry->d_inode; - struct fs_info *fs_info = inode->i_sb->s_fs_info; - - de = get_devfs_entry_from_vfs_inode(inode); - if (de == NULL) - return -ENODEV; - retval = inode_change_ok(inode, iattr); - if (retval != 0) - return retval; - retval = inode_setattr(inode, iattr); - if (retval != 0) - return retval; - DPRINTK(DEBUG_I_CHANGE, "(%d): VFS inode: %p devfs_entry: %p\n", - (int)inode->i_ino, inode, de); - DPRINTK(DEBUG_I_CHANGE, "(): mode: 0%o uid: %d gid: %d\n", - (int)inode->i_mode, (int)inode->i_uid, (int)inode->i_gid); - /* Inode is not on hash chains, thus must save permissions here rather - than in a write_inode() method */ - de->mode = inode->i_mode; - de->inode.uid = inode->i_uid; - de->inode.gid = inode->i_gid; - de->inode.atime = inode->i_atime; - de->inode.mtime = inode->i_mtime; - de->inode.ctime = inode->i_ctime; - if ((iattr->ia_valid & (ATTR_MODE | ATTR_UID | ATTR_GID)) && - !is_devfsd_or_child(fs_info)) - devfsd_notify_de(de, DEVFSD_NOTIFY_CHANGE, inode->i_mode, - inode->i_uid, inode->i_gid, fs_info); - return 0; -} /* End Function devfs_notify_change */ - -static struct super_operations devfs_sops = { - .drop_inode = generic_delete_inode, - .statfs = simple_statfs, -}; - -/** - * _devfs_get_vfs_inode - Get a VFS inode. - * @sb: The super block. - * @de: The devfs inode. - * @dentry: The dentry to register with the devfs inode. - * - * Returns the inode on success, else %NULL. An implicit devfs_get() is - * performed if the inode is created. - */ - -static struct inode *_devfs_get_vfs_inode(struct super_block *sb, - struct devfs_entry *de, - struct dentry *dentry) -{ - struct inode *inode; - - if (de->prev == de) - return NULL; /* Quick check to see if unhooked */ - if ((inode = new_inode(sb)) == NULL) { - PRINTK("(%s): new_inode() failed, de: %p\n", de->name, de); - return NULL; - } - if (de->parent) { - read_lock(&de->parent->u.dir.lock); - if (de->prev != de) - de->inode.dentry = dentry; /* Not unhooked */ - read_unlock(&de->parent->u.dir.lock); - } else - de->inode.dentry = dentry; /* Root: no locking needed */ - if (de->inode.dentry != dentry) { /* Must have been unhooked */ - iput(inode); - return NULL; - } - /* FIXME where is devfs_put? */ - inode->u.generic_ip = devfs_get(de); - inode->i_ino = de->inode.ino; - DPRINTK(DEBUG_I_GET, "(%d): VFS inode: %p devfs_entry: %p\n", - (int)inode->i_ino, inode, de); - inode->i_blocks = 0; - inode->i_blksize = FAKE_BLOCK_SIZE; - inode->i_op = &devfs_iops; - inode->i_mode = de->mode; - if (S_ISDIR(de->mode)) { - inode->i_op = &devfs_dir_iops; - inode->i_fop = &devfs_dir_fops; - } else if (S_ISLNK(de->mode)) { - inode->i_op = &devfs_symlink_iops; - inode->i_size = de->u.symlink.length; - } else if (S_ISCHR(de->mode) || S_ISBLK(de->mode)) { - init_special_inode(inode, de->mode, de->u.dev); - } else if (S_ISFIFO(de->mode) || S_ISSOCK(de->mode)) { - init_special_inode(inode, de->mode, 0); - } else { - PRINTK("(%s): unknown mode %o de: %p\n", - de->name, de->mode, de); - iput(inode); - devfs_put(de); - return NULL; - } - - inode->i_uid = de->inode.uid; - inode->i_gid = de->inode.gid; - inode->i_atime = de->inode.atime; - inode->i_mtime = de->inode.mtime; - inode->i_ctime = de->inode.ctime; - DPRINTK(DEBUG_I_GET, "(): mode: 0%o uid: %d gid: %d\n", - (int)inode->i_mode, (int)inode->i_uid, (int)inode->i_gid); - return inode; -} /* End Function _devfs_get_vfs_inode */ - -/* File operations for device entries follow */ - -static int devfs_readdir(struct file *file, void *dirent, filldir_t filldir) -{ - int err, count; - int stored = 0; - struct fs_info *fs_info; - struct devfs_entry *parent, *de, *next = NULL; - struct inode *inode = file->f_dentry->d_inode; - - fs_info = inode->i_sb->s_fs_info; - parent = get_devfs_entry_from_vfs_inode(file->f_dentry->d_inode); - if ((long)file->f_pos < 0) - return -EINVAL; - DPRINTK(DEBUG_F_READDIR, "(%s): fs_info: %p pos: %ld\n", - parent->name, fs_info, (long)file->f_pos); - switch ((long)file->f_pos) { - case 0: - err = (*filldir) (dirent, "..", 2, file->f_pos, - parent_ino(file->f_dentry), DT_DIR); - if (err == -EINVAL) - break; - if (err < 0) - return err; - file->f_pos++; - ++stored; - /* Fall through */ - case 1: - err = - (*filldir) (dirent, ".", 1, file->f_pos, inode->i_ino, - DT_DIR); - if (err == -EINVAL) - break; - if (err < 0) - return err; - file->f_pos++; - ++stored; - /* Fall through */ - default: - /* Skip entries */ - count = file->f_pos - 2; - read_lock(&parent->u.dir.lock); - for (de = parent->u.dir.first; de && (count > 0); de = de->next) - --count; - devfs_get(de); - read_unlock(&parent->u.dir.lock); - /* Now add all remaining entries */ - while (de) { - err = (*filldir) (dirent, de->name, de->namelen, - file->f_pos, de->inode.ino, - de->mode >> 12); - if (err < 0) - devfs_put(de); - else { - file->f_pos++; - ++stored; - } - if (err == -EINVAL) - break; - if (err < 0) - return err; - read_lock(&parent->u.dir.lock); - next = devfs_get(de->next); - read_unlock(&parent->u.dir.lock); - devfs_put(de); - de = next; - } - break; - } - return stored; -} /* End Function devfs_readdir */ - -/* Open devfs specific special files */ -static int devfs_open(struct inode *inode, struct file *file) -{ - int err; - int minor = MINOR(inode->i_rdev); - struct file_operations *old_fops, *new_fops; - - switch (minor) { - case 0: /* /dev/.devfsd */ - new_fops = fops_get(&devfsd_fops); - break; -#ifdef CONFIG_DEVFS_DEBUG - case 1: /* /dev/.stat */ - new_fops = fops_get(&stat_fops); - break; -#endif - default: - return -ENODEV; - } - - if (new_fops == NULL) - return -ENODEV; - old_fops = file->f_op; - file->f_op = new_fops; - err = new_fops->open ? new_fops->open(inode, file) : 0; - if (err) { - file->f_op = old_fops; - fops_put(new_fops); - } else - fops_put(old_fops); - return err; -} /* End Function devfs_open */ - -static const struct file_operations devfs_fops = { - .open = devfs_open, -}; - -static const struct file_operations devfs_dir_fops = { - .read = generic_read_dir, - .readdir = devfs_readdir, -}; - -/* Dentry operations for device entries follow */ - -/** - * devfs_d_release - Callback for when a dentry is freed. - * @dentry: The dentry. - */ - -static void devfs_d_release(struct dentry *dentry) -{ - DPRINTK(DEBUG_D_RELEASE, "(%p): inode: %p\n", dentry, dentry->d_inode); -} /* End Function devfs_d_release */ - -/** - * devfs_d_iput - Callback for when a dentry loses its inode. - * @dentry: The dentry. - * @inode: The inode. - */ - -static void devfs_d_iput(struct dentry *dentry, struct inode *inode) -{ - struct devfs_entry *de; - - de = get_devfs_entry_from_vfs_inode(inode); - DPRINTK(DEBUG_D_IPUT, - "(%s): dentry: %p inode: %p de: %p de->dentry: %p\n", de->name, - dentry, inode, de, de->inode.dentry); - if (de->inode.dentry && (de->inode.dentry != dentry)) - OOPS("(%s): de: %p dentry: %p de->dentry: %p\n", - de->name, de, dentry, de->inode.dentry); - de->inode.dentry = NULL; - iput(inode); - devfs_put(de); -} /* End Function devfs_d_iput */ - -static int devfs_d_delete(struct dentry *dentry); - -static struct dentry_operations devfs_dops = { - .d_delete = devfs_d_delete, - .d_release = devfs_d_release, - .d_iput = devfs_d_iput, -}; - -static int devfs_d_revalidate_wait(struct dentry *dentry, struct nameidata *); - -static struct dentry_operations devfs_wait_dops = { - .d_delete = devfs_d_delete, - .d_release = devfs_d_release, - .d_iput = devfs_d_iput, - .d_revalidate = devfs_d_revalidate_wait, -}; - -/** - * devfs_d_delete - Callback for when all files for a dentry are closed. - * @dentry: The dentry. - */ - -static int devfs_d_delete(struct dentry *dentry) -{ - struct inode *inode = dentry->d_inode; - - if (dentry->d_op == &devfs_wait_dops) - dentry->d_op = &devfs_dops; - /* Unhash dentry if negative (has no inode) */ - if (inode == NULL) { - DPRINTK(DEBUG_D_DELETE, "(%p): dropping negative dentry\n", - dentry); - return 1; - } - return 0; -} /* End Function devfs_d_delete */ - -struct devfs_lookup_struct { - devfs_handle_t de; - wait_queue_head_t wait_queue; -}; - -/* XXX: this doesn't handle the case where we got a negative dentry - but a devfs entry has been registered in the meanwhile */ -static int devfs_d_revalidate_wait(struct dentry *dentry, struct nameidata *nd) -{ - struct inode *dir = dentry->d_parent->d_inode; - struct fs_info *fs_info = dir->i_sb->s_fs_info; - devfs_handle_t parent = get_devfs_entry_from_vfs_inode(dir); - struct devfs_lookup_struct *lookup_info = dentry->d_fsdata; - DECLARE_WAITQUEUE(wait, current); - int need_lock; - - /* - * FIXME HACK - * - * make sure that - * d_instantiate always runs under lock - * we release i_mutex lock before going to sleep - * - * unfortunately sometimes d_revalidate is called with - * and sometimes without i_mutex lock held. The following checks - * attempt to deduce when we need to add (and drop resp.) lock - * here. This relies on current (2.6.2) calling coventions: - * - * lookup_hash is always run under i_mutex and is passing NULL - * as nd - * - * open(...,O_CREATE,...) calls _lookup_hash under i_mutex - * and sets flags to LOOKUP_OPEN|LOOKUP_CREATE - * - * all other invocations of ->d_revalidate seem to happen - * outside of i_mutex - */ - need_lock = nd && - (!(nd->flags & LOOKUP_CREATE) || (nd->flags & LOOKUP_PARENT)); - - if (need_lock) - mutex_lock(&dir->i_mutex); - - if (is_devfsd_or_child(fs_info)) { - devfs_handle_t de = lookup_info->de; - struct inode *inode; - - DPRINTK(DEBUG_I_LOOKUP, - "(%s): dentry: %p inode: %p de: %p by: \"%s\"\n", - dentry->d_name.name, dentry, dentry->d_inode, de, - current->comm); - if (dentry->d_inode) - goto out; - if (de == NULL) { - read_lock(&parent->u.dir.lock); - de = _devfs_search_dir(parent, dentry->d_name.name, - dentry->d_name.len); - read_unlock(&parent->u.dir.lock); - if (de == NULL) - goto out; - lookup_info->de = de; - } - /* Create an inode, now that the driver information is available */ - inode = _devfs_get_vfs_inode(dir->i_sb, de, dentry); - if (!inode) - goto out; - DPRINTK(DEBUG_I_LOOKUP, - "(%s): new VFS inode(%u): %p de: %p by: \"%s\"\n", - de->name, de->inode.ino, inode, de, current->comm); - d_instantiate(dentry, inode); - goto out; - } - if (lookup_info == NULL) - goto out; /* Early termination */ - read_lock(&parent->u.dir.lock); - if (dentry->d_fsdata) { - set_current_state(TASK_UNINTERRUPTIBLE); - add_wait_queue(&lookup_info->wait_queue, &wait); - read_unlock(&parent->u.dir.lock); - /* at this point it is always (hopefully) locked */ - mutex_unlock(&dir->i_mutex); - schedule(); - mutex_lock(&dir->i_mutex); - /* - * This does not need nor should remove wait from wait_queue. - * Wait queue head is never reused - nothing is ever added to it - * after all waiters have been waked up and head itself disappears - * very soon after it. Moreover it is local variable on stack that - * is likely to have already disappeared so any reference to it - * at this point is buggy. - */ - - } else - read_unlock(&parent->u.dir.lock); - - out: - if (need_lock) - mutex_unlock(&dir->i_mutex); - return 1; -} /* End Function devfs_d_revalidate_wait */ - -/* Inode operations for device entries follow */ - -static struct dentry *devfs_lookup(struct inode *dir, struct dentry *dentry, - struct nameidata *nd) -{ - struct devfs_entry tmp; /* Must stay in scope until devfsd idle again */ - struct devfs_lookup_struct lookup_info; - struct fs_info *fs_info = dir->i_sb->s_fs_info; - struct devfs_entry *parent, *de; - struct inode *inode; - struct dentry *retval = NULL; - - /* Set up the dentry operations before anything else, to ensure cleaning - up on any error */ - dentry->d_op = &devfs_dops; - /* First try to get the devfs entry for this directory */ - parent = get_devfs_entry_from_vfs_inode(dir); - DPRINTK(DEBUG_I_LOOKUP, "(%s): dentry: %p parent: %p by: \"%s\"\n", - dentry->d_name.name, dentry, parent, current->comm); - if (parent == NULL) - return ERR_PTR(-ENOENT); - read_lock(&parent->u.dir.lock); - de = _devfs_search_dir(parent, dentry->d_name.name, dentry->d_name.len); - read_unlock(&parent->u.dir.lock); - lookup_info.de = de; - init_waitqueue_head(&lookup_info.wait_queue); - dentry->d_fsdata = &lookup_info; - if (de == NULL) { /* Try with devfsd. For any kind of failure, leave a negative dentry - so someone else can deal with it (in the case where the sysadmin - does a mknod()). It's important to do this before hashing the - dentry, so that the devfsd queue is filled before revalidates - can start */ - if (try_modload(parent, fs_info, dentry->d_name.name, dentry->d_name.len, &tmp) < 0) { /* Lookup event was not queued to devfsd */ - d_add(dentry, NULL); - return NULL; - } - } - dentry->d_op = &devfs_wait_dops; - d_add(dentry, NULL); /* Open the floodgates */ - /* Unlock directory semaphore, which will release any waiters. They - will get the hashed dentry, and may be forced to wait for - revalidation */ - mutex_unlock(&dir->i_mutex); - wait_for_devfsd_finished(fs_info); /* If I'm not devfsd, must wait */ - mutex_lock(&dir->i_mutex); /* Grab it again because them's the rules */ - de = lookup_info.de; - /* If someone else has been so kind as to make the inode, we go home - early */ - if (dentry->d_inode) - goto out; - if (de == NULL) { - read_lock(&parent->u.dir.lock); - de = _devfs_search_dir(parent, dentry->d_name.name, - dentry->d_name.len); - read_unlock(&parent->u.dir.lock); - if (de == NULL) - goto out; - /* OK, there's an entry now, but no VFS inode yet */ - } - /* Create an inode, now that the driver information is available */ - inode = _devfs_get_vfs_inode(dir->i_sb, de, dentry); - if (!inode) { - retval = ERR_PTR(-ENOMEM); - goto out; - } - DPRINTK(DEBUG_I_LOOKUP, - "(%s): new VFS inode(%u): %p de: %p by: \"%s\"\n", de->name, - de->inode.ino, inode, de, current->comm); - d_instantiate(dentry, inode); - out: - write_lock(&parent->u.dir.lock); - dentry->d_op = &devfs_dops; - dentry->d_fsdata = NULL; - wake_up(&lookup_info.wait_queue); - write_unlock(&parent->u.dir.lock); - devfs_put(de); - return retval; -} /* End Function devfs_lookup */ - -static int devfs_unlink(struct inode *dir, struct dentry *dentry) -{ - int unhooked; - struct devfs_entry *de; - struct inode *inode = dentry->d_inode; - struct fs_info *fs_info = dir->i_sb->s_fs_info; - - de = get_devfs_entry_from_vfs_inode(inode); - DPRINTK(DEBUG_I_UNLINK, "(%s): de: %p\n", dentry->d_name.name, de); - if (de == NULL) - return -ENOENT; - if (!de->vfs) - return -EPERM; - write_lock(&de->parent->u.dir.lock); - unhooked = _devfs_unhook(de); - write_unlock(&de->parent->u.dir.lock); - if (!unhooked) - return -ENOENT; - if (!is_devfsd_or_child(fs_info)) - devfsd_notify_de(de, DEVFSD_NOTIFY_DELETE, inode->i_mode, - inode->i_uid, inode->i_gid, fs_info); - free_dentry(de); - devfs_put(de); - return 0; -} /* End Function devfs_unlink */ - -static int devfs_symlink(struct inode *dir, struct dentry *dentry, - const char *symname) -{ - int err; - struct fs_info *fs_info = dir->i_sb->s_fs_info; - struct devfs_entry *parent, *de; - struct inode *inode; - - /* First try to get the devfs entry for this directory */ - parent = get_devfs_entry_from_vfs_inode(dir); - if (parent == NULL) - return -ENOENT; - err = devfs_do_symlink(parent, dentry->d_name.name, symname, &de); - DPRINTK(DEBUG_DISABLED, "(%s): errcode from : %d\n", - dentry->d_name.name, err); - if (err < 0) - return err; - de->vfs = TRUE; - de->inode.uid = current->euid; - de->inode.gid = current->egid; - de->inode.atime = CURRENT_TIME; - de->inode.mtime = CURRENT_TIME; - de->inode.ctime = CURRENT_TIME; - if ((inode = _devfs_get_vfs_inode(dir->i_sb, de, dentry)) == NULL) - return -ENOMEM; - DPRINTK(DEBUG_DISABLED, "(%s): new VFS inode(%u): %p dentry: %p\n", - dentry->d_name.name, de->inode.ino, inode, dentry); - d_instantiate(dentry, inode); - if (!is_devfsd_or_child(fs_info)) - devfsd_notify_de(de, DEVFSD_NOTIFY_CREATE, inode->i_mode, - inode->i_uid, inode->i_gid, fs_info); - return 0; -} /* End Function devfs_symlink */ - -static int devfs_mkdir(struct inode *dir, struct dentry *dentry, int mode) -{ - int err; - struct fs_info *fs_info = dir->i_sb->s_fs_info; - struct devfs_entry *parent, *de; - struct inode *inode; - - mode = (mode & ~S_IFMT) | S_IFDIR; /* VFS doesn't pass S_IFMT part */ - parent = get_devfs_entry_from_vfs_inode(dir); - if (parent == NULL) - return -ENOENT; - de = _devfs_alloc_entry(dentry->d_name.name, dentry->d_name.len, mode); - if (!de) - return -ENOMEM; - de->vfs = TRUE; - if ((err = _devfs_append_entry(parent, de, NULL)) != 0) - return err; - de->inode.uid = current->euid; - de->inode.gid = current->egid; - de->inode.atime = CURRENT_TIME; - de->inode.mtime = CURRENT_TIME; - de->inode.ctime = CURRENT_TIME; - if ((inode = _devfs_get_vfs_inode(dir->i_sb, de, dentry)) == NULL) - return -ENOMEM; - DPRINTK(DEBUG_DISABLED, "(%s): new VFS inode(%u): %p dentry: %p\n", - dentry->d_name.name, de->inode.ino, inode, dentry); - d_instantiate(dentry, inode); - if (!is_devfsd_or_child(fs_info)) - devfsd_notify_de(de, DEVFSD_NOTIFY_CREATE, inode->i_mode, - inode->i_uid, inode->i_gid, fs_info); - return 0; -} /* End Function devfs_mkdir */ - -static int devfs_rmdir(struct inode *dir, struct dentry *dentry) -{ - int err = 0; - struct devfs_entry *de; - struct fs_info *fs_info = dir->i_sb->s_fs_info; - struct inode *inode = dentry->d_inode; - - if (dir->i_sb->s_fs_info != inode->i_sb->s_fs_info) - return -EINVAL; - de = get_devfs_entry_from_vfs_inode(inode); - if (de == NULL) - return -ENOENT; - if (!S_ISDIR(de->mode)) - return -ENOTDIR; - if (!de->vfs) - return -EPERM; - /* First ensure the directory is empty and will stay that way */ - write_lock(&de->u.dir.lock); - if (de->u.dir.first) - err = -ENOTEMPTY; - else - de->u.dir.no_more_additions = TRUE; - write_unlock(&de->u.dir.lock); - if (err) - return err; - /* Now unhook the directory from its parent */ - write_lock(&de->parent->u.dir.lock); - if (!_devfs_unhook(de)) - err = -ENOENT; - write_unlock(&de->parent->u.dir.lock); - if (err) - return err; - if (!is_devfsd_or_child(fs_info)) - devfsd_notify_de(de, DEVFSD_NOTIFY_DELETE, inode->i_mode, - inode->i_uid, inode->i_gid, fs_info); - free_dentry(de); - devfs_put(de); - return 0; -} /* End Function devfs_rmdir */ - -static int devfs_mknod(struct inode *dir, struct dentry *dentry, int mode, - dev_t rdev) -{ - int err; - struct fs_info *fs_info = dir->i_sb->s_fs_info; - struct devfs_entry *parent, *de; - struct inode *inode; - - DPRINTK(DEBUG_I_MKNOD, "(%s): mode: 0%o dev: %u:%u\n", - dentry->d_name.name, mode, MAJOR(rdev), MINOR(rdev)); - parent = get_devfs_entry_from_vfs_inode(dir); - if (parent == NULL) - return -ENOENT; - de = _devfs_alloc_entry(dentry->d_name.name, dentry->d_name.len, mode); - if (!de) - return -ENOMEM; - de->vfs = TRUE; - if (S_ISCHR(mode) || S_ISBLK(mode)) - de->u.dev = rdev; - if ((err = _devfs_append_entry(parent, de, NULL)) != 0) - return err; - de->inode.uid = current->euid; - de->inode.gid = current->egid; - de->inode.atime = CURRENT_TIME; - de->inode.mtime = CURRENT_TIME; - de->inode.ctime = CURRENT_TIME; - if ((inode = _devfs_get_vfs_inode(dir->i_sb, de, dentry)) == NULL) - return -ENOMEM; - DPRINTK(DEBUG_I_MKNOD, ": new VFS inode(%u): %p dentry: %p\n", - de->inode.ino, inode, dentry); - d_instantiate(dentry, inode); - if (!is_devfsd_or_child(fs_info)) - devfsd_notify_de(de, DEVFSD_NOTIFY_CREATE, inode->i_mode, - inode->i_uid, inode->i_gid, fs_info); - return 0; -} /* End Function devfs_mknod */ - -static void *devfs_follow_link(struct dentry *dentry, struct nameidata *nd) -{ - struct devfs_entry *p = get_devfs_entry_from_vfs_inode(dentry->d_inode); - nd_set_link(nd, p ? p->u.symlink.linkname : ERR_PTR(-ENODEV)); - return NULL; -} /* End Function devfs_follow_link */ - -static struct inode_operations devfs_iops = { - .setattr = devfs_notify_change, -}; - -static struct inode_operations devfs_dir_iops = { - .lookup = devfs_lookup, - .unlink = devfs_unlink, - .symlink = devfs_symlink, - .mkdir = devfs_mkdir, - .rmdir = devfs_rmdir, - .mknod = devfs_mknod, - .setattr = devfs_notify_change, -}; - -static struct inode_operations devfs_symlink_iops = { - .readlink = generic_readlink, - .follow_link = devfs_follow_link, - .setattr = devfs_notify_change, -}; - -static int devfs_fill_super(struct super_block *sb, void *data, int silent) -{ - struct inode *root_inode = NULL; - - if (_devfs_get_root_entry() == NULL) - goto out_no_root; - atomic_set(&fs_info.devfsd_overrun_count, 0); - init_waitqueue_head(&fs_info.devfsd_wait_queue); - init_waitqueue_head(&fs_info.revalidate_wait_queue); - fs_info.sb = sb; - sb->s_fs_info = &fs_info; - sb->s_blocksize = 1024; - sb->s_blocksize_bits = 10; - sb->s_magic = DEVFS_SUPER_MAGIC; - sb->s_op = &devfs_sops; - sb->s_time_gran = 1; - if ((root_inode = _devfs_get_vfs_inode(sb, root_entry, NULL)) == NULL) - goto out_no_root; - sb->s_root = d_alloc_root(root_inode); - if (!sb->s_root) - goto out_no_root; - DPRINTK(DEBUG_S_READ, "(): made devfs ptr: %p\n", sb->s_fs_info); - return 0; - - out_no_root: - PRINTK("(): get root inode failed\n"); - if (root_inode) - iput(root_inode); - return -EINVAL; -} /* End Function devfs_fill_super */ - -static int devfs_get_sb(struct file_system_type *fs_type, - int flags, const char *dev_name, - void *data, struct vfsmount *mnt) -{ - return get_sb_single(fs_type, flags, data, devfs_fill_super, mnt); -} - -static struct file_system_type devfs_fs_type = { - .name = DEVFS_NAME, - .get_sb = devfs_get_sb, - .kill_sb = kill_anon_super, -}; - -/* File operations for devfsd follow */ - -static ssize_t devfsd_read(struct file *file, char __user *buf, size_t len, - loff_t * ppos) -{ - int done = FALSE; - int ival; - loff_t pos, devname_offset, tlen, rpos; - devfs_handle_t de; - struct devfsd_buf_entry *entry; - struct fs_info *fs_info = file->f_dentry->d_inode->i_sb->s_fs_info; - struct devfsd_notify_struct *info = fs_info->devfsd_info; - DECLARE_WAITQUEUE(wait, current); - - /* Verify the task has grabbed the queue */ - if (fs_info->devfsd_task != current) - return -EPERM; - info->major = 0; - info->minor = 0; - /* Block for a new entry */ - set_current_state(TASK_INTERRUPTIBLE); - add_wait_queue(&fs_info->devfsd_wait_queue, &wait); - while (devfsd_queue_empty(fs_info)) { - fs_info->devfsd_sleeping = TRUE; - wake_up(&fs_info->revalidate_wait_queue); - schedule(); - fs_info->devfsd_sleeping = FALSE; - if (signal_pending(current)) { - remove_wait_queue(&fs_info->devfsd_wait_queue, &wait); - __set_current_state(TASK_RUNNING); - return -EINTR; - } - set_current_state(TASK_INTERRUPTIBLE); - } - remove_wait_queue(&fs_info->devfsd_wait_queue, &wait); - __set_current_state(TASK_RUNNING); - /* Now play with the data */ - ival = atomic_read(&fs_info->devfsd_overrun_count); - info->overrun_count = ival; - entry = fs_info->devfsd_first_event; - info->type = entry->type; - info->mode = entry->mode; - info->uid = entry->uid; - info->gid = entry->gid; - de = entry->de; - if (S_ISCHR(de->mode) || S_ISBLK(de->mode)) { - info->major = MAJOR(de->u.dev); - info->minor = MINOR(de->u.dev); - } - pos = devfs_generate_path(de, info->devname, DEVFS_PATHLEN); - if (pos < 0) - return pos; - info->namelen = DEVFS_PATHLEN - pos - 1; - if (info->mode == 0) - info->mode = de->mode; - devname_offset = info->devname - (char *)info; - rpos = *ppos; - if (rpos < devname_offset) { - /* Copy parts of the header */ - tlen = devname_offset - rpos; - if (tlen > len) - tlen = len; - if (copy_to_user(buf, (char *)info + rpos, tlen)) { - return -EFAULT; - } - rpos += tlen; - buf += tlen; - len -= tlen; - } - if ((rpos >= devname_offset) && (len > 0)) { - /* Copy the name */ - tlen = info->namelen + 1; - if (tlen > len) - tlen = len; - else - done = TRUE; - if (copy_to_user - (buf, info->devname + pos + rpos - devname_offset, tlen)) { - return -EFAULT; - } - rpos += tlen; - } - tlen = rpos - *ppos; - if (done) { - devfs_handle_t parent; - - spin_lock(&fs_info->devfsd_buffer_lock); - fs_info->devfsd_first_event = entry->next; - if (entry->next == NULL) - fs_info->devfsd_last_event = NULL; - spin_unlock(&fs_info->devfsd_buffer_lock); - for (; de != NULL; de = parent) { - parent = de->parent; - devfs_put(de); - } - kmem_cache_free(devfsd_buf_cache, entry); - if (ival > 0) - atomic_sub(ival, &fs_info->devfsd_overrun_count); - *ppos = 0; - } else - *ppos = rpos; - return tlen; -} /* End Function devfsd_read */ - -static int devfsd_ioctl(struct inode *inode, struct file *file, - unsigned int cmd, unsigned long arg) -{ - int ival; - struct fs_info *fs_info = inode->i_sb->s_fs_info; - - switch (cmd) { - case DEVFSDIOC_GET_PROTO_REV: - ival = DEVFSD_PROTOCOL_REVISION_KERNEL; - if (copy_to_user((void __user *)arg, &ival, sizeof ival)) - return -EFAULT; - break; - case DEVFSDIOC_SET_EVENT_MASK: - /* Ensure only one reader has access to the queue. This scheme will - work even if the global kernel lock were to be removed, because it - doesn't matter who gets in first, as long as only one gets it */ - if (fs_info->devfsd_task == NULL) { - static DEFINE_SPINLOCK(lock); - - if (!spin_trylock(&lock)) - return -EBUSY; - if (fs_info->devfsd_task != NULL) { /* We lost the race... */ - spin_unlock(&lock); - return -EBUSY; - } - fs_info->devfsd_task = current; - spin_unlock(&lock); - fs_info->devfsd_pgrp = - (process_group(current) == - current->pid) ? process_group(current) : 0; - fs_info->devfsd_file = file; - fs_info->devfsd_info = - kmalloc(sizeof *fs_info->devfsd_info, GFP_KERNEL); - if (!fs_info->devfsd_info) { - devfsd_close(inode, file); - return -ENOMEM; - } - } else if (fs_info->devfsd_task != current) - return -EBUSY; - fs_info->devfsd_event_mask = arg; /* Let the masses come forth */ - break; - case DEVFSDIOC_RELEASE_EVENT_QUEUE: - if (fs_info->devfsd_file != file) - return -EPERM; - return devfsd_close(inode, file); - /*break; */ -#ifdef CONFIG_DEVFS_DEBUG - case DEVFSDIOC_SET_DEBUG_MASK: - if (copy_from_user(&ival, (void __user *)arg, sizeof ival)) - return -EFAULT; - devfs_debug = ival; - break; -#endif - default: - return -ENOIOCTLCMD; - } - return 0; -} /* End Function devfsd_ioctl */ - -static int devfsd_close(struct inode *inode, struct file *file) -{ - struct devfsd_buf_entry *entry, *next; - struct fs_info *fs_info = inode->i_sb->s_fs_info; - - if (fs_info->devfsd_file != file) - return 0; - fs_info->devfsd_event_mask = 0; - fs_info->devfsd_file = NULL; - spin_lock(&fs_info->devfsd_buffer_lock); - entry = fs_info->devfsd_first_event; - fs_info->devfsd_first_event = NULL; - fs_info->devfsd_last_event = NULL; - kfree(fs_info->devfsd_info); - fs_info->devfsd_info = NULL; - spin_unlock(&fs_info->devfsd_buffer_lock); - fs_info->devfsd_pgrp = 0; - fs_info->devfsd_task = NULL; - wake_up(&fs_info->revalidate_wait_queue); - for (; entry; entry = next) { - next = entry->next; - kmem_cache_free(devfsd_buf_cache, entry); - } - return 0; -} /* End Function devfsd_close */ - -#ifdef CONFIG_DEVFS_DEBUG -static ssize_t stat_read(struct file *file, char __user *buf, size_t len, - loff_t * ppos) -{ - ssize_t num; - char txt[80]; - - num = sprintf(txt, "Number of entries: %u number of bytes: %u\n", - stat_num_entries, stat_num_bytes) + 1; - if (*ppos >= num) - return 0; - if (*ppos + len > num) - len = num - *ppos; - if (copy_to_user(buf, txt + *ppos, len)) - return -EFAULT; - *ppos += len; - return len; -} /* End Function stat_read */ -#endif - -static int __init init_devfs_fs(void) -{ - int err; - int major; - struct devfs_entry *devfsd; -#ifdef CONFIG_DEVFS_DEBUG - struct devfs_entry *stat; -#endif - - if (_devfs_get_root_entry() == NULL) - return -ENOMEM; - - printk(KERN_INFO "%s: %s Richard Gooch (rgooch@atnf.csiro.au)\n", - DEVFS_NAME, DEVFS_VERSION); - devfsd_buf_cache = kmem_cache_create("devfsd_event", - sizeof(struct devfsd_buf_entry), - 0, 0, NULL, NULL); - if (!devfsd_buf_cache) - OOPS("(): unable to allocate event slab\n"); -#ifdef CONFIG_DEVFS_DEBUG - devfs_debug = devfs_debug_init; - printk(KERN_INFO "%s: devfs_debug: 0x%0x\n", DEVFS_NAME, devfs_debug); -#endif - printk(KERN_INFO "%s: boot_options: 0x%0x\n", DEVFS_NAME, boot_options); - - /* register special device for devfsd communication */ - major = register_chrdev(0, "devfs", &devfs_fops); - if (major < 0) - return major; - - /* And create the entry for ".devfsd" */ - devfsd = _devfs_alloc_entry(".devfsd", 0, S_IFCHR | S_IRUSR | S_IWUSR); - if (devfsd == NULL) - return -ENOMEM; - devfsd->u.dev = MKDEV(major, 0); - _devfs_append_entry(root_entry, devfsd, NULL); - -#ifdef CONFIG_DEVFS_DEBUG - stat = _devfs_alloc_entry(".stat", 0, S_IFCHR | S_IRUGO); - if (stat == NULL) - return -ENOMEM; - stat->u.dev = MKDEV(major, 1); - _devfs_append_entry(root_entry, stat, NULL); -#endif - - err = register_filesystem(&devfs_fs_type); - return err; -} /* End Function init_devfs_fs */ - -void __init mount_devfs_fs(void) -{ - int err; - - if (!(boot_options & OPTION_MOUNT)) - return; - err = do_mount("none", "/dev", "devfs", 0, NULL); - if (err == 0) - printk(KERN_INFO "Mounted devfs on /dev\n"); - else - PRINTK("(): unable to mount devfs, err: %d\n", err); -} /* End Function mount_devfs_fs */ - -module_init(init_devfs_fs) diff --git a/fs/devfs/util.c b/fs/devfs/util.c deleted file mode 100644 index db06d388c9a..00000000000 --- a/fs/devfs/util.c +++ /dev/null @@ -1,97 +0,0 @@ -/* devfs (Device FileSystem) utilities. - - Copyright (C) 1999-2002 Richard Gooch - - This library is free software; you can redistribute it and/or - modify it under the terms of the GNU Library General Public - License as published by the Free Software Foundation; either - version 2 of the License, or (at your option) any later version. - - This library is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - Library General Public License for more details. - - You should have received a copy of the GNU Library General Public - License along with this library; if not, write to the Free - Software Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. - - Richard Gooch may be reached by email at rgooch@atnf.csiro.au - The postal address is: - Richard Gooch, c/o ATNF, P. O. Box 76, Epping, N.S.W., 2121, Australia. - - ChangeLog - - 19991031 Richard Gooch - Created. - 19991103 Richard Gooch - Created <_devfs_convert_name> and supported SCSI and IDE CD-ROMs - 20000203 Richard Gooch - Changed operations pointer type to void *. - 20000621 Richard Gooch - Changed interface to . - 20000622 Richard Gooch - Took account of interface change to . - Took account of interface change to . - 20010519 Richard Gooch - Documentation cleanup. - 20010709 Richard Gooch - Created and . - 20010710 Richard Gooch - Created . - 20010730 Richard Gooch - Documentation typo fix. - 20010806 Richard Gooch - Made and private. - 20010813 Richard Gooch - Fixed bug in : limited to 128 numbers - 20010818 Richard Gooch - Updated major masks up to Linus' "no new majors" proclamation. - Block: were 126 now 122 free, char: were 26 now 19 free. - 20020324 Richard Gooch - Fixed bug in : was clearing beyond - bitfield. - 20020326 Richard Gooch - Fixed bitfield data type for . - Made major bitfield type and initialiser 64 bit safe. - 20020413 Richard Gooch - Fixed shift warning on 64 bit machines. - 20020428 Richard Gooch - Copied and used macro for error messages from fs/devfs/base.c - 20021013 Richard Gooch - Documentation fix. - 20030101 Adam J. Richter - Eliminate DEVFS_SPECIAL_{CHR,BLK}. Use mode_t instead. - 20030106 Christoph Hellwig - Rewrite devfs_{,de}alloc_devnum to look like C code. -*/ -#include -#include -#include -#include -#include -#include -#include - -int devfs_register_tape(const char *name) -{ - char tname[32], dest[64]; - static unsigned int tape_counter; - unsigned int n = tape_counter++; - - sprintf(dest, "../%s", name); - sprintf(tname, "tapes/tape%u", n); - devfs_mk_symlink(tname, dest); - - return n; -} - -EXPORT_SYMBOL(devfs_register_tape); - -void devfs_unregister_tape(int num) -{ - if (num >= 0) - devfs_remove("tapes/tape%u", num); -} - -EXPORT_SYMBOL(devfs_unregister_tape); diff --git a/include/linux/compat_ioctl.h b/include/linux/compat_ioctl.h index 917d62e4148..269d000bb2a 100644 --- a/include/linux/compat_ioctl.h +++ b/include/linux/compat_ioctl.h @@ -567,11 +567,6 @@ COMPATIBLE_IOCTL(AUTOFS_IOC_PROTOSUBVER) COMPATIBLE_IOCTL(AUTOFS_IOC_ASKREGHOST) COMPATIBLE_IOCTL(AUTOFS_IOC_TOGGLEREGHOST) COMPATIBLE_IOCTL(AUTOFS_IOC_ASKUMOUNT) -/* DEVFS */ -COMPATIBLE_IOCTL(DEVFSDIOC_GET_PROTO_REV) -COMPATIBLE_IOCTL(DEVFSDIOC_SET_EVENT_MASK) -COMPATIBLE_IOCTL(DEVFSDIOC_RELEASE_EVENT_QUEUE) -COMPATIBLE_IOCTL(DEVFSDIOC_SET_DEBUG_MASK) /* Raw devices */ COMPATIBLE_IOCTL(RAW_SETBIND) COMPATIBLE_IOCTL(RAW_GETBIND) diff --git a/include/linux/devfs_fs.h b/include/linux/devfs_fs.h deleted file mode 100644 index de236f43187..00000000000 --- a/include/linux/devfs_fs.h +++ /dev/null @@ -1,41 +0,0 @@ -#ifndef _LINUX_DEVFS_FS_H -#define _LINUX_DEVFS_FS_H - -#include - -#define DEVFSD_PROTOCOL_REVISION_KERNEL 5 - -#define DEVFSD_IOCTL_BASE 'd' - -/* These are the various ioctls */ -#define DEVFSDIOC_GET_PROTO_REV _IOR(DEVFSD_IOCTL_BASE, 0, int) -#define DEVFSDIOC_SET_EVENT_MASK _IOW(DEVFSD_IOCTL_BASE, 2, int) -#define DEVFSDIOC_RELEASE_EVENT_QUEUE _IOW(DEVFSD_IOCTL_BASE, 3, int) -#define DEVFSDIOC_SET_DEBUG_MASK _IOW(DEVFSD_IOCTL_BASE, 4, int) - -#define DEVFSD_NOTIFY_REGISTERED 0 -#define DEVFSD_NOTIFY_UNREGISTERED 1 -#define DEVFSD_NOTIFY_ASYNC_OPEN 2 -#define DEVFSD_NOTIFY_CLOSE 3 -#define DEVFSD_NOTIFY_LOOKUP 4 -#define DEVFSD_NOTIFY_CHANGE 5 -#define DEVFSD_NOTIFY_CREATE 6 -#define DEVFSD_NOTIFY_DELETE 7 - -#define DEVFS_PATHLEN 1024 /* Never change this otherwise the - binary interface will change */ - -struct devfsd_notify_struct { /* Use native C types to ensure same types in kernel and user space */ - unsigned int type; /* DEVFSD_NOTIFY_* value */ - unsigned int mode; /* Mode of the inode or device entry */ - unsigned int major; /* Major number of device entry */ - unsigned int minor; /* Minor number of device entry */ - unsigned int uid; /* Uid of process, inode or device entry */ - unsigned int gid; /* Gid of process, inode or device entry */ - unsigned int overrun_count; /* Number of lost events */ - unsigned int namelen; /* Number of characters not including '\0' */ - /* The device name MUST come last */ - char devname[DEVFS_PATHLEN]; /* This will be '\0' terminated */ -}; - -#endif /* _LINUX_DEVFS_FS_H */ diff --git a/include/linux/devfs_fs_kernel.h b/include/linux/devfs_fs_kernel.h index 0d74a6f22ab..ef34d7ffe49 100644 --- a/include/linux/devfs_fs_kernel.h +++ b/include/linux/devfs_fs_kernel.h @@ -4,25 +4,8 @@ #include #include #include - #include -#define DEVFS_SUPER_MAGIC 0x1373 - -#ifdef CONFIG_DEVFS_FS -extern int devfs_mk_bdev(dev_t dev, umode_t mode, const char *fmt, ...) - __attribute__ ((format(printf, 3, 4))); -extern int devfs_mk_cdev(dev_t dev, umode_t mode, const char *fmt, ...) - __attribute__ ((format(printf, 3, 4))); -extern int devfs_mk_symlink(const char *name, const char *link); -extern int devfs_mk_dir(const char *fmt, ...) - __attribute__ ((format(printf, 1, 2))); -extern void devfs_remove(const char *fmt, ...) - __attribute__ ((format(printf, 1, 2))); -extern int devfs_register_tape(const char *name); -extern void devfs_unregister_tape(int num); -extern void mount_devfs_fs(void); -#else /* CONFIG_DEVFS_FS */ static inline int devfs_mk_bdev(dev_t dev, umode_t mode, const char *fmt, ...) { return 0; @@ -53,5 +36,4 @@ static inline void mount_devfs_fs(void) { return; } -#endif /* CONFIG_DEVFS_FS */ #endif /* _LINUX_DEVFS_FS_KERNEL_H */ -- cgit v1.2.3-70-g09d2 From bdaf8529385d5126ef791e8f1914afff8cd59bcf Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Mon, 20 Jun 2005 21:15:16 -0700 Subject: [PATCH] devfs: Remove devfs from the init code This patch removes the devfs code from the init/ directory. Signed-off-by: Greg Kroah-Hartman --- include/linux/devfs_fs_kernel.h | 4 -- init/Makefile | 1 - init/do_mounts.c | 8 +-- init/do_mounts.h | 15 +---- init/do_mounts_devfs.c | 137 ---------------------------------------- init/do_mounts_initrd.c | 6 +- init/do_mounts_md.c | 7 +- init/do_mounts_rd.c | 4 +- 8 files changed, 10 insertions(+), 172 deletions(-) delete mode 100644 init/do_mounts_devfs.c (limited to 'include') diff --git a/include/linux/devfs_fs_kernel.h b/include/linux/devfs_fs_kernel.h index ef34d7ffe49..3dd430e016b 100644 --- a/include/linux/devfs_fs_kernel.h +++ b/include/linux/devfs_fs_kernel.h @@ -32,8 +32,4 @@ static inline int devfs_register_tape(const char *name) static inline void devfs_unregister_tape(int num) { } -static inline void mount_devfs_fs(void) -{ - return; -} #endif /* _LINUX_DEVFS_FS_KERNEL_H */ diff --git a/init/Makefile b/init/Makefile index a2300078f2b..633a268d270 100644 --- a/init/Makefile +++ b/init/Makefile @@ -6,7 +6,6 @@ obj-y := main.o version.o mounts.o initramfs.o obj-$(CONFIG_GENERIC_CALIBRATE_DELAY) += calibrate.o mounts-y := do_mounts.o -mounts-$(CONFIG_DEVFS_FS) += do_mounts_devfs.o mounts-$(CONFIG_BLK_DEV_RAM) += do_mounts_rd.o mounts-$(CONFIG_BLK_DEV_INITRD) += do_mounts_initrd.o mounts-$(CONFIG_BLK_DEV_MD) += do_mounts_md.o diff --git a/init/do_mounts.c b/init/do_mounts.c index 21b3b8f33a7..94aeec7aa91 100644 --- a/init/do_mounts.c +++ b/init/do_mounts.c @@ -325,7 +325,7 @@ static int __init mount_nfs_root(void) { void *data = nfs_root_data(); - create_dev("/dev/root", ROOT_DEV, NULL); + create_dev("/dev/root", ROOT_DEV); if (data && do_mount_root("/dev/root", "nfs", root_mountflags, data) == 0) return 1; @@ -386,7 +386,7 @@ void __init mount_root(void) change_floppy("root floppy"); } #endif - create_dev("/dev/root", ROOT_DEV, root_device_name); + create_dev("/dev/root", ROOT_DEV); mount_block_root("/dev/root", root_mountflags); } @@ -397,8 +397,6 @@ void __init prepare_namespace(void) { int is_floppy; - mount_devfs(); - if (root_delay) { printk(KERN_INFO "Waiting %dsec before mounting root device...\n", root_delay); @@ -428,10 +426,8 @@ void __init prepare_namespace(void) mount_root(); out: - umount_devfs("/dev"); sys_mount(".", "/", NULL, MS_MOVE, NULL); sys_chroot("."); security_sb_post_mountroot(); - mount_devfs_fs (); } diff --git a/init/do_mounts.h b/init/do_mounts.h index e0a7ac9649e..66548efe4c3 100644 --- a/init/do_mounts.h +++ b/init/do_mounts.h @@ -15,25 +15,12 @@ void mount_root(void); extern int root_mountflags; extern char *root_device_name; -#ifdef CONFIG_DEVFS_FS - -void mount_devfs(void); -void umount_devfs(char *path); -int create_dev(char *name, dev_t dev, char *devfs_name); - -#else - -static inline void mount_devfs(void) {} -static inline void umount_devfs(const char *path) {} - -static inline int create_dev(char *name, dev_t dev, char *devfs_name) +static inline int create_dev(char *name, dev_t dev) { sys_unlink(name); return sys_mknod(name, S_IFBLK|0600, new_encode_dev(dev)); } -#endif - #if BITS_PER_LONG == 32 static inline u32 bstat(char *name) { diff --git a/init/do_mounts_devfs.c b/init/do_mounts_devfs.c deleted file mode 100644 index cc526474690..00000000000 --- a/init/do_mounts_devfs.c +++ /dev/null @@ -1,137 +0,0 @@ - -#include -#include -#include - -#include "do_mounts.h" - -void __init mount_devfs(void) -{ - sys_mount("devfs", "/dev", "devfs", 0, NULL); -} - -void __init umount_devfs(char *path) -{ - sys_umount(path, 0); -} - -/* - * If the dir will fit in *buf, return its length. If it won't fit, return - * zero. Return -ve on error. - */ -static int __init do_read_dir(int fd, void *buf, int len) -{ - long bytes, n; - char *p = buf; - sys_lseek(fd, 0, 0); - - for (bytes = 0; bytes < len; bytes += n) { - n = sys_getdents64(fd, (struct linux_dirent64 *)(p + bytes), - len - bytes); - if (n < 0) - return n; - if (n == 0) - return bytes; - } - return 0; -} - -/* - * Try to read all of a directory. Returns the contents at *p, which - * is kmalloced memory. Returns the number of bytes read at *len. Returns - * NULL on error. - */ -static void * __init read_dir(char *path, int *len) -{ - int size; - int fd = sys_open(path, 0, 0); - - *len = 0; - if (fd < 0) - return NULL; - - for (size = 1 << 9; size <= (PAGE_SIZE << MAX_ORDER); size <<= 1) { - void *p = kmalloc(size, GFP_KERNEL); - int n; - if (!p) - break; - n = do_read_dir(fd, p, size); - if (n > 0) { - sys_close(fd); - *len = n; - return p; - } - kfree(p); - if (n == -EINVAL) - continue; /* Try a larger buffer */ - if (n < 0) - break; - } - sys_close(fd); - return NULL; -} - -/* - * recursively scan , looking for a device node of type - */ -static int __init find_in_devfs(char *path, unsigned dev) -{ - char *end = path + strlen(path); - int rest = path + 64 - end; - int size; - char *p = read_dir(path, &size); - char *s; - - if (!p) - return -1; - for (s = p; s < p + size; s += ((struct linux_dirent64 *)s)->d_reclen) { - struct linux_dirent64 *d = (struct linux_dirent64 *)s; - if (strlen(d->d_name) + 2 > rest) - continue; - switch (d->d_type) { - case DT_BLK: - sprintf(end, "/%s", d->d_name); - if (bstat(path) != dev) - break; - kfree(p); - return 0; - case DT_DIR: - if (strcmp(d->d_name, ".") == 0) - break; - if (strcmp(d->d_name, "..") == 0) - break; - sprintf(end, "/%s", d->d_name); - if (find_in_devfs(path, dev) < 0) - break; - kfree(p); - return 0; - } - } - kfree(p); - return -1; -} - -/* - * create a device node called which points to - * if possible, otherwise find a device node - * which matches and make a symlink pointing to it. - */ -int __init create_dev(char *name, dev_t dev, char *devfs_name) -{ - char path[64]; - - sys_unlink(name); - if (devfs_name && devfs_name[0]) { - if (strncmp(devfs_name, "/dev/", 5) == 0) - devfs_name += 5; - sprintf(path, "/dev/%s", devfs_name); - if (sys_access(path, 0) == 0) - return sys_symlink(devfs_name, name); - } - if (!dev) - return -1; - strcpy(path, "/dev"); - if (find_in_devfs(path, new_encode_dev(dev)) < 0) - return -1; - return sys_symlink(path + 5, name); -} diff --git a/init/do_mounts_initrd.c b/init/do_mounts_initrd.c index 405f9031af8..a06f037fa00 100644 --- a/init/do_mounts_initrd.c +++ b/init/do_mounts_initrd.c @@ -44,7 +44,7 @@ static void __init handle_initrd(void) int pid; real_root_dev = new_encode_dev(ROOT_DEV); - create_dev("/dev/root.old", Root_RAM0, NULL); + create_dev("/dev/root.old", Root_RAM0); /* mount initrd on rootfs' /root */ mount_block_root("/dev/root.old", root_mountflags & ~MS_RDONLY); sys_mkdir("/old", 0700); @@ -54,7 +54,6 @@ static void __init handle_initrd(void) sys_chdir("/root"); sys_mount(".", "/", NULL, MS_MOVE, NULL); sys_chroot("."); - mount_devfs_fs (); current->flags |= PF_NOFREEZE; pid = kernel_thread(do_linuxrc, "/linuxrc", SIGCHLD); @@ -71,7 +70,6 @@ static void __init handle_initrd(void) sys_chroot("."); sys_close(old_fd); sys_close(root_fd); - umount_devfs("/old/dev"); if (new_decode_dev(real_root_dev) == Root_RAM0) { sys_chdir("/old"); @@ -107,7 +105,7 @@ static void __init handle_initrd(void) int __init initrd_load(void) { if (mount_initrd) { - create_dev("/dev/ram", Root_RAM0, NULL); + create_dev("/dev/ram", Root_RAM0); /* * Load the initrd data into /dev/ram0. Execute it as initrd * unless /dev/ram0 is supposed to be our actual root device, diff --git a/init/do_mounts_md.c b/init/do_mounts_md.c index f6f36806f84..2429e1bf8c6 100644 --- a/init/do_mounts_md.c +++ b/init/do_mounts_md.c @@ -125,19 +125,18 @@ static void __init md_setup_drive(void) int err = 0; char *devname; mdu_disk_info_t dinfo; - char name[16], devfs_name[16]; + char name[16]; minor = md_setup_args[ent].minor; partitioned = md_setup_args[ent].partitioned; devname = md_setup_args[ent].device_names; sprintf(name, "/dev/md%s%d", partitioned?"_d":"", minor); - sprintf(devfs_name, "/dev/md/%s%d", partitioned?"d":"", minor); if (partitioned) dev = MKDEV(mdp_major, minor << MdpMinorShift); else dev = MKDEV(MD_MAJOR, minor); - create_dev(name, dev, devfs_name); + create_dev(name, dev); for (i = 0; i < MD_SB_DISKS && devname != 0; i++) { char *p; char comp_name[64]; @@ -272,7 +271,7 @@ __setup("md=", md_setup); void __init md_run_setup(void) { - create_dev("/dev/md0", MKDEV(MD_MAJOR, 0), "md/0"); + create_dev("/dev/md0", MKDEV(MD_MAJOR, 0)); if (raid_noautodetect) printk(KERN_INFO "md: Skipping autodetection of RAID arrays. (raid=noautodetect)\n"); else { diff --git a/init/do_mounts_rd.c b/init/do_mounts_rd.c index c2683fcd792..ed652f40f07 100644 --- a/init/do_mounts_rd.c +++ b/init/do_mounts_rd.c @@ -262,8 +262,8 @@ int __init rd_load_disk(int n) { if (rd_prompt) change_floppy("root floppy disk to be loaded into RAM disk"); - create_dev("/dev/root", ROOT_DEV, root_device_name); - create_dev("/dev/ram", MKDEV(RAMDISK_MAJOR, n), NULL); + create_dev("/dev/root", ROOT_DEV); + create_dev("/dev/ram", MKDEV(RAMDISK_MAJOR, n)); return rd_load_image("/dev/root"); } -- cgit v1.2.3-70-g09d2 From aa4148cfc7b3b93eeaf755a7d14f10afaffe9a96 Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Mon, 20 Jun 2005 21:15:16 -0700 Subject: [PATCH] devfs: Remove devfs support from the serial subsystem Also fixes all serial drivers. Signed-off-by: Greg Kroah-Hartman --- drivers/serial/21285.c | 1 - drivers/serial/8250.c | 1 - drivers/serial/at91_serial.c | 1 - drivers/serial/dz.c | 4 ---- drivers/serial/imx.c | 1 - drivers/serial/ip22zilog.c | 1 - drivers/serial/m32r_sio.c | 1 - drivers/serial/mcfserial.c | 1 - drivers/serial/mpc52xx_uart.c | 1 - drivers/serial/mpsc.c | 2 -- drivers/serial/pmac_zilog.c | 1 - drivers/serial/pxa.c | 1 - drivers/serial/s3c2410.c | 2 -- drivers/serial/sa1100.c | 1 - drivers/serial/serial_core.c | 3 +-- drivers/serial/serial_txx9.c | 3 --- drivers/serial/sh-sci.c | 3 --- drivers/serial/sunhv.c | 1 - drivers/serial/sunsab.c | 1 - drivers/serial/sunsu.c | 1 - drivers/serial/sunzilog.c | 1 - drivers/serial/v850e_uart.c | 1 - drivers/serial/vr41xx_siu.c | 1 - include/linux/serial_core.h | 1 - 24 files changed, 1 insertion(+), 34 deletions(-) (limited to 'include') diff --git a/drivers/serial/21285.c b/drivers/serial/21285.c index 7572665a885..9fd0de4b7af 100644 --- a/drivers/serial/21285.c +++ b/drivers/serial/21285.c @@ -479,7 +479,6 @@ static struct uart_driver serial21285_reg = { .owner = THIS_MODULE, .driver_name = "ttyFB", .dev_name = "ttyFB", - .devfs_name = "ttyFB", .major = SERIAL_21285_MAJOR, .minor = SERIAL_21285_MINOR, .nr = 1, diff --git a/drivers/serial/8250.c b/drivers/serial/8250.c index bbf78aaf9e0..f361b356bd1 100644 --- a/drivers/serial/8250.c +++ b/drivers/serial/8250.c @@ -2354,7 +2354,6 @@ int __init serial8250_start_console(struct uart_port *port, char *options) static struct uart_driver serial8250_reg = { .owner = THIS_MODULE, .driver_name = "serial", - .devfs_name = "tts/", .dev_name = "ttyS", .major = TTY_MAJOR, .minor = 64, diff --git a/drivers/serial/at91_serial.c b/drivers/serial/at91_serial.c index db5b25fafed..df9500bdade 100644 --- a/drivers/serial/at91_serial.c +++ b/drivers/serial/at91_serial.c @@ -863,7 +863,6 @@ static struct uart_driver at91_uart = { .owner = THIS_MODULE, .driver_name = "at91_serial", .dev_name = AT91_DEVICENAME, - .devfs_name = AT91_DEVICENAME, .major = SERIAL_AT91_MAJOR, .minor = MINOR_START, .nr = AT91_NR_UART, diff --git a/drivers/serial/dz.c b/drivers/serial/dz.c index bf71bad5c34..466d06cc7d3 100644 --- a/drivers/serial/dz.c +++ b/drivers/serial/dz.c @@ -768,11 +768,7 @@ void __init dz_serial_console_init(void) static struct uart_driver dz_reg = { .owner = THIS_MODULE, .driver_name = "serial", -#ifdef CONFIG_DEVFS - .dev_name = "tts/%d", -#else .dev_name = "ttyS%d", -#endif .major = TTY_MAJOR, .minor = 64, .nr = DZ_NB_PORT, diff --git a/drivers/serial/imx.c b/drivers/serial/imx.c index d202eb4f384..da85bafa094 100644 --- a/drivers/serial/imx.c +++ b/drivers/serial/imx.c @@ -888,7 +888,6 @@ static struct uart_driver imx_reg = { .owner = THIS_MODULE, .driver_name = DRIVER_NAME, .dev_name = "ttySMX", - .devfs_name = "ttsmx/", .major = SERIAL_IMX_MAJOR, .minor = MINOR_START, .nr = ARRAY_SIZE(imx_ports), diff --git a/drivers/serial/ip22zilog.c b/drivers/serial/ip22zilog.c index 651772474ac..56b093ecd77 100644 --- a/drivers/serial/ip22zilog.c +++ b/drivers/serial/ip22zilog.c @@ -1085,7 +1085,6 @@ static struct console ip22zilog_console = { static struct uart_driver ip22zilog_reg = { .owner = THIS_MODULE, .driver_name = "serial", - .devfs_name = "tts/", .dev_name = "ttyS", .major = TTY_MAJOR, .minor = 64, diff --git a/drivers/serial/m32r_sio.c b/drivers/serial/m32r_sio.c index 321a40f33b5..6a2a25d9b59 100644 --- a/drivers/serial/m32r_sio.c +++ b/drivers/serial/m32r_sio.c @@ -1131,7 +1131,6 @@ console_initcall(m32r_sio_console_init); static struct uart_driver m32r_sio_reg = { .owner = THIS_MODULE, .driver_name = "sio", - .devfs_name = "tts/", .dev_name = "ttyS", .major = TTY_MAJOR, .minor = 64, diff --git a/drivers/serial/mcfserial.c b/drivers/serial/mcfserial.c index 8ad24293436..29c0630e3e6 100644 --- a/drivers/serial/mcfserial.c +++ b/drivers/serial/mcfserial.c @@ -1713,7 +1713,6 @@ mcfrs_init(void) /* Initialize the tty_driver structure */ mcfrs_serial_driver->owner = THIS_MODULE; mcfrs_serial_driver->name = "ttyS"; - mcfrs_serial_driver->devfs_name = "ttys/"; mcfrs_serial_driver->driver_name = "serial"; mcfrs_serial_driver->major = TTY_MAJOR; mcfrs_serial_driver->minor_start = 64; diff --git a/drivers/serial/mpc52xx_uart.c b/drivers/serial/mpc52xx_uart.c index 6459edc7f5c..1aa34844218 100644 --- a/drivers/serial/mpc52xx_uart.c +++ b/drivers/serial/mpc52xx_uart.c @@ -693,7 +693,6 @@ static struct uart_driver mpc52xx_uart_driver = { .owner = THIS_MODULE, .driver_name = "mpc52xx_psc_uart", .dev_name = "ttyPSC", - .devfs_name = "ttyPSC", .major = SERIAL_PSC_MAJOR, .minor = SERIAL_PSC_MINOR, .nr = MPC52xx_PSC_MAXNUM, diff --git a/drivers/serial/mpsc.c b/drivers/serial/mpsc.c index 94681922ea0..1cd102f84bf 100644 --- a/drivers/serial/mpsc.c +++ b/drivers/serial/mpsc.c @@ -315,7 +315,6 @@ struct mpsc_port_info *mpsc_device_remove(int index); #define MPSC_MAJOR 204 #define MPSC_MINOR_START 44 #define MPSC_DRIVER_NAME "MPSC" -#define MPSC_DEVFS_NAME "ttymm/" #define MPSC_DEV_NAME "ttyMM" #define MPSC_VERSION "1.00" @@ -1863,7 +1862,6 @@ static struct platform_driver mpsc_shared_driver = { static struct uart_driver mpsc_reg = { .owner = THIS_MODULE, .driver_name = MPSC_DRIVER_NAME, - .devfs_name = MPSC_DEVFS_NAME, .dev_name = MPSC_DEV_NAME, .major = MPSC_MAJOR, .minor = MPSC_MINOR_START, diff --git a/drivers/serial/pmac_zilog.c b/drivers/serial/pmac_zilog.c index 513ff859770..e3ba7e17a24 100644 --- a/drivers/serial/pmac_zilog.c +++ b/drivers/serial/pmac_zilog.c @@ -101,7 +101,6 @@ static DEFINE_MUTEX(pmz_irq_mutex); static struct uart_driver pmz_uart_reg = { .owner = THIS_MODULE, .driver_name = "ttyS", - .devfs_name = "tts/", .dev_name = "ttyS", .major = TTY_MAJOR, }; diff --git a/drivers/serial/pxa.c b/drivers/serial/pxa.c index ae364956854..0fa0ccc9ed2 100644 --- a/drivers/serial/pxa.c +++ b/drivers/serial/pxa.c @@ -780,7 +780,6 @@ static struct uart_pxa_port serial_pxa_ports[] = { static struct uart_driver serial_pxa_reg = { .owner = THIS_MODULE, .driver_name = "PXA serial", - .devfs_name = "tts/", .dev_name = "ttyS", .major = TTY_MAJOR, .minor = 64, diff --git a/drivers/serial/s3c2410.c b/drivers/serial/s3c2410.c index 837b6da520b..4c62ab949ec 100644 --- a/drivers/serial/s3c2410.c +++ b/drivers/serial/s3c2410.c @@ -149,7 +149,6 @@ s3c24xx_serial_dbg(const char *fmt, ...) /* UART name and device definitions */ #define S3C24XX_SERIAL_NAME "ttySAC" -#define S3C24XX_SERIAL_DEVFS "tts/" #define S3C24XX_SERIAL_MAJOR 204 #define S3C24XX_SERIAL_MINOR 64 @@ -952,7 +951,6 @@ static struct uart_driver s3c24xx_uart_drv = { .nr = 3, .cons = S3C24XX_SERIAL_CONSOLE, .driver_name = S3C24XX_SERIAL_NAME, - .devfs_name = S3C24XX_SERIAL_DEVFS, .major = S3C24XX_SERIAL_MAJOR, .minor = S3C24XX_SERIAL_MINOR, }; diff --git a/drivers/serial/sa1100.c b/drivers/serial/sa1100.c index c2d9068b491..8bbd8567669 100644 --- a/drivers/serial/sa1100.c +++ b/drivers/serial/sa1100.c @@ -816,7 +816,6 @@ static struct uart_driver sa1100_reg = { .owner = THIS_MODULE, .driver_name = "ttySA", .dev_name = "ttySA", - .devfs_name = "ttySA", .major = SERIAL_SA1100_MAJOR, .minor = MINOR_START, .nr = NR_PORTS, diff --git a/drivers/serial/serial_core.c b/drivers/serial/serial_core.c index 17839e753e4..84d70d85559 100644 --- a/drivers/serial/serial_core.c +++ b/drivers/serial/serial_core.c @@ -2153,7 +2153,6 @@ int uart_register_driver(struct uart_driver *drv) normal->owner = drv->owner; normal->driver_name = drv->driver_name; - normal->devfs_name = drv->devfs_name; normal->name = drv->dev_name; normal->major = drv->major; normal->minor_start = drv->minor; @@ -2312,7 +2311,7 @@ int uart_remove_one_port(struct uart_driver *drv, struct uart_port *port) mutex_unlock(&state->mutex); /* - * Remove the devices from devfs + * Remove the devices from the tty layer */ tty_unregister_device(drv->tty_driver, port->line); diff --git a/drivers/serial/serial_txx9.c b/drivers/serial/serial_txx9.c index 3bdee64d1a9..a901a7e446f 100644 --- a/drivers/serial/serial_txx9.c +++ b/drivers/serial/serial_txx9.c @@ -69,12 +69,10 @@ static char *serial_name = "TX39/49 Serial driver"; #if !defined(CONFIG_SERIAL_TXX9_STDSERIAL) /* "ttyS" is used for standard serial driver */ #define TXX9_TTY_NAME "ttyTX" -#define TXX9_TTY_DEVFS_NAME "tttx/" #define TXX9_TTY_MINOR_START (64 + 64) /* ttyTX0(128), ttyTX1(129) */ #else /* acts like standard serial driver */ #define TXX9_TTY_NAME "ttyS" -#define TXX9_TTY_DEVFS_NAME "tts/" #define TXX9_TTY_MINOR_START 64 #endif #define TXX9_TTY_MAJOR TTY_MAJOR @@ -971,7 +969,6 @@ console_initcall(serial_txx9_console_init); static struct uart_driver serial_txx9_reg = { .owner = THIS_MODULE, .driver_name = "serial_txx9", - .devfs_name = TXX9_TTY_DEVFS_NAME, .dev_name = TXX9_TTY_NAME, .major = TXX9_TTY_MAJOR, .minor = TXX9_TTY_MINOR_START, diff --git a/drivers/serial/sh-sci.c b/drivers/serial/sh-sci.c index 44f6bf79bbe..d97f3ca6cc2 100644 --- a/drivers/serial/sh-sci.c +++ b/drivers/serial/sh-sci.c @@ -1699,9 +1699,6 @@ static char banner[] __initdata = static struct uart_driver sci_uart_driver = { .owner = THIS_MODULE, .driver_name = "sci", -#ifdef CONFIG_DEVFS_FS - .devfs_name = "ttsc/", -#endif .dev_name = "ttySC", .major = SCI_MAJOR, .minor = SCI_MINOR_START, diff --git a/drivers/serial/sunhv.c b/drivers/serial/sunhv.c index ba22e256c6f..d36bc400339 100644 --- a/drivers/serial/sunhv.c +++ b/drivers/serial/sunhv.c @@ -353,7 +353,6 @@ static struct uart_ops sunhv_pops = { static struct uart_driver sunhv_reg = { .owner = THIS_MODULE, .driver_name = "serial", - .devfs_name = "tts/", .dev_name = "ttyS", .major = TTY_MAJOR, }; diff --git a/drivers/serial/sunsab.c b/drivers/serial/sunsab.c index e4c0fd2d6a9..7da02d11c36 100644 --- a/drivers/serial/sunsab.c +++ b/drivers/serial/sunsab.c @@ -851,7 +851,6 @@ static struct uart_ops sunsab_pops = { static struct uart_driver sunsab_reg = { .owner = THIS_MODULE, .driver_name = "serial", - .devfs_name = "tts/", .dev_name = "ttyS", .major = TTY_MAJOR, }; diff --git a/drivers/serial/sunsu.c b/drivers/serial/sunsu.c index 0268b307c01..6e28c25138c 100644 --- a/drivers/serial/sunsu.c +++ b/drivers/serial/sunsu.c @@ -1265,7 +1265,6 @@ out: static struct uart_driver sunsu_reg = { .owner = THIS_MODULE, .driver_name = "serial", - .devfs_name = "tts/", .dev_name = "ttyS", .major = TTY_MAJOR, }; diff --git a/drivers/serial/sunzilog.c b/drivers/serial/sunzilog.c index 76c9bac9271..9f42677287a 100644 --- a/drivers/serial/sunzilog.c +++ b/drivers/serial/sunzilog.c @@ -1017,7 +1017,6 @@ static int zilog_irq = -1; static struct uart_driver sunzilog_reg = { .owner = THIS_MODULE, .driver_name = "ttyS", - .devfs_name = "tts/", .dev_name = "ttyS", .major = TTY_MAJOR, }; diff --git a/drivers/serial/v850e_uart.c b/drivers/serial/v850e_uart.c index df705fda424..a0da2aaf71c 100644 --- a/drivers/serial/v850e_uart.c +++ b/drivers/serial/v850e_uart.c @@ -468,7 +468,6 @@ static struct uart_ops v850e_uart_ops = { static struct uart_driver v850e_uart_driver = { .owner = THIS_MODULE, .driver_name = "v850e_uart", - .devfs_name = "tts/", .dev_name = "ttyS", .major = TTY_MAJOR, .minor = V850E_UART_MINOR_BASE, diff --git a/drivers/serial/vr41xx_siu.c b/drivers/serial/vr41xx_siu.c index df5e8713fa3..017571ffa19 100644 --- a/drivers/serial/vr41xx_siu.c +++ b/drivers/serial/vr41xx_siu.c @@ -911,7 +911,6 @@ static struct uart_driver siu_uart_driver = { .owner = THIS_MODULE, .driver_name = "SIU", .dev_name = "ttyVR", - .devfs_name = "ttvr/", .major = SIU_MAJOR, .minor = SIU_MINOR_BASE, .cons = SERIAL_VR41XX_CONSOLE, diff --git a/include/linux/serial_core.h b/include/linux/serial_core.h index 951c4e85827..fc1104a2cfa 100644 --- a/include/linux/serial_core.h +++ b/include/linux/serial_core.h @@ -336,7 +336,6 @@ struct uart_driver { struct module *owner; const char *driver_name; const char *dev_name; - const char *devfs_name; int major; int minor; int nr; -- cgit v1.2.3-70-g09d2 From 94f6c59dcf16f10a20fbe3d1f098b159433f94bd Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Mon, 20 Jun 2005 21:15:16 -0700 Subject: [PATCH] devfs: Remove devfs support from the ide subsystem. Also removes the ide drive devfs_name field as it's no longer needed Signed-off-by: Greg Kroah-Hartman --- drivers/ide/ide-cd.c | 2 -- drivers/ide/ide-disk.c | 2 -- drivers/ide/ide-floppy.c | 1 - drivers/ide/ide-probe.c | 11 ----------- drivers/ide/ide-tape.c | 12 ------------ drivers/ide/ide.c | 10 +--------- include/linux/ide.h | 1 - 7 files changed, 1 insertion(+), 38 deletions(-) (limited to 'include') diff --git a/drivers/ide/ide-cd.c b/drivers/ide/ide-cd.c index 99fa42402e7..bfafd4846a0 100644 --- a/drivers/ide/ide-cd.c +++ b/drivers/ide/ide-cd.c @@ -3527,8 +3527,6 @@ static int ide_cd_probe(ide_drive_t *drive) drive->driver_data = info; g->minors = 1; - snprintf(g->devfs_name, sizeof(g->devfs_name), - "%s/cd", drive->devfs_name); g->driverfs_dev = &drive->gendev; g->flags = GENHD_FL_CD | GENHD_FL_REMOVABLE; if (ide_cdrom_setup(drive)) { diff --git a/drivers/ide/ide-disk.c b/drivers/ide/ide-disk.c index a5017de72da..d7952632071 100644 --- a/drivers/ide/ide-disk.c +++ b/drivers/ide/ide-disk.c @@ -1018,7 +1018,6 @@ static void ide_disk_release(struct kref *kref) struct gendisk *g = idkp->disk; drive->driver_data = NULL; - drive->devfs_name[0] = '\0'; g->private_data = NULL; put_disk(g); kfree(idkp); @@ -1222,7 +1221,6 @@ static int ide_disk_probe(ide_drive_t *drive) drive->attach = 1; g->minors = 1 << PARTN_BITS; - strcpy(g->devfs_name, drive->devfs_name); g->driverfs_dev = &drive->gendev; g->flags = drive->removable ? GENHD_FL_REMOVABLE : 0; set_capacity(g, idedisk_capacity(drive)); diff --git a/drivers/ide/ide-floppy.c b/drivers/ide/ide-floppy.c index 4656587aa2f..68628327c0f 100644 --- a/drivers/ide/ide-floppy.c +++ b/drivers/ide/ide-floppy.c @@ -2176,7 +2176,6 @@ static int ide_floppy_probe(ide_drive_t *drive) g->minors = 1 << PARTN_BITS; g->driverfs_dev = &drive->gendev; - strcpy(g->devfs_name, drive->devfs_name); g->flags = drive->removable ? GENHD_FL_REMOVABLE : 0; g->fops = &idefloppy_ops; drive->attach = 1; diff --git a/drivers/ide/ide-probe.c b/drivers/ide/ide-probe.c index 9ebf8ae2a5e..0d5038a2856 100644 --- a/drivers/ide/ide-probe.c +++ b/drivers/ide/ide-probe.c @@ -47,7 +47,6 @@ #include #include #include -#include #include #include #include @@ -1279,10 +1278,6 @@ static void drive_release_dev (struct device *dev) ide_drive_t *drive = container_of(dev, ide_drive_t, gendev); spin_lock_irq(&ide_lock); - if (drive->devfs_name[0] != '\0') { - devfs_remove(drive->devfs_name); - drive->devfs_name[0] = '\0'; - } ide_remove_drive_from_hwgroup(drive); kfree(drive->id); drive->id = NULL; @@ -1316,12 +1311,6 @@ static void init_gendisk (ide_hwif_t *hwif) drive->gendev.bus = &ide_bus_type; drive->gendev.driver_data = drive; drive->gendev.release = drive_release_dev; - if (drive->present) { - sprintf(drive->devfs_name, "ide/host%d/bus%d/target%d/lun%d", - (hwif->channel && hwif->mate) ? - hwif->mate->index : hwif->index, - hwif->channel, unit, drive->lun); - } } blk_register_region(MKDEV(hwif->major, 0), MAX_DRIVES << PARTN_BITS, THIS_MODULE, ata_probe, ata_lock, hwif); diff --git a/drivers/ide/ide-tape.c b/drivers/ide/ide-tape.c index 09f3a7dab28..4b91101e12b 100644 --- a/drivers/ide/ide-tape.c +++ b/drivers/ide/ide-tape.c @@ -435,7 +435,6 @@ #include #include #include -#include #include #include #include @@ -4726,9 +4725,6 @@ static void ide_tape_release(struct kref *kref) MKDEV(IDETAPE_MAJOR, tape->minor)); class_device_destroy(idetape_sysfs_class, MKDEV(IDETAPE_MAJOR, tape->minor + 128)); - devfs_remove("%s/mt", drive->devfs_name); - devfs_remove("%s/mtn", drive->devfs_name); - devfs_unregister_tape(g->number); idetape_devs[tape->minor] = NULL; g->private_data = NULL; put_disk(g); @@ -4902,14 +4898,6 @@ static int ide_tape_probe(ide_drive_t *drive) class_device_create(idetape_sysfs_class, NULL, MKDEV(IDETAPE_MAJOR, minor + 128), &drive->gendev, "n%s", tape->name); - devfs_mk_cdev(MKDEV(HWIF(drive)->major, minor), - S_IFCHR | S_IRUGO | S_IWUGO, - "%s/mt", drive->devfs_name); - devfs_mk_cdev(MKDEV(HWIF(drive)->major, minor + 128), - S_IFCHR | S_IRUGO | S_IWUGO, - "%s/mtn", drive->devfs_name); - - g->number = devfs_register_tape(drive->devfs_name); g->fops = &idetape_block_ops; ide_register_region(g); diff --git a/drivers/ide/ide.c b/drivers/ide/ide.c index 59fe358048b..1cdf4420516 100644 --- a/drivers/ide/ide.c +++ b/drivers/ide/ide.c @@ -147,7 +147,6 @@ #include #include #include -#include #include #include #include @@ -592,13 +591,8 @@ void ide_unregister(unsigned int index) goto abort; for (unit = 0; unit < MAX_DRIVES; ++unit) { drive = &hwif->drives[unit]; - if (!drive->present) { - if (drive->devfs_name[0] != '\0') { - devfs_remove(drive->devfs_name); - drive->devfs_name[0] = '\0'; - } + if (!drive->present) continue; - } spin_unlock_irq(&ide_lock); device_unregister(&drive->gendev); wait_for_completion(&drive->gendev_rel_comp); @@ -1996,7 +1990,6 @@ EXPORT_SYMBOL_GPL(ide_bus_type); static int __init ide_init(void) { printk(KERN_INFO "Uniform Multi-Platform E-IDE driver " REVISION "\n"); - devfs_mk_dir("ide"); system_bus_speed = ide_system_bus_speed(); bus_register(&ide_bus_type); @@ -2074,7 +2067,6 @@ void cleanup_module (void) #ifdef CONFIG_PROC_FS proc_ide_destroy(); #endif - devfs_remove("ide"); bus_unregister(&ide_bus_type); } diff --git a/include/linux/ide.h b/include/linux/ide.h index ef7bef207f4..eeb52f4b813 100644 --- a/include/linux/ide.h +++ b/include/linux/ide.h @@ -552,7 +552,6 @@ typedef struct ide_drive_s { struct hd_driveid *id; /* drive model identification info */ struct proc_dir_entry *proc; /* /proc/ide/ directory entry */ struct ide_settings_s *settings;/* /proc/ide/ drive settings */ - char devfs_name[64]; /* devfs crap */ struct hwif_s *hwif; /* actually (ide_hwif_t *) */ -- cgit v1.2.3-70-g09d2 From 0e6c62da7cd929b0389fc4a7e41464bb738647dc Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Mon, 20 Jun 2005 21:15:16 -0700 Subject: [PATCH] devfs: Remove devfs_*_tape() functions from the kernel tree Removes the devfs_register_tape() and devfs_unregister_tape() functions and all callers of them. Signed-off-by: Greg Kroah-Hartman --- drivers/char/viotape.c | 3 --- include/linux/devfs_fs_kernel.h | 7 ------- 2 files changed, 10 deletions(-) (limited to 'include') diff --git a/drivers/char/viotape.c b/drivers/char/viotape.c index 11c7e9de595..ebc337acca1 100644 --- a/drivers/char/viotape.c +++ b/drivers/char/viotape.c @@ -246,7 +246,6 @@ static struct device *tape_device[VIOTAPE_MAX_TAPE]; */ static struct { unsigned char cur_part; - int dev_handle; unsigned char part_stat_rwi[MAX_PARTITIONS]; } state[VIOTAPE_MAX_TAPE]; @@ -964,7 +963,6 @@ static int viotape_probe(struct vio_dev *vdev, const struct vio_device_id *id) devfs_mk_cdev(MKDEV(VIOTAPE_MAJOR, i | 0x80), S_IFCHR | S_IRUSR | S_IWUSR, "iseries/nvt%d", i); sprintf(tapename, "iseries/vt%d", i); - state[i].dev_handle = devfs_register_tape(tapename); printk(VIOTAPE_KERN_INFO "tape %s is iSeries " "resource %10.10s type %4.4s, model %3.3s\n", tapename, viotape_unitinfo[i].rsrcname, @@ -978,7 +976,6 @@ static int viotape_remove(struct vio_dev *vdev) devfs_remove("iseries/nvt%d", i); devfs_remove("iseries/vt%d", i); - devfs_unregister_tape(state[i].dev_handle); class_device_destroy(tape_class, MKDEV(VIOTAPE_MAJOR, i | 0x80)); class_device_destroy(tape_class, MKDEV(VIOTAPE_MAJOR, i)); return 0; diff --git a/include/linux/devfs_fs_kernel.h b/include/linux/devfs_fs_kernel.h index 3dd430e016b..d1dd876bc1a 100644 --- a/include/linux/devfs_fs_kernel.h +++ b/include/linux/devfs_fs_kernel.h @@ -25,11 +25,4 @@ static inline int devfs_mk_dir(const char *fmt, ...) static inline void devfs_remove(const char *fmt, ...) { } -static inline int devfs_register_tape(const char *name) -{ - return -1; -} -static inline void devfs_unregister_tape(int num) -{ -} #endif /* _LINUX_DEVFS_FS_KERNEL_H */ -- cgit v1.2.3-70-g09d2 From 95dc112a5770dc670a1b45a3d9ee346fdd2b2697 Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Mon, 20 Jun 2005 21:15:16 -0700 Subject: [PATCH] devfs: Remove devfs_mk_dir() function from the kernel tree Removes the devfs_mk_dir() function and all callers of it. Signed-off-by: Greg Kroah-Hartman --- arch/um/drivers/ubd_kern.c | 2 -- drivers/block/acsi_slm.c | 1 - drivers/block/floppy.c | 2 -- drivers/block/loop.c | 2 -- drivers/block/nbd.c | 1 - drivers/block/paride/pg.c | 1 - drivers/block/paride/pt.c | 1 - drivers/block/rd.c | 2 -- drivers/block/swim3.c | 2 -- drivers/block/sx8.c | 2 -- drivers/block/ub.c | 1 - drivers/cdrom/sbpcd.c | 2 -- drivers/char/ipmi/ipmi_devintf.c | 2 -- drivers/char/istallion.c | 1 - drivers/char/lp.c | 1 - drivers/char/ppdev.c | 1 - drivers/char/pty.c | 1 - drivers/char/stallion.c | 1 - drivers/char/tipar.c | 3 --- drivers/md/dm-ioctl.c | 1 - drivers/md/md.c | 1 - drivers/media/dvb/dvb-core/dvbdev.c | 3 --- drivers/mmc/mmc_block.c | 1 - drivers/net/wan/cosa.c | 1 - drivers/s390/block/dasd.c | 3 --- drivers/s390/block/xpram.c | 2 -- drivers/sbus/char/bpp.c | 1 - drivers/sbus/char/vfc_dev.c | 1 - drivers/video/fbmem.c | 1 - fs/coda/psdev.c | 1 - include/linux/devfs_fs_kernel.h | 4 ---- mm/shmem.c | 4 +--- mm/tiny-shmem.c | 3 --- 33 files changed, 1 insertion(+), 55 deletions(-) (limited to 'include') diff --git a/arch/um/drivers/ubd_kern.c b/arch/um/drivers/ubd_kern.c index 0897852b09a..8f6bb937519 100644 --- a/arch/um/drivers/ubd_kern.c +++ b/arch/um/drivers/ubd_kern.c @@ -841,7 +841,6 @@ int ubd_init(void) { int i; - devfs_mk_dir("ubd"); if (register_blkdev(MAJOR_NR, "ubd")) return -1; @@ -855,7 +854,6 @@ int ubd_init(void) char name[sizeof("ubd_nnn\0")]; snprintf(name, sizeof(name), "ubd_%d", fake_major); - devfs_mk_dir(name); if (register_blkdev(fake_major, "ubd")) return -1; } diff --git a/drivers/block/acsi_slm.c b/drivers/block/acsi_slm.c index 4cb9c133628..1ecbcc3da14 100644 --- a/drivers/block/acsi_slm.c +++ b/drivers/block/acsi_slm.c @@ -1005,7 +1005,6 @@ int slm_init( void ) BufferP = SLMBuffer; SLMState = IDLE; - devfs_mk_dir("slm"); for (i = 0; i < MAX_SLM; i++) { devfs_mk_cdev(MKDEV(ACSI_MAJOR, i), S_IFCHR|S_IRUSR|S_IWUSR, "slm/%d", i); diff --git a/drivers/block/floppy.c b/drivers/block/floppy.c index dff1e67b1dd..03a4dff05f2 100644 --- a/drivers/block/floppy.c +++ b/drivers/block/floppy.c @@ -4261,8 +4261,6 @@ static int __init floppy_init(void) motor_off_timer[dr].function = motor_off_callback; } - devfs_mk_dir("floppy"); - err = register_blkdev(FLOPPY_MAJOR, "fd"); if (err) goto out_devfs_remove; diff --git a/drivers/block/loop.c b/drivers/block/loop.c index 3c74ea729fc..f650d20f2a1 100644 --- a/drivers/block/loop.c +++ b/drivers/block/loop.c @@ -1277,8 +1277,6 @@ static int __init loop_init(void) goto out_mem3; } - devfs_mk_dir("loop"); - for (i = 0; i < max_loop; i++) { struct loop_device *lo = &loop_dev[i]; struct gendisk *disk = disks[i]; diff --git a/drivers/block/nbd.c b/drivers/block/nbd.c index 7f554f2ed07..618e258ca5d 100644 --- a/drivers/block/nbd.c +++ b/drivers/block/nbd.c @@ -642,7 +642,6 @@ static int __init nbd_init(void) printk(KERN_INFO "nbd: registered device at major %d\n", NBD_MAJOR); dprintk(DBG_INIT, "nbd: debugflags=0x%x\n", debugflags); - devfs_mk_dir("nbd"); for (i = 0; i < nbds_max; i++) { struct gendisk *disk = nbd_dev[i].disk; nbd_dev[i].file = NULL; diff --git a/drivers/block/paride/pg.c b/drivers/block/paride/pg.c index 79b86825403..9f72bdd2e20 100644 --- a/drivers/block/paride/pg.c +++ b/drivers/block/paride/pg.c @@ -674,7 +674,6 @@ static int __init pg_init(void) err = PTR_ERR(pg_class); goto out_chrdev; } - devfs_mk_dir("pg"); for (unit = 0; unit < PG_UNITS; unit++) { struct pg *dev = &devices[unit]; if (dev->present) { diff --git a/drivers/block/paride/pt.c b/drivers/block/paride/pt.c index d2013d36240..7adadd5b29f 100644 --- a/drivers/block/paride/pt.c +++ b/drivers/block/paride/pt.c @@ -971,7 +971,6 @@ static int __init pt_init(void) goto out_chrdev; } - devfs_mk_dir("pt"); for (unit = 0; unit < PT_UNITS; unit++) if (pt[unit].present) { class_device_create(pt_class, NULL, MKDEV(major, unit), diff --git a/drivers/block/rd.c b/drivers/block/rd.c index 940bfd7951e..67821a8c621 100644 --- a/drivers/block/rd.c +++ b/drivers/block/rd.c @@ -442,8 +442,6 @@ static int __init rd_init(void) goto out; } - devfs_mk_dir("rd"); - for (i = 0; i < CONFIG_BLK_DEV_RAM_COUNT; i++) { struct gendisk *disk = rd_disks[i]; diff --git a/drivers/block/swim3.c b/drivers/block/swim3.c index 01f042f6f1c..018b2921e8f 100644 --- a/drivers/block/swim3.c +++ b/drivers/block/swim3.c @@ -1019,8 +1019,6 @@ int swim3_init(void) int err = -ENOMEM; int i; - devfs_mk_dir("floppy"); - swim = find_devices("floppy"); while (swim && (floppy_count < MAX_FLOPPIES)) { diff --git a/drivers/block/sx8.c b/drivers/block/sx8.c index 2ae08b343b9..9fb3f51e574 100644 --- a/drivers/block/sx8.c +++ b/drivers/block/sx8.c @@ -1672,8 +1672,6 @@ static int carm_init_one (struct pci_dev *pdev, const struct pci_device_id *ent) if (host->flags & FL_DYN_MAJOR) host->major = rc; - devfs_mk_dir(DRV_NAME); - rc = carm_init_disks(host); if (rc) goto err_out_blkdev_disks; diff --git a/drivers/block/ub.c b/drivers/block/ub.c index 60e9a9457c6..088e5ba15c2 100644 --- a/drivers/block/ub.c +++ b/drivers/block/ub.c @@ -2445,7 +2445,6 @@ static int __init ub_init(void) if ((rc = register_blkdev(UB_MAJOR, DRV_NAME)) != 0) goto err_regblkdev; - devfs_mk_dir(DEVFS_NAME); if ((rc = usb_register(&ub_driver)) != 0) goto err_register; diff --git a/drivers/cdrom/sbpcd.c b/drivers/cdrom/sbpcd.c index 05c9e865eca..ee4499acf1a 100644 --- a/drivers/cdrom/sbpcd.c +++ b/drivers/cdrom/sbpcd.c @@ -5808,8 +5808,6 @@ int __init sbpcd_init(void) return -ENOMEM; } - devfs_mk_dir("sbp"); - for (j=0;jnum = num; adap->name = name; adap->module = module; @@ -410,8 +409,6 @@ static int __init init_dvbdev(void) goto error; } - devfs_mk_dir("dvb"); - dvb_class = class_create(THIS_MODULE, "dvb"); if (IS_ERR(dvb_class)) { retval = PTR_ERR(dvb_class); diff --git a/drivers/mmc/mmc_block.c b/drivers/mmc/mmc_block.c index 587458b370b..82c660bf165 100644 --- a/drivers/mmc/mmc_block.c +++ b/drivers/mmc/mmc_block.c @@ -555,7 +555,6 @@ static int __init mmc_blk_init(void) if (major == 0) major = res; - devfs_mk_dir("mmc"); return mmc_register_driver(&mmc_driver); out: diff --git a/drivers/net/wan/cosa.c b/drivers/net/wan/cosa.c index e392ee8b37a..62b7087e44d 100644 --- a/drivers/net/wan/cosa.c +++ b/drivers/net/wan/cosa.c @@ -393,7 +393,6 @@ static int __init cosa_init(void) err = -ENODEV; goto out; } - devfs_mk_dir("cosa"); cosa_class = class_create(THIS_MODULE, "cosa"); if (IS_ERR(cosa_class)) { err = PTR_ERR(cosa_class); diff --git a/drivers/s390/block/dasd.c b/drivers/s390/block/dasd.c index cfb1fff3787..09bf44fdb1d 100644 --- a/drivers/s390/block/dasd.c +++ b/drivers/s390/block/dasd.c @@ -2111,9 +2111,6 @@ dasd_init(void) dasd_diag_discipline_pointer = NULL; - rc = devfs_mk_dir("dasd"); - if (rc) - goto failed; rc = dasd_devmap_init(); if (rc) goto failed; diff --git a/drivers/s390/block/xpram.c b/drivers/s390/block/xpram.c index 54ecd548c31..36bc01f83a5 100644 --- a/drivers/s390/block/xpram.c +++ b/drivers/s390/block/xpram.c @@ -439,8 +439,6 @@ static int __init xpram_setup_blkdev(void) if (rc < 0) goto out; - devfs_mk_dir("slram"); - /* * Assign the other needed values: make request function, sizes and * hardsect size. All the minor devices feature the same value. diff --git a/drivers/sbus/char/bpp.c b/drivers/sbus/char/bpp.c index ccb20a6f5f3..03f6e988069 100644 --- a/drivers/sbus/char/bpp.c +++ b/drivers/sbus/char/bpp.c @@ -1031,7 +1031,6 @@ static int __init bpp_init(void) instances[idx].opened = 0; probeLptPort(idx); } - devfs_mk_dir("bpp"); for (idx = 0; idx < BPP_NO; idx++) { devfs_mk_cdev(MKDEV(BPP_MAJOR, idx), S_IFCHR | S_IRUSR | S_IWUSR, "bpp/%d", idx); diff --git a/drivers/sbus/char/vfc_dev.c b/drivers/sbus/char/vfc_dev.c index ddcd330b9e8..31232ce7207 100644 --- a/drivers/sbus/char/vfc_dev.c +++ b/drivers/sbus/char/vfc_dev.c @@ -677,7 +677,6 @@ static int vfc_probe(void) kfree(vfc_dev_lst); return -EIO; } - devfs_mk_dir("vfc"); instance = 0; for_all_sbusdev(sdev, sbus) { if (strcmp(sdev->prom_name, "vfc") == 0) { diff --git a/drivers/video/fbmem.c b/drivers/video/fbmem.c index 31143afe7c9..93546a848f4 100644 --- a/drivers/video/fbmem.c +++ b/drivers/video/fbmem.c @@ -1432,7 +1432,6 @@ fbmem_init(void) { create_proc_read_entry("fb", 0, NULL, fbmem_read_proc, NULL); - devfs_mk_dir("fb"); if (register_chrdev(FB_MAJOR,"fb",&fb_fops)) printk("unable to get major %d for fb devs\n", FB_MAJOR); diff --git a/fs/coda/psdev.c b/fs/coda/psdev.c index 7caee8d8ea3..a193823f905 100644 --- a/fs/coda/psdev.c +++ b/fs/coda/psdev.c @@ -365,7 +365,6 @@ static int init_coda_psdev(void) err = PTR_ERR(coda_psdev_class); goto out_chrdev; } - devfs_mk_dir ("coda"); for (i = 0; i < MAX_CODADEVS; i++) { class_device_create(coda_psdev_class, NULL, MKDEV(CODA_PSDEV_MAJOR,i), NULL, "cfs%d", i); diff --git a/include/linux/devfs_fs_kernel.h b/include/linux/devfs_fs_kernel.h index d1dd876bc1a..ef8766e172e 100644 --- a/include/linux/devfs_fs_kernel.h +++ b/include/linux/devfs_fs_kernel.h @@ -18,10 +18,6 @@ static inline int devfs_mk_symlink(const char *name, const char *link) { return 0; } -static inline int devfs_mk_dir(const char *fmt, ...) -{ - return 0; -} static inline void devfs_remove(const char *fmt, ...) { } diff --git a/mm/shmem.c b/mm/shmem.c index 38bc3334f26..a4161abb9e7 100644 --- a/mm/shmem.c +++ b/mm/shmem.c @@ -2252,9 +2252,7 @@ static int __init init_tmpfs(void) printk(KERN_ERR "Could not register tmpfs\n"); goto out2; } -#ifdef CONFIG_TMPFS - devfs_mk_dir("shm"); -#endif + shm_mnt = vfs_kern_mount(&tmpfs_fs_type, MS_NOUSER, tmpfs_fs_type.name, NULL); if (IS_ERR(shm_mnt)) { diff --git a/mm/tiny-shmem.c b/mm/tiny-shmem.c index f9d6a9cc91c..11ab99b8c61 100644 --- a/mm/tiny-shmem.c +++ b/mm/tiny-shmem.c @@ -33,9 +33,6 @@ static int __init init_tmpfs(void) { BUG_ON(register_filesystem(&tmpfs_fs_type) != 0); -#ifdef CONFIG_TMPFS - devfs_mk_dir("shm"); -#endif shm_mnt = kern_mount(&tmpfs_fs_type); BUG_ON(IS_ERR(shm_mnt)); -- cgit v1.2.3-70-g09d2 From 79021a625c36162d24c852bbbdb04f0c1cb32db3 Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Mon, 20 Jun 2005 21:15:16 -0700 Subject: [PATCH] devfs: Remove devfs_mk_symlink() function from the kernel tree Removes the devfs_mk_symlink() function and all callers of it. Signed-off-by: Greg Kroah-Hartman --- arch/um/drivers/ubd_kern.c | 7 ------- include/linux/devfs_fs_kernel.h | 4 ---- 2 files changed, 11 deletions(-) (limited to 'include') diff --git a/arch/um/drivers/ubd_kern.c b/arch/um/drivers/ubd_kern.c index 8f6bb937519..3883c5e6a7d 100644 --- a/arch/um/drivers/ubd_kern.c +++ b/arch/um/drivers/ubd_kern.c @@ -628,7 +628,6 @@ static int ubd_new_disk(int major, u64 size, int unit, { struct gendisk *disk; - char from[sizeof("ubd/nnnnn\0")], to[sizeof("discnnnnn/disc\0")]; int err; disk = alloc_disk(1 << UBD_SHIFT); @@ -642,12 +641,6 @@ static int ubd_new_disk(int major, u64 size, int unit, if(major == MAJOR_NR){ sprintf(disk->disk_name, "ubd%c", 'a' + unit); sprintf(disk->devfs_name, "ubd/disc%d", unit); - sprintf(from, "ubd/%d", unit); - sprintf(to, "disc%d/disc", unit); - err = devfs_mk_symlink(from, to); - if(err) - printk("ubd_new_disk failed to make link from %s to " - "%s, error = %d\n", from, to, err); } else { sprintf(disk->disk_name, "ubd_fake%d", unit); diff --git a/include/linux/devfs_fs_kernel.h b/include/linux/devfs_fs_kernel.h index ef8766e172e..da336a3c6c7 100644 --- a/include/linux/devfs_fs_kernel.h +++ b/include/linux/devfs_fs_kernel.h @@ -14,10 +14,6 @@ static inline int devfs_mk_cdev(dev_t dev, umode_t mode, const char *fmt, ...) { return 0; } -static inline int devfs_mk_symlink(const char *name, const char *link) -{ - return 0; -} static inline void devfs_remove(const char *fmt, ...) { } -- cgit v1.2.3-70-g09d2 From 1a715c5cf917326a285533d1116d725f5f2593c2 Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Mon, 20 Jun 2005 21:15:16 -0700 Subject: [PATCH] devfs: Remove devfs_mk_bdev() function from the kernel tree Removes the devfs_mk_bdev() function and all callers of it. Signed-off-by: Greg Kroah-Hartman --- drivers/block/floppy.c | 19 ------------------- drivers/md/dm-ioctl.c | 14 -------------- drivers/md/md.c | 13 ------------- fs/partitions/check.c | 4 ---- include/linux/devfs_fs_kernel.h | 4 ---- 5 files changed, 54 deletions(-) (limited to 'include') diff --git a/drivers/block/floppy.c b/drivers/block/floppy.c index 03a4dff05f2..ec14e6e6bcc 100644 --- a/drivers/block/floppy.c +++ b/drivers/block/floppy.c @@ -224,7 +224,6 @@ static struct completion device_release; static unsigned short virtual_dma_port = 0x3f0; irqreturn_t floppy_interrupt(int irq, void *dev_id, struct pt_regs *regs); static int set_dor(int fdc, char mask, char data); -static void register_devfs_entries(int drive) __init; #define K_64 0x10000 /* 64KB */ @@ -3676,7 +3675,6 @@ static void __init config_types(void) first = 0; } printk("%s fd%d is %s", prepend, drive, name); - register_devfs_entries(drive); } *UDP = *params; } @@ -3969,23 +3967,6 @@ static int t360[] = { 1, 0 }, static int *table_sup[] = { NULL, t360, t1200, t3in + 5 + 8, t3in + 5, t3in, t3in }; -static void __init register_devfs_entries(int drive) -{ - int base_minor = (drive < 4) ? drive : (124 + drive); - - if (UDP->cmos < ARRAY_SIZE(default_drive_params)) { - int i = 0; - do { - int minor = base_minor + (table_sup[UDP->cmos][i] << 2); - - devfs_mk_bdev(MKDEV(FLOPPY_MAJOR, minor), - S_IFBLK | S_IRUSR | S_IWUSR | S_IRGRP | - S_IWGRP, "floppy/%d%s", drive, - table[table_sup[UDP->cmos][i]]); - } while (table_sup[UDP->cmos][i++]); - } -} - /* * Floppy Driver initialization * ============================= diff --git a/drivers/md/dm-ioctl.c b/drivers/md/dm-ioctl.c index 181971a978b..6e1b51d748a 100644 --- a/drivers/md/dm-ioctl.c +++ b/drivers/md/dm-ioctl.c @@ -173,16 +173,6 @@ static void free_cell(struct hash_cell *hc) /* * devfs stuff. */ -static int register_with_devfs(struct hash_cell *hc) -{ - struct gendisk *disk = dm_disk(hc->md); - - devfs_mk_bdev(MKDEV(disk->major, disk->first_minor), - S_IFBLK | S_IRUSR | S_IWUSR | S_IRGRP, - DM_DIR "/%s", hc->name); - return 0; -} - static int unregister_with_devfs(struct hash_cell *hc) { devfs_remove(DM_DIR"/%s", hc->name); @@ -225,7 +215,6 @@ static int dm_hash_insert(const char *name, const char *uuid, struct mapped_devi } list_add(&cell->uuid_list, _uuid_buckets + hash_str(uuid)); } - register_with_devfs(cell); dm_get(md); dm_set_mdptr(md, cell); up_write(&_hash_lock); @@ -348,9 +337,6 @@ static int dm_hash_rename(const char *old, const char *new) hc->name = new_name; list_add(&hc->name_list, _name_buckets + hash_str(new_name)); - /* rename the device node in devfs */ - register_with_devfs(hc); - /* * Wake up any dm event waiters. */ diff --git a/drivers/md/md.c b/drivers/md/md.c index 33b1a8943e4..ee7320b4a30 100644 --- a/drivers/md/md.c +++ b/drivers/md/md.c @@ -5538,8 +5538,6 @@ static void md_geninit(void) static int __init md_init(void) { - int minor; - printk(KERN_INFO "md: md driver %d.%d.%d MAX_MD_DEVS=%d," " MD_SB_DISKS=%d\n", MD_MAJOR_VERSION, MD_MINOR_VERSION, @@ -5558,17 +5556,6 @@ static int __init md_init(void) blk_register_region(MKDEV(mdp_major, 0), MAX_MD_DEVS<nr_sects = len; p->partno = part; - devfs_mk_bdev(MKDEV(disk->major, disk->first_minor + part), - S_IFBLK|S_IRUSR|S_IWUSR, - "%s/part%d", disk->devfs_name, part); - if (isdigit(disk->kobj.name[strlen(disk->kobj.name)-1])) snprintf(p->kobj.name,KOBJ_NAME_LEN,"%sp%d",disk->kobj.name,part); else diff --git a/include/linux/devfs_fs_kernel.h b/include/linux/devfs_fs_kernel.h index da336a3c6c7..6d9f003dbce 100644 --- a/include/linux/devfs_fs_kernel.h +++ b/include/linux/devfs_fs_kernel.h @@ -6,10 +6,6 @@ #include #include -static inline int devfs_mk_bdev(dev_t dev, umode_t mode, const char *fmt, ...) -{ - return 0; -} static inline int devfs_mk_cdev(dev_t dev, umode_t mode, const char *fmt, ...) { return 0; -- cgit v1.2.3-70-g09d2 From 7c69ef79741910883d5543caafa06aca3ebadbd1 Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Mon, 20 Jun 2005 21:15:16 -0700 Subject: [PATCH] devfs: Remove devfs_mk_cdev() function from the kernel tree Removes the devfs_mk_cdev() function and all callers of it. Signed-off-by: Greg Kroah-Hartman --- arch/sparc64/solaris/socksys.c | 2 -- drivers/block/acsi_slm.c | 4 ---- drivers/block/paride/pg.c | 11 +---------- drivers/block/paride/pt.c | 16 ---------------- drivers/char/dsp56k.c | 8 -------- drivers/char/dtlk.c | 3 --- drivers/char/ftape/zftape/zftape-init.c | 18 ------------------ drivers/char/ip2/ip2main.c | 17 ----------------- drivers/char/ipmi/ipmi_devintf.c | 3 --- drivers/char/istallion.c | 6 +----- drivers/char/lp.c | 2 -- drivers/char/mem.c | 5 +---- drivers/char/misc.c | 9 +-------- drivers/char/ppdev.c | 4 ---- drivers/char/raw.c | 8 -------- drivers/char/stallion.c | 6 +----- drivers/char/tipar.c | 10 ---------- drivers/char/tty_io.c | 7 ------- drivers/char/vc_screen.c | 8 -------- drivers/char/viotape.c | 4 ---- drivers/isdn/capi/capi.c | 2 -- drivers/isdn/hardware/eicon/divamnt.c | 1 - drivers/isdn/hardware/eicon/divasi.c | 1 - drivers/isdn/hardware/eicon/divasmain.c | 1 - drivers/macintosh/adb.c | 2 -- drivers/media/dvb/dvb-core/dvbdev.c | 4 ---- drivers/media/video/videodev.c | 7 ++----- drivers/net/ppp_generic.c | 7 ------- drivers/net/wan/cosa.c | 7 ------- drivers/sbus/char/bpp.c | 4 ---- drivers/sbus/char/vfc_dev.c | 4 ---- drivers/telephony/phonedev.c | 2 -- drivers/video/fbmem.c | 2 -- fs/coda/psdev.c | 11 +---------- include/linux/devfs_fs_kernel.h | 4 ---- 35 files changed, 8 insertions(+), 202 deletions(-) (limited to 'include') diff --git a/arch/sparc64/solaris/socksys.c b/arch/sparc64/solaris/socksys.c index fc6669e8dde..30ba11682ca 100644 --- a/arch/sparc64/solaris/socksys.c +++ b/arch/sparc64/solaris/socksys.c @@ -190,8 +190,6 @@ init_socksys(void) return ret; } - devfs_mk_cdev(MKDEV(30, 0), S_IFCHR|S_IRUSR|S_IWUSR, "socksys"); - file = fcheck(ret); /* N.B. Is this valid? Suppose the f_ops are in a module ... */ socksys_file_ops = *file->f_op; diff --git a/drivers/block/acsi_slm.c b/drivers/block/acsi_slm.c index 1ecbcc3da14..b087b3a6f06 100644 --- a/drivers/block/acsi_slm.c +++ b/drivers/block/acsi_slm.c @@ -1005,10 +1005,6 @@ int slm_init( void ) BufferP = SLMBuffer; SLMState = IDLE; - for (i = 0; i < MAX_SLM; i++) { - devfs_mk_cdev(MKDEV(ACSI_MAJOR, i), - S_IFCHR|S_IRUSR|S_IWUSR, "slm/%d", i); - } return 0; } diff --git a/drivers/block/paride/pg.c b/drivers/block/paride/pg.c index 9f72bdd2e20..3d464f767ea 100644 --- a/drivers/block/paride/pg.c +++ b/drivers/block/paride/pg.c @@ -676,22 +676,13 @@ static int __init pg_init(void) } for (unit = 0; unit < PG_UNITS; unit++) { struct pg *dev = &devices[unit]; - if (dev->present) { + if (dev->present) class_device_create(pg_class, NULL, MKDEV(major, unit), NULL, "pg%u", unit); - err = devfs_mk_cdev(MKDEV(major, unit), - S_IFCHR | S_IRUSR | S_IWUSR, "pg/%u", - unit); - if (err) - goto out_class; - } } err = 0; goto out; -out_class: - class_device_destroy(pg_class, MKDEV(major, unit)); - class_destroy(pg_class); out_chrdev: unregister_chrdev(major, "pg"); out: diff --git a/drivers/block/paride/pt.c b/drivers/block/paride/pt.c index 7adadd5b29f..c85bdcb384a 100644 --- a/drivers/block/paride/pt.c +++ b/drivers/block/paride/pt.c @@ -975,27 +975,11 @@ static int __init pt_init(void) if (pt[unit].present) { class_device_create(pt_class, NULL, MKDEV(major, unit), NULL, "pt%d", unit); - err = devfs_mk_cdev(MKDEV(major, unit), - S_IFCHR | S_IRUSR | S_IWUSR, - "pt/%d", unit); - if (err) { - class_device_destroy(pt_class, MKDEV(major, unit)); - goto out_class; - } class_device_create(pt_class, NULL, MKDEV(major, unit + 128), NULL, "pt%dn", unit); - err = devfs_mk_cdev(MKDEV(major, unit + 128), - S_IFCHR | S_IRUSR | S_IWUSR, - "pt/%dn", unit); - if (err) { - class_device_destroy(pt_class, MKDEV(major, unit + 128)); - goto out_class; - } } goto out; -out_class: - class_destroy(pt_class); out_chrdev: unregister_chrdev(major, "pt"); out: diff --git a/drivers/char/dsp56k.c b/drivers/char/dsp56k.c index e233cf280bc..45caf778578 100644 --- a/drivers/char/dsp56k.c +++ b/drivers/char/dsp56k.c @@ -518,17 +518,9 @@ static int __init dsp56k_init_driver(void) } class_device_create(dsp56k_class, NULL, MKDEV(DSP56K_MAJOR, 0), NULL, "dsp56k"); - err = devfs_mk_cdev(MKDEV(DSP56K_MAJOR, 0), - S_IFCHR | S_IRUSR | S_IWUSR, "dsp56k"); - if(err) - goto out_class; - printk(banner); goto out; -out_class: - class_device_destroy(dsp56k_class, MKDEV(DSP56K_MAJOR, 0)); - class_destroy(dsp56k_class); out_chrdev: unregister_chrdev(DSP56K_MAJOR, "dsp56k"); out: diff --git a/drivers/char/dtlk.c b/drivers/char/dtlk.c index 87dcaa237f0..e34b4f4d1c2 100644 --- a/drivers/char/dtlk.c +++ b/drivers/char/dtlk.c @@ -337,9 +337,6 @@ static int __init dtlk_init(void) if (dtlk_dev_probe() == 0) printk(", MAJOR %d\n", dtlk_major); - devfs_mk_cdev(MKDEV(dtlk_major, DTLK_MINOR), - S_IFCHR | S_IRUSR | S_IWUSR, "dtlk"); - init_timer(&dtlk_timer); dtlk_timer.function = dtlk_timer_tick; init_waitqueue_head(&dtlk_process_list); diff --git a/drivers/char/ftape/zftape/zftape-init.c b/drivers/char/ftape/zftape/zftape-init.c index 821357ce7e0..8c6090c3a42 100644 --- a/drivers/char/ftape/zftape/zftape-init.c +++ b/drivers/char/ftape/zftape/zftape-init.c @@ -332,29 +332,11 @@ KERN_INFO zft_class = class_create(THIS_MODULE, "zft"); for (i = 0; i < 4; i++) { class_device_create(zft_class, NULL, MKDEV(QIC117_TAPE_MAJOR, i), NULL, "qft%i", i); - devfs_mk_cdev(MKDEV(QIC117_TAPE_MAJOR, i), - S_IFCHR | S_IRUSR | S_IWUSR, - "qft%i", i); class_device_create(zft_class, NULL, MKDEV(QIC117_TAPE_MAJOR, i + 4), NULL, "nqft%i", i); - devfs_mk_cdev(MKDEV(QIC117_TAPE_MAJOR, i + 4), - S_IFCHR | S_IRUSR | S_IWUSR, - "nqft%i", i); class_device_create(zft_class, NULL, MKDEV(QIC117_TAPE_MAJOR, i + 16), NULL, "zqft%i", i); - devfs_mk_cdev(MKDEV(QIC117_TAPE_MAJOR, i + 16), - S_IFCHR | S_IRUSR | S_IWUSR, - "zqft%i", i); class_device_create(zft_class, NULL, MKDEV(QIC117_TAPE_MAJOR, i + 20), NULL, "nzqft%i", i); - devfs_mk_cdev(MKDEV(QIC117_TAPE_MAJOR, i + 20), - S_IFCHR | S_IRUSR | S_IWUSR, - "nzqft%i", i); class_device_create(zft_class, NULL, MKDEV(QIC117_TAPE_MAJOR, i + 32), NULL, "rawqft%i", i); - devfs_mk_cdev(MKDEV(QIC117_TAPE_MAJOR, i + 32), - S_IFCHR | S_IRUSR | S_IWUSR, - "rawqft%i", i); class_device_create(zft_class, NULL, MKDEV(QIC117_TAPE_MAJOR, i + 36), NULL, "nrawrawqft%i", i); - devfs_mk_cdev(MKDEV(QIC117_TAPE_MAJOR, i + 36), - S_IFCHR | S_IRUSR | S_IWUSR, - "nrawqft%i", i); } #ifdef CONFIG_ZFT_COMPRESSOR diff --git a/drivers/char/ip2/ip2main.c b/drivers/char/ip2/ip2main.c index 9ab33c3d359..afc25db4975 100644 --- a/drivers/char/ip2/ip2main.c +++ b/drivers/char/ip2/ip2main.c @@ -724,26 +724,9 @@ ip2_loadmain(int *iop, int *irqp, unsigned char *firmware, int firmsize) class_device_create(ip2_class, NULL, MKDEV(IP2_IPL_MAJOR, 4 * i), NULL, "ipl%d", i); - err = devfs_mk_cdev(MKDEV(IP2_IPL_MAJOR, 4 * i), - S_IRUSR | S_IWUSR | S_IRGRP | S_IFCHR, - "ip2/ipl%d", i); - if (err) { - class_device_destroy(ip2_class, - MKDEV(IP2_IPL_MAJOR, 4 * i)); - goto out_class; - } - class_device_create(ip2_class, NULL, MKDEV(IP2_IPL_MAJOR, 4 * i + 1), NULL, "stat%d", i); - err = devfs_mk_cdev(MKDEV(IP2_IPL_MAJOR, 4 * i + 1), - S_IRUSR | S_IWUSR | S_IRGRP | S_IFCHR, - "ip2/stat%d", i); - if (err) { - class_device_destroy(ip2_class, - MKDEV(IP2_IPL_MAJOR, 4 * i + 1)); - goto out_class; - } for ( box = 0; box < ABS_MAX_BOXES; ++box ) { diff --git a/drivers/char/ipmi/ipmi_devintf.c b/drivers/char/ipmi/ipmi_devintf.c index 55c57451ee1..ec22aba12be 100644 --- a/drivers/char/ipmi/ipmi_devintf.c +++ b/drivers/char/ipmi/ipmi_devintf.c @@ -804,9 +804,6 @@ static void ipmi_new_smi(int if_num, struct device *device) dev_t dev = MKDEV(ipmi_major, if_num); struct ipmi_reg_list *entry; - devfs_mk_cdev(dev, S_IFCHR | S_IRUSR | S_IWUSR, - "ipmidev/%d", if_num); - entry = kmalloc(sizeof(*entry), GFP_KERNEL); if (!entry) { printk(KERN_ERR "ipmi_devintf: Unable to create the" diff --git a/drivers/char/istallion.c b/drivers/char/istallion.c index d5054ea4747..1395d4a5204 100644 --- a/drivers/char/istallion.c +++ b/drivers/char/istallion.c @@ -5214,14 +5214,10 @@ int __init stli_init(void) "device\n"); istallion_class = class_create(THIS_MODULE, "staliomem"); - for (i = 0; i < 4; i++) { - devfs_mk_cdev(MKDEV(STL_SIOMEMMAJOR, i), - S_IFCHR | S_IRUSR | S_IWUSR, - "staliomem/%d", i); + for (i = 0; i < 4; i++) class_device_create(istallion_class, NULL, MKDEV(STL_SIOMEMMAJOR, i), NULL, "staliomem%d", i); - } /* * Set up the tty driver structure and register us as a driver. diff --git a/drivers/char/lp.c b/drivers/char/lp.c index f07a9e361f3..1a08e8ea238 100644 --- a/drivers/char/lp.c +++ b/drivers/char/lp.c @@ -807,8 +807,6 @@ static int lp_register(int nr, struct parport *port) class_device_create(lp_class, NULL, MKDEV(LP_MAJOR, nr), NULL, "lp%d", nr); - devfs_mk_cdev(MKDEV(LP_MAJOR, nr), S_IFCHR | S_IRUGO | S_IWUGO, - "printers/%d", nr); printk(KERN_INFO "lp%d: using %s (%s).\n", nr, port->name, (port->irq == PARPORT_IRQ_NONE)?"polling":"interrupt-driven"); diff --git a/drivers/char/mem.c b/drivers/char/mem.c index 1fa9fa157c1..e7aa71575aa 100644 --- a/drivers/char/mem.c +++ b/drivers/char/mem.c @@ -941,13 +941,10 @@ static int __init chr_dev_init(void) printk("unable to get major %d for memory devs\n", MEM_MAJOR); mem_class = class_create(THIS_MODULE, "mem"); - for (i = 0; i < ARRAY_SIZE(devlist); i++) { + for (i = 0; i < ARRAY_SIZE(devlist); i++) class_device_create(mem_class, NULL, MKDEV(MEM_MAJOR, devlist[i].minor), NULL, devlist[i].name); - devfs_mk_cdev(MKDEV(MEM_MAJOR, devlist[i].minor), - S_IFCHR | devlist[i].mode, devlist[i].name); - } return 0; } diff --git a/drivers/char/misc.c b/drivers/char/misc.c index 96eb2a709e2..71e51377104 100644 --- a/drivers/char/misc.c +++ b/drivers/char/misc.c @@ -204,7 +204,7 @@ int misc_register(struct miscdevice * misc) { struct miscdevice *c; dev_t dev; - int err; + int err = 0; down(&misc_sem); list_for_each_entry(c, &misc_list, list) { @@ -241,13 +241,6 @@ int misc_register(struct miscdevice * misc) goto out; } - err = devfs_mk_cdev(dev, S_IFCHR|S_IRUSR|S_IWUSR|S_IRGRP, - misc->devfs_name); - if (err) { - class_device_destroy(misc_class, dev); - goto out; - } - /* * Add it to the front, so that later devices can "override" * earlier defaults diff --git a/drivers/char/ppdev.c b/drivers/char/ppdev.c index d437a4c9e63..ba350799405 100644 --- a/drivers/char/ppdev.c +++ b/drivers/char/ppdev.c @@ -782,10 +782,6 @@ static int __init ppdev_init (void) err = PTR_ERR(ppdev_class); goto out_chrdev; } - for (i = 0; i < PARPORT_MAX; i++) { - devfs_mk_cdev(MKDEV(PP_MAJOR, i), - S_IFCHR | S_IRUGO | S_IWUGO, "parports/%d", i); - } if (parport_register_driver(&pp_driver)) { printk (KERN_WARNING CHRDEV ": unable to register with parport\n"); goto out_class; diff --git a/drivers/char/raw.c b/drivers/char/raw.c index 15a7b408652..72c8ee4c006 100644 --- a/drivers/char/raw.c +++ b/drivers/char/raw.c @@ -288,7 +288,6 @@ static struct cdev raw_cdev = { static int __init raw_init(void) { - int i; dev_t dev = MKDEV(RAW_MAJOR, 0); if (register_chrdev_region(dev, MAX_RAW_MINORS, "raw")) @@ -310,13 +309,6 @@ static int __init raw_init(void) } class_device_create(raw_class, NULL, MKDEV(RAW_MAJOR, 0), NULL, "rawctl"); - devfs_mk_cdev(MKDEV(RAW_MAJOR, 0), - S_IFCHR | S_IRUGO | S_IWUGO, - "raw/rawctl"); - for (i = 1; i < MAX_RAW_MINORS; i++) - devfs_mk_cdev(MKDEV(RAW_MAJOR, i), - S_IFCHR | S_IRUGO | S_IWUGO, - "raw/raw%d", i); return 0; error: diff --git a/drivers/char/stallion.c b/drivers/char/stallion.c index c090a63fbde..a1a68f90d80 100644 --- a/drivers/char/stallion.c +++ b/drivers/char/stallion.c @@ -3071,14 +3071,10 @@ static int __init stl_init(void) printk("STALLION: failed to register serial board device\n"); stallion_class = class_create(THIS_MODULE, "staliomem"); - for (i = 0; i < 4; i++) { - devfs_mk_cdev(MKDEV(STL_SIOMEMMAJOR, i), - S_IFCHR|S_IRUSR|S_IWUSR, - "staliomem/%d", i); + for (i = 0; i < 4; i++) class_device_create(stallion_class, NULL, MKDEV(STL_SIOMEMMAJOR, i), NULL, "staliomem%d", i); - } stl_serial->owner = THIS_MODULE; stl_serial->driver_name = stl_drvname; diff --git a/drivers/char/tipar.c b/drivers/char/tipar.c index 70cfe07ab99..0dc83a5bd4c 100644 --- a/drivers/char/tipar.c +++ b/drivers/char/tipar.c @@ -443,12 +443,6 @@ tipar_register(int nr, struct parport *port) class_device_create(tipar_class, NULL, MKDEV(TIPAR_MAJOR, TIPAR_MINOR + nr), NULL, "par%d", nr); - /* Use devfs, tree: /dev/ticables/par/[0..2] */ - err = devfs_mk_cdev(MKDEV(TIPAR_MAJOR, TIPAR_MINOR + nr), - S_IFCHR | S_IRUGO | S_IWUGO, - "ticables/par/%d", nr); - if (err) - goto out_class; /* Display informations */ pr_info("tipar%d: using %s (%s)\n", nr, port->name, (port->irq == @@ -460,11 +454,7 @@ tipar_register(int nr, struct parport *port) pr_info("tipar%d: link cable not found\n", nr); err = 0; - goto out; -out_class: - class_device_destroy(tipar_class, MKDEV(TIPAR_MAJOR, TIPAR_MINOR + nr)); - class_destroy(tipar_class); out: return err; } diff --git a/drivers/char/tty_io.c b/drivers/char/tty_io.c index 8b2a5996986..0b9a33c0f1b 100644 --- a/drivers/char/tty_io.c +++ b/drivers/char/tty_io.c @@ -2979,9 +2979,6 @@ struct class_device *tty_register_device(struct tty_driver *driver, return ERR_PTR(-EINVAL); } - devfs_mk_cdev(dev, S_IFCHR | S_IRUSR | S_IWUSR, - "%s%d", driver->devfs_name, index + driver->name_base); - if (driver->type == TTY_DRIVER_TYPE_PTY) pty_line_name(driver, index, name); else @@ -3241,14 +3238,12 @@ static int __init tty_init(void) if (cdev_add(&tty_cdev, MKDEV(TTYAUX_MAJOR, 0), 1) || register_chrdev_region(MKDEV(TTYAUX_MAJOR, 0), 1, "/dev/tty") < 0) panic("Couldn't register /dev/tty driver\n"); - devfs_mk_cdev(MKDEV(TTYAUX_MAJOR, 0), S_IFCHR|S_IRUGO|S_IWUGO, "tty"); class_device_create(tty_class, NULL, MKDEV(TTYAUX_MAJOR, 0), NULL, "tty"); cdev_init(&console_cdev, &console_fops); if (cdev_add(&console_cdev, MKDEV(TTYAUX_MAJOR, 1), 1) || register_chrdev_region(MKDEV(TTYAUX_MAJOR, 1), 1, "/dev/console") < 0) panic("Couldn't register /dev/console driver\n"); - devfs_mk_cdev(MKDEV(TTYAUX_MAJOR, 1), S_IFCHR|S_IRUSR|S_IWUSR, "console"); class_device_create(tty_class, NULL, MKDEV(TTYAUX_MAJOR, 1), NULL, "console"); #ifdef CONFIG_UNIX98_PTYS @@ -3256,7 +3251,6 @@ static int __init tty_init(void) if (cdev_add(&ptmx_cdev, MKDEV(TTYAUX_MAJOR, 2), 1) || register_chrdev_region(MKDEV(TTYAUX_MAJOR, 2), 1, "/dev/ptmx") < 0) panic("Couldn't register /dev/ptmx driver\n"); - devfs_mk_cdev(MKDEV(TTYAUX_MAJOR, 2), S_IFCHR|S_IRUGO|S_IWUGO, "ptmx"); class_device_create(tty_class, NULL, MKDEV(TTYAUX_MAJOR, 2), NULL, "ptmx"); #endif @@ -3265,7 +3259,6 @@ static int __init tty_init(void) if (cdev_add(&vc0_cdev, MKDEV(TTY_MAJOR, 0), 1) || register_chrdev_region(MKDEV(TTY_MAJOR, 0), 1, "/dev/vc/0") < 0) panic("Couldn't register /dev/tty0 driver\n"); - devfs_mk_cdev(MKDEV(TTY_MAJOR, 0), S_IFCHR|S_IRUSR|S_IWUSR, "vc/0"); class_device_create(tty_class, NULL, MKDEV(TTY_MAJOR, 0), NULL, "tty0"); vty_init(); diff --git a/drivers/char/vc_screen.c b/drivers/char/vc_screen.c index 3c1dafaa344..1633d820652 100644 --- a/drivers/char/vc_screen.c +++ b/drivers/char/vc_screen.c @@ -478,12 +478,6 @@ static struct class *vc_class; void vcs_make_devfs(struct tty_struct *tty) { - devfs_mk_cdev(MKDEV(VCS_MAJOR, tty->index + 1), - S_IFCHR|S_IRUSR|S_IWUSR, - "vcc/%u", tty->index + 1); - devfs_mk_cdev(MKDEV(VCS_MAJOR, tty->index + 129), - S_IFCHR|S_IRUSR|S_IWUSR, - "vcc/a%u", tty->index + 1); class_device_create(vc_class, NULL, MKDEV(VCS_MAJOR, tty->index + 1), NULL, "vcs%u", tty->index + 1); class_device_create(vc_class, NULL, MKDEV(VCS_MAJOR, tty->index + 129), @@ -503,8 +497,6 @@ int __init vcs_init(void) panic("unable to get major %d for vcs device", VCS_MAJOR); vc_class = class_create(THIS_MODULE, "vc"); - devfs_mk_cdev(MKDEV(VCS_MAJOR, 0), S_IFCHR|S_IRUSR|S_IWUSR, "vcc/0"); - devfs_mk_cdev(MKDEV(VCS_MAJOR, 128), S_IFCHR|S_IRUSR|S_IWUSR, "vcc/a0"); class_device_create(vc_class, NULL, MKDEV(VCS_MAJOR, 0), NULL, "vcs"); class_device_create(vc_class, NULL, MKDEV(VCS_MAJOR, 128), NULL, "vcsa"); return 0; diff --git a/drivers/char/viotape.c b/drivers/char/viotape.c index ebc337acca1..4c1a128fd77 100644 --- a/drivers/char/viotape.c +++ b/drivers/char/viotape.c @@ -958,10 +958,6 @@ static int viotape_probe(struct vio_dev *vdev, const struct vio_device_id *id) "iseries!vt%d", i); class_device_create(tape_class, NULL, MKDEV(VIOTAPE_MAJOR, i | 0x80), NULL, "iseries!nvt%d", i); - devfs_mk_cdev(MKDEV(VIOTAPE_MAJOR, i), S_IFCHR | S_IRUSR | S_IWUSR, - "iseries/vt%d", i); - devfs_mk_cdev(MKDEV(VIOTAPE_MAJOR, i | 0x80), - S_IFCHR | S_IRUSR | S_IWUSR, "iseries/nvt%d", i); sprintf(tapename, "iseries/vt%d", i); printk(VIOTAPE_KERN_INFO "tape %s is iSeries " "resource %10.10s type %4.4s, model %3.3s\n", diff --git a/drivers/isdn/capi/capi.c b/drivers/isdn/capi/capi.c index 2e541fa0202..1c8fe77270a 100644 --- a/drivers/isdn/capi/capi.c +++ b/drivers/isdn/capi/capi.c @@ -1516,8 +1516,6 @@ static int __init capi_init(void) } class_device_create(capi_class, NULL, MKDEV(capi_major, 0), NULL, "capi"); - devfs_mk_cdev(MKDEV(capi_major, 0), S_IFCHR | S_IRUSR | S_IWUSR, - "isdn/capi20"); #ifdef CONFIG_ISDN_CAPI_MIDDLEWARE if (capinc_tty_init() < 0) { diff --git a/drivers/isdn/hardware/eicon/divamnt.c b/drivers/isdn/hardware/eicon/divamnt.c index 6146f7633be..12465b3c2b4 100644 --- a/drivers/isdn/hardware/eicon/divamnt.c +++ b/drivers/isdn/hardware/eicon/divamnt.c @@ -190,7 +190,6 @@ static int DIVA_INIT_FUNCTION divas_maint_register_chrdev(void) DRIVERLNAME); return (0); } - devfs_mk_cdev(MKDEV(major, 0), S_IFCHR|S_IRUSR|S_IWUSR, DEVNAME); return (1); } diff --git a/drivers/isdn/hardware/eicon/divasi.c b/drivers/isdn/hardware/eicon/divasi.c index df715b47e2b..2d80d94b5d0 100644 --- a/drivers/isdn/hardware/eicon/divasi.c +++ b/drivers/isdn/hardware/eicon/divasi.c @@ -157,7 +157,6 @@ static int DIVA_INIT_FUNCTION divas_idi_register_chrdev(void) DRIVERLNAME); return (0); } - devfs_mk_cdev(MKDEV(major, 0), S_IFCHR|S_IRUSR|S_IWUSR, DEVNAME); return (1); } diff --git a/drivers/isdn/hardware/eicon/divasmain.c b/drivers/isdn/hardware/eicon/divasmain.c index c9b26e86d18..71b3e06cbe3 100644 --- a/drivers/isdn/hardware/eicon/divasmain.c +++ b/drivers/isdn/hardware/eicon/divasmain.c @@ -690,7 +690,6 @@ static int DIVA_INIT_FUNCTION divas_register_chrdev(void) DRIVERLNAME); return (0); } - devfs_mk_cdev(MKDEV(major, 0), S_IFCHR|S_IRUSR|S_IWUSR, DEVNAME); return (1); } diff --git a/drivers/macintosh/adb.c b/drivers/macintosh/adb.c index 259fd8973ce..a60e3abaa8e 100644 --- a/drivers/macintosh/adb.c +++ b/drivers/macintosh/adb.c @@ -904,8 +904,6 @@ adbdev_init(void) return; } - devfs_mk_cdev(MKDEV(ADB_MAJOR, 0), S_IFCHR | S_IRUSR | S_IWUSR, "adb"); - adb_dev_class = class_create(THIS_MODULE, "adb"); if (IS_ERR(adb_dev_class)) return; diff --git a/drivers/media/dvb/dvb-core/dvbdev.c b/drivers/media/dvb/dvb-core/dvbdev.c index cf78a72320e..8e926d98bd9 100644 --- a/drivers/media/dvb/dvb-core/dvbdev.c +++ b/drivers/media/dvb/dvb-core/dvbdev.c @@ -231,10 +231,6 @@ int dvb_register_device(struct dvb_adapter *adap, struct dvb_device **pdvbdev, mutex_unlock(&dvbdev_register_lock); - devfs_mk_cdev(MKDEV(DVB_MAJOR, nums2minor(adap->num, type, id)), - S_IFCHR | S_IRUSR | S_IWUSR, - "dvb/adapter%d/%s%d", adap->num, dnames[type], id); - class_device_create(dvb_class, NULL, MKDEV(DVB_MAJOR, nums2minor(adap->num, type, id)), adap->device, "dvb%d.%s%d", adap->num, dnames[type], id); diff --git a/drivers/media/video/videodev.c b/drivers/media/video/videodev.c index 2dfa7f23d0c..cc67ee95283 100644 --- a/drivers/media/video/videodev.c +++ b/drivers/media/video/videodev.c @@ -1563,10 +1563,6 @@ int video_register_device(struct video_device *vfd, int type, int nr) video_device[i]=vfd; vfd->minor=i; mutex_unlock(&videodev_lock); - - sprintf(vfd->devfs_name, "v4l/%s%d", name_base, i - base); - devfs_mk_cdev(MKDEV(VIDEO_MAJOR, vfd->minor), - S_IFCHR | S_IRUSR | S_IWUSR, vfd->devfs_name); mutex_init(&vfd->lock); /* sysfs class */ @@ -1575,7 +1571,8 @@ int video_register_device(struct video_device *vfd, int type, int nr) vfd->class_dev.dev = vfd->dev; vfd->class_dev.class = &video_class; vfd->class_dev.devt = MKDEV(VIDEO_MAJOR, vfd->minor); - strlcpy(vfd->class_dev.class_id, vfd->devfs_name + 4, BUS_ID_SIZE); + sprintf(vfd->devfs_name, "%s%d", name_base, i - base); + strlcpy(vfd->class_dev.class_id, vfd->devfs_name, BUS_ID_SIZE); class_device_register(&vfd->class_dev); class_device_create_file(&vfd->class_dev, &class_device_attr_name); diff --git a/drivers/net/ppp_generic.c b/drivers/net/ppp_generic.c index d643a097faa..52bc5154513 100644 --- a/drivers/net/ppp_generic.c +++ b/drivers/net/ppp_generic.c @@ -863,10 +863,6 @@ static int __init ppp_init(void) goto out_chrdev; } class_device_create(ppp_class, NULL, MKDEV(PPP_MAJOR, 0), NULL, "ppp"); - err = devfs_mk_cdev(MKDEV(PPP_MAJOR, 0), - S_IFCHR|S_IRUSR|S_IWUSR, "ppp"); - if (err) - goto out_class; } out: @@ -874,9 +870,6 @@ out: printk(KERN_ERR "failed to register PPP device (%d)\n", err); return err; -out_class: - class_device_destroy(ppp_class, MKDEV(PPP_MAJOR,0)); - class_destroy(ppp_class); out_chrdev: unregister_chrdev(PPP_MAJOR, "ppp"); goto out; diff --git a/drivers/net/wan/cosa.c b/drivers/net/wan/cosa.c index 62b7087e44d..c7b530628c8 100644 --- a/drivers/net/wan/cosa.c +++ b/drivers/net/wan/cosa.c @@ -401,13 +401,6 @@ static int __init cosa_init(void) for (i=0; iminor = i; - devfs_mk_cdev(MKDEV(PHONE_MAJOR,i), - S_IFCHR|S_IRUSR|S_IWUSR, "phone/%d", i); mutex_unlock(&phone_lock); return 0; } diff --git a/drivers/video/fbmem.c b/drivers/video/fbmem.c index 93546a848f4..be2479ab8bc 100644 --- a/drivers/video/fbmem.c +++ b/drivers/video/fbmem.c @@ -1331,8 +1331,6 @@ register_framebuffer(struct fb_info *fb_info) fb_add_videomode(&mode, &fb_info->modelist); registered_fb[i] = fb_info; - devfs_mk_cdev(MKDEV(FB_MAJOR, i), - S_IFCHR | S_IRUGO | S_IWUGO, "fb/%d", i); event.info = fb_info; blocking_notifier_call_chain(&fb_notifier_list, FB_EVENT_FB_REGISTERED, &event); diff --git a/fs/coda/psdev.c b/fs/coda/psdev.c index a193823f905..f478222cb5c 100644 --- a/fs/coda/psdev.c +++ b/fs/coda/psdev.c @@ -365,21 +365,12 @@ static int init_coda_psdev(void) err = PTR_ERR(coda_psdev_class); goto out_chrdev; } - for (i = 0; i < MAX_CODADEVS; i++) { + for (i = 0; i < MAX_CODADEVS; i++) class_device_create(coda_psdev_class, NULL, MKDEV(CODA_PSDEV_MAJOR,i), NULL, "cfs%d", i); - err = devfs_mk_cdev(MKDEV(CODA_PSDEV_MAJOR, i), - S_IFCHR|S_IRUSR|S_IWUSR, "coda/%d", i); - if (err) - goto out_class; - } coda_sysctl_init(); goto out; -out_class: - for (i = 0; i < MAX_CODADEVS; i++) - class_device_destroy(coda_psdev_class, MKDEV(CODA_PSDEV_MAJOR, i)); - class_destroy(coda_psdev_class); out_chrdev: unregister_chrdev(CODA_PSDEV_MAJOR, "coda"); out: diff --git a/include/linux/devfs_fs_kernel.h b/include/linux/devfs_fs_kernel.h index 6d9f003dbce..c7d05dbc2af 100644 --- a/include/linux/devfs_fs_kernel.h +++ b/include/linux/devfs_fs_kernel.h @@ -6,10 +6,6 @@ #include #include -static inline int devfs_mk_cdev(dev_t dev, umode_t mode, const char *fmt, ...) -{ - return 0; -} static inline void devfs_remove(const char *fmt, ...) { } -- cgit v1.2.3-70-g09d2 From 8ab5e4c15b53e147c08031a959d9f776823dbe73 Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Mon, 20 Jun 2005 21:15:16 -0700 Subject: [PATCH] devfs: Remove devfs_remove() function from the kernel tree Removes the devfs_remove() function and all callers of it. Signed-off-by: Greg Kroah-Hartman --- arch/sparc64/solaris/socksys.c | 1 - drivers/block/acsi_slm.c | 4 ---- drivers/block/cpqarray.c | 2 -- drivers/block/floppy.c | 33 +-------------------------------- drivers/block/loop.c | 2 -- drivers/block/nbd.c | 1 - drivers/block/paride/pg.c | 5 +---- drivers/block/paride/pt.c | 3 --- drivers/block/rd.c | 1 - drivers/block/sx8.c | 1 - drivers/block/ub.c | 2 -- drivers/cdrom/sbpcd.c | 2 -- drivers/char/dsp56k.c | 1 - drivers/char/dtlk.c | 1 - drivers/char/ftape/zftape/zftape-init.c | 6 ------ drivers/char/ip2/ip2main.c | 3 --- drivers/char/ipmi/ipmi_devintf.c | 2 -- drivers/char/istallion.c | 5 +---- drivers/char/lp.c | 3 --- drivers/char/misc.c | 1 - drivers/char/ppdev.c | 9 +-------- drivers/char/raw.c | 6 ------ drivers/char/stallion.c | 5 +---- drivers/char/tipar.c | 3 --- drivers/char/tty_io.c | 1 - drivers/char/vc_screen.c | 2 -- drivers/char/viotape.c | 2 -- drivers/isdn/capi/capi.c | 1 - drivers/isdn/hardware/eicon/divamnt.c | 1 - drivers/isdn/hardware/eicon/divasi.c | 1 - drivers/isdn/hardware/eicon/divasmain.c | 1 - drivers/md/dm-ioctl.c | 13 ------------- drivers/md/md.c | 8 +------- drivers/media/dvb/dvb-core/dvbdev.c | 6 ------ drivers/media/video/videodev.c | 1 - drivers/mmc/mmc_block.c | 1 - drivers/net/ppp_generic.c | 1 - drivers/net/wan/cosa.c | 5 +---- drivers/s390/block/dasd.c | 1 - drivers/s390/block/xpram.c | 2 -- drivers/sbus/char/bpp.c | 3 --- drivers/sbus/char/vfc_dev.c | 2 -- drivers/telephony/phonedev.c | 1 - drivers/video/fbmem.c | 1 - fs/coda/psdev.c | 10 ++-------- fs/partitions/check.c | 1 - include/linux/devfs_fs_kernel.h | 3 --- 47 files changed, 9 insertions(+), 161 deletions(-) (limited to 'include') diff --git a/arch/sparc64/solaris/socksys.c b/arch/sparc64/solaris/socksys.c index 30ba11682ca..09664ab124e 100644 --- a/arch/sparc64/solaris/socksys.c +++ b/arch/sparc64/solaris/socksys.c @@ -205,5 +205,4 @@ cleanup_socksys(void) { if (unregister_chrdev(30, "socksys")) printk ("Couldn't unregister socksys character device\n"); - devfs_remove ("socksys"); } diff --git a/drivers/block/acsi_slm.c b/drivers/block/acsi_slm.c index b087b3a6f06..7495fef2418 100644 --- a/drivers/block/acsi_slm.c +++ b/drivers/block/acsi_slm.c @@ -1027,10 +1027,6 @@ int init_module(void) void cleanup_module(void) { - int i; - for (i = 0; i < MAX_SLM; i++) - devfs_remove("slm/%d", i); - devfs_remove("slm"); if (unregister_chrdev( ACSI_MAJOR, "slm" ) != 0) printk( KERN_ERR "acsi_slm: cleanup_module failed\n"); atari_stram_free( SLMBuffer ); diff --git a/drivers/block/cpqarray.c b/drivers/block/cpqarray.c index 5eb6fb7b5cf..2701ed734b2 100644 --- a/drivers/block/cpqarray.c +++ b/drivers/block/cpqarray.c @@ -348,7 +348,6 @@ static void __devexit cpqarray_remove_one(int i) for(j = 0; j < NWD; j++) { if (ida_gendisk[i][j]->flags & GENHD_FL_UP) del_gendisk(ida_gendisk[i][j]); - devfs_remove("ida/c%dd%d",i,j); put_disk(ida_gendisk[i][j]); } blk_cleanup_queue(hba[i]->queue); @@ -1844,7 +1843,6 @@ static void __exit cpqarray_exit(void) } } - devfs_remove("ida"); remove_proc_entry("cpqarray", proc_root_driver); } diff --git a/drivers/block/floppy.c b/drivers/block/floppy.c index ec14e6e6bcc..8e101a9655b 100644 --- a/drivers/block/floppy.c +++ b/drivers/block/floppy.c @@ -3952,20 +3952,6 @@ static struct block_device_operations floppy_fops = { .media_changed = check_floppy_change, .revalidate_disk = floppy_revalidate, }; -static char *table[] = { - "", "d360", "h1200", "u360", "u720", "h360", "h720", - "u1440", "u2880", "CompaQ", "h1440", "u1680", "h410", - "u820", "h1476", "u1722", "h420", "u830", "h1494", "u1743", - "h880", "u1040", "u1120", "h1600", "u1760", "u1920", - "u3200", "u3520", "u3840", "u1840", "u800", "u1600", - NULL -}; -static int t360[] = { 1, 0 }, - t1200[] = { 2, 5, 6, 10, 12, 14, 16, 18, 20, 23, 0 }, - t3in[] = { 8, 9, 26, 27, 28, 7, 11, 15, 19, 24, 25, 29, 31, 3, 4, 13, - 17, 21, 22, 30, 0 }; -static int *table_sup[] = - { NULL, t360, t1200, t3in + 5 + 8, t3in + 5, t3in, t3in }; /* * Floppy Driver initialization @@ -4244,7 +4230,7 @@ static int __init floppy_init(void) err = register_blkdev(FLOPPY_MAJOR, "fd"); if (err) - goto out_devfs_remove; + goto out_put_disk; floppy_queue = blk_init_queue(do_fd_request, &floppy_lock); if (!floppy_queue) { @@ -4403,8 +4389,6 @@ out_unreg_region: blk_cleanup_queue(floppy_queue); out_unreg_blkdev: unregister_blkdev(FLOPPY_MAJOR, "fd"); -out_devfs_remove: - devfs_remove("floppy"); out_put_disk: while (dr--) { del_timer(&motor_off_timer[dr]); @@ -4565,19 +4549,6 @@ static void floppy_release_irq_and_dma(void) static char *floppy; -static void unregister_devfs_entries(int drive) -{ - int i; - - if (UDP->cmos < ARRAY_SIZE(default_drive_params)) { - i = 0; - do { - devfs_remove("floppy/%d%s", drive, - table[table_sup[UDP->cmos][i]]); - } while (table_sup[UDP->cmos][i++]); - } -} - static void __init parse_floppy_cfg_string(char *cfg) { char *ptr; @@ -4614,13 +4585,11 @@ void cleanup_module(void) if ((allowed_drive_mask & (1 << drive)) && fdc_state[FDC(drive)].version != FDC_NONE) { del_gendisk(disks[drive]); - unregister_devfs_entries(drive); device_remove_file(&floppy_device[drive].dev, &dev_attr_cmos); platform_device_unregister(&floppy_device[drive]); } put_disk(disks[drive]); } - devfs_remove("floppy"); del_timer_sync(&fd_timeout); del_timer_sync(&fd_timer); diff --git a/drivers/block/loop.c b/drivers/block/loop.c index f650d20f2a1..0319b096b55 100644 --- a/drivers/block/loop.c +++ b/drivers/block/loop.c @@ -1308,7 +1308,6 @@ static int __init loop_init(void) out_mem4: while (i--) blk_cleanup_queue(loop_dev[i].lo_queue); - devfs_remove("loop"); i = max_loop; out_mem3: while (i--) @@ -1331,7 +1330,6 @@ static void loop_exit(void) blk_cleanup_queue(loop_dev[i].lo_queue); put_disk(disks[i]); } - devfs_remove("loop"); if (unregister_blkdev(LOOP_MAJOR, "loop")) printk(KERN_WARNING "loop: cannot unregister blkdev\n"); diff --git a/drivers/block/nbd.c b/drivers/block/nbd.c index 618e258ca5d..c24b349ae57 100644 --- a/drivers/block/nbd.c +++ b/drivers/block/nbd.c @@ -685,7 +685,6 @@ static void __exit nbd_cleanup(void) put_disk(disk); } } - devfs_remove("nbd"); unregister_blkdev(NBD_MAJOR, "nbd"); printk(KERN_INFO "nbd: unregistered device at major %d\n", NBD_MAJOR); } diff --git a/drivers/block/paride/pg.c b/drivers/block/paride/pg.c index 3d464f767ea..e6f1614b69c 100644 --- a/drivers/block/paride/pg.c +++ b/drivers/block/paride/pg.c @@ -695,13 +695,10 @@ static void __exit pg_exit(void) for (unit = 0; unit < PG_UNITS; unit++) { struct pg *dev = &devices[unit]; - if (dev->present) { + if (dev->present) class_device_destroy(pg_class, MKDEV(major, unit)); - devfs_remove("pg/%u", unit); - } } class_destroy(pg_class); - devfs_remove("pg"); unregister_chrdev(major, name); for (unit = 0; unit < PG_UNITS; unit++) { diff --git a/drivers/block/paride/pt.c b/drivers/block/paride/pt.c index c85bdcb384a..ea12073c5fb 100644 --- a/drivers/block/paride/pt.c +++ b/drivers/block/paride/pt.c @@ -992,12 +992,9 @@ static void __exit pt_exit(void) for (unit = 0; unit < PT_UNITS; unit++) if (pt[unit].present) { class_device_destroy(pt_class, MKDEV(major, unit)); - devfs_remove("pt/%d", unit); class_device_destroy(pt_class, MKDEV(major, unit + 128)); - devfs_remove("pt/%dn", unit); } class_destroy(pt_class); - devfs_remove("pt"); unregister_chrdev(major, name); for (unit = 0; unit < PT_UNITS; unit++) if (pt[unit].present) diff --git a/drivers/block/rd.c b/drivers/block/rd.c index 67821a8c621..acaa2abf05a 100644 --- a/drivers/block/rd.c +++ b/drivers/block/rd.c @@ -412,7 +412,6 @@ static void __exit rd_cleanup(void) put_disk(rd_disks[i]); blk_cleanup_queue(rd_queue[i]); } - devfs_remove("rd"); unregister_blkdev(RAMDISK_MAJOR, "ramdisk"); } diff --git a/drivers/block/sx8.c b/drivers/block/sx8.c index 9fb3f51e574..a366ea8c6a4 100644 --- a/drivers/block/sx8.c +++ b/drivers/block/sx8.c @@ -1736,7 +1736,6 @@ static void carm_remove_one (struct pci_dev *pdev) free_irq(pdev->irq, host); carm_free_disks(host); - devfs_remove(DRV_NAME); unregister_blkdev(host->major, host->name); if (host->major == 160) clear_bit(0, &carm_major_alloc); diff --git a/drivers/block/ub.c b/drivers/block/ub.c index 088e5ba15c2..b68abc9d4c1 100644 --- a/drivers/block/ub.c +++ b/drivers/block/ub.c @@ -2453,7 +2453,6 @@ static int __init ub_init(void) return 0; err_register: - devfs_remove(DEVFS_NAME); unregister_blkdev(UB_MAJOR, DRV_NAME); err_regblkdev: return rc; @@ -2463,7 +2462,6 @@ static void __exit ub_exit(void) { usb_deregister(&ub_driver); - devfs_remove(DEVFS_NAME); unregister_blkdev(UB_MAJOR, DRV_NAME); usb_usual_clear_present(USB_US_TYPE_UB); } diff --git a/drivers/cdrom/sbpcd.c b/drivers/cdrom/sbpcd.c index ee4499acf1a..55cbd23fd9c 100644 --- a/drivers/cdrom/sbpcd.c +++ b/drivers/cdrom/sbpcd.c @@ -5904,7 +5904,6 @@ static void sbpcd_exit(void) if (D_S[j].drv_id==-1) continue; del_gendisk(D_S[j].disk); put_disk(D_S[j].disk); - devfs_remove("sbp/c0t%d", j); vfree(D_S[j].sbp_buf); if (D_S[j].sbp_audsiz>0) vfree(D_S[j].aud_buf); @@ -5915,7 +5914,6 @@ static void sbpcd_exit(void) } vfree(D_S[j].sbpcd_infop); } - devfs_remove("sbp"); msg(DBG_INF, "%s module released.\n", major_name); } diff --git a/drivers/char/dsp56k.c b/drivers/char/dsp56k.c index 45caf778578..ee32e58328a 100644 --- a/drivers/char/dsp56k.c +++ b/drivers/char/dsp56k.c @@ -533,7 +533,6 @@ static void __exit dsp56k_cleanup_driver(void) class_device_destroy(dsp56k_class, MKDEV(DSP56K_MAJOR, 0)); class_destroy(dsp56k_class); unregister_chrdev(DSP56K_MAJOR, "dsp56k"); - devfs_remove("dsp56k"); } module_exit(dsp56k_cleanup_driver); diff --git a/drivers/char/dtlk.c b/drivers/char/dtlk.c index e34b4f4d1c2..6da88e292dc 100644 --- a/drivers/char/dtlk.c +++ b/drivers/char/dtlk.c @@ -354,7 +354,6 @@ static void __exit dtlk_cleanup (void) dtlk_write_tts(DTLK_CLEAR); unregister_chrdev(dtlk_major, "dtlk"); - devfs_remove("dtlk"); release_region(dtlk_port_lpc, DTLK_IO_EXTENT); } diff --git a/drivers/char/ftape/zftape/zftape-init.c b/drivers/char/ftape/zftape/zftape-init.c index 8c6090c3a42..e09ad2b100b 100644 --- a/drivers/char/ftape/zftape/zftape-init.c +++ b/drivers/char/ftape/zftape/zftape-init.c @@ -362,17 +362,11 @@ static void zft_exit(void) TRACE(ft_t_info, "successful"); } for (i = 0; i < 4; i++) { - devfs_remove("qft%i", i); class_device_destroy(zft_class, MKDEV(QIC117_TAPE_MAJOR, i)); - devfs_remove("nqft%i", i); class_device_destroy(zft_class, MKDEV(QIC117_TAPE_MAJOR, i + 4)); - devfs_remove("zqft%i", i); class_device_destroy(zft_class, MKDEV(QIC117_TAPE_MAJOR, i + 16)); - devfs_remove("nzqft%i", i); class_device_destroy(zft_class, MKDEV(QIC117_TAPE_MAJOR, i + 20)); - devfs_remove("rawqft%i", i); class_device_destroy(zft_class, MKDEV(QIC117_TAPE_MAJOR, i + 32)); - devfs_remove("nrawqft%i", i); class_device_destroy(zft_class, MKDEV(QIC117_TAPE_MAJOR, i + 36)); } class_destroy(zft_class); diff --git a/drivers/char/ip2/ip2main.c b/drivers/char/ip2/ip2main.c index afc25db4975..5d2d5732d82 100644 --- a/drivers/char/ip2/ip2main.c +++ b/drivers/char/ip2/ip2main.c @@ -414,9 +414,7 @@ cleanup_module(void) /* free io addresses and Tibet */ release_region( ip2config.addr[i], 8 ); class_device_destroy(ip2_class, MKDEV(IP2_IPL_MAJOR, 4 * i)); - devfs_remove("ip2/ipl%d", i); class_device_destroy(ip2_class, MKDEV(IP2_IPL_MAJOR, 4 * i + 1)); - devfs_remove("ip2/stat%d", i); } /* Disable and remove interrupt handler. */ if ( (ip2config.irq[i] > 0) && have_requested_irq(ip2config.irq[i]) ) { @@ -425,7 +423,6 @@ cleanup_module(void) } } class_destroy(ip2_class); - devfs_remove("ip2"); if ( ( err = tty_unregister_driver ( ip2_tty_driver ) ) ) { printk(KERN_ERR "IP2: failed to unregister tty driver (%d)\n", err); } diff --git a/drivers/char/ipmi/ipmi_devintf.c b/drivers/char/ipmi/ipmi_devintf.c index ec22aba12be..3103ed8e3ad 100644 --- a/drivers/char/ipmi/ipmi_devintf.c +++ b/drivers/char/ipmi/ipmi_devintf.c @@ -833,7 +833,6 @@ static void ipmi_smi_gone(int if_num) } class_device_destroy(ipmi_class, dev); mutex_unlock(®_list_mutex); - devfs_remove("ipmidev/%d", if_num); } static struct ipmi_smi_watcher smi_watcher = @@ -893,7 +892,6 @@ static __exit void cleanup_ipmi(void) mutex_unlock(®_list_mutex); class_destroy(ipmi_class); ipmi_smi_watcher_unregister(&smi_watcher); - devfs_remove(DEVICE_NAME); unregister_chrdev(ipmi_major, DEVICE_NAME); } module_exit(cleanup_ipmi); diff --git a/drivers/char/istallion.c b/drivers/char/istallion.c index 1395d4a5204..0a5808e15b7 100644 --- a/drivers/char/istallion.c +++ b/drivers/char/istallion.c @@ -849,11 +849,8 @@ static void __exit istallion_module_exit(void) return; } put_tty_driver(stli_serial); - for (i = 0; i < 4; i++) { - devfs_remove("staliomem/%d", i); + for (i = 0; i < 4; i++) class_device_destroy(istallion_class, MKDEV(STL_SIOMEMMAJOR, i)); - } - devfs_remove("staliomem"); class_destroy(istallion_class); if ((i = unregister_chrdev(STL_SIOMEMMAJOR, "staliomem"))) printk("STALLION: failed to un-register serial memory device, " diff --git a/drivers/char/lp.c b/drivers/char/lp.c index 1a08e8ea238..530b19de4a3 100644 --- a/drivers/char/lp.c +++ b/drivers/char/lp.c @@ -930,7 +930,6 @@ static int __init lp_init (void) out_class: class_destroy(lp_class); out_devfs: - devfs_remove("printers"); unregister_chrdev(LP_MAJOR, "lp"); return err; } @@ -978,10 +977,8 @@ static void lp_cleanup_module (void) if (lp_table[offset].dev == NULL) continue; parport_unregister_device(lp_table[offset].dev); - devfs_remove("printers/%d", offset); class_device_destroy(lp_class, MKDEV(LP_MAJOR, offset)); } - devfs_remove("printers"); class_destroy(lp_class); } diff --git a/drivers/char/misc.c b/drivers/char/misc.c index 71e51377104..bf2e49e11cc 100644 --- a/drivers/char/misc.c +++ b/drivers/char/misc.c @@ -271,7 +271,6 @@ int misc_deregister(struct miscdevice * misc) down(&misc_sem); list_del(&misc->list); class_device_destroy(misc_class, MKDEV(MISC_MAJOR, misc->minor)); - devfs_remove(misc->devfs_name); if (i < DYNAMIC_MINORS && i>0) { misc_minors[i>>3] &= ~(1 << (misc->minor & 7)); } diff --git a/drivers/char/ppdev.c b/drivers/char/ppdev.c index ba350799405..c8a2eb1a558 100644 --- a/drivers/char/ppdev.c +++ b/drivers/char/ppdev.c @@ -770,7 +770,7 @@ static struct parport_driver pp_driver = { static int __init ppdev_init (void) { - int i, err = 0; + int err = 0; if (register_chrdev (PP_MAJOR, CHRDEV, &pp_fops)) { printk (KERN_WARNING CHRDEV ": unable to get major %d\n", @@ -791,9 +791,6 @@ static int __init ppdev_init (void) goto out; out_class: - for (i = 0; i < PARPORT_MAX; i++) - devfs_remove("parports/%d", i); - devfs_remove("parports"); class_destroy(ppdev_class); out_chrdev: unregister_chrdev(PP_MAJOR, CHRDEV); @@ -803,12 +800,8 @@ out: static void __exit ppdev_cleanup (void) { - int i; /* Clean up all parport stuff */ - for (i = 0; i < PARPORT_MAX; i++) - devfs_remove("parports/%d", i); parport_unregister_driver(&pp_driver); - devfs_remove("parports"); class_destroy(ppdev_class); unregister_chrdev (PP_MAJOR, CHRDEV); } diff --git a/drivers/char/raw.c b/drivers/char/raw.c index 72c8ee4c006..941a17ab219 100644 --- a/drivers/char/raw.c +++ b/drivers/char/raw.c @@ -318,12 +318,6 @@ error: static void __exit raw_exit(void) { - int i; - - for (i = 1; i < MAX_RAW_MINORS; i++) - devfs_remove("raw/raw%d", i); - devfs_remove("raw/rawctl"); - devfs_remove("raw"); class_device_destroy(raw_class, MKDEV(RAW_MAJOR, 0)); class_destroy(raw_class); cdev_del(&raw_cdev); diff --git a/drivers/char/stallion.c b/drivers/char/stallion.c index a1a68f90d80..7840eb16f17 100644 --- a/drivers/char/stallion.c +++ b/drivers/char/stallion.c @@ -773,11 +773,8 @@ static void __exit stallion_module_exit(void) restore_flags(flags); return; } - for (i = 0; i < 4; i++) { - devfs_remove("staliomem/%d", i); + for (i = 0; i < 4; i++) class_device_destroy(stallion_class, MKDEV(STL_SIOMEMMAJOR, i)); - } - devfs_remove("staliomem"); if ((i = unregister_chrdev(STL_SIOMEMMAJOR, "staliomem"))) printk("STALLION: failed to un-register serial memory device, " "errno=%d\n", -i); diff --git a/drivers/char/tipar.c b/drivers/char/tipar.c index 0dc83a5bd4c..48476f8fcda 100644 --- a/drivers/char/tipar.c +++ b/drivers/char/tipar.c @@ -515,7 +515,6 @@ out_class: class_destroy(tipar_class); out_chrdev: - devfs_remove("ticables/par"); unregister_chrdev(TIPAR_MAJOR, "tipar"); out: return err; @@ -536,10 +535,8 @@ tipar_cleanup_module(void) continue; parport_unregister_device(table[i].dev); class_device_destroy(tipar_class, MKDEV(TIPAR_MAJOR, i)); - devfs_remove("ticables/par/%d", i); } class_destroy(tipar_class); - devfs_remove("ticables/par"); pr_info("tipar: module unloaded\n"); } diff --git a/drivers/char/tty_io.c b/drivers/char/tty_io.c index 0b9a33c0f1b..b8e03a31a50 100644 --- a/drivers/char/tty_io.c +++ b/drivers/char/tty_io.c @@ -2997,7 +2997,6 @@ struct class_device *tty_register_device(struct tty_driver *driver, */ void tty_unregister_device(struct tty_driver *driver, unsigned index) { - devfs_remove("%s%d", driver->devfs_name, index + driver->name_base); class_device_destroy(tty_class, MKDEV(driver->major, driver->minor_start) + index); } diff --git a/drivers/char/vc_screen.c b/drivers/char/vc_screen.c index 1633d820652..2266fbad62c 100644 --- a/drivers/char/vc_screen.c +++ b/drivers/char/vc_screen.c @@ -485,8 +485,6 @@ void vcs_make_devfs(struct tty_struct *tty) } void vcs_remove_devfs(struct tty_struct *tty) { - devfs_remove("vcc/%u", tty->index + 1); - devfs_remove("vcc/a%u", tty->index + 1); class_device_destroy(vc_class, MKDEV(VCS_MAJOR, tty->index + 1)); class_device_destroy(vc_class, MKDEV(VCS_MAJOR, tty->index + 129)); } diff --git a/drivers/char/viotape.c b/drivers/char/viotape.c index 4c1a128fd77..b101134a844 100644 --- a/drivers/char/viotape.c +++ b/drivers/char/viotape.c @@ -970,8 +970,6 @@ static int viotape_remove(struct vio_dev *vdev) { int i = vdev->unit_address; - devfs_remove("iseries/nvt%d", i); - devfs_remove("iseries/vt%d", i); class_device_destroy(tape_class, MKDEV(VIOTAPE_MAJOR, i | 0x80)); class_device_destroy(tape_class, MKDEV(VIOTAPE_MAJOR, i)); return 0; diff --git a/drivers/isdn/capi/capi.c b/drivers/isdn/capi/capi.c index 1c8fe77270a..1200d1de1de 100644 --- a/drivers/isdn/capi/capi.c +++ b/drivers/isdn/capi/capi.c @@ -1550,7 +1550,6 @@ static void __exit capi_exit(void) class_device_destroy(capi_class, MKDEV(capi_major, 0)); class_destroy(capi_class); unregister_chrdev(capi_major, "capi20"); - devfs_remove("isdn/capi20"); #ifdef CONFIG_ISDN_CAPI_MIDDLEWARE capinc_tty_exit(); diff --git a/drivers/isdn/hardware/eicon/divamnt.c b/drivers/isdn/hardware/eicon/divamnt.c index 12465b3c2b4..83841d16c35 100644 --- a/drivers/isdn/hardware/eicon/divamnt.c +++ b/drivers/isdn/hardware/eicon/divamnt.c @@ -178,7 +178,6 @@ static struct file_operations divas_maint_fops = { static void divas_maint_unregister_chrdev(void) { - devfs_remove(DEVNAME); unregister_chrdev(major, DEVNAME); } diff --git a/drivers/isdn/hardware/eicon/divasi.c b/drivers/isdn/hardware/eicon/divasi.c index 2d80d94b5d0..dcd5b650a00 100644 --- a/drivers/isdn/hardware/eicon/divasi.c +++ b/drivers/isdn/hardware/eicon/divasi.c @@ -145,7 +145,6 @@ static struct file_operations divas_idi_fops = { static void divas_idi_unregister_chrdev(void) { - devfs_remove(DEVNAME); unregister_chrdev(major, DEVNAME); } diff --git a/drivers/isdn/hardware/eicon/divasmain.c b/drivers/isdn/hardware/eicon/divasmain.c index 71b3e06cbe3..bbb7d8fc207 100644 --- a/drivers/isdn/hardware/eicon/divasmain.c +++ b/drivers/isdn/hardware/eicon/divasmain.c @@ -678,7 +678,6 @@ static struct file_operations divas_fops = { static void divas_unregister_chrdev(void) { - devfs_remove(DEVNAME); unregister_chrdev(major, DEVNAME); } diff --git a/drivers/md/dm-ioctl.c b/drivers/md/dm-ioctl.c index 6e1b51d748a..361d50b6469 100644 --- a/drivers/md/dm-ioctl.c +++ b/drivers/md/dm-ioctl.c @@ -74,7 +74,6 @@ static int dm_hash_init(void) static void dm_hash_exit(void) { dm_hash_remove_all(0); - devfs_remove(DM_DIR); } /*----------------------------------------------------------------- @@ -170,15 +169,6 @@ static void free_cell(struct hash_cell *hc) } } -/* - * devfs stuff. - */ -static int unregister_with_devfs(struct hash_cell *hc) -{ - devfs_remove(DM_DIR"/%s", hc->name); - return 0; -} - /* * The kdev_t and uuid of a device can never change once it is * initially inserted. @@ -234,7 +224,6 @@ static void __hash_remove(struct hash_cell *hc) /* remove from the dev hash */ list_del(&hc->uuid_list); list_del(&hc->name_list); - unregister_with_devfs(hc); dm_set_mdptr(hc->md, NULL); table = dm_get_table(hc->md); @@ -330,8 +319,6 @@ static int dm_hash_rename(const char *old, const char *new) /* * rename and move the name cell. */ - unregister_with_devfs(hc); - list_del(&hc->name_list); old_name = hc->name; hc->name = new_name; diff --git a/drivers/md/md.c b/drivers/md/md.c index ee7320b4a30..9fc2314b58d 100644 --- a/drivers/md/md.c +++ b/drivers/md/md.c @@ -5611,15 +5611,9 @@ static __exit void md_exit(void) { mddev_t *mddev; struct list_head *tmp; - int i; + blk_unregister_region(MKDEV(MAJOR_NR,0), MAX_MD_DEVS); blk_unregister_region(MKDEV(mdp_major,0), MAX_MD_DEVS << MdpMinorShift); - for (i=0; i < MAX_MD_DEVS; i++) - devfs_remove("md/%d", i); - for (i=0; i < MAX_MD_DEVS; i++) - devfs_remove("md/d%d", i); - - devfs_remove("md"); unregister_blkdev(MAJOR_NR,"md"); unregister_blkdev(mdp_major, "mdp"); diff --git a/drivers/media/dvb/dvb-core/dvbdev.c b/drivers/media/dvb/dvb-core/dvbdev.c index 8e926d98bd9..40774feb895 100644 --- a/drivers/media/dvb/dvb-core/dvbdev.c +++ b/drivers/media/dvb/dvb-core/dvbdev.c @@ -248,9 +248,6 @@ void dvb_unregister_device(struct dvb_device *dvbdev) if (!dvbdev) return; - devfs_remove("dvb/adapter%d/%s%d", dvbdev->adapter->num, - dnames[dvbdev->type], dvbdev->id); - class_device_destroy(dvb_class, MKDEV(DVB_MAJOR, nums2minor(dvbdev->adapter->num, dvbdev->type, dvbdev->id))); @@ -314,8 +311,6 @@ EXPORT_SYMBOL(dvb_register_adapter); int dvb_unregister_adapter(struct dvb_adapter *adap) { - devfs_remove("dvb/adapter%d", adap->num); - if (mutex_lock_interruptible(&dvbdev_register_lock)) return -ERESTARTSYS; list_del (&adap->list_head); @@ -421,7 +416,6 @@ error: static void __exit exit_dvbdev(void) { - devfs_remove("dvb"); class_destroy(dvb_class); cdev_del(&dvb_device_cdev); unregister_chrdev_region(MKDEV(DVB_MAJOR, 0), MAX_DVB_MINORS); diff --git a/drivers/media/video/videodev.c b/drivers/media/video/videodev.c index cc67ee95283..3178353787d 100644 --- a/drivers/media/video/videodev.c +++ b/drivers/media/video/videodev.c @@ -1601,7 +1601,6 @@ void video_unregister_device(struct video_device *vfd) if(video_device[vfd->minor]!=vfd) panic("videodev: bad unregister"); - devfs_remove(vfd->devfs_name); video_device[vfd->minor]=NULL; class_device_unregister(&vfd->class_dev); mutex_unlock(&videodev_lock); diff --git a/drivers/mmc/mmc_block.c b/drivers/mmc/mmc_block.c index 82c660bf165..2e5218b8d4a 100644 --- a/drivers/mmc/mmc_block.c +++ b/drivers/mmc/mmc_block.c @@ -564,7 +564,6 @@ static int __init mmc_blk_init(void) static void __exit mmc_blk_exit(void) { mmc_unregister_driver(&mmc_driver); - devfs_remove("mmc"); unregister_blkdev(major, "mmc"); } diff --git a/drivers/net/ppp_generic.c b/drivers/net/ppp_generic.c index 52bc5154513..16a59cafb8d 100644 --- a/drivers/net/ppp_generic.c +++ b/drivers/net/ppp_generic.c @@ -2674,7 +2674,6 @@ static void __exit ppp_cleanup(void) cardmap_destroy(&all_ppp_units); if (unregister_chrdev(PPP_MAJOR, "ppp") != 0) printk(KERN_ERR "PPP: failed to unregister PPP device\n"); - devfs_remove("ppp"); class_device_destroy(ppp_class, MKDEV(PPP_MAJOR, 0)); class_destroy(ppp_class); } diff --git a/drivers/net/wan/cosa.c b/drivers/net/wan/cosa.c index c7b530628c8..3d1e2b6700a 100644 --- a/drivers/net/wan/cosa.c +++ b/drivers/net/wan/cosa.c @@ -418,12 +418,9 @@ static void __exit cosa_exit(void) int i; printk(KERN_INFO "Unloading the cosa module\n"); - for (i=0; inchannels; i++) { diff --git a/drivers/s390/block/dasd.c b/drivers/s390/block/dasd.c index 09bf44fdb1d..8b67ce00652 100644 --- a/drivers/s390/block/dasd.c +++ b/drivers/s390/block/dasd.c @@ -1834,7 +1834,6 @@ dasd_exit(void) } dasd_gendisk_exit(); dasd_devmap_exit(); - devfs_remove("dasd"); if (dasd_debug_area != NULL) { debug_unregister(dasd_debug_area); dasd_debug_area = NULL; diff --git a/drivers/s390/block/xpram.c b/drivers/s390/block/xpram.c index 36bc01f83a5..15aad01bba0 100644 --- a/drivers/s390/block/xpram.c +++ b/drivers/s390/block/xpram.c @@ -474,7 +474,6 @@ static int __init xpram_setup_blkdev(void) return 0; out_unreg: - devfs_remove("slram"); unregister_blkdev(XPRAM_MAJOR, XPRAM_NAME); out: while (i--) @@ -493,7 +492,6 @@ static void __exit xpram_exit(void) put_disk(xpram_disks[i]); } unregister_blkdev(XPRAM_MAJOR, XPRAM_NAME); - devfs_remove("slram"); blk_cleanup_queue(xpram_queue); sysdev_unregister(&xpram_sys_device); sysdev_class_unregister(&xpram_sysclass); diff --git a/drivers/sbus/char/bpp.c b/drivers/sbus/char/bpp.c index e680de41d76..03ff5a3b121 100644 --- a/drivers/sbus/char/bpp.c +++ b/drivers/sbus/char/bpp.c @@ -1039,9 +1039,6 @@ static void __exit bpp_cleanup(void) { unsigned idx; - for (idx = 0; idx < BPP_NO; idx++) - devfs_remove("bpp/%d", idx); - devfs_remove("bpp"); unregister_chrdev(BPP_MAJOR, dev_name); for (idx = 0; idx < BPP_NO; idx++) { diff --git a/drivers/sbus/char/vfc_dev.c b/drivers/sbus/char/vfc_dev.c index 5d32384ad72..55b2b31bd7a 100644 --- a/drivers/sbus/char/vfc_dev.c +++ b/drivers/sbus/char/vfc_dev.c @@ -712,7 +712,6 @@ static void deinit_vfc_device(struct vfc_dev *dev) { if(dev == NULL) return; - devfs_remove("vfc/%d", dev->instance); sbus_iounmap(dev->regs, sizeof(struct vfc_regs)); kfree(dev); } @@ -726,7 +725,6 @@ void cleanup_module(void) for (devp = vfc_dev_lst; *devp; devp++) deinit_vfc_device(*devp); - devfs_remove("vfc"); kfree(vfc_dev_lst); return; } diff --git a/drivers/telephony/phonedev.c b/drivers/telephony/phonedev.c index 80f9fe40527..bc5b1d245df 100644 --- a/drivers/telephony/phonedev.c +++ b/drivers/telephony/phonedev.c @@ -123,7 +123,6 @@ void phone_unregister_device(struct phone_device *pfd) mutex_lock(&phone_lock); if (phone_device[pfd->minor] != pfd) panic("phone: bad unregister"); - devfs_remove("phone/%d", pfd->minor); phone_device[pfd->minor] = NULL; mutex_unlock(&phone_lock); } diff --git a/drivers/video/fbmem.c b/drivers/video/fbmem.c index be2479ab8bc..afa56a887fa 100644 --- a/drivers/video/fbmem.c +++ b/drivers/video/fbmem.c @@ -1357,7 +1357,6 @@ unregister_framebuffer(struct fb_info *fb_info) i = fb_info->node; if (!registered_fb[i]) return -EINVAL; - devfs_remove("fb/%d", i); if (fb_info->pixmap.addr && (fb_info->pixmap.flags & FB_PIXMAP_DEFAULT)) diff --git a/fs/coda/psdev.c b/fs/coda/psdev.c index f478222cb5c..aaf6462a942 100644 --- a/fs/coda/psdev.c +++ b/fs/coda/psdev.c @@ -409,12 +409,9 @@ static int __init init_coda(void) } return 0; out: - for (i = 0; i < MAX_CODADEVS; i++) { + for (i = 0; i < MAX_CODADEVS; i++) class_device_destroy(coda_psdev_class, MKDEV(CODA_PSDEV_MAJOR, i)); - devfs_remove("coda/%d", i); - } class_destroy(coda_psdev_class); - devfs_remove("coda"); unregister_chrdev(CODA_PSDEV_MAJOR, "coda"); coda_sysctl_clean(); out1: @@ -431,12 +428,9 @@ static void __exit exit_coda(void) if ( err != 0 ) { printk("coda: failed to unregister filesystem\n"); } - for (i = 0; i < MAX_CODADEVS; i++) { + for (i = 0; i < MAX_CODADEVS; i++) class_device_destroy(coda_psdev_class, MKDEV(CODA_PSDEV_MAJOR, i)); - devfs_remove("coda/%d", i); - } class_destroy(coda_psdev_class); - devfs_remove("coda"); unregister_chrdev(CODA_PSDEV_MAJOR, "coda"); coda_sysctl_clean(); coda_destroy_inodecache(); diff --git a/fs/partitions/check.c b/fs/partitions/check.c index d7a5078b3d7..202f50acea8 100644 --- a/fs/partitions/check.c +++ b/fs/partitions/check.c @@ -320,7 +320,6 @@ void delete_partition(struct gendisk *disk, int part) p->nr_sects = 0; p->ios[0] = p->ios[1] = 0; p->sectors[0] = p->sectors[1] = 0; - devfs_remove("%s/part%d", disk->devfs_name, part); sysfs_remove_link(&p->kobj, "subsystem"); if (p->holder_dir) kobject_unregister(p->holder_dir); diff --git a/include/linux/devfs_fs_kernel.h b/include/linux/devfs_fs_kernel.h index c7d05dbc2af..2036d2e9098 100644 --- a/include/linux/devfs_fs_kernel.h +++ b/include/linux/devfs_fs_kernel.h @@ -6,7 +6,4 @@ #include #include -static inline void devfs_remove(const char *fmt, ...) -{ -} #endif /* _LINUX_DEVFS_FS_KERNEL_H */ -- cgit v1.2.3-70-g09d2 From ff23eca3e8f613034e0d20ff86f6a89b62f5a14e Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Mon, 20 Jun 2005 21:15:16 -0700 Subject: [PATCH] devfs: Remove the devfs_fs_kernel.h file from the tree Also fixes up all files that #include it. Signed-off-by: Greg Kroah-Hartman --- arch/sparc64/solaris/socksys.c | 1 - arch/um/drivers/line.c | 1 - arch/um/drivers/ubd_kern.c | 1 - drivers/block/acsi_slm.c | 1 - drivers/block/cpqarray.c | 1 - drivers/block/floppy.c | 1 - drivers/block/loop.c | 1 - drivers/block/nbd.c | 2 -- drivers/block/paride/pg.c | 1 - drivers/block/paride/pt.c | 1 - drivers/block/rd.c | 1 - drivers/block/swim3.c | 1 - drivers/block/sx8.c | 1 - drivers/block/ub.c | 1 - drivers/cdrom/cdu31a.c | 1 - drivers/cdrom/cm206.c | 1 - drivers/cdrom/mcdx.c | 1 - drivers/cdrom/sbpcd.c | 1 - drivers/char/dsp56k.c | 1 - drivers/char/dtlk.c | 1 - drivers/char/ftape/zftape/zftape-init.c | 1 - drivers/char/ip2/ip2main.c | 1 - drivers/char/ipmi/ipmi_devintf.c | 1 - drivers/char/istallion.c | 1 - drivers/char/lp.c | 1 - drivers/char/mem.c | 1 - drivers/char/misc.c | 1 - drivers/char/mmtimer.c | 1 - drivers/char/ppdev.c | 1 - drivers/char/pty.c | 1 - drivers/char/raw.c | 1 - drivers/char/stallion.c | 1 - drivers/char/tipar.c | 1 - drivers/char/tty_io.c | 1 - drivers/char/vc_screen.c | 1 - drivers/char/viotape.c | 1 - drivers/char/vt.c | 1 - drivers/input/serio/serio_raw.c | 1 - drivers/isdn/capi/capi.c | 1 - drivers/isdn/hardware/eicon/divamnt.c | 1 - drivers/isdn/hardware/eicon/divasi.c | 1 - drivers/isdn/hardware/eicon/divasmain.c | 1 - drivers/macintosh/adb.c | 1 - drivers/md/dm-ioctl.c | 1 - drivers/md/md.c | 1 - drivers/media/dvb/dvb-core/dvbdev.h | 1 - drivers/media/dvb/ttpci/av7110.h | 4 ---- drivers/media/video/arv.c | 1 - drivers/media/video/videodev.c | 1 - drivers/mmc/mmc_block.c | 1 - drivers/net/ppp_generic.c | 1 - drivers/net/wan/cosa.c | 1 - drivers/s390/block/dasd_int.h | 1 - drivers/s390/block/xpram.c | 1 - drivers/sbus/char/bpp.c | 1 - drivers/sbus/char/vfc.h | 2 -- drivers/telephony/phonedev.c | 1 - drivers/video/fbmem.c | 1 - fs/block_dev.c | 1 - fs/char_dev.c | 1 - fs/coda/psdev.c | 1 - fs/partitions/check.c | 1 - include/asm-ppc/ocp.h | 1 - include/linux/devfs_fs_kernel.h | 9 --------- include/linux/fb.h | 1 - init/do_mounts.h | 1 - init/main.c | 1 - mm/shmem.c | 1 - mm/tiny-shmem.c | 1 - net/irda/irnet/irnet.h | 1 - 70 files changed, 83 deletions(-) delete mode 100644 include/linux/devfs_fs_kernel.h (limited to 'include') diff --git a/arch/sparc64/solaris/socksys.c b/arch/sparc64/solaris/socksys.c index 09664ab124e..bc3df95bc05 100644 --- a/arch/sparc64/solaris/socksys.c +++ b/arch/sparc64/solaris/socksys.c @@ -26,7 +26,6 @@ #include #include #include -#include #include diff --git a/arch/um/drivers/line.c b/arch/um/drivers/line.c index 6c2d4ccaf20..ee1fe6a2490 100644 --- a/arch/um/drivers/line.c +++ b/arch/um/drivers/line.c @@ -8,7 +8,6 @@ #include "linux/list.h" #include "linux/kd.h" #include "linux/interrupt.h" -#include "linux/devfs_fs_kernel.h" #include "asm/uaccess.h" #include "chan_kern.h" #include "irq_user.h" diff --git a/arch/um/drivers/ubd_kern.c b/arch/um/drivers/ubd_kern.c index 3883c5e6a7d..5856a02fff5 100644 --- a/arch/um/drivers/ubd_kern.c +++ b/arch/um/drivers/ubd_kern.c @@ -25,7 +25,6 @@ #include "linux/blkdev.h" #include "linux/hdreg.h" #include "linux/init.h" -#include "linux/devfs_fs_kernel.h" #include "linux/cdrom.h" #include "linux/proc_fs.h" #include "linux/ctype.h" diff --git a/drivers/block/acsi_slm.c b/drivers/block/acsi_slm.c index 7495fef2418..4030a8fd118 100644 --- a/drivers/block/acsi_slm.c +++ b/drivers/block/acsi_slm.c @@ -65,7 +65,6 @@ not be guaranteed. There are several ways to assure this: #include #include #include -#include #include #include diff --git a/drivers/block/cpqarray.c b/drivers/block/cpqarray.c index 2701ed734b2..f96e6c03593 100644 --- a/drivers/block/cpqarray.c +++ b/drivers/block/cpqarray.c @@ -33,7 +33,6 @@ #include #include #include -#include #include #include #include diff --git a/drivers/block/floppy.c b/drivers/block/floppy.c index 8e101a9655b..0242cbb86a8 100644 --- a/drivers/block/floppy.c +++ b/drivers/block/floppy.c @@ -177,7 +177,6 @@ static int print_unex = 1; #include #include #include -#include #include #include /* for invalidate_buffers() */ #include diff --git a/drivers/block/loop.c b/drivers/block/loop.c index 0319b096b55..dbf905ce024 100644 --- a/drivers/block/loop.c +++ b/drivers/block/loop.c @@ -63,7 +63,6 @@ #include #include #include -#include #include #include #include diff --git a/drivers/block/nbd.c b/drivers/block/nbd.c index c24b349ae57..b164ee0ac79 100644 --- a/drivers/block/nbd.c +++ b/drivers/block/nbd.c @@ -29,8 +29,6 @@ #include #include -#include - #include #include #include diff --git a/drivers/block/paride/pg.c b/drivers/block/paride/pg.c index e6f1614b69c..13f998aa1cd 100644 --- a/drivers/block/paride/pg.c +++ b/drivers/block/paride/pg.c @@ -156,7 +156,6 @@ enum {D_PRT, D_PRO, D_UNI, D_MOD, D_SLV, D_DLY}; #include #include #include -#include #include #include #include diff --git a/drivers/block/paride/pt.c b/drivers/block/paride/pt.c index ea12073c5fb..35fb2663672 100644 --- a/drivers/block/paride/pt.c +++ b/drivers/block/paride/pt.c @@ -141,7 +141,6 @@ static int (*drives[4])[6] = {&drive0, &drive1, &drive2, &drive3}; #include #include #include -#include #include #include #include diff --git a/drivers/block/rd.c b/drivers/block/rd.c index acaa2abf05a..536be2de989 100644 --- a/drivers/block/rd.c +++ b/drivers/block/rd.c @@ -50,7 +50,6 @@ #include #include #include -#include #include #include #include diff --git a/drivers/block/swim3.c b/drivers/block/swim3.c index 018b2921e8f..ef557b1043d 100644 --- a/drivers/block/swim3.c +++ b/drivers/block/swim3.c @@ -25,7 +25,6 @@ #include #include #include -#include #include #include #include diff --git a/drivers/block/sx8.c b/drivers/block/sx8.c index a366ea8c6a4..3a481455156 100644 --- a/drivers/block/sx8.c +++ b/drivers/block/sx8.c @@ -18,7 +18,6 @@ #include #include #include -#include #include #include #include diff --git a/drivers/block/ub.c b/drivers/block/ub.c index b68abc9d4c1..87ec6458bd1 100644 --- a/drivers/block/ub.c +++ b/drivers/block/ub.c @@ -24,7 +24,6 @@ #include #include #include -#include #include #include diff --git a/drivers/cdrom/cdu31a.c b/drivers/cdrom/cdu31a.c index 72ffd64e8b1..5f0f2027f29 100644 --- a/drivers/cdrom/cdu31a.c +++ b/drivers/cdrom/cdu31a.c @@ -161,7 +161,6 @@ #include #include #include -#include #include #include #include diff --git a/drivers/cdrom/cm206.c b/drivers/cdrom/cm206.c index f43a988dd41..4ee288688fe 100644 --- a/drivers/cdrom/cm206.c +++ b/drivers/cdrom/cm206.c @@ -187,7 +187,6 @@ History: #include #include #include -#include #include #include #include diff --git a/drivers/cdrom/mcdx.c b/drivers/cdrom/mcdx.c index 0f6e7aab8d2..788c7a0b2fe 100644 --- a/drivers/cdrom/mcdx.c +++ b/drivers/cdrom/mcdx.c @@ -74,7 +74,6 @@ static const char *mcdx_c_version #include #define MAJOR_NR MITSUMI_X_CDROM_MAJOR #include -#include #include "mcdx.h" diff --git a/drivers/cdrom/sbpcd.c b/drivers/cdrom/sbpcd.c index 55cbd23fd9c..a75b363b19c 100644 --- a/drivers/cdrom/sbpcd.c +++ b/drivers/cdrom/sbpcd.c @@ -371,7 +371,6 @@ #include #include #include -#include #include #include #include diff --git a/drivers/char/dsp56k.c b/drivers/char/dsp56k.c index ee32e58328a..09b413618b5 100644 --- a/drivers/char/dsp56k.c +++ b/drivers/char/dsp56k.c @@ -33,7 +33,6 @@ #include #include #include -#include #include #include diff --git a/drivers/char/dtlk.c b/drivers/char/dtlk.c index 6da88e292dc..da2c89f1b8b 100644 --- a/drivers/char/dtlk.c +++ b/drivers/char/dtlk.c @@ -62,7 +62,6 @@ #include /* for __init, module_{init,exit} */ #include /* for POLLIN, etc. */ #include /* local header file for DoubleTalk values */ -#include #include #ifdef TRACING diff --git a/drivers/char/ftape/zftape/zftape-init.c b/drivers/char/ftape/zftape/zftape-init.c index e09ad2b100b..3eeb869a9a1 100644 --- a/drivers/char/ftape/zftape/zftape-init.c +++ b/drivers/char/ftape/zftape/zftape-init.c @@ -33,7 +33,6 @@ #endif #include #include -#include #include #include diff --git a/drivers/char/ip2/ip2main.c b/drivers/char/ip2/ip2main.c index 5d2d5732d82..a273eef69a8 100644 --- a/drivers/char/ip2/ip2main.c +++ b/drivers/char/ip2/ip2main.c @@ -91,7 +91,6 @@ #include #include #include -#include #include #include #include diff --git a/drivers/char/ipmi/ipmi_devintf.c b/drivers/char/ipmi/ipmi_devintf.c index 3103ed8e3ad..da637adbbfa 100644 --- a/drivers/char/ipmi/ipmi_devintf.c +++ b/drivers/char/ipmi/ipmi_devintf.c @@ -40,7 +40,6 @@ #include #include #include -#include #include #include #include diff --git a/drivers/char/istallion.c b/drivers/char/istallion.c index 0a5808e15b7..1386d6a519d 100644 --- a/drivers/char/istallion.c +++ b/drivers/char/istallion.c @@ -39,7 +39,6 @@ #include #include #include -#include #include #include diff --git a/drivers/char/lp.c b/drivers/char/lp.c index 530b19de4a3..b11a390581b 100644 --- a/drivers/char/lp.c +++ b/drivers/char/lp.c @@ -120,7 +120,6 @@ #include #include #include -#include #include #include #include diff --git a/drivers/char/mem.c b/drivers/char/mem.c index e7aa71575aa..6fe7b6c6c46 100644 --- a/drivers/char/mem.c +++ b/drivers/char/mem.c @@ -20,7 +20,6 @@ #include #include #include -#include #include #include #include diff --git a/drivers/char/misc.c b/drivers/char/misc.c index bf2e49e11cc..f1fdd84c5dc 100644 --- a/drivers/char/misc.c +++ b/drivers/char/misc.c @@ -44,7 +44,6 @@ #include #include #include -#include #include #include #include diff --git a/drivers/char/mmtimer.c b/drivers/char/mmtimer.c index d65b3109318..f25168a9676 100644 --- a/drivers/char/mmtimer.c +++ b/drivers/char/mmtimer.c @@ -25,7 +25,6 @@ #include #include #include -#include #include #include #include diff --git a/drivers/char/ppdev.c b/drivers/char/ppdev.c index c8a2eb1a558..24231d9743d 100644 --- a/drivers/char/ppdev.c +++ b/drivers/char/ppdev.c @@ -60,7 +60,6 @@ #include #include #include -#include #include #include #include diff --git a/drivers/char/pty.c b/drivers/char/pty.c index 64d3d1d0b31..345f8b8d57c 100644 --- a/drivers/char/pty.c +++ b/drivers/char/pty.c @@ -24,7 +24,6 @@ #include #include #include -#include #include #include diff --git a/drivers/char/raw.c b/drivers/char/raw.c index 941a17ab219..9bf97c5e38c 100644 --- a/drivers/char/raw.c +++ b/drivers/char/raw.c @@ -10,7 +10,6 @@ #include #include -#include #include #include #include diff --git a/drivers/char/stallion.c b/drivers/char/stallion.c index 7840eb16f17..36df18cd96d 100644 --- a/drivers/char/stallion.c +++ b/drivers/char/stallion.c @@ -40,7 +40,6 @@ #include #include #include -#include #include #include diff --git a/drivers/char/tipar.c b/drivers/char/tipar.c index 48476f8fcda..f7802e5bd7c 100644 --- a/drivers/char/tipar.c +++ b/drivers/char/tipar.c @@ -56,7 +56,6 @@ #include #include #include -#include /* DevFs support */ #include /* Our code depend on parport */ #include diff --git a/drivers/char/tty_io.c b/drivers/char/tty_io.c index b8e03a31a50..ebe1158e6c6 100644 --- a/drivers/char/tty_io.c +++ b/drivers/char/tty_io.c @@ -102,7 +102,6 @@ #include #include #include -#include #include diff --git a/drivers/char/vc_screen.c b/drivers/char/vc_screen.c index 2266fbad62c..234d7f3fb11 100644 --- a/drivers/char/vc_screen.c +++ b/drivers/char/vc_screen.c @@ -26,7 +26,6 @@ #include #include #include -#include #include #include #include diff --git a/drivers/char/viotape.c b/drivers/char/viotape.c index b101134a844..198f1505ae2 100644 --- a/drivers/char/viotape.c +++ b/drivers/char/viotape.c @@ -43,7 +43,6 @@ #include #include #include -#include #include #include #include diff --git a/drivers/char/vt.c b/drivers/char/vt.c index 714d95ff2f1..42afe7cea04 100644 --- a/drivers/char/vt.c +++ b/drivers/char/vt.c @@ -79,7 +79,6 @@ #include #include #include -#include #include #include #include diff --git a/drivers/input/serio/serio_raw.c b/drivers/input/serio/serio_raw.c index 5a2703b536d..71a8eea816c 100644 --- a/drivers/input/serio/serio_raw.c +++ b/drivers/input/serio/serio_raw.c @@ -16,7 +16,6 @@ #include #include #include -#include #include #include #include diff --git a/drivers/isdn/capi/capi.c b/drivers/isdn/capi/capi.c index 1200d1de1de..334402e96e1 100644 --- a/drivers/isdn/capi/capi.c +++ b/drivers/isdn/capi/capi.c @@ -39,7 +39,6 @@ #include #include #include -#include #include #include #if defined(CONFIG_ISDN_CAPI_CAPIFS) || defined(CONFIG_ISDN_CAPI_CAPIFS_MODULE) diff --git a/drivers/isdn/hardware/eicon/divamnt.c b/drivers/isdn/hardware/eicon/divamnt.c index 83841d16c35..b163c590918 100644 --- a/drivers/isdn/hardware/eicon/divamnt.c +++ b/drivers/isdn/hardware/eicon/divamnt.c @@ -17,7 +17,6 @@ #include #include #include -#include #include #include "platform.h" diff --git a/drivers/isdn/hardware/eicon/divasi.c b/drivers/isdn/hardware/eicon/divasi.c index dcd5b650a00..6e7d89a31c1 100644 --- a/drivers/isdn/hardware/eicon/divasi.c +++ b/drivers/isdn/hardware/eicon/divasi.c @@ -19,7 +19,6 @@ #include #include #include -#include #include #include "platform.h" diff --git a/drivers/isdn/hardware/eicon/divasmain.c b/drivers/isdn/hardware/eicon/divasmain.c index bbb7d8fc207..9dee6a39104 100644 --- a/drivers/isdn/hardware/eicon/divasmain.c +++ b/drivers/isdn/hardware/eicon/divasmain.c @@ -14,7 +14,6 @@ #include #include #include -#include #include #include #include diff --git a/drivers/macintosh/adb.c b/drivers/macintosh/adb.c index a60e3abaa8e..9f1a049dc22 100644 --- a/drivers/macintosh/adb.c +++ b/drivers/macintosh/adb.c @@ -36,7 +36,6 @@ #include #include #include -#include #include #include diff --git a/drivers/md/dm-ioctl.c b/drivers/md/dm-ioctl.c index 361d50b6469..3ad60caea77 100644 --- a/drivers/md/dm-ioctl.c +++ b/drivers/md/dm-ioctl.c @@ -13,7 +13,6 @@ #include #include #include -#include #include #include diff --git a/drivers/md/md.c b/drivers/md/md.c index 9fc2314b58d..95c275d1931 100644 --- a/drivers/md/md.c +++ b/drivers/md/md.c @@ -39,7 +39,6 @@ #include #include #include -#include #include /* for invalidate_bdev */ #include #include diff --git a/drivers/media/dvb/dvb-core/dvbdev.h b/drivers/media/dvb/dvb-core/dvbdev.h index d7a976d040d..7a7f75fd168 100644 --- a/drivers/media/dvb/dvb-core/dvbdev.h +++ b/drivers/media/dvb/dvb-core/dvbdev.h @@ -27,7 +27,6 @@ #include #include #include -#include #include #define DVB_MAJOR 212 diff --git a/drivers/media/dvb/ttpci/av7110.h b/drivers/media/dvb/ttpci/av7110.h index 3e2e12124ba..9c79696da08 100644 --- a/drivers/media/dvb/ttpci/av7110.h +++ b/drivers/media/dvb/ttpci/av7110.h @@ -6,10 +6,6 @@ #include #include -#ifdef CONFIG_DEVFS_FS -#include -#endif - #include #include #include diff --git a/drivers/media/video/arv.c b/drivers/media/video/arv.c index 6e08e32346e..ae14f5f3203 100644 --- a/drivers/media/video/arv.c +++ b/drivers/media/video/arv.c @@ -20,7 +20,6 @@ #include #include -#include #include #include #include diff --git a/drivers/media/video/videodev.c b/drivers/media/video/videodev.c index 3178353787d..a802a7e0df7 100644 --- a/drivers/media/video/videodev.c +++ b/drivers/media/video/videodev.c @@ -37,7 +37,6 @@ #include #include #include -#include #include #include diff --git a/drivers/mmc/mmc_block.c b/drivers/mmc/mmc_block.c index 2e5218b8d4a..46138f4e208 100644 --- a/drivers/mmc/mmc_block.c +++ b/drivers/mmc/mmc_block.c @@ -27,7 +27,6 @@ #include #include #include -#include #include #include diff --git a/drivers/net/ppp_generic.c b/drivers/net/ppp_generic.c index 16a59cafb8d..425ff5b117f 100644 --- a/drivers/net/ppp_generic.c +++ b/drivers/net/ppp_generic.c @@ -28,7 +28,6 @@ #include #include #include -#include #include #include #include diff --git a/drivers/net/wan/cosa.c b/drivers/net/wan/cosa.c index 3d1e2b6700a..be5e33814cb 100644 --- a/drivers/net/wan/cosa.c +++ b/drivers/net/wan/cosa.c @@ -85,7 +85,6 @@ #include #include #include -#include #include #include #include diff --git a/drivers/s390/block/dasd_int.h b/drivers/s390/block/dasd_int.h index d4b13e300a7..1893797f2bf 100644 --- a/drivers/s390/block/dasd_int.h +++ b/drivers/s390/block/dasd_int.h @@ -54,7 +54,6 @@ #include #include #include -#include #include #include #include diff --git a/drivers/s390/block/xpram.c b/drivers/s390/block/xpram.c index 15aad01bba0..f57a6c8816b 100644 --- a/drivers/s390/block/xpram.c +++ b/drivers/s390/block/xpram.c @@ -36,7 +36,6 @@ #include /* HDIO_GETGEO */ #include #include -#include #include #define XPRAM_NAME "xpram" diff --git a/drivers/sbus/char/bpp.c b/drivers/sbus/char/bpp.c index 03ff5a3b121..385f4f76831 100644 --- a/drivers/sbus/char/bpp.c +++ b/drivers/sbus/char/bpp.c @@ -20,7 +20,6 @@ #include #include #include -#include #include #include diff --git a/drivers/sbus/char/vfc.h b/drivers/sbus/char/vfc.h index 8045cd5e7cb..63941a259b9 100644 --- a/drivers/sbus/char/vfc.h +++ b/drivers/sbus/char/vfc.h @@ -1,8 +1,6 @@ #ifndef _LINUX_VFC_H_ #define _LINUX_VFC_H_ -#include - /* * The control register for the vfc is at offset 0x4000 * The first field ram bank is located at offset 0x5000 diff --git a/drivers/telephony/phonedev.c b/drivers/telephony/phonedev.c index bc5b1d245df..e41f49afd0f 100644 --- a/drivers/telephony/phonedev.c +++ b/drivers/telephony/phonedev.c @@ -28,7 +28,6 @@ #include #include -#include #include #define PHONE_NUM_DEVICES 256 diff --git a/drivers/video/fbmem.c b/drivers/video/fbmem.c index afa56a887fa..a171daab0ad 100644 --- a/drivers/video/fbmem.c +++ b/drivers/video/fbmem.c @@ -32,7 +32,6 @@ #ifdef CONFIG_KMOD #include #endif -#include #include #include #include diff --git a/fs/block_dev.c b/fs/block_dev.c index 028d9fb9c2d..ddb305eebf9 100644 --- a/fs/block_dev.c +++ b/fs/block_dev.c @@ -12,7 +12,6 @@ #include #include #include -#include #include #include #include diff --git a/fs/char_dev.c b/fs/char_dev.c index f3418f7a6e9..97986635b64 100644 --- a/fs/char_dev.c +++ b/fs/char_dev.c @@ -14,7 +14,6 @@ #include #include #include -#include #include #include diff --git a/fs/coda/psdev.c b/fs/coda/psdev.c index aaf6462a942..803aacf0d49 100644 --- a/fs/coda/psdev.c +++ b/fs/coda/psdev.c @@ -28,7 +28,6 @@ #include #include #include -#include #include #include #include diff --git a/fs/partitions/check.c b/fs/partitions/check.c index 202f50acea8..839634026eb 100644 --- a/fs/partitions/check.c +++ b/fs/partitions/check.c @@ -18,7 +18,6 @@ #include #include #include -#include #include "check.h" diff --git a/include/asm-ppc/ocp.h b/include/asm-ppc/ocp.h index 3be5d760ffc..16dbc7d1745 100644 --- a/include/asm-ppc/ocp.h +++ b/include/asm-ppc/ocp.h @@ -26,7 +26,6 @@ #include #include -#include #include #include diff --git a/include/linux/devfs_fs_kernel.h b/include/linux/devfs_fs_kernel.h deleted file mode 100644 index 2036d2e9098..00000000000 --- a/include/linux/devfs_fs_kernel.h +++ /dev/null @@ -1,9 +0,0 @@ -#ifndef _LINUX_DEVFS_FS_KERNEL_H -#define _LINUX_DEVFS_FS_KERNEL_H - -#include -#include -#include -#include - -#endif /* _LINUX_DEVFS_FS_KERNEL_H */ diff --git a/include/linux/fb.h b/include/linux/fb.h index 07a08e92bc7..b45928f5c63 100644 --- a/include/linux/fb.h +++ b/include/linux/fb.h @@ -380,7 +380,6 @@ struct fb_cursor { #include #include #include -#include #include #include #include diff --git a/init/do_mounts.h b/init/do_mounts.h index 66548efe4c3..e7f2e7fa066 100644 --- a/init/do_mounts.h +++ b/init/do_mounts.h @@ -1,6 +1,5 @@ #include #include -#include #include #include #include diff --git a/init/main.c b/init/main.c index 80af1a52485..acbb0b74913 100644 --- a/init/main.c +++ b/init/main.c @@ -15,7 +15,6 @@ #include #include #include -#include #include #include #include diff --git a/mm/shmem.c b/mm/shmem.c index a4161abb9e7..355904712a8 100644 --- a/mm/shmem.c +++ b/mm/shmem.c @@ -26,7 +26,6 @@ #include #include #include -#include #include #include #include diff --git a/mm/tiny-shmem.c b/mm/tiny-shmem.c index 11ab99b8c61..5f2cbf0f153 100644 --- a/mm/tiny-shmem.c +++ b/mm/tiny-shmem.c @@ -12,7 +12,6 @@ #include #include -#include #include #include #include diff --git a/net/irda/irnet/irnet.h b/net/irda/irnet/irnet.h index e4fe1e80029..ad6caba02a7 100644 --- a/net/irda/irnet/irnet.h +++ b/net/irda/irnet/irnet.h @@ -244,7 +244,6 @@ #include #include #include -#include #include #include #include -- cgit v1.2.3-70-g09d2 From 96192ff1a9d0c6ef365d21667080259d83ea2f5b Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Mon, 20 Jun 2005 21:15:16 -0700 Subject: [PATCH] devfs: Remove the miscdevice devfs_name field as it's no longer needed Also fixes all drivers that set this field. Signed-off-by: Greg Kroah-Hartman --- arch/i386/kernel/microcode.c | 1 - arch/sh/kernel/cpu/sh4/sq.c | 1 - drivers/block/pktcdvd.c | 1 - drivers/char/misc.c | 4 ---- drivers/char/mmtimer.c | 1 - drivers/md/dm-ioctl.c | 1 - drivers/media/radio/miropcm20-rds.c | 1 - drivers/net/tun.c | 1 - drivers/s390/char/monreader.c | 1 - drivers/s390/crypto/z90main.c | 1 - include/linux/miscdevice.h | 1 - 11 files changed, 14 deletions(-) (limited to 'include') diff --git a/arch/i386/kernel/microcode.c b/arch/i386/kernel/microcode.c index 0a865889b2a..40b44cc0d14 100644 --- a/arch/i386/kernel/microcode.c +++ b/arch/i386/kernel/microcode.c @@ -493,7 +493,6 @@ static struct file_operations microcode_fops = { static struct miscdevice microcode_dev = { .minor = MICROCODE_MINOR, .name = "microcode", - .devfs_name = "cpu/microcode", .fops = µcode_fops, }; diff --git a/arch/sh/kernel/cpu/sh4/sq.c b/arch/sh/kernel/cpu/sh4/sq.c index 8437ea7430f..83a4f91bce5 100644 --- a/arch/sh/kernel/cpu/sh4/sq.c +++ b/arch/sh/kernel/cpu/sh4/sq.c @@ -417,7 +417,6 @@ static struct file_operations sq_fops = { static struct miscdevice sq_dev = { .minor = STORE_QUEUE_MINOR, .name = "sq", - .devfs_name = "cpu/sq", .fops = &sq_fops, }; diff --git a/drivers/block/pktcdvd.c b/drivers/block/pktcdvd.c index a04f60693c3..3e4cce5e473 100644 --- a/drivers/block/pktcdvd.c +++ b/drivers/block/pktcdvd.c @@ -2612,7 +2612,6 @@ static struct file_operations pkt_ctl_fops = { static struct miscdevice pkt_misc = { .minor = MISC_DYNAMIC_MINOR, .name = "pktcdvd", - .devfs_name = "pktcdvd/control", .fops = &pkt_ctl_fops }; diff --git a/drivers/char/misc.c b/drivers/char/misc.c index f1fdd84c5dc..dfe1cede391 100644 --- a/drivers/char/misc.c +++ b/drivers/char/misc.c @@ -227,10 +227,6 @@ int misc_register(struct miscdevice * misc) if (misc->minor < DYNAMIC_MINORS) misc_minors[misc->minor >> 3] |= 1 << (misc->minor & 7); - if (misc->devfs_name[0] == '\0') { - snprintf(misc->devfs_name, sizeof(misc->devfs_name), - "misc/%s", misc->name); - } dev = MKDEV(MISC_MAJOR, misc->minor); misc->class = class_device_create(misc_class, NULL, dev, misc->dev, diff --git a/drivers/char/mmtimer.c b/drivers/char/mmtimer.c index f25168a9676..95e8122b806 100644 --- a/drivers/char/mmtimer.c +++ b/drivers/char/mmtimer.c @@ -693,7 +693,6 @@ static int __init mmtimer_init(void) return -1; } - strcpy(mmtimer_miscdev.devfs_name, MMTIMER_NAME); if (misc_register(&mmtimer_miscdev)) { printk(KERN_ERR "%s: failed to register device\n", MMTIMER_NAME); diff --git a/drivers/md/dm-ioctl.c b/drivers/md/dm-ioctl.c index 3ad60caea77..d13bb15a8a0 100644 --- a/drivers/md/dm-ioctl.c +++ b/drivers/md/dm-ioctl.c @@ -1472,7 +1472,6 @@ static struct file_operations _ctl_fops = { static struct miscdevice _dm_misc = { .minor = MISC_DYNAMIC_MINOR, .name = DM_NAME, - .devfs_name = "mapper/control", .fops = &_ctl_fops }; diff --git a/drivers/media/radio/miropcm20-rds.c b/drivers/media/radio/miropcm20-rds.c index 87b37b7691d..c1b1db65e66 100644 --- a/drivers/media/radio/miropcm20-rds.c +++ b/drivers/media/radio/miropcm20-rds.c @@ -115,7 +115,6 @@ static struct file_operations rds_fops = { static struct miscdevice rds_miscdev = { .minor = MISC_DYNAMIC_MINOR, .name = "radiotext", - .devfs_name = "v4l/rds/radiotext", .fops = &rds_fops, }; diff --git a/drivers/net/tun.c b/drivers/net/tun.c index 6c62d5c8826..732c5edec2e 100644 --- a/drivers/net/tun.c +++ b/drivers/net/tun.c @@ -780,7 +780,6 @@ static struct miscdevice tun_miscdev = { .minor = TUN_MINOR, .name = "tun", .fops = &tun_fops, - .devfs_name = "net/tun", }; /* ethtool interface */ diff --git a/drivers/s390/char/monreader.c b/drivers/s390/char/monreader.c index fb7bc9e5eeb..a138b151009 100644 --- a/drivers/s390/char/monreader.c +++ b/drivers/s390/char/monreader.c @@ -586,7 +586,6 @@ static struct file_operations mon_fops = { static struct miscdevice mon_dev = { .name = "monreader", - .devfs_name = "monreader", .fops = &mon_fops, .minor = MISC_DYNAMIC_MINOR, }; diff --git a/drivers/s390/crypto/z90main.c b/drivers/s390/crypto/z90main.c index 982acc7303e..b2f20ab8431 100644 --- a/drivers/s390/crypto/z90main.c +++ b/drivers/s390/crypto/z90main.c @@ -411,7 +411,6 @@ static struct miscdevice z90crypt_misc_device = { .minor = Z90CRYPT_MINOR, .name = DEV_NAME, .fops = &z90crypt_fops, - .devfs_name = DEV_NAME }; /** diff --git a/include/linux/miscdevice.h b/include/linux/miscdevice.h index 5b584dafb5a..b03cfb91e22 100644 --- a/include/linux/miscdevice.h +++ b/include/linux/miscdevice.h @@ -40,7 +40,6 @@ struct miscdevice { struct list_head list; struct device *dev; struct class_device *class; - char devfs_name[64]; }; extern int misc_register(struct miscdevice * misc); -- cgit v1.2.3-70-g09d2 From ce7b0f46bbf4bff8daab2dd3d878b9e72a623d09 Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Mon, 20 Jun 2005 21:15:16 -0700 Subject: [PATCH] devfs: Remove the gendisk devfs_name field as it's no longer needed And remove the now unneeded number field. Also fixes all drivers that set these fields. Signed-off-by: Greg Kroah-Hartman --- arch/um/drivers/ubd_kern.c | 8 ++------ drivers/block/DAC960.c | 1 - drivers/block/acsi.c | 5 +---- drivers/block/cciss.c | 1 - drivers/block/cpqarray.c | 2 -- drivers/block/loop.c | 1 - drivers/block/nbd.c | 1 - drivers/block/ps2esdi.c | 1 - drivers/block/rd.c | 1 - drivers/block/swim3.c | 1 - drivers/block/sx8.c | 1 - drivers/block/ub.c | 2 -- drivers/block/umem.c | 1 - drivers/block/viodasd.c | 2 -- drivers/block/xd.c | 1 - drivers/block/z2ram.c | 1 - drivers/cdrom/aztcd.c | 1 - drivers/cdrom/gscd.c | 1 - drivers/cdrom/optcd.c | 1 - drivers/cdrom/sbpcd.c | 1 - drivers/cdrom/sjcd.c | 1 - drivers/cdrom/sonycd535.c | 1 - drivers/cdrom/viocd.c | 2 -- drivers/md/md.c | 7 ++----- drivers/message/i2o/i2o_block.c | 1 - drivers/mmc/mmc_block.c | 1 - drivers/s390/block/dasd_genhd.c | 2 -- drivers/s390/block/xpram.c | 1 - include/linux/genhd.h | 2 -- 29 files changed, 5 insertions(+), 47 deletions(-) (limited to 'include') diff --git a/arch/um/drivers/ubd_kern.c b/arch/um/drivers/ubd_kern.c index 5856a02fff5..c3e205b6a4c 100644 --- a/arch/um/drivers/ubd_kern.c +++ b/arch/um/drivers/ubd_kern.c @@ -637,14 +637,10 @@ static int ubd_new_disk(int major, u64 size, int unit, disk->first_minor = unit << UBD_SHIFT; disk->fops = &ubd_blops; set_capacity(disk, size / 512); - if(major == MAJOR_NR){ + if(major == MAJOR_NR) sprintf(disk->disk_name, "ubd%c", 'a' + unit); - sprintf(disk->devfs_name, "ubd/disc%d", unit); - } - else { + else sprintf(disk->disk_name, "ubd_fake%d", unit); - sprintf(disk->devfs_name, "ubd_fake/disc%d", unit); - } /* sysfs register (not for ide fake devices) */ if (major == MAJOR_NR) { diff --git a/drivers/block/DAC960.c b/drivers/block/DAC960.c index dd8a1501142..50ca1aa4ee3 100644 --- a/drivers/block/DAC960.c +++ b/drivers/block/DAC960.c @@ -2530,7 +2530,6 @@ static boolean DAC960_RegisterBlockDevice(DAC960_Controller_T *Controller) blk_queue_max_sectors(RequestQueue, Controller->MaxBlocksPerCommand); disk->queue = RequestQueue; sprintf(disk->disk_name, "rd/c%dd%d", Controller->ControllerNumber, n); - sprintf(disk->devfs_name, "rd/host%d/target%d", Controller->ControllerNumber, n); disk->major = MajorNumber; disk->first_minor = n << DAC960_MaxPartitionsBits; disk->fops = &DAC960_BlockDeviceOperations; diff --git a/drivers/block/acsi.c b/drivers/block/acsi.c index 196c0ec9cd5..a317e430395 100644 --- a/drivers/block/acsi.c +++ b/drivers/block/acsi.c @@ -1732,13 +1732,10 @@ int acsi_init( void ) struct gendisk *disk = acsi_gendisk[i]; sprintf(disk->disk_name, "ad%c", 'a'+i); aip = &acsi_info[NDevices]; - sprintf(disk->devfs_name, "ad/target%d/lun%d", aip->target, aip->lun); disk->major = ACSI_MAJOR; disk->first_minor = i << 4; - if (acsi_info[i].type != HARDDISK) { + if (acsi_info[i].type != HARDDISK) disk->minors = 1; - strcat(disk->devfs_name, "/disc"); - } disk->fops = &acsi_fops; disk->private_data = &acsi_info[i]; set_capacity(disk, acsi_info[i].size); diff --git a/drivers/block/cciss.c b/drivers/block/cciss.c index 39b0f53186e..05fb08312c0 100644 --- a/drivers/block/cciss.c +++ b/drivers/block/cciss.c @@ -3248,7 +3248,6 @@ static int __devinit cciss_init_one(struct pci_dev *pdev, q->queuedata = hba[i]; sprintf(disk->disk_name, "cciss/c%dd%d", i, j); - sprintf(disk->devfs_name, "cciss/host%d/target%d", i, j); disk->major = hba[i]->major; disk->first_minor = j << NWD_SHIFT; disk->fops = &cciss_fops; diff --git a/drivers/block/cpqarray.c b/drivers/block/cpqarray.c index f96e6c03593..bfd245df0a8 100644 --- a/drivers/block/cpqarray.c +++ b/drivers/block/cpqarray.c @@ -1805,8 +1805,6 @@ static void getgeometry(int ctlr) } - sprintf(disk->devfs_name, "ida/c%dd%d", ctlr, log_unit); - info_p->phys_drives = sense_config_buf->ctlr_phys_drv; info_p->drv_assign_map diff --git a/drivers/block/loop.c b/drivers/block/loop.c index dbf905ce024..9983e720021 100644 --- a/drivers/block/loop.c +++ b/drivers/block/loop.c @@ -1293,7 +1293,6 @@ static int __init loop_init(void) disk->first_minor = i; disk->fops = &lo_fops; sprintf(disk->disk_name, "loop%d", i); - sprintf(disk->devfs_name, "loop/%d", i); disk->private_data = lo; disk->queue = lo->lo_queue; } diff --git a/drivers/block/nbd.c b/drivers/block/nbd.c index b164ee0ac79..39662f0c9cc 100644 --- a/drivers/block/nbd.c +++ b/drivers/block/nbd.c @@ -657,7 +657,6 @@ static int __init nbd_init(void) disk->private_data = &nbd_dev[i]; disk->flags |= GENHD_FL_SUPPRESS_PARTITION_INFO; sprintf(disk->disk_name, "nbd%d", i); - sprintf(disk->devfs_name, "nbd/%d", i); set_capacity(disk, 0x7ffffc00ULL << 1); /* 2 TB */ add_disk(disk); } diff --git a/drivers/block/ps2esdi.c b/drivers/block/ps2esdi.c index bea75f2cb21..a729013a397 100644 --- a/drivers/block/ps2esdi.c +++ b/drivers/block/ps2esdi.c @@ -421,7 +421,6 @@ static int __init ps2esdi_geninit(void) disk->major = PS2ESDI_MAJOR; disk->first_minor = i<<6; sprintf(disk->disk_name, "ed%c", 'a'+i); - sprintf(disk->devfs_name, "ed/target%d", i); disk->fops = &ps2esdi_fops; ps2esdi_gendisk[i] = disk; } diff --git a/drivers/block/rd.c b/drivers/block/rd.c index 536be2de989..e64f26ec614 100644 --- a/drivers/block/rd.c +++ b/drivers/block/rd.c @@ -457,7 +457,6 @@ static int __init rd_init(void) disk->queue = rd_queue[i]; disk->flags |= GENHD_FL_SUPPRESS_PARTITION_INFO; sprintf(disk->disk_name, "ram%d", i); - sprintf(disk->devfs_name, "rd/%d", i); set_capacity(disk, rd_size * 2); add_disk(rd_disks[i]); } diff --git a/drivers/block/swim3.c b/drivers/block/swim3.c index ef557b1043d..628877945f9 100644 --- a/drivers/block/swim3.c +++ b/drivers/block/swim3.c @@ -1061,7 +1061,6 @@ int swim3_init(void) disk->queue = swim3_queue; disk->flags |= GENHD_FL_REMOVABLE; sprintf(disk->disk_name, "fd%d", i); - sprintf(disk->devfs_name, "floppy/%d", i); set_capacity(disk, 2880); add_disk(disk); } diff --git a/drivers/block/sx8.c b/drivers/block/sx8.c index 3a481455156..fbddeb4b04a 100644 --- a/drivers/block/sx8.c +++ b/drivers/block/sx8.c @@ -1509,7 +1509,6 @@ static int carm_init_disks(struct carm_host *host) port->disk = disk; sprintf(disk->disk_name, DRV_NAME "/%u", (unsigned int) (host->id * CARM_MAX_PORTS) + i); - sprintf(disk->devfs_name, DRV_NAME "/%u_%u", host->id, i); disk->major = host->major; disk->first_minor = i * CARM_MINORS_PER_MAJOR; disk->fops = &carm_bd_ops; diff --git a/drivers/block/ub.c b/drivers/block/ub.c index 87ec6458bd1..d62b49fbf10 100644 --- a/drivers/block/ub.c +++ b/drivers/block/ub.c @@ -28,7 +28,6 @@ #include #define DRV_NAME "ub" -#define DEVFS_NAME DRV_NAME #define UB_MAJOR 180 @@ -2290,7 +2289,6 @@ static int ub_probe_lun(struct ub_dev *sc, int lnum) goto err_diskalloc; sprintf(disk->disk_name, DRV_NAME "%c", lun->id + 'a'); - sprintf(disk->devfs_name, DEVFS_NAME "/%c", lun->id + 'a'); disk->major = UB_MAJOR; disk->first_minor = lun->id * UB_PARTS_PER_LUN; disk->fops = &ub_bd_fops; diff --git a/drivers/block/umem.c b/drivers/block/umem.c index f7d4c65a7b8..585197b95af 100644 --- a/drivers/block/umem.c +++ b/drivers/block/umem.c @@ -1192,7 +1192,6 @@ static int __init mm_init(void) for (i = 0; i < num_cards; i++) { struct gendisk *disk = mm_gendisk[i]; sprintf(disk->disk_name, "umem%c", 'a'+i); - sprintf(disk->devfs_name, "umem/card%d", i); spin_lock_init(&cards[i].lock); disk->major = major_nr; disk->first_minor = i << MM_SHIFT; diff --git a/drivers/block/viodasd.c b/drivers/block/viodasd.c index b0df4f5ab97..5cc7e671b03 100644 --- a/drivers/block/viodasd.c +++ b/drivers/block/viodasd.c @@ -523,8 +523,6 @@ retry: else snprintf(g->disk_name, sizeof(g->disk_name), VIOD_GENHD_NAME "%c", 'a' + (dev_no % 26)); - snprintf(g->devfs_name, sizeof(g->devfs_name), - "%s%d", VIOD_GENHD_DEVFS_NAME, dev_no); g->fops = &viodasd_fops; g->queue = q; g->private_data = d; diff --git a/drivers/block/xd.c b/drivers/block/xd.c index cbce7c5e944..e828e4cbd3e 100644 --- a/drivers/block/xd.c +++ b/drivers/block/xd.c @@ -215,7 +215,6 @@ static int __init xd_init(void) disk->major = XT_DISK_MAJOR; disk->first_minor = i<<6; sprintf(disk->disk_name, "xd%c", i+'a'); - sprintf(disk->devfs_name, "xd/target%d", i); disk->fops = &xd_fops; disk->private_data = p; disk->queue = xd_queue; diff --git a/drivers/block/z2ram.c b/drivers/block/z2ram.c index bb5e8d665a2..82ddbdd7bd4 100644 --- a/drivers/block/z2ram.c +++ b/drivers/block/z2ram.c @@ -354,7 +354,6 @@ z2_init(void) z2ram_gendisk->first_minor = 0; z2ram_gendisk->fops = &z2_fops; sprintf(z2ram_gendisk->disk_name, "z2ram"); - strcpy(z2ram_gendisk->devfs_name, z2ram_gendisk->disk_name); z2ram_gendisk->queue = z2_queue; add_disk(z2ram_gendisk); diff --git a/drivers/cdrom/aztcd.c b/drivers/cdrom/aztcd.c index ec004897b63..ec469497c10 100644 --- a/drivers/cdrom/aztcd.c +++ b/drivers/cdrom/aztcd.c @@ -1918,7 +1918,6 @@ static int __init aztcd_init(void) azt_disk->first_minor = 0; azt_disk->fops = &azt_fops; sprintf(azt_disk->disk_name, "aztcd"); - sprintf(azt_disk->devfs_name, "aztcd"); azt_disk->queue = azt_queue; add_disk(azt_disk); azt_invalidate_buffers(); diff --git a/drivers/cdrom/gscd.c b/drivers/cdrom/gscd.c index ad5464ab99b..b6ee50a2916 100644 --- a/drivers/cdrom/gscd.c +++ b/drivers/cdrom/gscd.c @@ -955,7 +955,6 @@ static int __init gscd_init(void) gscd_disk->first_minor = 0; gscd_disk->fops = &gscd_fops; sprintf(gscd_disk->disk_name, "gscd"); - sprintf(gscd_disk->devfs_name, "gscd"); if (register_blkdev(MAJOR_NR, "gscd")) { ret = -EIO; diff --git a/drivers/cdrom/optcd.c b/drivers/cdrom/optcd.c index 0b0eab4f40f..25032d7edc5 100644 --- a/drivers/cdrom/optcd.c +++ b/drivers/cdrom/optcd.c @@ -2033,7 +2033,6 @@ static int __init optcd_init(void) optcd_disk->first_minor = 0; optcd_disk->fops = &opt_fops; sprintf(optcd_disk->disk_name, "optcd"); - sprintf(optcd_disk->devfs_name, "optcd"); if (!request_region(optcd_port, 4, "optcd")) { printk(KERN_ERR "optcd: conflict, I/O port 0x%x already used\n", diff --git a/drivers/cdrom/sbpcd.c b/drivers/cdrom/sbpcd.c index a75b363b19c..2fc966c65a0 100644 --- a/drivers/cdrom/sbpcd.c +++ b/drivers/cdrom/sbpcd.c @@ -5868,7 +5868,6 @@ int __init sbpcd_init(void) disk->fops = &sbpcd_bdops; strcpy(disk->disk_name, sbpcd_infop->name); disk->flags = GENHD_FL_CD; - sprintf(disk->devfs_name, "sbp/c0t%d", p->drv_id); p->disk = disk; if (register_cdrom(sbpcd_infop)) { diff --git a/drivers/cdrom/sjcd.c b/drivers/cdrom/sjcd.c index 74b1cadbf16..bf5aef4e555 100644 --- a/drivers/cdrom/sjcd.c +++ b/drivers/cdrom/sjcd.c @@ -1695,7 +1695,6 @@ static int __init sjcd_init(void) sjcd_disk->first_minor = 0, sjcd_disk->fops = &sjcd_fops, sprintf(sjcd_disk->disk_name, "sjcd"); - sprintf(sjcd_disk->devfs_name, "sjcd"); if (!request_region(sjcd_base, 4,"sjcd")) { printk diff --git a/drivers/cdrom/sonycd535.c b/drivers/cdrom/sonycd535.c index e6565992643..8f7cc452af8 100644 --- a/drivers/cdrom/sonycd535.c +++ b/drivers/cdrom/sonycd535.c @@ -1589,7 +1589,6 @@ static int __init sony535_init(void) cdu_disk->first_minor = 0; cdu_disk->fops = &cdu_fops; sprintf(cdu_disk->disk_name, "cdu"); - sprintf(cdu_disk->devfs_name, "cdu535"); if (!request_region(sony535_cd_base_io, 4, CDU535_HANDLE)) { printk(KERN_WARNING"sonycd535: Unable to request region 0x%x\n", diff --git a/drivers/cdrom/viocd.c b/drivers/cdrom/viocd.c index af6b3bfd169..f6559708117 100644 --- a/drivers/cdrom/viocd.c +++ b/drivers/cdrom/viocd.c @@ -688,8 +688,6 @@ static int viocd_probe(struct vio_dev *vdev, const struct vio_device_id *id) gendisk->first_minor = deviceno; strncpy(gendisk->disk_name, c->name, sizeof(gendisk->disk_name)); - snprintf(gendisk->devfs_name, sizeof(gendisk->devfs_name), - VIOCD_DEVICE_DEVFS "%d", deviceno); blk_queue_max_hw_segments(q, 1); blk_queue_max_phys_segments(q, 1); blk_queue_max_sectors(q, 4096 / 512); diff --git a/drivers/md/md.c b/drivers/md/md.c index 95c275d1931..2ec1b3520a0 100644 --- a/drivers/md/md.c +++ b/drivers/md/md.c @@ -2910,13 +2910,10 @@ static struct kobject *md_probe(dev_t dev, int *part, void *data) } disk->major = MAJOR(dev); disk->first_minor = unit << shift; - if (partitioned) { + if (partitioned) sprintf(disk->disk_name, "md_d%d", unit); - sprintf(disk->devfs_name, "md/d%d", unit); - } else { + else sprintf(disk->disk_name, "md%d", unit); - sprintf(disk->devfs_name, "md/%d", unit); - } disk->fops = &md_fops; disk->private_data = mddev; disk->queue = mddev->queue; diff --git a/drivers/message/i2o/i2o_block.c b/drivers/message/i2o/i2o_block.c index 7d4c5497785..1ddc2fb429d 100644 --- a/drivers/message/i2o/i2o_block.c +++ b/drivers/message/i2o/i2o_block.c @@ -1089,7 +1089,6 @@ static int i2o_block_probe(struct device *dev) gd = i2o_blk_dev->gd; gd->first_minor = unit << 4; sprintf(gd->disk_name, "i2o/hd%c", 'a' + unit); - sprintf(gd->devfs_name, "i2o/hd%c", 'a' + unit); gd->driverfs_dev = &i2o_dev->device; /* setup request queue */ diff --git a/drivers/mmc/mmc_block.c b/drivers/mmc/mmc_block.c index 46138f4e208..115cc21094b 100644 --- a/drivers/mmc/mmc_block.c +++ b/drivers/mmc/mmc_block.c @@ -408,7 +408,6 @@ static struct mmc_blk_data *mmc_blk_alloc(struct mmc_card *card) */ sprintf(md->disk->disk_name, "mmcblk%d", devidx); - sprintf(md->disk->devfs_name, "mmc/blk%d", devidx); blk_queue_hardsect_size(md->queue.queue, 1 << md->block_bits); diff --git a/drivers/s390/block/dasd_genhd.c b/drivers/s390/block/dasd_genhd.c index fce2835e7d1..61ffde718a7 100644 --- a/drivers/s390/block/dasd_genhd.c +++ b/drivers/s390/block/dasd_genhd.c @@ -68,8 +68,6 @@ dasd_gendisk_alloc(struct dasd_device *device) } len += sprintf(gdp->disk_name + len, "%c", 'a'+(device->devindex%26)); - sprintf(gdp->devfs_name, "dasd/%s", device->cdev->dev.bus_id); - if (device->features & DASD_FEATURE_READONLY) set_disk_ro(gdp, 1); gdp->private_data = device; diff --git a/drivers/s390/block/xpram.c b/drivers/s390/block/xpram.c index f57a6c8816b..4c1e56b9b98 100644 --- a/drivers/s390/block/xpram.c +++ b/drivers/s390/block/xpram.c @@ -466,7 +466,6 @@ static int __init xpram_setup_blkdev(void) disk->private_data = &xpram_devices[i]; disk->queue = xpram_queue; sprintf(disk->disk_name, "slram%d", i); - sprintf(disk->devfs_name, "slram/%d", i); set_capacity(disk, xpram_sizes[i] << 1); add_disk(disk); } diff --git a/include/linux/genhd.h b/include/linux/genhd.h index 3498a0c6818..e4af57e87c1 100644 --- a/include/linux/genhd.h +++ b/include/linux/genhd.h @@ -112,8 +112,6 @@ struct gendisk { sector_t capacity; int flags; - char devfs_name[64]; /* devfs crap */ - int number; /* more of the same */ struct device *driverfs_dev; struct kobject kobj; struct kobject *holder_dir; -- cgit v1.2.3-70-g09d2 From 5e483075a1f9862a65af648d1d939499969c7b5d Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Mon, 20 Jun 2005 21:15:16 -0700 Subject: [PATCH] devfs: Remove the videodevice devfs_name field as it's no longer needed Also fixes all drivers that set this field. Signed-off-by: Greg Kroah-Hartman --- drivers/media/video/videodev.c | 3 +-- include/media/v4l2-dev.h | 1 - 2 files changed, 1 insertion(+), 3 deletions(-) (limited to 'include') diff --git a/drivers/media/video/videodev.c b/drivers/media/video/videodev.c index a802a7e0df7..b26ebaff226 100644 --- a/drivers/media/video/videodev.c +++ b/drivers/media/video/videodev.c @@ -1570,8 +1570,7 @@ int video_register_device(struct video_device *vfd, int type, int nr) vfd->class_dev.dev = vfd->dev; vfd->class_dev.class = &video_class; vfd->class_dev.devt = MKDEV(VIDEO_MAJOR, vfd->minor); - sprintf(vfd->devfs_name, "%s%d", name_base, i - base); - strlcpy(vfd->class_dev.class_id, vfd->devfs_name, BUS_ID_SIZE); + sprintf(vfd->class_dev.class_id, "%s%d", name_base, i - base); class_device_register(&vfd->class_dev); class_device_create_file(&vfd->class_dev, &class_device_attr_name); diff --git a/include/media/v4l2-dev.h b/include/media/v4l2-dev.h index a1b473190e6..62dae1a8c44 100644 --- a/include/media/v4l2-dev.h +++ b/include/media/v4l2-dev.h @@ -314,7 +314,6 @@ void *priv; /* for videodev.c intenal usage -- please don't touch */ int users; /* video_exclusive_{open|close} ... */ struct mutex lock; /* ... helper function uses these */ - char devfs_name[64]; /* devfs */ struct class_device class_dev; /* sysfs */ }; -- cgit v1.2.3-70-g09d2 From f4eaa37017a5a68f67ef86729508022c13fb8e6d Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Mon, 20 Jun 2005 21:15:16 -0700 Subject: [PATCH] devfs: Remove the tty_driver devfs_name field as it's no longer needed Also fixes all drivers that set this field. Signed-off-by: Greg Kroah-Hartman --- drivers/char/cyclades.c | 1 - drivers/char/epca.c | 1 - drivers/char/esp.c | 1 - drivers/char/hvc_console.c | 1 - drivers/char/hvcs.c | 1 - drivers/char/hvsi.c | 1 - drivers/char/ip2/ip2main.c | 1 - drivers/char/isicom.c | 1 - drivers/char/moxa.c | 1 - drivers/char/pty.c | 2 -- drivers/char/riscom8.c | 1 - drivers/char/rocket.c | 1 - drivers/char/serial167.c | 1 - drivers/char/stallion.c | 1 - drivers/char/viocons.c | 1 - drivers/char/vme_scc.c | 1 - drivers/char/vt.c | 1 - drivers/isdn/capi/capi.c | 1 - drivers/isdn/gigaset/bas-gigaset.c | 4 +--- drivers/isdn/gigaset/common.c | 4 +--- drivers/isdn/gigaset/gigaset.h | 3 +-- drivers/isdn/gigaset/interface.c | 4 +--- drivers/isdn/gigaset/usb-gigaset.c | 4 +--- drivers/isdn/i4l/isdn_tty.c | 1 - drivers/s390/char/tty3270.c | 1 - drivers/tc/zs.c | 1 - drivers/usb/class/cdc-acm.c | 1 - drivers/usb/gadget/serial.c | 1 - drivers/usb/serial/usb-serial.c | 1 - include/linux/tty_driver.h | 1 - net/bluetooth/rfcomm/tty.c | 1 - net/irda/ircomm/ircomm_tty.c | 1 - 32 files changed, 5 insertions(+), 42 deletions(-) (limited to 'include') diff --git a/drivers/char/cyclades.c b/drivers/char/cyclades.c index 122e7a72a4e..2657eeba7da 100644 --- a/drivers/char/cyclades.c +++ b/drivers/char/cyclades.c @@ -5250,7 +5250,6 @@ cy_init(void) cy_serial_driver->owner = THIS_MODULE; cy_serial_driver->driver_name = "cyclades"; cy_serial_driver->name = "ttyC"; - cy_serial_driver->devfs_name = "tts/C"; cy_serial_driver->major = CYCLADES_MAJOR; cy_serial_driver->minor_start = 0; cy_serial_driver->type = TTY_DRIVER_TYPE_SERIAL; diff --git a/drivers/char/epca.c b/drivers/char/epca.c index 9cad8501d62..881d7426fda 100644 --- a/drivers/char/epca.c +++ b/drivers/char/epca.c @@ -1232,7 +1232,6 @@ static int __init pc_init(void) pc_driver->owner = THIS_MODULE; pc_driver->name = "ttyD"; - pc_driver->devfs_name = "tts/D"; pc_driver->major = DIGI_MAJOR; pc_driver->minor_start = 0; pc_driver->type = TTY_DRIVER_TYPE_SERIAL; diff --git a/drivers/char/esp.c b/drivers/char/esp.c index 922174d527a..9827d170ca1 100644 --- a/drivers/char/esp.c +++ b/drivers/char/esp.c @@ -2449,7 +2449,6 @@ static int __init espserial_init(void) esp_driver->owner = THIS_MODULE; esp_driver->name = "ttyP"; - esp_driver->devfs_name = "tts/P"; esp_driver->major = ESP_IN_MAJOR; esp_driver->minor_start = 0; esp_driver->type = TTY_DRIVER_TYPE_SERIAL; diff --git a/drivers/char/hvc_console.c b/drivers/char/hvc_console.c index a5c6a9d7ff0..6e380aecea6 100644 --- a/drivers/char/hvc_console.c +++ b/drivers/char/hvc_console.c @@ -820,7 +820,6 @@ int __init hvc_init(void) return -ENOMEM; drv->owner = THIS_MODULE; - drv->devfs_name = "hvc/"; drv->driver_name = "hvc"; drv->name = "hvc"; drv->major = HVC_MAJOR; diff --git a/drivers/char/hvcs.c b/drivers/char/hvcs.c index 8d97b391129..2019175a222 100644 --- a/drivers/char/hvcs.c +++ b/drivers/char/hvcs.c @@ -1364,7 +1364,6 @@ static int __init hvcs_module_init(void) hvcs_tty_driver->driver_name = hvcs_driver_name; hvcs_tty_driver->name = hvcs_device_node; - hvcs_tty_driver->devfs_name = hvcs_device_node; /* * We'll let the system assign us a major number, indicated by leaving diff --git a/drivers/char/hvsi.c b/drivers/char/hvsi.c index a0370ed752c..7b04eb15320 100644 --- a/drivers/char/hvsi.c +++ b/drivers/char/hvsi.c @@ -1154,7 +1154,6 @@ static int __init hvsi_init(void) return -ENOMEM; hvsi_driver->owner = THIS_MODULE; - hvsi_driver->devfs_name = "hvsi/"; hvsi_driver->driver_name = "hvsi"; hvsi_driver->name = "hvsi"; hvsi_driver->major = HVSI_MAJOR; diff --git a/drivers/char/ip2/ip2main.c b/drivers/char/ip2/ip2main.c index a273eef69a8..2db55d1f040 100644 --- a/drivers/char/ip2/ip2main.c +++ b/drivers/char/ip2/ip2main.c @@ -671,7 +671,6 @@ ip2_loadmain(int *iop, int *irqp, unsigned char *firmware, int firmsize) ip2_tty_driver->owner = THIS_MODULE; ip2_tty_driver->name = "ttyF"; - ip2_tty_driver->devfs_name = "tts/F"; ip2_tty_driver->driver_name = pcDriver_name; ip2_tty_driver->major = IP2_TTY_MAJOR; ip2_tty_driver->minor_start = 0; diff --git a/drivers/char/isicom.c b/drivers/char/isicom.c index efaaa1937ab..478bf4d7d06 100644 --- a/drivers/char/isicom.c +++ b/drivers/char/isicom.c @@ -1581,7 +1581,6 @@ static int __devinit isicom_register_tty_driver(void) isicom_normal->owner = THIS_MODULE; isicom_normal->name = "ttyM"; - isicom_normal->devfs_name = "isicom/"; isicom_normal->major = ISICOM_NMAJOR; isicom_normal->minor_start = 0; isicom_normal->type = TTY_DRIVER_TYPE_SERIAL; diff --git a/drivers/char/moxa.c b/drivers/char/moxa.c index f43c2e04ead..afa2cda5e35 100644 --- a/drivers/char/moxa.c +++ b/drivers/char/moxa.c @@ -342,7 +342,6 @@ static int __init moxa_init(void) init_MUTEX(&moxaBuffSem); moxaDriver->owner = THIS_MODULE; moxaDriver->name = "ttyMX"; - moxaDriver->devfs_name = "tts/a"; moxaDriver->major = ttymajor; moxaDriver->minor_start = 0; moxaDriver->type = TTY_DRIVER_TYPE_SERIAL; diff --git a/drivers/char/pty.c b/drivers/char/pty.c index 345f8b8d57c..d3a5092402c 100644 --- a/drivers/char/pty.c +++ b/drivers/char/pty.c @@ -264,7 +264,6 @@ static void __init legacy_pty_init(void) pty_driver->owner = THIS_MODULE; pty_driver->driver_name = "pty_master"; pty_driver->name = "pty"; - pty_driver->devfs_name = "pty/m"; pty_driver->major = PTY_MASTER_MAJOR; pty_driver->minor_start = 0; pty_driver->type = TTY_DRIVER_TYPE_PTY; @@ -282,7 +281,6 @@ static void __init legacy_pty_init(void) pty_slave_driver->owner = THIS_MODULE; pty_slave_driver->driver_name = "pty_slave"; pty_slave_driver->name = "ttyp"; - pty_slave_driver->devfs_name = "pty/s"; pty_slave_driver->major = PTY_SLAVE_MAJOR; pty_slave_driver->minor_start = 0; pty_slave_driver->type = TTY_DRIVER_TYPE_PTY; diff --git a/drivers/char/riscom8.c b/drivers/char/riscom8.c index 657c0d88f48..c84c3c3f10c 100644 --- a/drivers/char/riscom8.c +++ b/drivers/char/riscom8.c @@ -1634,7 +1634,6 @@ static inline int rc_init_drivers(void) memset(IRQ_to_board, 0, sizeof(IRQ_to_board)); riscom_driver->owner = THIS_MODULE; riscom_driver->name = "ttyL"; - riscom_driver->devfs_name = "tts/L"; riscom_driver->major = RISCOM8_NORMAL_MAJOR; riscom_driver->type = TTY_DRIVER_TYPE_SERIAL; riscom_driver->subtype = SERIAL_TYPE_NORMAL; diff --git a/drivers/char/rocket.c b/drivers/char/rocket.c index 0708c5164c8..ed511ec9b57 100644 --- a/drivers/char/rocket.c +++ b/drivers/char/rocket.c @@ -2427,7 +2427,6 @@ static int __init rp_init(void) rocket_driver->owner = THIS_MODULE; rocket_driver->flags = TTY_DRIVER_NO_DEVFS; - rocket_driver->devfs_name = "tts/R"; rocket_driver->name = "ttyR"; rocket_driver->driver_name = "Comtrol RocketPort"; rocket_driver->major = TTY_ROCKET_MAJOR; diff --git a/drivers/char/serial167.c b/drivers/char/serial167.c index 037c940ac71..c851eeaa406 100644 --- a/drivers/char/serial167.c +++ b/drivers/char/serial167.c @@ -2235,7 +2235,6 @@ scrn[1] = '\0'; /* Initialize the tty_driver structure */ cy_serial_driver->owner = THIS_MODULE; - cy_serial_driver->devfs_name = "tts/"; cy_serial_driver->name = "ttyS"; cy_serial_driver->major = TTY_MAJOR; cy_serial_driver->minor_start = 64; diff --git a/drivers/char/stallion.c b/drivers/char/stallion.c index 36df18cd96d..38fdd4da242 100644 --- a/drivers/char/stallion.c +++ b/drivers/char/stallion.c @@ -3075,7 +3075,6 @@ static int __init stl_init(void) stl_serial->owner = THIS_MODULE; stl_serial->driver_name = stl_drvname; stl_serial->name = "ttyE"; - stl_serial->devfs_name = "tts/E"; stl_serial->major = STL_SERIALMAJOR; stl_serial->minor_start = 0; stl_serial->type = TTY_DRIVER_TYPE_SERIAL; diff --git a/drivers/char/viocons.c b/drivers/char/viocons.c index 4e536038874..07f5ce4b28e 100644 --- a/drivers/char/viocons.c +++ b/drivers/char/viocons.c @@ -1152,7 +1152,6 @@ static int __init viocons_init2(void) viotty_driver = alloc_tty_driver(VTTY_PORTS); viotty_driver->owner = THIS_MODULE; viotty_driver->driver_name = "vioconsole"; - viotty_driver->devfs_name = "vcs/"; viotty_driver->name = "tty"; viotty_driver->name_base = 1; viotty_driver->major = TTY_MAJOR; diff --git a/drivers/char/vme_scc.c b/drivers/char/vme_scc.c index fd00822ac14..fe99fc1aba4 100644 --- a/drivers/char/vme_scc.c +++ b/drivers/char/vme_scc.c @@ -147,7 +147,6 @@ static int scc_init_drivers(void) scc_driver->owner = THIS_MODULE; scc_driver->driver_name = "scc"; scc_driver->name = "ttyS"; - scc_driver->devfs_name = "tts/"; scc_driver->major = TTY_MAJOR; scc_driver->minor_start = SCC_MINOR_BASE; scc_driver->type = TTY_DRIVER_TYPE_SERIAL; diff --git a/drivers/char/vt.c b/drivers/char/vt.c index 42afe7cea04..d6f65032649 100644 --- a/drivers/char/vt.c +++ b/drivers/char/vt.c @@ -2662,7 +2662,6 @@ int __init vty_init(void) if (!console_driver) panic("Couldn't allocate console driver\n"); console_driver->owner = THIS_MODULE; - console_driver->devfs_name = "vc/"; console_driver->name = "tty"; console_driver->name_base = 1; console_driver->major = TTY_MAJOR; diff --git a/drivers/isdn/capi/capi.c b/drivers/isdn/capi/capi.c index 334402e96e1..a518ec53102 100644 --- a/drivers/isdn/capi/capi.c +++ b/drivers/isdn/capi/capi.c @@ -1336,7 +1336,6 @@ static int capinc_tty_init(void) drv->owner = THIS_MODULE; drv->driver_name = "capi_nc"; - drv->devfs_name = "capi/"; drv->name = "capi"; drv->major = capi_ttymajor; drv->minor_start = 0; diff --git a/drivers/isdn/gigaset/bas-gigaset.c b/drivers/isdn/gigaset/bas-gigaset.c index 8a45715dd4c..3845defd490 100644 --- a/drivers/isdn/gigaset/bas-gigaset.c +++ b/drivers/isdn/gigaset/bas-gigaset.c @@ -41,7 +41,6 @@ MODULE_PARM_DESC(cidmode, "Call-ID mode"); #define GIGASET_MINORS 1 #define GIGASET_MINOR 16 #define GIGASET_MODULENAME "bas_gigaset" -#define GIGASET_DEVFSNAME "gig/bas/" #define GIGASET_DEVNAME "ttyGB" /* length limit according to Siemens 3070usb-protokoll.doc ch. 2.1 */ @@ -2349,8 +2348,7 @@ static int __init bas_gigaset_init(void) /* allocate memory for our driver state and intialize it */ if ((driver = gigaset_initdriver(GIGASET_MINOR, GIGASET_MINORS, GIGASET_MODULENAME, GIGASET_DEVNAME, - GIGASET_DEVFSNAME, &gigops, - THIS_MODULE)) == NULL) + &gigops, THIS_MODULE)) == NULL) goto error; /* allocate memory for our device state and intialize it */ diff --git a/drivers/isdn/gigaset/common.c b/drivers/isdn/gigaset/common.c index acb7e265678..578e6797ef0 100644 --- a/drivers/isdn/gigaset/common.c +++ b/drivers/isdn/gigaset/common.c @@ -1092,14 +1092,12 @@ EXPORT_SYMBOL_GPL(gigaset_freedriver); * minors Number of minors this driver can handle * procname Name of the driver * devname Name of the device files (prefix without minor number) - * devfsname Devfs name of the device files without %d * return value: * Pointer to the gigaset_driver structure on success, NULL on failure. */ struct gigaset_driver *gigaset_initdriver(unsigned minor, unsigned minors, const char *procname, const char *devname, - const char *devfsname, const struct gigaset_ops *ops, struct module *owner) { @@ -1139,7 +1137,7 @@ struct gigaset_driver *gigaset_initdriver(unsigned minor, unsigned minors, drv->cs[i].minor_index = i; } - gigaset_if_initdriver(drv, procname, devname, devfsname); + gigaset_if_initdriver(drv, procname, devname); spin_lock_irqsave(&driver_lock, flags); list_add(&drv->list, &drivers); diff --git a/drivers/isdn/gigaset/gigaset.h b/drivers/isdn/gigaset/gigaset.h index 8d63d822104..1ca3bfdef51 100644 --- a/drivers/isdn/gigaset/gigaset.h +++ b/drivers/isdn/gigaset/gigaset.h @@ -769,7 +769,6 @@ void gigaset_block_channels(struct cardstate *cs); struct gigaset_driver *gigaset_initdriver(unsigned minor, unsigned minors, const char *procname, const char *devname, - const char *devfsname, const struct gigaset_ops *ops, struct module *owner); @@ -892,7 +891,7 @@ int gigaset_fill_inbuf(struct inbuf_t *inbuf, const unsigned char *src, /* initialize interface */ void gigaset_if_initdriver(struct gigaset_driver *drv, const char *procname, - const char *devname, const char *devfsname); + const char *devname); /* release interface */ void gigaset_if_freedriver(struct gigaset_driver *drv); /* add minor */ diff --git a/drivers/isdn/gigaset/interface.c b/drivers/isdn/gigaset/interface.c index 74fd234956c..9f8628bbdf5 100644 --- a/drivers/isdn/gigaset/interface.c +++ b/drivers/isdn/gigaset/interface.c @@ -673,10 +673,9 @@ EXPORT_SYMBOL_GPL(gigaset_if_receive); * drv Driver * procname Name of the driver (e.g. for /proc/tty/drivers) * devname Name of the device files (prefix without minor number) - * devfsname Devfs name of the device files without %d */ void gigaset_if_initdriver(struct gigaset_driver *drv, const char *procname, - const char *devname, const char *devfsname) + const char *devname) { unsigned minors = drv->minors; int ret; @@ -700,7 +699,6 @@ void gigaset_if_initdriver(struct gigaset_driver *drv, const char *procname, tty->num = drv->minors; tty->owner = THIS_MODULE; - tty->devfs_name = devfsname; tty->init_termios = tty_std_termios; //FIXME tty->init_termios.c_cflag = B9600 | CS8 | CREAD | HUPCL | CLOCAL; //FIXME diff --git a/drivers/isdn/gigaset/usb-gigaset.c b/drivers/isdn/gigaset/usb-gigaset.c index d86ab68114b..6e05d9d4a51 100644 --- a/drivers/isdn/gigaset/usb-gigaset.c +++ b/drivers/isdn/gigaset/usb-gigaset.c @@ -41,7 +41,6 @@ MODULE_PARM_DESC(cidmode, "Call-ID mode"); #define GIGASET_MINORS 1 #define GIGASET_MINOR 8 #define GIGASET_MODULENAME "usb_gigaset" -#define GIGASET_DEVFSNAME "gig/usb/" #define GIGASET_DEVNAME "ttyGU" #define IF_WRITEBUF 2000 //FIXME // WAKEUP_CHARS: 256 @@ -896,8 +895,7 @@ static int __init usb_gigaset_init(void) /* allocate memory for our driver state and intialize it */ if ((driver = gigaset_initdriver(GIGASET_MINOR, GIGASET_MINORS, GIGASET_MODULENAME, GIGASET_DEVNAME, - GIGASET_DEVFSNAME, &ops, - THIS_MODULE)) == NULL) + &ops, THIS_MODULE)) == NULL) goto error; /* allocate memory for our device state and intialize it */ diff --git a/drivers/isdn/i4l/isdn_tty.c b/drivers/isdn/i4l/isdn_tty.c index 2ac90242d26..e7f0f15c735 100644 --- a/drivers/isdn/i4l/isdn_tty.c +++ b/drivers/isdn/i4l/isdn_tty.c @@ -1890,7 +1890,6 @@ isdn_tty_modem_init(void) if (!m->tty_modem) return -ENOMEM; m->tty_modem->name = "ttyI"; - m->tty_modem->devfs_name = "isdn/ttyI"; m->tty_modem->major = ISDN_TTY_MAJOR; m->tty_modem->minor_start = 0; m->tty_modem->type = TTY_DRIVER_TYPE_SERIAL; diff --git a/drivers/s390/char/tty3270.c b/drivers/s390/char/tty3270.c index 9a141776873..43c8c9f5f30 100644 --- a/drivers/s390/char/tty3270.c +++ b/drivers/s390/char/tty3270.c @@ -1785,7 +1785,6 @@ tty3270_init(void) * proc_entry, set_termios, flush_buffer, set_ldisc, write_proc */ driver->owner = THIS_MODULE; - driver->devfs_name = "ttyTUB/"; driver->driver_name = "ttyTUB"; driver->name = "ttyTUB"; driver->major = IBM_TTY3270_MAJOR; diff --git a/drivers/tc/zs.c b/drivers/tc/zs.c index 2dffa8e303b..582d926cd00 100644 --- a/drivers/tc/zs.c +++ b/drivers/tc/zs.c @@ -1745,7 +1745,6 @@ int __init zs_init(void) /* Not all of this is exactly right for us. */ serial_driver->owner = THIS_MODULE; - serial_driver->devfs_name = "tts/"; serial_driver->name = "ttyS"; serial_driver->major = TTY_MAJOR; serial_driver->minor_start = 64; diff --git a/drivers/usb/class/cdc-acm.c b/drivers/usb/class/cdc-acm.c index d41dc67ba4c..a0b9eddd98e 100644 --- a/drivers/usb/class/cdc-acm.c +++ b/drivers/usb/class/cdc-acm.c @@ -1145,7 +1145,6 @@ static int __init acm_init(void) acm_tty_driver->owner = THIS_MODULE, acm_tty_driver->driver_name = "acm", acm_tty_driver->name = "ttyACM", - acm_tty_driver->devfs_name = "usb/acm/", acm_tty_driver->major = ACM_TTY_MAJOR, acm_tty_driver->minor_start = 0, acm_tty_driver->type = TTY_DRIVER_TYPE_SERIAL, diff --git a/drivers/usb/gadget/serial.c b/drivers/usb/gadget/serial.c index 9d6e1d29552..2c8e5c10eb9 100644 --- a/drivers/usb/gadget/serial.c +++ b/drivers/usb/gadget/serial.c @@ -588,7 +588,6 @@ static int __init gs_module_init(void) gs_tty_driver->owner = THIS_MODULE; gs_tty_driver->driver_name = GS_SHORT_NAME; gs_tty_driver->name = "ttygs"; - gs_tty_driver->devfs_name = "usb/ttygs/"; gs_tty_driver->major = GS_MAJOR; gs_tty_driver->minor_start = GS_MINOR_START; gs_tty_driver->type = TTY_DRIVER_TYPE_SERIAL; diff --git a/drivers/usb/serial/usb-serial.c b/drivers/usb/serial/usb-serial.c index a30135c7cfe..db7f54b0d85 100644 --- a/drivers/usb/serial/usb-serial.c +++ b/drivers/usb/serial/usb-serial.c @@ -1059,7 +1059,6 @@ static int __init usb_serial_init(void) usb_serial_tty_driver->owner = THIS_MODULE; usb_serial_tty_driver->driver_name = "usbserial"; - usb_serial_tty_driver->devfs_name = "usb/tts/"; usb_serial_tty_driver->name = "ttyUSB"; usb_serial_tty_driver->major = SERIAL_TTY_MAJOR; usb_serial_tty_driver->minor_start = 0; diff --git a/include/linux/tty_driver.h b/include/linux/tty_driver.h index b368b296d03..0bd31f0830f 100644 --- a/include/linux/tty_driver.h +++ b/include/linux/tty_driver.h @@ -157,7 +157,6 @@ struct tty_driver { struct cdev cdev; struct module *owner; const char *driver_name; - const char *devfs_name; const char *name; int name_base; /* offset of printed name */ int major; /* major device number */ diff --git a/net/bluetooth/rfcomm/tty.c b/net/bluetooth/rfcomm/tty.c index 74368f79ee5..afb3c524db3 100644 --- a/net/bluetooth/rfcomm/tty.c +++ b/net/bluetooth/rfcomm/tty.c @@ -1025,7 +1025,6 @@ int rfcomm_init_ttys(void) rfcomm_tty_driver->owner = THIS_MODULE; rfcomm_tty_driver->driver_name = "rfcomm"; - rfcomm_tty_driver->devfs_name = "bluetooth/rfcomm/"; rfcomm_tty_driver->name = "rfcomm"; rfcomm_tty_driver->major = RFCOMM_TTY_MAJOR; rfcomm_tty_driver->minor_start = RFCOMM_TTY_MINOR; diff --git a/net/irda/ircomm/ircomm_tty.c b/net/irda/ircomm/ircomm_tty.c index 6f20b4206e0..b592c4bc333 100644 --- a/net/irda/ircomm/ircomm_tty.c +++ b/net/irda/ircomm/ircomm_tty.c @@ -124,7 +124,6 @@ static int __init ircomm_tty_init(void) driver->owner = THIS_MODULE; driver->driver_name = "ircomm"; driver->name = "ircomm"; - driver->devfs_name = "ircomm"; driver->major = IRCOMM_TTY_MAJOR; driver->minor_start = IRCOMM_TTY_MINOR; driver->type = TTY_DRIVER_TYPE_SERIAL; -- cgit v1.2.3-70-g09d2 From 331b831983f9d706f4a40d08a996d5c2c7a6ea7b Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Mon, 20 Jun 2005 21:15:16 -0700 Subject: [PATCH] devfs: Rename TTY_DRIVER_NO_DEVFS to TTY_DRIVER_DYNAMIC_DEV I've always found this flag confusing. Now that devfs is no longer around, it has been renamed, and the documentation for when this flag should be used has been updated. Also fixes all drivers that use this flag. Signed-off-by: Greg Kroah-Hartman --- arch/cris/arch-v10/kernel/debugport.c | 2 +- arch/cris/arch-v32/kernel/debugport.c | 2 +- arch/ppc/4xx_io/serial_sicc.c | 2 +- drivers/char/ip2/ip2main.c | 2 +- drivers/char/pty.c | 4 ++-- drivers/char/rocket.c | 4 ++-- drivers/char/tty_io.c | 8 ++++---- drivers/isdn/gigaset/interface.c | 2 +- drivers/isdn/i4l/isdn_tty.c | 2 +- drivers/s390/char/tty3270.c | 2 +- drivers/serial/crisv10.c | 2 +- drivers/serial/serial_core.c | 2 +- drivers/tc/zs.c | 2 +- drivers/usb/class/cdc-acm.c | 2 +- drivers/usb/gadget/serial.c | 2 +- drivers/usb/serial/usb-serial.c | 2 +- include/linux/tty_driver.h | 13 ++++++++++--- net/bluetooth/rfcomm/tty.c | 2 +- 18 files changed, 32 insertions(+), 25 deletions(-) (limited to 'include') diff --git a/arch/cris/arch-v10/kernel/debugport.c b/arch/cris/arch-v10/kernel/debugport.c index f3a85b77c17..dde813e1629 100644 --- a/arch/cris/arch-v10/kernel/debugport.c +++ b/arch/cris/arch-v10/kernel/debugport.c @@ -541,7 +541,7 @@ init_dummy_console(void) dummy_driver.init_termios = tty_std_termios; dummy_driver.init_termios.c_cflag = B115200 | CS8 | CREAD | HUPCL | CLOCAL; /* is normally B9600 default... */ - dummy_driver.flags = TTY_DRIVER_REAL_RAW | TTY_DRIVER_NO_DEVFS; + dummy_driver.flags = TTY_DRIVER_REAL_RAW | TTY_DRIVER_DYNAMIC_DEV; dummy_driver.open = dummy_open; dummy_driver.close = dummy_close; diff --git a/arch/cris/arch-v32/kernel/debugport.c b/arch/cris/arch-v32/kernel/debugport.c index ffc1ebf2dfe..3dc587e6201 100644 --- a/arch/cris/arch-v32/kernel/debugport.c +++ b/arch/cris/arch-v32/kernel/debugport.c @@ -353,7 +353,7 @@ init_dummy_console(void) dummy_driver.init_termios = tty_std_termios; dummy_driver.init_termios.c_cflag = B115200 | CS8 | CREAD | HUPCL | CLOCAL; /* is normally B9600 default... */ - dummy_driver.flags = TTY_DRIVER_REAL_RAW | TTY_DRIVER_NO_DEVFS; + dummy_driver.flags = TTY_DRIVER_REAL_RAW | TTY_DRIVER_DYNAMIC_DEV; dummy_driver.open = dummy_open; dummy_driver.close = dummy_close; diff --git a/arch/ppc/4xx_io/serial_sicc.c b/arch/ppc/4xx_io/serial_sicc.c index 98b25fa0049..1e113d0f59b 100644 --- a/arch/ppc/4xx_io/serial_sicc.c +++ b/arch/ppc/4xx_io/serial_sicc.c @@ -1758,7 +1758,7 @@ int __init siccuart_init(void) siccnormal_driver->subtype = SERIAL_TYPE_NORMAL; siccnormal_driver->init_termios = tty_std_termios; siccnormal_driver->init_termios.c_cflag = B9600 | CS8 | CREAD | HUPCL | CLOCAL; - siccnormal_driver->flags = TTY_DRIVER_REAL_RAW | TTY_DRIVER_NO_DEVFS; + siccnormal_driver->flags = TTY_DRIVER_REAL_RAW | TTY_DRIVER_DYNAMIC_DEV; tty_set_operations(siccnormal_driver, &sicc_ops); if (tty_register_driver(siccnormal_driver)) diff --git a/drivers/char/ip2/ip2main.c b/drivers/char/ip2/ip2main.c index 2db55d1f040..8619542766c 100644 --- a/drivers/char/ip2/ip2main.c +++ b/drivers/char/ip2/ip2main.c @@ -678,7 +678,7 @@ ip2_loadmain(int *iop, int *irqp, unsigned char *firmware, int firmsize) ip2_tty_driver->subtype = SERIAL_TYPE_NORMAL; ip2_tty_driver->init_termios = tty_std_termios; ip2_tty_driver->init_termios.c_cflag = B9600|CS8|CREAD|HUPCL|CLOCAL; - ip2_tty_driver->flags = TTY_DRIVER_REAL_RAW | TTY_DRIVER_NO_DEVFS; + ip2_tty_driver->flags = TTY_DRIVER_REAL_RAW | TTY_DRIVER_DYNAMIC_DEV; tty_set_operations(ip2_tty_driver, &ip2_ops); ip2trace (ITRC_NO_PORT, ITRC_INIT, 3, 0 ); diff --git a/drivers/char/pty.c b/drivers/char/pty.c index d3a5092402c..f373d0155a7 100644 --- a/drivers/char/pty.c +++ b/drivers/char/pty.c @@ -368,7 +368,7 @@ static void __init unix98_pty_init(void) ptm_driver->init_termios.c_cflag = B38400 | CS8 | CREAD; ptm_driver->init_termios.c_lflag = 0; ptm_driver->flags = TTY_DRIVER_RESET_TERMIOS | TTY_DRIVER_REAL_RAW | - TTY_DRIVER_NO_DEVFS | TTY_DRIVER_DEVPTS_MEM; + TTY_DRIVER_DYNAMIC_DEV | TTY_DRIVER_DEVPTS_MEM; ptm_driver->other = pts_driver; tty_set_operations(ptm_driver, &pty_ops); ptm_driver->ioctl = pty_unix98_ioctl; @@ -383,7 +383,7 @@ static void __init unix98_pty_init(void) pts_driver->init_termios = tty_std_termios; pts_driver->init_termios.c_cflag = B38400 | CS8 | CREAD; pts_driver->flags = TTY_DRIVER_RESET_TERMIOS | TTY_DRIVER_REAL_RAW | - TTY_DRIVER_NO_DEVFS | TTY_DRIVER_DEVPTS_MEM; + TTY_DRIVER_DYNAMIC_DEV | TTY_DRIVER_DEVPTS_MEM; pts_driver->other = ptm_driver; tty_set_operations(pts_driver, &pty_ops); diff --git a/drivers/char/rocket.c b/drivers/char/rocket.c index ed511ec9b57..0ac13188132 100644 --- a/drivers/char/rocket.c +++ b/drivers/char/rocket.c @@ -2426,7 +2426,7 @@ static int __init rp_init(void) */ rocket_driver->owner = THIS_MODULE; - rocket_driver->flags = TTY_DRIVER_NO_DEVFS; + rocket_driver->flags = TTY_DRIVER_DYNAMIC_DEV; rocket_driver->name = "ttyR"; rocket_driver->driver_name = "Comtrol RocketPort"; rocket_driver->major = TTY_ROCKET_MAJOR; @@ -2437,7 +2437,7 @@ static int __init rp_init(void) rocket_driver->init_termios.c_cflag = B9600 | CS8 | CREAD | HUPCL | CLOCAL; #ifdef ROCKET_SOFT_FLOW - rocket_driver->flags |= TTY_DRIVER_REAL_RAW | TTY_DRIVER_NO_DEVFS; + rocket_driver->flags |= TTY_DRIVER_REAL_RAW | TTY_DRIVER_DYNAMIC_DEV; #endif tty_set_operations(rocket_driver, &rocket_ops); diff --git a/drivers/char/tty_io.c b/drivers/char/tty_io.c index ebe1158e6c6..6c79ff3030f 100644 --- a/drivers/char/tty_io.c +++ b/drivers/char/tty_io.c @@ -2963,8 +2963,8 @@ static struct class *tty_class; * Returns a pointer to the class device (or ERR_PTR(-EFOO) on error). * * This call is required to be made to register an individual tty device if - * the tty driver's flags have the TTY_DRIVER_NO_DEVFS bit set. If that - * bit is not set, this function should not be called. + * the tty driver's flags have the TTY_DRIVER_DYNAMIC_DEV bit set. If that + * bit is not set, this function should not be called by a tty driver. */ struct class_device *tty_register_device(struct tty_driver *driver, unsigned index, struct device *device) @@ -3117,7 +3117,7 @@ int tty_register_driver(struct tty_driver *driver) list_add(&driver->tty_drivers, &tty_drivers); - if ( !(driver->flags & TTY_DRIVER_NO_DEVFS) ) { + if ( !(driver->flags & TTY_DRIVER_DYNAMIC_DEV) ) { for(i = 0; i < driver->num; i++) tty_register_device(driver, i, NULL); } @@ -3160,7 +3160,7 @@ int tty_unregister_driver(struct tty_driver *driver) driver->termios_locked[i] = NULL; kfree(tp); } - if (!(driver->flags & TTY_DRIVER_NO_DEVFS)) + if (!(driver->flags & TTY_DRIVER_DYNAMIC_DEV)) tty_unregister_device(driver, i); } p = driver->ttys; diff --git a/drivers/isdn/gigaset/interface.c b/drivers/isdn/gigaset/interface.c index 9f8628bbdf5..bd2e4267528 100644 --- a/drivers/isdn/gigaset/interface.c +++ b/drivers/isdn/gigaset/interface.c @@ -691,7 +691,7 @@ void gigaset_if_initdriver(struct gigaset_driver *drv, const char *procname, tty->major = GIG_MAJOR, tty->type = TTY_DRIVER_TYPE_SERIAL, tty->subtype = SERIAL_TYPE_NORMAL, - tty->flags = TTY_DRIVER_REAL_RAW | TTY_DRIVER_NO_DEVFS, + tty->flags = TTY_DRIVER_REAL_RAW | TTY_DRIVER_DYNAMIC_DEV; tty->driver_name = procname; tty->name = devname; diff --git a/drivers/isdn/i4l/isdn_tty.c b/drivers/isdn/i4l/isdn_tty.c index e7f0f15c735..77579f82aed 100644 --- a/drivers/isdn/i4l/isdn_tty.c +++ b/drivers/isdn/i4l/isdn_tty.c @@ -1896,7 +1896,7 @@ isdn_tty_modem_init(void) m->tty_modem->subtype = SERIAL_TYPE_NORMAL; m->tty_modem->init_termios = tty_std_termios; m->tty_modem->init_termios.c_cflag = B9600 | CS8 | CREAD | HUPCL | CLOCAL; - m->tty_modem->flags = TTY_DRIVER_REAL_RAW | TTY_DRIVER_NO_DEVFS; + m->tty_modem->flags = TTY_DRIVER_REAL_RAW | TTY_DRIVER_DYNAMIC_DEV; m->tty_modem->driver_name = "isdn_tty"; tty_set_operations(m->tty_modem, &modem_ops); retval = tty_register_driver(m->tty_modem); diff --git a/drivers/s390/char/tty3270.c b/drivers/s390/char/tty3270.c index 43c8c9f5f30..7d26a3e4cb8 100644 --- a/drivers/s390/char/tty3270.c +++ b/drivers/s390/char/tty3270.c @@ -1792,7 +1792,7 @@ tty3270_init(void) driver->type = TTY_DRIVER_TYPE_SYSTEM; driver->subtype = SYSTEM_TYPE_TTY; driver->init_termios = tty_std_termios; - driver->flags = TTY_DRIVER_RESET_TERMIOS | TTY_DRIVER_NO_DEVFS; + driver->flags = TTY_DRIVER_RESET_TERMIOS | TTY_DRIVER_DYNAMIC_DEV; tty_set_operations(driver, &tty3270_ops); ret = tty_register_driver(driver); if (ret) { diff --git a/drivers/serial/crisv10.c b/drivers/serial/crisv10.c index 89700141f87..2364c394998 100644 --- a/drivers/serial/crisv10.c +++ b/drivers/serial/crisv10.c @@ -4884,7 +4884,7 @@ rs_init(void) driver->init_termios = tty_std_termios; driver->init_termios.c_cflag = B115200 | CS8 | CREAD | HUPCL | CLOCAL; /* is normally B9600 default... */ - driver->flags = TTY_DRIVER_REAL_RAW | TTY_DRIVER_NO_DEVFS; + driver->flags = TTY_DRIVER_REAL_RAW | TTY_DRIVER_DYNAMIC_DEV; driver->termios = serial_termios; driver->termios_locked = serial_termios_locked; diff --git a/drivers/serial/serial_core.c b/drivers/serial/serial_core.c index 84d70d85559..7dc1e67b685 100644 --- a/drivers/serial/serial_core.c +++ b/drivers/serial/serial_core.c @@ -2160,7 +2160,7 @@ int uart_register_driver(struct uart_driver *drv) normal->subtype = SERIAL_TYPE_NORMAL; normal->init_termios = tty_std_termios; normal->init_termios.c_cflag = B9600 | CS8 | CREAD | HUPCL | CLOCAL; - normal->flags = TTY_DRIVER_REAL_RAW | TTY_DRIVER_NO_DEVFS; + normal->flags = TTY_DRIVER_REAL_RAW | TTY_DRIVER_DYNAMIC_DEV; normal->driver_state = drv; tty_set_operations(normal, &uart_ops); diff --git a/drivers/tc/zs.c b/drivers/tc/zs.c index 582d926cd00..7f27b356eaf 100644 --- a/drivers/tc/zs.c +++ b/drivers/tc/zs.c @@ -1753,7 +1753,7 @@ int __init zs_init(void) serial_driver->init_termios = tty_std_termios; serial_driver->init_termios.c_cflag = B9600 | CS8 | CREAD | HUPCL | CLOCAL; - serial_driver->flags = TTY_DRIVER_REAL_RAW | TTY_DRIVER_NO_DEVFS; + serial_driver->flags = TTY_DRIVER_REAL_RAW | TTY_DRIVER_DYNAMIC_DEV; tty_set_operations(serial_driver, &serial_ops); if (tty_register_driver(serial_driver)) diff --git a/drivers/usb/class/cdc-acm.c b/drivers/usb/class/cdc-acm.c index a0b9eddd98e..3670d77e912 100644 --- a/drivers/usb/class/cdc-acm.c +++ b/drivers/usb/class/cdc-acm.c @@ -1149,7 +1149,7 @@ static int __init acm_init(void) acm_tty_driver->minor_start = 0, acm_tty_driver->type = TTY_DRIVER_TYPE_SERIAL, acm_tty_driver->subtype = SERIAL_TYPE_NORMAL, - acm_tty_driver->flags = TTY_DRIVER_REAL_RAW | TTY_DRIVER_NO_DEVFS, + acm_tty_driver->flags = TTY_DRIVER_REAL_RAW | TTY_DRIVER_DYNAMIC_DEV; acm_tty_driver->init_termios = tty_std_termios; acm_tty_driver->init_termios.c_cflag = B9600 | CS8 | CREAD | HUPCL | CLOCAL; tty_set_operations(acm_tty_driver, &acm_ops); diff --git a/drivers/usb/gadget/serial.c b/drivers/usb/gadget/serial.c index 2c8e5c10eb9..416acac879d 100644 --- a/drivers/usb/gadget/serial.c +++ b/drivers/usb/gadget/serial.c @@ -592,7 +592,7 @@ static int __init gs_module_init(void) gs_tty_driver->minor_start = GS_MINOR_START; gs_tty_driver->type = TTY_DRIVER_TYPE_SERIAL; gs_tty_driver->subtype = SERIAL_TYPE_NORMAL; - gs_tty_driver->flags = TTY_DRIVER_REAL_RAW | TTY_DRIVER_NO_DEVFS; + gs_tty_driver->flags = TTY_DRIVER_REAL_RAW | TTY_DRIVER_DYNAMIC_DEV; gs_tty_driver->init_termios = tty_std_termios; gs_tty_driver->init_termios.c_cflag = B9600 | CS8 | CREAD | HUPCL | CLOCAL; tty_set_operations(gs_tty_driver, &gs_tty_ops); diff --git a/drivers/usb/serial/usb-serial.c b/drivers/usb/serial/usb-serial.c index db7f54b0d85..f466f89eeb6 100644 --- a/drivers/usb/serial/usb-serial.c +++ b/drivers/usb/serial/usb-serial.c @@ -1064,7 +1064,7 @@ static int __init usb_serial_init(void) usb_serial_tty_driver->minor_start = 0; usb_serial_tty_driver->type = TTY_DRIVER_TYPE_SERIAL; usb_serial_tty_driver->subtype = SERIAL_TYPE_NORMAL; - usb_serial_tty_driver->flags = TTY_DRIVER_REAL_RAW | TTY_DRIVER_NO_DEVFS; + usb_serial_tty_driver->flags = TTY_DRIVER_REAL_RAW | TTY_DRIVER_DYNAMIC_DEV; usb_serial_tty_driver->init_termios = tty_std_termios; usb_serial_tty_driver->init_termios.c_cflag = B9600 | CS8 | CREAD | HUPCL | CLOCAL; tty_set_operations(usb_serial_tty_driver, &serial_ops); diff --git a/include/linux/tty_driver.h b/include/linux/tty_driver.h index 0bd31f0830f..58c961c9e17 100644 --- a/include/linux/tty_driver.h +++ b/include/linux/tty_driver.h @@ -241,8 +241,15 @@ void tty_set_operations(struct tty_driver *driver, struct tty_operations *op); * is also a promise, if the above case is true, not to signal * overruns, either.) * - * TTY_DRIVER_NO_DEVFS --- if set, do not create devfs entries. This - * is only used by tty_register_driver(). + * TTY_DRIVER_DYNAMIC_DEV --- if set, the individual tty devices need + * to be registered with a call to tty_register_driver() when the + * device is found in the system and unregistered with a call to + * tty_unregister_device() so the devices will be show up + * properly in sysfs. If not set, driver->num entries will be + * created by the tty core in sysfs when tty_register_driver() is + * called. This is to be used by drivers that have tty devices + * that can appear and disappear while the main tty driver is + * registered with the tty core. * * TTY_DRIVER_DEVPTS_MEM -- don't use the standard arrays, instead * use dynamic memory keyed through the devpts filesystem. This @@ -251,7 +258,7 @@ void tty_set_operations(struct tty_driver *driver, struct tty_operations *op); #define TTY_DRIVER_INSTALLED 0x0001 #define TTY_DRIVER_RESET_TERMIOS 0x0002 #define TTY_DRIVER_REAL_RAW 0x0004 -#define TTY_DRIVER_NO_DEVFS 0x0008 +#define TTY_DRIVER_DYNAMIC_DEV 0x0008 #define TTY_DRIVER_DEVPTS_MEM 0x0010 /* tty driver types */ diff --git a/net/bluetooth/rfcomm/tty.c b/net/bluetooth/rfcomm/tty.c index afb3c524db3..5fdc3667862 100644 --- a/net/bluetooth/rfcomm/tty.c +++ b/net/bluetooth/rfcomm/tty.c @@ -1030,7 +1030,7 @@ int rfcomm_init_ttys(void) rfcomm_tty_driver->minor_start = RFCOMM_TTY_MINOR; rfcomm_tty_driver->type = TTY_DRIVER_TYPE_SERIAL; rfcomm_tty_driver->subtype = SERIAL_TYPE_NORMAL; - rfcomm_tty_driver->flags = TTY_DRIVER_REAL_RAW | TTY_DRIVER_NO_DEVFS; + rfcomm_tty_driver->flags = TTY_DRIVER_REAL_RAW | TTY_DRIVER_DYNAMIC_DEV; rfcomm_tty_driver->init_termios = tty_std_termios; rfcomm_tty_driver->init_termios.c_cflag = B9600 | CS8 | CREAD | HUPCL | CLOCAL; tty_set_operations(rfcomm_tty_driver, &rfcomm_ops); -- cgit v1.2.3-70-g09d2 From 438bc9c3dec27ab37f0ff78471d0b8b91addd2dd Mon Sep 17 00:00:00 2001 From: Jeff Garzik Date: Mon, 26 Jun 2006 20:52:17 -0400 Subject: [libata] sata_vsc: partially revert a PCI ID-related commit Partially revert 74d0a988d3aa359b6b8a8536c8cb92cce02ca5d5: [PATCH] PCI: Move various PCI IDs to header file libata policy is to avoid use of named PCI device ID constants. These are often single-use constants, which have little value over direct numeric constants save for constant include/linux/pci_ids.h patching/merging headaches. Signed-off-by: Jeff Garzik --- drivers/scsi/sata_vsc.c | 10 +++------- include/linux/pci_ids.h | 1 - 2 files changed, 3 insertions(+), 8 deletions(-) (limited to 'include') diff --git a/drivers/scsi/sata_vsc.c b/drivers/scsi/sata_vsc.c index 01893a0cf1a..616fd9634b4 100644 --- a/drivers/scsi/sata_vsc.c +++ b/drivers/scsi/sata_vsc.c @@ -443,16 +443,12 @@ err_out: } -/* - * Intel 31244 is supposed to be identical. - * Compatibility is untested as of yet. - */ static const struct pci_device_id vsc_sata_pci_tbl[] = { - { PCI_VENDOR_ID_VITESSE, PCI_DEVICE_ID_VITESSE_VSC7174, + { PCI_VENDOR_ID_VITESSE, 0x7174, PCI_ANY_ID, PCI_ANY_ID, 0x10600, 0xFFFFFF, 0 }, - { PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_GD31244, + { PCI_VENDOR_ID_INTEL, 0x3200, PCI_ANY_ID, PCI_ANY_ID, 0x10600, 0xFFFFFF, 0 }, - { } + { } /* terminate list */ }; diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h index c2fd2d19938..6abd2761e63 100644 --- a/include/linux/pci_ids.h +++ b/include/linux/pci_ids.h @@ -2170,7 +2170,6 @@ #define PCI_DEVICE_ID_INTEL_ICH8_4 0x2815 #define PCI_DEVICE_ID_INTEL_ICH8_5 0x283e #define PCI_DEVICE_ID_INTEL_ICH8_6 0x2850 -#define PCI_DEVICE_ID_INTEL_GD31244 0x3200 #define PCI_DEVICE_ID_INTEL_82855PM_HB 0x3340 #define PCI_DEVICE_ID_INTEL_82830_HB 0x3575 #define PCI_DEVICE_ID_INTEL_82830_CGC 0x3577 -- cgit v1.2.3-70-g09d2 From 41542dbe12e34165e586de1e3fe0a245707aa39e Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Fri, 23 Jun 2006 04:18:31 -0700 Subject: [PATCH] libata.h needs scatterlist.h From: Andrew Morton s390: In file included from drivers/scsi/libata-bmdma.c:39: include/linux/libata.h:391: error: field 'sgent' has incomplete type include/linux/libata.h:392: error: field 'pad_sgent' has incomplete type include/linux/libata.h: In function 'ata_sg_is_last': include/linux/libata.h:849: error: arithmetic on pointer to an incomplete type include/linux/libata.h:849: error: arithmetic on pointer to an incomplete type include/linux/libata.h: In function 'ata_qc_next_sg': include/linux/libata.h:869: error: increment of pointer to unknown structure include/linux/libata.h:869: error: arithmetic on pointer to an incomplete type include/linux/libata.h:869: error: arithmetic on pointer to an incomplete type include/linux/libata.h:869: error: arithmetic on pointer to an incomplete type Cc: Jeff Garzik Signed-off-by: Andrew Morton Signed-off-by: Jeff Garzik --- include/linux/libata.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include') diff --git a/include/linux/libata.h b/include/linux/libata.h index 20b1cf527c6..c227dace1d2 100644 --- a/include/linux/libata.h +++ b/include/linux/libata.h @@ -30,6 +30,7 @@ #include #include #include +#include #include #include #include -- cgit v1.2.3-70-g09d2 From 5806db22cffc7557b675d3c9229f327980aee797 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Sat, 24 Jun 2006 20:30:19 +0900 Subject: [PATCH] libata: implement ata_port_max_devices() Implement ata_port_max_devices(). This function returns the number of possible devices on a port. This will be used by new PM implementation. Signed-off-by: Tejun Heo Signed-off-by: Jeff Garzik --- include/linux/libata.h | 14 ++++++++++++++ 1 file changed, 14 insertions(+) (limited to 'include') diff --git a/include/linux/libata.h b/include/linux/libata.h index c227dace1d2..f4284bf8975 100644 --- a/include/linux/libata.h +++ b/include/linux/libata.h @@ -888,6 +888,9 @@ static inline unsigned int ata_tag_internal(unsigned int tag) return tag == ATA_MAX_QUEUE - 1; } +/* + * device helpers + */ static inline unsigned int ata_class_enabled(unsigned int class) { return class == ATA_DEV_ATA || class == ATA_DEV_ATAPI; @@ -918,6 +921,17 @@ static inline unsigned int ata_dev_absent(const struct ata_device *dev) return ata_class_absent(dev->class); } +/* + * port helpers + */ +static inline int ata_port_max_devices(const struct ata_port *ap) +{ + if (ap->flags & ATA_FLAG_SLAVE_POSS) + return 2; + return 1; +} + + static inline u8 ata_chk_status(struct ata_port *ap) { return ap->ops->check_status(ap); -- cgit v1.2.3-70-g09d2 From 353dcf7c89519a2cb9b7699f3bcf70bb685f22e1 Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Sun, 25 Jun 2006 01:36:55 -0700 Subject: [PATCH] ata: add some NVIDIA chipset IDs From: Randy Dunlap Add some nVidia chipset ID's support. http://www.kernel.org/git/?p=linux/kernel/git/bcollins/ubuntu-dapper.git;a=commitdiff;h=b407680553280f9999a20706d5ab2a3be65312c1;hp=ce4cb48010ab2cca537432b5ccb47d4b1fb489e5 Snagged from lkml. Cc: Jeff Garzik Cc: Alan Cox Cc: Tejun Heo Signed-off-by: Andrew Morton Signed-off-by: Jeff Garzik --- drivers/ide/pci/amd74xx.c | 7 +++++-- include/linux/pci_ids.h | 1 + 2 files changed, 6 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/drivers/ide/pci/amd74xx.c b/drivers/ide/pci/amd74xx.c index 6e9dbf4d807..85007cb12c5 100644 --- a/drivers/ide/pci/amd74xx.c +++ b/drivers/ide/pci/amd74xx.c @@ -75,6 +75,7 @@ static struct amd_ide_chip { { PCI_DEVICE_ID_NVIDIA_NFORCE_MCP51_IDE, 0x50, AMD_UDMA_133 }, { PCI_DEVICE_ID_NVIDIA_NFORCE_MCP55_IDE, 0x50, AMD_UDMA_133 }, { PCI_DEVICE_ID_NVIDIA_NFORCE_MCP61_IDE, 0x50, AMD_UDMA_133 }, + { PCI_DEVICE_ID_NVIDIA_NFORCE_MCP65_IDE, 0x50, AMD_UDMA_133 }, { PCI_DEVICE_ID_AMD_CS5536_IDE, 0x40, AMD_UDMA_100 }, { 0 } }; @@ -490,7 +491,8 @@ static ide_pci_device_t amd74xx_chipsets[] __devinitdata = { /* 15 */ DECLARE_NV_DEV("NFORCE-MCP51"), /* 16 */ DECLARE_NV_DEV("NFORCE-MCP55"), /* 17 */ DECLARE_NV_DEV("NFORCE-MCP61"), - /* 18 */ DECLARE_AMD_DEV("AMD5536"), + /* 18 */ DECLARE_NV_DEV("NFORCE-MCP65"), + /* 19 */ DECLARE_AMD_DEV("AMD5536"), }; static int __devinit amd74xx_probe(struct pci_dev *dev, const struct pci_device_id *id) @@ -528,7 +530,8 @@ static struct pci_device_id amd74xx_pci_tbl[] = { { PCI_VENDOR_ID_NVIDIA, PCI_DEVICE_ID_NVIDIA_NFORCE_MCP51_IDE, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 15 }, { PCI_VENDOR_ID_NVIDIA, PCI_DEVICE_ID_NVIDIA_NFORCE_MCP55_IDE, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 16 }, { PCI_VENDOR_ID_NVIDIA, PCI_DEVICE_ID_NVIDIA_NFORCE_MCP61_IDE, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 17 }, - { PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_CS5536_IDE, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 18 }, + { PCI_VENDOR_ID_NVIDIA, PCI_DEVICE_ID_NVIDIA_NFORCE_MCP65_IDE, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 18 }, + { PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_CS5536_IDE, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 19 }, { 0, }, }; MODULE_DEVICE_TABLE(pci, amd74xx_pci_tbl); diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h index 6abd2761e63..9ae6b1a7536 100644 --- a/include/linux/pci_ids.h +++ b/include/linux/pci_ids.h @@ -1202,6 +1202,7 @@ #define PCI_DEVICE_ID_NVIDIA_NVENET_19 0x03EF #define PCI_DEVICE_ID_NVIDIA_NFORCE_MCP61_SATA2 0x03F6 #define PCI_DEVICE_ID_NVIDIA_NFORCE_MCP61_SATA3 0x03F7 +#define PCI_DEVICE_ID_NVIDIA_NFORCE_MCP65_IDE 0x0448 #define PCI_DEVICE_ID_NVIDIA_NVENET_20 0x0450 #define PCI_DEVICE_ID_NVIDIA_NVENET_21 0x0451 #define PCI_DEVICE_ID_NVIDIA_NVENET_22 0x0452 -- cgit v1.2.3-70-g09d2 From 4d6b5aee9d19d1a982d9db0b5ffea20ee94816c6 Mon Sep 17 00:00:00 2001 From: Hans Verkuil Date: Mon, 26 Jun 2006 09:31:18 -0300 Subject: V4L/DVB (4257): Fix 64-bit compile warnings. Signed-off-by: Hans Verkuil Signed-off-by: Mauro Carvalho Chehab --- drivers/media/video/cx2341x.c | 2 +- drivers/media/video/saa7134/saa6752hs.c | 4 ++-- include/media/cx2341x.h | 2 +- 3 files changed, 4 insertions(+), 4 deletions(-) (limited to 'include') diff --git a/drivers/media/video/cx2341x.c b/drivers/media/video/cx2341x.c index 01b22eab572..c98740979b6 100644 --- a/drivers/media/video/cx2341x.c +++ b/drivers/media/video/cx2341x.c @@ -601,7 +601,7 @@ static void cx2341x_calc_audio_properties(struct cx2341x_mpeg_params *params) } int cx2341x_ext_ctrls(struct cx2341x_mpeg_params *params, - struct v4l2_ext_controls *ctrls, int cmd) + struct v4l2_ext_controls *ctrls, unsigned int cmd) { int err = 0; int i; diff --git a/drivers/media/video/saa7134/saa6752hs.c b/drivers/media/video/saa7134/saa6752hs.c index de7b9e6e932..afc8f352b8e 100644 --- a/drivers/media/video/saa7134/saa6752hs.c +++ b/drivers/media/video/saa7134/saa6752hs.c @@ -432,10 +432,10 @@ static void saa6752hs_old_set_params(struct i2c_client* client, } static int handle_ctrl(struct saa6752hs_mpeg_params *params, - struct v4l2_ext_control *ctrl, int cmd) + struct v4l2_ext_control *ctrl, unsigned int cmd) { int old = 0, new; - int set = cmd == VIDIOC_S_EXT_CTRLS; + int set = (cmd == VIDIOC_S_EXT_CTRLS); new = ctrl->value; switch (ctrl->id) { diff --git a/include/media/cx2341x.h b/include/media/cx2341x.h index 074c4008ad5..e1cbf79741b 100644 --- a/include/media/cx2341x.h +++ b/include/media/cx2341x.h @@ -89,7 +89,7 @@ int cx2341x_ctrl_query(struct cx2341x_mpeg_params *params, struct v4l2_queryctrl *qctrl); const char **cx2341x_ctrl_get_menu(u32 id); int cx2341x_ext_ctrls(struct cx2341x_mpeg_params *params, - struct v4l2_ext_controls *ctrls, int cmd); + struct v4l2_ext_controls *ctrls, unsigned int cmd); void cx2341x_fill_defaults(struct cx2341x_mpeg_params *p); void cx2341x_log_status(struct cx2341x_mpeg_params *p, int cardid); -- cgit v1.2.3-70-g09d2 From 99eb44fe941b35b28914881774515005fa2ab7f6 Mon Sep 17 00:00:00 2001 From: Hans Verkuil Date: Mon, 26 Jun 2006 18:24:05 -0300 Subject: V4L/DVB (4259): Pass an explicit log prefix to cx2341x_log_status A card number is not unique enough. Instead, let the caller specify the prefix of the status messages. Signed-off-by: Hans Verkuil Signed-off-by: Mauro Carvalho Chehab --- drivers/media/video/cx2341x.c | 38 +++++++++++++++---------------- drivers/media/video/cx88/cx88-blackbird.c | 7 +++++- drivers/media/video/pvrusb2/pvrusb2-hdw.c | 2 +- include/media/cx2341x.h | 2 +- 4 files changed, 27 insertions(+), 22 deletions(-) (limited to 'include') diff --git a/drivers/media/video/cx2341x.c b/drivers/media/video/cx2341x.c index c98740979b6..65f00fc08fa 100644 --- a/drivers/media/video/cx2341x.c +++ b/drivers/media/video/cx2341x.c @@ -847,22 +847,22 @@ invalid: return ""; } -void cx2341x_log_status(struct cx2341x_mpeg_params *p, int card_id) +void cx2341x_log_status(struct cx2341x_mpeg_params *p, const char *prefix) { int is_mpeg1 = p->video_encoding == V4L2_MPEG_VIDEO_ENCODING_MPEG_1; /* Stream */ - printk(KERN_INFO "cx2341x-%d: Stream: %s\n", - card_id, + printk(KERN_INFO "%s: Stream: %s\n", + prefix, cx2341x_menu_item(p, V4L2_CID_MPEG_STREAM_TYPE)); /* Video */ - printk(KERN_INFO "cx2341x-%d: Video: %dx%d, %d fps\n", - card_id, + printk(KERN_INFO "%s: Video: %dx%d, %d fps\n", + prefix, p->width / (is_mpeg1 ? 2 : 1), p->height / (is_mpeg1 ? 2 : 1), p->is_50hz ? 25 : 30); - printk(KERN_INFO "cx2341x-%d: Video: %s, %s, %s, %d", - card_id, + printk(KERN_INFO "%s: Video: %s, %s, %s, %d", + prefix, cx2341x_menu_item(p, V4L2_CID_MPEG_VIDEO_ENCODING), cx2341x_menu_item(p, V4L2_CID_MPEG_VIDEO_ASPECT), cx2341x_menu_item(p, V4L2_CID_MPEG_VIDEO_BITRATE_MODE), @@ -871,19 +871,19 @@ void cx2341x_log_status(struct cx2341x_mpeg_params *p, int card_id) printk(", Peak %d", p->video_bitrate_peak); } printk("\n"); - printk(KERN_INFO "cx2341x-%d: Video: GOP Size %d, %d B-Frames, %sGOP Closure, %s3:2 Pulldown\n", - card_id, + printk(KERN_INFO "%s: Video: GOP Size %d, %d B-Frames, %sGOP Closure, %s3:2 Pulldown\n", + prefix, p->video_gop_size, p->video_b_frames, p->video_gop_closure ? "" : "No ", p->video_pulldown ? "" : "No "); if (p->video_temporal_decimation) { - printk(KERN_INFO "cx2341x-%d: Video: Temporal Decimation %d\n", - card_id, p->video_temporal_decimation); + printk(KERN_INFO "%s: Video: Temporal Decimation %d\n", + prefix, p->video_temporal_decimation); } /* Audio */ - printk(KERN_INFO "cx2341x-%d: Audio: %s, %s, %s, %s", - card_id, + printk(KERN_INFO "%s: Audio: %s, %s, %s, %s", + prefix, cx2341x_menu_item(p, V4L2_CID_MPEG_AUDIO_SAMPLING_FREQ), cx2341x_menu_item(p, V4L2_CID_MPEG_AUDIO_ENCODING), cx2341x_menu_item(p, V4L2_CID_MPEG_AUDIO_L2_BITRATE), @@ -897,18 +897,18 @@ void cx2341x_log_status(struct cx2341x_mpeg_params *p, int card_id) cx2341x_menu_item(p, V4L2_CID_MPEG_AUDIO_CRC)); /* Encoding filters */ - printk(KERN_INFO "cx2341x-%d: Spatial Filter: %s, Luma %s, Chroma %s, %d\n", - card_id, + printk(KERN_INFO "%s: Spatial Filter: %s, Luma %s, Chroma %s, %d\n", + prefix, cx2341x_menu_item(p, V4L2_CID_MPEG_CX2341X_VIDEO_SPATIAL_FILTER_MODE), cx2341x_menu_item(p, V4L2_CID_MPEG_CX2341X_VIDEO_LUMA_SPATIAL_FILTER_TYPE), cx2341x_menu_item(p, V4L2_CID_MPEG_CX2341X_VIDEO_CHROMA_SPATIAL_FILTER_TYPE), p->video_spatial_filter); - printk(KERN_INFO "cx2341x-%d: Temporal Filter: %s, %d\n", - card_id, + printk(KERN_INFO "%s: Temporal Filter: %s, %d\n", + prefix, cx2341x_menu_item(p, V4L2_CID_MPEG_CX2341X_VIDEO_TEMPORAL_FILTER_MODE), p->video_temporal_filter); - printk(KERN_INFO "cx2341x-%d: Median Filter: %s, Luma [%d, %d], Chroma [%d, %d]\n", - card_id, + printk(KERN_INFO "%s: Median Filter: %s, Luma [%d, %d], Chroma [%d, %d]\n", + prefix, cx2341x_menu_item(p, V4L2_CID_MPEG_CX2341X_VIDEO_MEDIAN_FILTER_TYPE), p->video_luma_median_filter_bottom, p->video_luma_median_filter_top, diff --git a/drivers/media/video/cx88/cx88-blackbird.c b/drivers/media/video/cx88/cx88-blackbird.c index a7893a6eb58..4ff81582ec5 100644 --- a/drivers/media/video/cx88/cx88-blackbird.c +++ b/drivers/media/video/cx88/cx88-blackbird.c @@ -854,13 +854,18 @@ static int mpeg_do_ioctl(struct inode *inode, struct file *file, return 0; } case VIDIOC_LOG_STATUS: + { + char name[32 + 2]; + + snprintf(name, sizeof(name), "%s/2", core->name); printk("%s/2: ============ START LOG STATUS ============\n", core->name); cx88_call_i2c_clients(core, VIDIOC_LOG_STATUS, 0); - cx2341x_log_status(&dev->params, dev->mpeg_dev->minor); + cx2341x_log_status(&dev->params, name); printk("%s/2: ============= END LOG STATUS =============\n", core->name); return 0; + } default: return cx88_do_ioctl(inode, file, 0, dev->core, cmd, arg, mpeg_do_ioctl); diff --git a/drivers/media/video/pvrusb2/pvrusb2-hdw.c b/drivers/media/video/pvrusb2/pvrusb2-hdw.c index 2fd40e061cc..643c471375d 100644 --- a/drivers/media/video/pvrusb2/pvrusb2-hdw.c +++ b/drivers/media/video/pvrusb2/pvrusb2-hdw.c @@ -2407,7 +2407,7 @@ void pvr2_hdw_trigger_module_log(struct pvr2_hdw *hdw) hdw->log_requested = 0; pvr2_i2c_core_sync(hdw); pvr2_trace(PVR2_TRACE_INFO,"cx2341x config:"); - cx2341x_log_status(&hdw->enc_ctl_state,0); + cx2341x_log_status(&hdw->enc_ctl_state, "pvrusb2"); printk(KERN_INFO "pvrusb2: ================== END STATUS CARD #%d ==================\n", nr); } while (0); LOCK_GIVE(hdw->big_lock); } diff --git a/include/media/cx2341x.h b/include/media/cx2341x.h index e1cbf79741b..d91d88f93c8 100644 --- a/include/media/cx2341x.h +++ b/include/media/cx2341x.h @@ -91,7 +91,7 @@ const char **cx2341x_ctrl_get_menu(u32 id); int cx2341x_ext_ctrls(struct cx2341x_mpeg_params *params, struct v4l2_ext_controls *ctrls, unsigned int cmd); void cx2341x_fill_defaults(struct cx2341x_mpeg_params *p); -void cx2341x_log_status(struct cx2341x_mpeg_params *p, int cardid); +void cx2341x_log_status(struct cx2341x_mpeg_params *p, const char *prefix); /* Firmware names */ #define CX2341X_FIRM_ENC_FILENAME "v4l-cx2341x-enc.fw" -- cgit v1.2.3-70-g09d2 From 2c06a0821711a53d51a3d0492a9be0671b7152e5 Mon Sep 17 00:00:00 2001 From: Ben Dooks Date: Tue, 27 Jun 2006 14:35:46 +0100 Subject: [MTD NAND] s3c2412 support in s3c2410.c Add support for both the S3C2412 and S3C2412 Samsung SoCs to the increasingly mis-named s3c2410.c driver. This currently only supports SLC ECCs, and a chip on nFCE0. Signed-off-by: Ben Dooks Signed-off-by: David Woodhouse --- drivers/mtd/nand/s3c2410.c | 154 ++++++++++++++++++++++--------- include/asm-arm/arch-s3c2410/regs-nand.h | 48 +++++++++- 2 files changed, 158 insertions(+), 44 deletions(-) (limited to 'include') diff --git a/drivers/mtd/nand/s3c2410.c b/drivers/mtd/nand/s3c2410.c index 5219bd212cf..ff5cef24d5b 100644 --- a/drivers/mtd/nand/s3c2410.c +++ b/drivers/mtd/nand/s3c2410.c @@ -97,6 +97,12 @@ struct s3c2410_nand_mtd { int scan_res; }; +enum s3c_cpu_type { + TYPE_S3C2410, + TYPE_S3C2412, + TYPE_S3C2440, +}; + /* overview of the s3c2410 nand state */ struct s3c2410_nand_info { @@ -110,9 +116,11 @@ struct s3c2410_nand_info { struct resource *area; struct clk *clk; void __iomem *regs; + void __iomem *sel_reg; + int sel_bit; int mtd_count; - unsigned char is_s3c2440; + enum s3c_cpu_type cpu_type; }; /* conversion functions */ @@ -146,7 +154,7 @@ static inline int allow_clk_stop(struct s3c2410_nand_info *info) #define NS_IN_KHZ 1000000 -static int s3c2410_nand_calc_rate(int wanted, unsigned long clk, int max) +static int s3c_nand_calc_rate(int wanted, unsigned long clk, int max) { int result; @@ -170,24 +178,26 @@ static int s3c2410_nand_calc_rate(int wanted, unsigned long clk, int max) /* controller setup */ -static int s3c2410_nand_inithw(struct s3c2410_nand_info *info, struct platform_device *pdev) +static int s3c2410_nand_inithw(struct s3c2410_nand_info *info, + struct platform_device *pdev) { struct s3c2410_platform_nand *plat = to_nand_plat(pdev); unsigned long clkrate = clk_get_rate(info->clk); + int tacls_max = (info->cpu_type == TYPE_S3C2412) ? 8 : 4; int tacls, twrph0, twrph1; - unsigned long cfg; + unsigned long cfg = 0; /* calculate the timing information for the controller */ clkrate /= 1000; /* turn clock into kHz for ease of use */ if (plat != NULL) { - tacls = s3c2410_nand_calc_rate(plat->tacls, clkrate, 4); - twrph0 = s3c2410_nand_calc_rate(plat->twrph0, clkrate, 8); - twrph1 = s3c2410_nand_calc_rate(plat->twrph1, clkrate, 8); + tacls = s3c_nand_calc_rate(plat->tacls, clkrate, tacls_max); + twrph0 = s3c_nand_calc_rate(plat->twrph0, clkrate, 8); + twrph1 = s3c_nand_calc_rate(plat->twrph1, clkrate, 8); } else { /* default timings */ - tacls = 4; + tacls = tacls_max; twrph0 = 8; twrph1 = 8; } @@ -200,20 +210,23 @@ static int s3c2410_nand_inithw(struct s3c2410_nand_info *info, struct platform_d dev_info(info->device, "Tacls=%d, %dns Twrph0=%d %dns, Twrph1=%d %dns\n", tacls, to_ns(tacls, clkrate), twrph0, to_ns(twrph0, clkrate), twrph1, to_ns(twrph1, clkrate)); - if (!info->is_s3c2440) { + switch (info->cpu_type) { + case TYPE_S3C2410: cfg = S3C2410_NFCONF_EN; cfg |= S3C2410_NFCONF_TACLS(tacls - 1); cfg |= S3C2410_NFCONF_TWRPH0(twrph0 - 1); cfg |= S3C2410_NFCONF_TWRPH1(twrph1 - 1); - } else { + break; + + case TYPE_S3C2440: + case TYPE_S3C2412: cfg = S3C2440_NFCONF_TACLS(tacls - 1); cfg |= S3C2440_NFCONF_TWRPH0(twrph0 - 1); cfg |= S3C2440_NFCONF_TWRPH1(twrph1 - 1); /* enable the controller and de-assert nFCE */ - writel(S3C2440_NFCONT_ENABLE | S3C2440_NFCONT_ENABLE, - info->regs + S3C2440_NFCONT); + writel(S3C2440_NFCONT_ENABLE, info->regs + S3C2440_NFCONT); } dev_dbg(info->device, "NF_CONF is 0x%lx\n", cfg); @@ -229,23 +242,18 @@ static void s3c2410_nand_select_chip(struct mtd_info *mtd, int chip) struct s3c2410_nand_info *info; struct s3c2410_nand_mtd *nmtd; struct nand_chip *this = mtd->priv; - void __iomem *reg; unsigned long cur; - unsigned long bit; nmtd = this->priv; info = nmtd->info; - bit = (info->is_s3c2440) ? S3C2440_NFCONT_nFCE : S3C2410_NFCONF_nFCE; - reg = info->regs + ((info->is_s3c2440) ? S3C2440_NFCONT : S3C2410_NFCONF); - if (chip != -1 && allow_clk_stop(info)) clk_enable(info->clk); - cur = readl(reg); + cur = readl(info->sel_reg); if (chip == -1) { - cur |= bit; + cur |= info->sel_bit; } else { if (nmtd->set != NULL && chip > nmtd->set->nr_chips) { dev_err(info->device, "invalid chip %d\n", chip); @@ -257,10 +265,10 @@ static void s3c2410_nand_select_chip(struct mtd_info *mtd, int chip) (info->platform->select_chip) (nmtd->set, chip); } - cur &= ~bit; + cur &= ~info->sel_bit; } - writel(cur, reg); + writel(cur, info->sel_reg); if (chip == -1 && allow_clk_stop(info)) clk_disable(info->clk); @@ -309,15 +317,25 @@ static void s3c2440_nand_hwcontrol(struct mtd_info *mtd, int cmd, static int s3c2410_nand_devready(struct mtd_info *mtd) { struct s3c2410_nand_info *info = s3c2410_nand_mtd_toinfo(mtd); - - if (info->is_s3c2440) - return readb(info->regs + S3C2440_NFSTAT) & S3C2440_NFSTAT_READY; return readb(info->regs + S3C2410_NFSTAT) & S3C2410_NFSTAT_BUSY; } +static int s3c2440_nand_devready(struct mtd_info *mtd) +{ + struct s3c2410_nand_info *info = s3c2410_nand_mtd_toinfo(mtd); + return readb(info->regs + S3C2440_NFSTAT) & S3C2440_NFSTAT_READY; +} + +static int s3c2412_nand_devready(struct mtd_info *mtd) +{ + struct s3c2410_nand_info *info = s3c2410_nand_mtd_toinfo(mtd); + return readb(info->regs + S3C2412_NFSTAT) & S3C2412_NFSTAT_READY; +} + /* ECC handling functions */ -static int s3c2410_nand_correct_data(struct mtd_info *mtd, u_char *dat, u_char *read_ecc, u_char *calc_ecc) +static int s3c2410_nand_correct_data(struct mtd_info *mtd, u_char *dat, + u_char *read_ecc, u_char *calc_ecc) { pr_debug("s3c2410_nand_correct_data(%p,%p,%p,%p)\n", mtd, dat, read_ecc, calc_ecc); @@ -485,11 +503,8 @@ static void s3c2410_nand_init_chip(struct s3c2410_nand_info *info, struct s3c2410_nand_set *set) { struct nand_chip *chip = &nmtd->chip; + void __iomem *regs = info->regs; - chip->IO_ADDR_R = info->regs + S3C2410_NFDATA; - chip->IO_ADDR_W = info->regs + S3C2410_NFDATA; - chip->cmd_ctrl = s3c2410_nand_hwcontrol; - chip->dev_ready = s3c2410_nand_devready; chip->write_buf = s3c2410_nand_write_buf; chip->read_buf = s3c2410_nand_read_buf; chip->select_chip = s3c2410_nand_select_chip; @@ -498,11 +513,37 @@ static void s3c2410_nand_init_chip(struct s3c2410_nand_info *info, chip->options = 0; chip->controller = &info->controller; - if (info->is_s3c2440) { - chip->IO_ADDR_R = info->regs + S3C2440_NFDATA; - chip->IO_ADDR_W = info->regs + S3C2440_NFDATA; - chip->cmd_ctrl = s3c2440_nand_hwcontrol; - } + switch (info->cpu_type) { + case TYPE_S3C2410: + chip->IO_ADDR_W = regs + S3C2410_NFDATA; + info->sel_reg = regs + S3C2410_NFCONF; + info->sel_bit = S3C2410_NFCONF_nFCE; + chip->cmd_ctrl = s3c2410_nand_hwcontrol; + chip->dev_ready = s3c2410_nand_devready; + break; + + case TYPE_S3C2440: + chip->IO_ADDR_W = regs + S3C2440_NFDATA; + info->sel_reg = regs + S3C2440_NFCONT; + info->sel_bit = S3C2440_NFCONT_nFCE; + chip->cmd_ctrl = s3c2440_nand_hwcontrol; + chip->dev_ready = s3c2440_nand_devready; + break; + + case TYPE_S3C2412: + chip->IO_ADDR_W = regs + S3C2440_NFDATA; + info->sel_reg = regs + S3C2440_NFCONT; + info->sel_bit = S3C2412_NFCONT_nFCE0; + chip->cmd_ctrl = s3c2440_nand_hwcontrol; + chip->dev_ready = s3c2412_nand_devready; + + if (readl(regs + S3C2410_NFCONF) & S3C2412_NFCONF_NANDBOOT) + dev_info(info->device, "System booted from NAND\n"); + + break; + } + + chip->IO_ADDR_R = chip->IO_ADDR_W; nmtd->info = info; nmtd->mtd.priv = chip; @@ -510,17 +551,25 @@ static void s3c2410_nand_init_chip(struct s3c2410_nand_info *info, nmtd->set = set; if (hardware_ecc) { - chip->ecc.correct = s3c2410_nand_correct_data; - chip->ecc.hwctl = s3c2410_nand_enable_hwecc; chip->ecc.calculate = s3c2410_nand_calculate_ecc; + chip->ecc.correct = s3c2410_nand_correct_data; chip->ecc.mode = NAND_ECC_HW; chip->ecc.size = 512; chip->ecc.bytes = 3; chip->ecc.layout = &nand_hw_eccoob; - if (info->is_s3c2440) { - chip->ecc.hwctl = s3c2440_nand_enable_hwecc; - chip->ecc.calculate = s3c2440_nand_calculate_ecc; + switch (info->cpu_type) { + case TYPE_S3C2410: + chip->ecc.hwctl = s3c2410_nand_enable_hwecc; + chip->ecc.calculate = s3c2410_nand_calculate_ecc; + break; + + case TYPE_S3C2412: + case TYPE_S3C2440: + chip->ecc.hwctl = s3c2440_nand_enable_hwecc; + chip->ecc.calculate = s3c2440_nand_calculate_ecc; + break; + } } else { chip->ecc.mode = NAND_ECC_SOFT; @@ -535,7 +584,8 @@ static void s3c2410_nand_init_chip(struct s3c2410_nand_info *info, * nand layer to look for devices */ -static int s3c24xx_nand_probe(struct platform_device *pdev, int is_s3c2440) +static int s3c24xx_nand_probe(struct platform_device *pdev, + enum s3c_cpu_type cpu_type) { struct s3c2410_platform_nand *plat = to_nand_plat(pdev); struct s3c2410_nand_info *info; @@ -590,7 +640,7 @@ static int s3c24xx_nand_probe(struct platform_device *pdev, int is_s3c2440) info->device = &pdev->dev; info->platform = plat; info->regs = ioremap(res->start, size); - info->is_s3c2440 = is_s3c2440; + info->cpu_type = cpu_type; if (info->regs == NULL) { dev_err(&pdev->dev, "cannot reserve register region\n"); @@ -697,12 +747,17 @@ static int s3c24xx_nand_resume(struct platform_device *dev) static int s3c2410_nand_probe(struct platform_device *dev) { - return s3c24xx_nand_probe(dev, 0); + return s3c24xx_nand_probe(dev, TYPE_S3C2410); } static int s3c2440_nand_probe(struct platform_device *dev) { - return s3c24xx_nand_probe(dev, 1); + return s3c24xx_nand_probe(dev, TYPE_S3C2440); +} + +static int s3c2412_nand_probe(struct platform_device *dev) +{ + return s3c24xx_nand_probe(dev, TYPE_S3C2412); } static struct platform_driver s3c2410_nand_driver = { @@ -727,16 +782,29 @@ static struct platform_driver s3c2440_nand_driver = { }, }; +static struct platform_driver s3c2412_nand_driver = { + .probe = s3c2412_nand_probe, + .remove = s3c2410_nand_remove, + .suspend = s3c24xx_nand_suspend, + .resume = s3c24xx_nand_resume, + .driver = { + .name = "s3c2412-nand", + .owner = THIS_MODULE, + }, +}; + static int __init s3c2410_nand_init(void) { printk("S3C24XX NAND Driver, (c) 2004 Simtec Electronics\n"); + platform_driver_register(&s3c2412_nand_driver); platform_driver_register(&s3c2440_nand_driver); return platform_driver_register(&s3c2410_nand_driver); } static void __exit s3c2410_nand_exit(void) { + platform_driver_unregister(&s3c2412_nand_driver); platform_driver_unregister(&s3c2440_nand_driver); platform_driver_unregister(&s3c2410_nand_driver); } diff --git a/include/asm-arm/arch-s3c2410/regs-nand.h b/include/asm-arm/arch-s3c2410/regs-nand.h index 7cff235e667..c1470c695c3 100644 --- a/include/asm-arm/arch-s3c2410/regs-nand.h +++ b/include/asm-arm/arch-s3c2410/regs-nand.h @@ -39,10 +39,19 @@ #define S3C2440_NFESTAT1 S3C2410_NFREG(0x28) #define S3C2440_NFMECC0 S3C2410_NFREG(0x2C) #define S3C2440_NFMECC1 S3C2410_NFREG(0x30) -#define S3C2440_NFSECC S3C2410_NFREG(0x34) +#define S3C2440_NFSECC S3C24E10_NFREG(0x34) #define S3C2440_NFSBLK S3C2410_NFREG(0x38) #define S3C2440_NFEBLK S3C2410_NFREG(0x3C) +#define S3C2412_NFSBLK S3C2410_NFREG(0x20) +#define S3C2412_NFEBLK S3C2410_NFREG(0x24) +#define S3C2412_NFSTAT S3C2410_NFREG(0x28) +#define S3C2412_NFMECC_ERR0 S3C2410_NFREG(0x2C) +#define S3C2412_NFMECC_ERR1 S3C2410_NFREG(0x30) +#define S3C2412_NFMECC0 S3C2410_NFREG(0x34) +#define S3C2412_NFMECC1 S3C2410_NFREG(0x38) +#define S3C2412_NFSECC S3C2410_NFREG(0x3C) + #define S3C2410_NFCONF_EN (1<<15) #define S3C2410_NFCONF_512BYTE (1<<14) #define S3C2410_NFCONF_4STEP (1<<13) @@ -77,5 +86,42 @@ #define S3C2440_NFSTAT_RnB_CHANGE (1<<2) #define S3C2440_NFSTAT_ILLEGAL_ACCESS (1<<3) +#define S3C2412_NFCONF_NANDBOOT (1<<31) +#define S3C2412_NFCONF_ECCCLKCON (1<<30) +#define S3C2412_NFCONF_ECC_MLC (1<<24) +#define S3C2412_NFCONF_TACLS_MASK (7<<12) /* 1 extra bit of Tacls */ + +#define S3C2412_NFCONT_ECC4_DIRWR (1<<18) +#define S3C2412_NFCONT_LOCKTIGHT (1<<17) +#define S3C2412_NFCONT_SOFTLOCK (1<<16) +#define S3C2412_NFCONT_ECC4_ENCINT (1<<13) +#define S3C2412_NFCONT_ECC4_DECINT (1<<12) +#define S3C2412_NFCONT_MAIN_ECC_LOCK (1<<7) +#define S3C2412_NFCONT_INIT_MAIN_ECC (1<<5) +#define S3C2412_NFCONT_nFCE1 (1<<2) +#define S3C2412_NFCONT_nFCE0 (1<<1) + +#define S3C2412_NFSTAT_ECC_ENCDONE (1<<7) +#define S3C2412_NFSTAT_ECC_DECDONE (1<<6) +#define S3C2412_NFSTAT_ILLEGAL_ACCESS (1<<5) +#define S3C2412_NFSTAT_RnB_CHANGE (1<<4) +#define S3C2412_NFSTAT_nFCE1 (1<<3) +#define S3C2412_NFSTAT_nFCE0 (1<<2) +#define S3C2412_NFSTAT_Res1 (1<<1) +#define S3C2412_NFSTAT_READY (1<<0) + +#define S3C2412_NFECCERR_SERRDATA(x) (((x) >> 21) & 0xf) +#define S3C2412_NFECCERR_SERRBIT(x) (((x) >> 18) & 0x7) +#define S3C2412_NFECCERR_MERRDATA(x) (((x) >> 7) & 0x3ff) +#define S3C2412_NFECCERR_MERRBIT(x) (((x) >> 4) & 0x7) +#define S3C2412_NFECCERR_SPARE_ERR(x) (((x) >> 2) & 0x3) +#define S3C2412_NFECCERR_MAIN_ERR(x) (((x) >> 2) & 0x3) +#define S3C2412_NFECCERR_NONE (0) +#define S3C2412_NFECCERR_1BIT (1) +#define S3C2412_NFECCERR_MULTIBIT (2) +#define S3C2412_NFECCERR_ECCAREA (3) + + + #endif /* __ASM_ARM_REGS_NAND */ -- cgit v1.2.3-70-g09d2 From c9f700f840bd481b3e01fcad1ba8da01794a6726 Mon Sep 17 00:00:00 2001 From: KaiGai Kohei Date: Sun, 11 Jun 2006 10:35:15 +0900 Subject: [JFFS2][XATTR] using 'delete marker' for xdatum/xref deletion - When xdatum is removed, a new xdatum with 'delete marker' is written. (version==0xffffffff means 'delete marker') - When xref is removed, a new xref with 'delete marker' is written. (odd-numbered xseqno means 'delete marker') - delete_xattr_(datum/xref)_delay() are new deletion functions are added. We can only use them if we can detect the target obsolete xdatum/xref as a orphan or errir one. (e.g when inode deletion, or detecting crc error) [1/3] jffs2-xattr-v6-01-delete_marker.patch Signed-off-by: KaiGai Kohei Signed-off-by: David Woodhouse --- fs/jffs2/erase.c | 19 +- fs/jffs2/gc.c | 5 +- fs/jffs2/jffs2_fs_sb.h | 3 + fs/jffs2/malloc.c | 2 + fs/jffs2/nodemgmt.c | 21 +- fs/jffs2/scan.c | 57 ++-- fs/jffs2/summary.c | 39 +-- fs/jffs2/xattr.c | 736 +++++++++++++++++++++++++++++++------------------ fs/jffs2/xattr.h | 40 ++- include/linux/jffs2.h | 1 + 10 files changed, 584 insertions(+), 339 deletions(-) (limited to 'include') diff --git a/fs/jffs2/erase.c b/fs/jffs2/erase.c index 1862e8bc101..1644e340885 100644 --- a/fs/jffs2/erase.c +++ b/fs/jffs2/erase.c @@ -230,7 +230,6 @@ static inline void jffs2_remove_node_refs_from_ino_list(struct jffs2_sb_info *c, at the end of the linked list. Stash it and continue from the beginning of the list */ ic = (struct jffs2_inode_cache *)(*prev); - BUG_ON(ic->class != RAWNODE_CLASS_INODE_CACHE); prev = &ic->nodes; continue; } @@ -254,7 +253,8 @@ static inline void jffs2_remove_node_refs_from_ino_list(struct jffs2_sb_info *c, /* PARANOIA */ if (!ic) { - printk(KERN_WARNING "inode_cache not found in remove_node_refs()!!\n"); + JFFS2_WARNING("inode_cache/xattr_datum/xattr_ref" + " not found in remove_node_refs()!!\n"); return; } @@ -279,8 +279,19 @@ static inline void jffs2_remove_node_refs_from_ino_list(struct jffs2_sb_info *c, printk("\n"); }); - if (ic->nodes == (void *)ic && ic->nlink == 0) - jffs2_del_ino_cache(c, ic); + switch (ic->class) { +#ifdef CONFIG_JFFS2_FS_XATTR + case RAWNODE_CLASS_XATTR_DATUM: + jffs2_release_xattr_datum(c, (struct jffs2_xattr_datum *)ic); + break; + case RAWNODE_CLASS_XATTR_REF: + jffs2_release_xattr_ref(c, (struct jffs2_xattr_ref *)ic); + break; +#endif + default: + if (ic->nodes == (void *)ic && ic->nlink == 0) + jffs2_del_ino_cache(c, ic); + } } void jffs2_free_jeb_node_refs(struct jffs2_sb_info *c, struct jffs2_eraseblock *jeb) diff --git a/fs/jffs2/gc.c b/fs/jffs2/gc.c index 477c526d638..f59b147661c 100644 --- a/fs/jffs2/gc.c +++ b/fs/jffs2/gc.c @@ -275,13 +275,12 @@ int jffs2_garbage_collect_pass(struct jffs2_sb_info *c) * We can decide whether this node is inode or xattr by ic->class. */ if (ic->class == RAWNODE_CLASS_XATTR_DATUM || ic->class == RAWNODE_CLASS_XATTR_REF) { - BUG_ON(raw->next_in_ino != (void *)ic); spin_unlock(&c->erase_completion_lock); if (ic->class == RAWNODE_CLASS_XATTR_DATUM) { - ret = jffs2_garbage_collect_xattr_datum(c, (struct jffs2_xattr_datum *)ic); + ret = jffs2_garbage_collect_xattr_datum(c, (struct jffs2_xattr_datum *)ic, raw); } else { - ret = jffs2_garbage_collect_xattr_ref(c, (struct jffs2_xattr_ref *)ic); + ret = jffs2_garbage_collect_xattr_ref(c, (struct jffs2_xattr_ref *)ic, raw); } goto release_sem; } diff --git a/fs/jffs2/jffs2_fs_sb.h b/fs/jffs2/jffs2_fs_sb.h index 935fec1b120..b98594992ee 100644 --- a/fs/jffs2/jffs2_fs_sb.h +++ b/fs/jffs2/jffs2_fs_sb.h @@ -119,8 +119,11 @@ struct jffs2_sb_info { #ifdef CONFIG_JFFS2_FS_XATTR #define XATTRINDEX_HASHSIZE (57) uint32_t highest_xid; + uint32_t highest_xseqno; struct list_head xattrindex[XATTRINDEX_HASHSIZE]; struct list_head xattr_unchecked; + struct list_head xattr_dead_list; + struct jffs2_xattr_ref *xref_dead_list; struct jffs2_xattr_ref *xref_temp; struct rw_semaphore xattr_sem; uint32_t xdatum_mem_usage; diff --git a/fs/jffs2/malloc.c b/fs/jffs2/malloc.c index 4889d0700c0..8310c95478e 100644 --- a/fs/jffs2/malloc.c +++ b/fs/jffs2/malloc.c @@ -291,6 +291,7 @@ struct jffs2_xattr_datum *jffs2_alloc_xattr_datum(void) memset(xd, 0, sizeof(struct jffs2_xattr_datum)); xd->class = RAWNODE_CLASS_XATTR_DATUM; + xd->node = (void *)xd; INIT_LIST_HEAD(&xd->xindex); return xd; } @@ -309,6 +310,7 @@ struct jffs2_xattr_ref *jffs2_alloc_xattr_ref(void) memset(ref, 0, sizeof(struct jffs2_xattr_ref)); ref->class = RAWNODE_CLASS_XATTR_REF; + ref->node = (void *)ref; return ref; } diff --git a/fs/jffs2/nodemgmt.c b/fs/jffs2/nodemgmt.c index 8bedfd2ff68..01594a2256e 100644 --- a/fs/jffs2/nodemgmt.c +++ b/fs/jffs2/nodemgmt.c @@ -684,19 +684,26 @@ void jffs2_mark_node_obsolete(struct jffs2_sb_info *c, struct jffs2_raw_node_ref spin_lock(&c->erase_completion_lock); ic = jffs2_raw_ref_to_ic(ref); - /* It seems we should never call jffs2_mark_node_obsolete() for - XATTR nodes.... yet. Make sure we notice if/when we change - that :) */ - BUG_ON(ic->class != RAWNODE_CLASS_INODE_CACHE); for (p = &ic->nodes; (*p) != ref; p = &((*p)->next_in_ino)) ; *p = ref->next_in_ino; ref->next_in_ino = NULL; - if (ic->nodes == (void *)ic && ic->nlink == 0) - jffs2_del_ino_cache(c, ic); - + switch (ic->class) { +#ifdef CONFIG_JFFS2_FS_XATTR + case RAWNODE_CLASS_XATTR_DATUM: + jffs2_release_xattr_datum(c, (struct jffs2_xattr_datum *)ic); + break; + case RAWNODE_CLASS_XATTR_REF: + jffs2_release_xattr_ref(c, (struct jffs2_xattr_ref *)ic); + break; +#endif + default: + if (ic->nodes == (void *)ic && ic->nlink == 0) + jffs2_del_ino_cache(c, ic); + break; + } spin_unlock(&c->erase_completion_lock); } diff --git a/fs/jffs2/scan.c b/fs/jffs2/scan.c index 61618080b86..79638f56c5e 100644 --- a/fs/jffs2/scan.c +++ b/fs/jffs2/scan.c @@ -317,20 +317,25 @@ static int jffs2_scan_xattr_node(struct jffs2_sb_info *c, struct jffs2_erasebloc struct jffs2_summary *s) { struct jffs2_xattr_datum *xd; - uint32_t totlen, crc; + uint32_t xid, version, totlen, crc; int err; crc = crc32(0, rx, sizeof(struct jffs2_raw_xattr) - 4); if (crc != je32_to_cpu(rx->node_crc)) { - if (je32_to_cpu(rx->node_crc) != 0xffffffff) - JFFS2_WARNING("node CRC failed at %#08x, read=%#08x, calc=%#08x\n", - ofs, je32_to_cpu(rx->node_crc), crc); + JFFS2_WARNING("node CRC failed at %#08x, read=%#08x, calc=%#08x\n", + ofs, je32_to_cpu(rx->node_crc), crc); if ((err = jffs2_scan_dirty_space(c, jeb, je32_to_cpu(rx->totlen)))) return err; return 0; } - totlen = PAD(sizeof(*rx) + rx->name_len + 1 + je16_to_cpu(rx->value_len)); + xid = je32_to_cpu(rx->xid); + version = je32_to_cpu(rx->version); + + totlen = sizeof(struct jffs2_raw_xattr); + if (version != XDATUM_DELETE_MARKER) + totlen += rx->name_len + 1 + je16_to_cpu(rx->value_len); + totlen = PAD(totlen); if (totlen != je32_to_cpu(rx->totlen)) { JFFS2_WARNING("node length mismatch at %#08x, read=%u, calc=%u\n", ofs, je32_to_cpu(rx->totlen), totlen); @@ -339,22 +344,24 @@ static int jffs2_scan_xattr_node(struct jffs2_sb_info *c, struct jffs2_erasebloc return 0; } - xd = jffs2_setup_xattr_datum(c, je32_to_cpu(rx->xid), je32_to_cpu(rx->version)); - if (IS_ERR(xd)) { - if (PTR_ERR(xd) == -EEXIST) { - if ((err = jffs2_scan_dirty_space(c, jeb, PAD(je32_to_cpu(rx->totlen))))) - return err; - return 0; - } + xd = jffs2_setup_xattr_datum(c, xid, version); + if (IS_ERR(xd)) return PTR_ERR(xd); - } - xd->xprefix = rx->xprefix; - xd->name_len = rx->name_len; - xd->value_len = je16_to_cpu(rx->value_len); - xd->data_crc = je32_to_cpu(rx->data_crc); - xd->node = jffs2_link_node_ref(c, jeb, ofs | REF_PRISTINE, totlen, NULL); - /* FIXME */ xd->node->next_in_ino = (void *)xd; + if (xd->version > version) { + struct jffs2_raw_node_ref *raw + = jffs2_link_node_ref(c, jeb, ofs | REF_PRISTINE, totlen, NULL); + raw->next_in_ino = xd->node->next_in_ino; + xd->node->next_in_ino = raw; + } else { + xd->version = version; + xd->xprefix = rx->xprefix; + xd->name_len = rx->name_len; + xd->value_len = je16_to_cpu(rx->value_len); + xd->data_crc = je32_to_cpu(rx->data_crc); + + jffs2_link_node_ref(c, jeb, ofs | REF_PRISTINE, totlen, (void *)xd); + } if (jffs2_sum_active()) jffs2_sum_add_xattr_mem(s, rx, ofs - jeb->offset); @@ -373,9 +380,8 @@ static int jffs2_scan_xref_node(struct jffs2_sb_info *c, struct jffs2_eraseblock crc = crc32(0, rr, sizeof(*rr) - 4); if (crc != je32_to_cpu(rr->node_crc)) { - if (je32_to_cpu(rr->node_crc) != 0xffffffff) - JFFS2_WARNING("node CRC failed at %#08x, read=%#08x, calc=%#08x\n", - ofs, je32_to_cpu(rr->node_crc), crc); + JFFS2_WARNING("node CRC failed at %#08x, read=%#08x, calc=%#08x\n", + ofs, je32_to_cpu(rr->node_crc), crc); if ((err = jffs2_scan_dirty_space(c, jeb, PAD(je32_to_cpu(rr->totlen))))) return err; return 0; @@ -395,6 +401,7 @@ static int jffs2_scan_xref_node(struct jffs2_sb_info *c, struct jffs2_eraseblock return -ENOMEM; /* BEFORE jffs2_build_xattr_subsystem() called, + * and AFTER xattr_ref is marked as a dead xref, * ref->xid is used to store 32bit xid, xd is not used * ref->ino is used to store 32bit inode-number, ic is not used * Thoes variables are declared as union, thus using those @@ -404,11 +411,13 @@ static int jffs2_scan_xref_node(struct jffs2_sb_info *c, struct jffs2_eraseblock */ ref->ino = je32_to_cpu(rr->ino); ref->xid = je32_to_cpu(rr->xid); + ref->xseqno = je32_to_cpu(rr->xseqno); + if (ref->xseqno > c->highest_xseqno) + c->highest_xseqno = (ref->xseqno & ~XREF_DELETE_MARKER); ref->next = c->xref_temp; c->xref_temp = ref; - ref->node = jffs2_link_node_ref(c, jeb, ofs | REF_PRISTINE, PAD(je32_to_cpu(rr->totlen)), NULL); - /* FIXME */ ref->node->next_in_ino = (void *)ref; + jffs2_link_node_ref(c, jeb, ofs | REF_PRISTINE, PAD(je32_to_cpu(rr->totlen)), (void *)ref); if (jffs2_sum_active()) jffs2_sum_add_xref_mem(s, rr, ofs - jeb->offset); diff --git a/fs/jffs2/summary.c b/fs/jffs2/summary.c index 0b02fc79e4d..c430f1d217e 100644 --- a/fs/jffs2/summary.c +++ b/fs/jffs2/summary.c @@ -310,8 +310,6 @@ int jffs2_sum_add_kvec(struct jffs2_sb_info *c, const struct kvec *invecs, #ifdef CONFIG_JFFS2_FS_XATTR case JFFS2_NODETYPE_XATTR: { struct jffs2_sum_xattr_mem *temp; - if (je32_to_cpu(node->x.version) == 0xffffffff) - return 0; temp = kmalloc(sizeof(struct jffs2_sum_xattr_mem), GFP_KERNEL); if (!temp) goto no_mem; @@ -327,10 +325,6 @@ int jffs2_sum_add_kvec(struct jffs2_sb_info *c, const struct kvec *invecs, } case JFFS2_NODETYPE_XREF: { struct jffs2_sum_xref_mem *temp; - - if (je32_to_cpu(node->r.ino) == 0xffffffff - && je32_to_cpu(node->r.xid) == 0xffffffff) - return 0; temp = kmalloc(sizeof(struct jffs2_sum_xref_mem), GFP_KERNEL); if (!temp) goto no_mem; @@ -483,22 +477,20 @@ static int jffs2_sum_process_sum_data(struct jffs2_sb_info *c, struct jffs2_eras xd = jffs2_setup_xattr_datum(c, je32_to_cpu(spx->xid), je32_to_cpu(spx->version)); - if (IS_ERR(xd)) { - if (PTR_ERR(xd) == -EEXIST) { - /* a newer version of xd exists */ - if ((err = jffs2_scan_dirty_space(c, jeb, je32_to_cpu(spx->totlen)))) - return err; - sp += JFFS2_SUMMARY_XATTR_SIZE; - break; - } - JFFS2_NOTICE("allocation of xattr_datum failed\n"); + if (IS_ERR(xd)) return PTR_ERR(xd); + if (xd->version > je32_to_cpu(spx->version)) { + /* node is not the newest one */ + struct jffs2_raw_node_ref *raw + = sum_link_node_ref(c, jeb, je32_to_cpu(spx->offset) | REF_UNCHECKED, + PAD(je32_to_cpu(spx->totlen)), NULL); + raw->next_in_ino = xd->node->next_in_ino; + xd->node->next_in_ino = raw; + } else { + xd->version = je32_to_cpu(spx->version); + sum_link_node_ref(c, jeb, je32_to_cpu(spx->offset) | REF_UNCHECKED, + PAD(je32_to_cpu(spx->totlen)), (void *)xd); } - - xd->node = sum_link_node_ref(c, jeb, je32_to_cpu(spx->offset) | REF_UNCHECKED, - PAD(je32_to_cpu(spx->totlen)), NULL); - /* FIXME */ xd->node->next_in_ino = (void *)xd; - *pseudo_random += je32_to_cpu(spx->xid); sp += JFFS2_SUMMARY_XATTR_SIZE; @@ -519,14 +511,11 @@ static int jffs2_sum_process_sum_data(struct jffs2_sb_info *c, struct jffs2_eras JFFS2_NOTICE("allocation of xattr_datum failed\n"); return -ENOMEM; } - ref->ino = 0xfffffffe; - ref->xid = 0xfffffffd; ref->next = c->xref_temp; c->xref_temp = ref; - ref->node = sum_link_node_ref(c, jeb, je32_to_cpu(spr->offset) | REF_UNCHECKED, - PAD(sizeof(struct jffs2_raw_xref)), NULL); - /* FIXME */ ref->node->next_in_ino = (void *)ref; + sum_link_node_ref(c, jeb, je32_to_cpu(spr->offset) | REF_UNCHECKED, + PAD(sizeof(struct jffs2_raw_xref)), (void *)ref); *pseudo_random += ref->node->flash_offset; sp += JFFS2_SUMMARY_XREF_SIZE; diff --git a/fs/jffs2/xattr.c b/fs/jffs2/xattr.c index 2d82e250be3..03871ab7c26 100644 --- a/fs/jffs2/xattr.c +++ b/fs/jffs2/xattr.c @@ -23,18 +23,15 @@ * xattr_datum_hashkey(xprefix, xname, xvalue, xsize) * is used to calcurate xdatum hashkey. The reminder of hashkey into XATTRINDEX_HASHSIZE is * the index of the xattr name/value pair cache (c->xattrindex). + * is_xattr_datum_unchecked(c, xd) + * returns 1, if xdatum contains any unchecked raw nodes. if all raw nodes are not + * unchecked, it returns 0. * unload_xattr_datum(c, xd) * is used to release xattr name/value pair and detach from c->xattrindex. * reclaim_xattr_datum(c) * is used to reclaim xattr name/value pairs on the xattr name/value pair cache when * memory usage by cache is over c->xdatum_mem_threshold. Currentry, this threshold * is hard coded as 32KiB. - * delete_xattr_datum_node(c, xd) - * is used to delete a jffs2 node is dominated by xdatum. When EBS(Erase Block Summary) is - * enabled, it overwrites the obsolete node by myself. - * delete_xattr_datum(c, xd) - * is used to delete jffs2_xattr_datum object. It must be called with 0-value of reference - * counter. (It means how many jffs2_xattr_ref object refers this xdatum.) * do_verify_xattr_datum(c, xd) * is used to load the xdatum informations without name/value pair from the medium. * It's necessary once, because those informations are not collected during mounting @@ -53,8 +50,13 @@ * is used to write xdatum to medium. xd->version will be incremented. * create_xattr_datum(c, xprefix, xname, xvalue, xsize) * is used to create new xdatum and write to medium. + * delete_xattr_datum_delay(c, xd) + * is used to delete a xdatum without 'delete marker'. It has a possibility to detect + * orphan xdatum on next mounting. + * delete_xattr_datum(c, xd) + * is used to delete a xdatum with 'delete marker'. Calling jffs2_reserve_space() is + * necessary before this function. * -------------------------------------------------- */ - static uint32_t xattr_datum_hashkey(int xprefix, const char *xname, const char *xvalue, int xsize) { int name_len = strlen(xname); @@ -62,6 +64,22 @@ static uint32_t xattr_datum_hashkey(int xprefix, const char *xname, const char * return crc32(xprefix, xname, name_len) ^ crc32(xprefix, xvalue, xsize); } +static int is_xattr_datum_unchecked(struct jffs2_sb_info *c, struct jffs2_xattr_datum *xd) +{ + struct jffs2_raw_node_ref *raw; + int rc = 0; + + spin_lock(&c->erase_completion_lock); + for (raw=xd->node; raw != (void *)xd; raw=raw->next_in_ino) { + if (ref_flags(raw) == REF_UNCHECKED) { + rc = 1; + break; + } + } + spin_unlock(&c->erase_completion_lock); + return rc; +} + static void unload_xattr_datum(struct jffs2_sb_info *c, struct jffs2_xattr_datum *xd) { /* must be called under down_write(xattr_sem) */ @@ -107,80 +125,39 @@ static void reclaim_xattr_datum(struct jffs2_sb_info *c) before, c->xdatum_mem_usage, before - c->xdatum_mem_usage); } -static void delete_xattr_datum_node(struct jffs2_sb_info *c, struct jffs2_xattr_datum *xd) -{ - /* must be called under down_write(xattr_sem) */ - struct jffs2_raw_xattr rx; - size_t length; - int rc; - - if (!xd->node) { - JFFS2_WARNING("xdatum (xid=%u) is removed twice.\n", xd->xid); - return; - } - if (jffs2_sum_active()) { - memset(&rx, 0xff, sizeof(struct jffs2_raw_xattr)); - rc = jffs2_flash_read(c, ref_offset(xd->node), - sizeof(struct jffs2_unknown_node), - &length, (char *)&rx); - if (rc || length != sizeof(struct jffs2_unknown_node)) { - JFFS2_ERROR("jffs2_flash_read()=%d, req=%zu, read=%zu at %#08x\n", - rc, sizeof(struct jffs2_unknown_node), - length, ref_offset(xd->node)); - } - rc = jffs2_flash_write(c, ref_offset(xd->node), sizeof(rx), - &length, (char *)&rx); - if (rc || length != sizeof(struct jffs2_raw_xattr)) { - JFFS2_ERROR("jffs2_flash_write()=%d, req=%zu, wrote=%zu ar %#08x\n", - rc, sizeof(rx), length, ref_offset(xd->node)); - } - } - spin_lock(&c->erase_completion_lock); - xd->node->next_in_ino = NULL; - spin_unlock(&c->erase_completion_lock); - jffs2_mark_node_obsolete(c, xd->node); - xd->node = NULL; -} - -static void delete_xattr_datum(struct jffs2_sb_info *c, struct jffs2_xattr_datum *xd) -{ - /* must be called under down_write(xattr_sem) */ - BUG_ON(xd->refcnt); - - unload_xattr_datum(c, xd); - if (xd->node) { - delete_xattr_datum_node(c, xd); - xd->node = NULL; - } - jffs2_free_xattr_datum(xd); -} - static int do_verify_xattr_datum(struct jffs2_sb_info *c, struct jffs2_xattr_datum *xd) { /* must be called under down_write(xattr_sem) */ struct jffs2_eraseblock *jeb; + struct jffs2_raw_node_ref *raw; struct jffs2_raw_xattr rx; size_t readlen; - uint32_t crc, totlen; + uint32_t crc, offset, totlen; int rc; - BUG_ON(!xd->node); - BUG_ON(ref_flags(xd->node) != REF_UNCHECKED); + spin_lock(&c->erase_completion_lock); + offset = ref_offset(xd->node); + if (ref_flags(xd->node) == REF_PRISTINE) + goto complete; + spin_unlock(&c->erase_completion_lock); - rc = jffs2_flash_read(c, ref_offset(xd->node), sizeof(rx), &readlen, (char *)&rx); + rc = jffs2_flash_read(c, offset, sizeof(rx), &readlen, (char *)&rx); if (rc || readlen != sizeof(rx)) { JFFS2_WARNING("jffs2_flash_read()=%d, req=%zu, read=%zu at %#08x\n", - rc, sizeof(rx), readlen, ref_offset(xd->node)); + rc, sizeof(rx), readlen, offset); return rc ? rc : -EIO; } crc = crc32(0, &rx, sizeof(rx) - 4); if (crc != je32_to_cpu(rx.node_crc)) { - if (je32_to_cpu(rx.node_crc) != 0xffffffff) - JFFS2_ERROR("node CRC failed at %#08x, read=%#08x, calc=%#08x\n", - ref_offset(xd->node), je32_to_cpu(rx.hdr_crc), crc); + JFFS2_ERROR("node CRC failed at %#08x, read=%#08x, calc=%#08x\n", + offset, je32_to_cpu(rx.hdr_crc), crc); + xd->flags |= JFFS2_XFLAGS_INVALID; return EIO; } - totlen = PAD(sizeof(rx) + rx.name_len + 1 + je16_to_cpu(rx.value_len)); + totlen = sizeof(rx); + if (xd->version != XDATUM_DELETE_MARKER) + totlen += rx.name_len + 1 + je16_to_cpu(rx.value_len); + totlen = PAD(totlen); if (je16_to_cpu(rx.magic) != JFFS2_MAGIC_BITMASK || je16_to_cpu(rx.nodetype) != JFFS2_NODETYPE_XATTR || je32_to_cpu(rx.totlen) != totlen @@ -188,11 +165,12 @@ static int do_verify_xattr_datum(struct jffs2_sb_info *c, struct jffs2_xattr_dat || je32_to_cpu(rx.version) != xd->version) { JFFS2_ERROR("inconsistent xdatum at %#08x, magic=%#04x/%#04x, " "nodetype=%#04x/%#04x, totlen=%u/%u, xid=%u/%u, version=%u/%u\n", - ref_offset(xd->node), je16_to_cpu(rx.magic), JFFS2_MAGIC_BITMASK, + offset, je16_to_cpu(rx.magic), JFFS2_MAGIC_BITMASK, je16_to_cpu(rx.nodetype), JFFS2_NODETYPE_XATTR, je32_to_cpu(rx.totlen), totlen, je32_to_cpu(rx.xid), xd->xid, je32_to_cpu(rx.version), xd->version); + xd->flags |= JFFS2_XFLAGS_INVALID; return EIO; } xd->xprefix = rx.xprefix; @@ -200,14 +178,17 @@ static int do_verify_xattr_datum(struct jffs2_sb_info *c, struct jffs2_xattr_dat xd->value_len = je16_to_cpu(rx.value_len); xd->data_crc = je32_to_cpu(rx.data_crc); - /* This JFFS2_NODETYPE_XATTR node is checked */ - jeb = &c->blocks[ref_offset(xd->node) / c->sector_size]; - totlen = PAD(je32_to_cpu(rx.totlen)); - spin_lock(&c->erase_completion_lock); - c->unchecked_size -= totlen; c->used_size += totlen; - jeb->unchecked_size -= totlen; jeb->used_size += totlen; - xd->node->flash_offset = ref_offset(xd->node) | REF_PRISTINE; + complete: + for (raw=xd->node; raw != (void *)xd; raw=raw->next_in_ino) { + jeb = &c->blocks[ref_offset(raw) / c->sector_size]; + totlen = PAD(ref_totlen(c, jeb, raw)); + if (ref_flags(raw) == REF_UNCHECKED) { + c->unchecked_size -= totlen; c->used_size += totlen; + jeb->unchecked_size -= totlen; jeb->used_size += totlen; + } + raw->flash_offset = ref_offset(raw) | ((xd->node==raw) ? REF_PRISTINE : REF_NORMAL); + } spin_unlock(&c->erase_completion_lock); /* unchecked xdatum is chained with c->xattr_unchecked */ @@ -227,7 +208,6 @@ static int do_load_xattr_datum(struct jffs2_sb_info *c, struct jffs2_xattr_datum uint32_t crc, length; int i, ret, retry = 0; - BUG_ON(!xd->node); BUG_ON(ref_flags(xd->node) != REF_PRISTINE); BUG_ON(!list_empty(&xd->xindex)); retry: @@ -253,6 +233,7 @@ static int do_load_xattr_datum(struct jffs2_sb_info *c, struct jffs2_xattr_datum " at %#08x, read: 0x%08x calculated: 0x%08x\n", ref_offset(xd->node), xd->data_crc, crc); kfree(data); + xd->flags |= JFFS2_XFLAGS_INVALID; return EIO; } @@ -286,16 +267,13 @@ static int load_xattr_datum(struct jffs2_sb_info *c, struct jffs2_xattr_datum *x * rc > 0 : Unrecoverable error, this node should be deleted. */ int rc = 0; - BUG_ON(xd->xname); - if (!xd->node) + + if (xd->xname) + return 0; + if (xd->flags & JFFS2_XFLAGS_INVALID) return EIO; - if (unlikely(ref_flags(xd->node) != REF_PRISTINE)) { + if (unlikely(is_xattr_datum_unchecked(c, xd))) rc = do_verify_xattr_datum(c, xd); - if (rc > 0) { - list_del_init(&xd->xindex); - delete_xattr_datum_node(c, xd); - } - } if (!rc) rc = do_load_xattr_datum(c, xd); return rc; @@ -304,36 +282,43 @@ static int load_xattr_datum(struct jffs2_sb_info *c, struct jffs2_xattr_datum *x static int save_xattr_datum(struct jffs2_sb_info *c, struct jffs2_xattr_datum *xd) { /* must be called under down_write(xattr_sem) */ - struct jffs2_raw_node_ref *raw; struct jffs2_raw_xattr rx; struct kvec vecs[2]; size_t length; - int rc, totlen; + int rc, totlen, nvecs = 1; uint32_t phys_ofs = write_ofs(c); - BUG_ON(!xd->xname); + BUG_ON(is_xattr_datum_dead(xd) || (xd->flags & JFFS2_XFLAGS_INVALID) + ? !!xd->xname : !xd->xname); vecs[0].iov_base = ℞ - vecs[0].iov_len = PAD(sizeof(rx)); - vecs[1].iov_base = xd->xname; - vecs[1].iov_len = xd->name_len + 1 + xd->value_len; - totlen = vecs[0].iov_len + vecs[1].iov_len; - + vecs[0].iov_len = totlen = sizeof(rx); + if (!is_xattr_datum_dead(xd) && !(xd->flags & JFFS2_XFLAGS_INVALID)) { + nvecs++; + vecs[1].iov_base = xd->xname; + vecs[1].iov_len = xd->name_len + 1 + xd->value_len; + totlen += vecs[1].iov_len; + } /* Setup raw-xattr */ + memset(&rx, 0, sizeof(rx)); rx.magic = cpu_to_je16(JFFS2_MAGIC_BITMASK); rx.nodetype = cpu_to_je16(JFFS2_NODETYPE_XATTR); rx.totlen = cpu_to_je32(PAD(totlen)); rx.hdr_crc = cpu_to_je32(crc32(0, &rx, sizeof(struct jffs2_unknown_node) - 4)); rx.xid = cpu_to_je32(xd->xid); - rx.version = cpu_to_je32(++xd->version); - rx.xprefix = xd->xprefix; - rx.name_len = xd->name_len; - rx.value_len = cpu_to_je16(xd->value_len); - rx.data_crc = cpu_to_je32(crc32(0, vecs[1].iov_base, vecs[1].iov_len)); + if (!is_xattr_datum_dead(xd) && !(xd->flags & JFFS2_XFLAGS_INVALID)) { + rx.version = cpu_to_je32(++xd->version); + rx.xprefix = xd->xprefix; + rx.name_len = xd->name_len; + rx.value_len = cpu_to_je16(xd->value_len); + rx.data_crc = cpu_to_je32(crc32(0, vecs[1].iov_base, vecs[1].iov_len)); + } else { + rx.version = cpu_to_je32(XDATUM_DELETE_MARKER); + } rx.node_crc = cpu_to_je32(crc32(0, &rx, sizeof(struct jffs2_raw_xattr) - 4)); - rc = jffs2_flash_writev(c, vecs, 2, phys_ofs, &length, 0); + rc = jffs2_flash_writev(c, vecs, nvecs, phys_ofs, &length, 0); if (rc || totlen != length) { JFFS2_WARNING("jffs2_flash_writev()=%d, req=%u, wrote=%zu, at %#08x\n", rc, totlen, length, phys_ofs); @@ -343,14 +328,8 @@ static int save_xattr_datum(struct jffs2_sb_info *c, struct jffs2_xattr_datum *x return rc; } - /* success */ - raw = jffs2_add_physical_node_ref(c, phys_ofs | REF_PRISTINE, PAD(totlen), NULL); - /* FIXME */ raw->next_in_ino = (void *)xd; - - if (xd->node) - delete_xattr_datum_node(c, xd); - xd->node = raw; + jffs2_add_physical_node_ref(c, phys_ofs | REF_PRISTINE, PAD(totlen), (void *)xd); dbg_xattr("success on saving xdatum (xid=%u, version=%u, xprefix=%u, xname='%s')\n", xd->xid, xd->version, xd->xprefix, xd->xname); @@ -426,6 +405,39 @@ static struct jffs2_xattr_datum *create_xattr_datum(struct jffs2_sb_info *c, return xd; } +static void delete_xattr_datum_delay(struct jffs2_sb_info *c, struct jffs2_xattr_datum *xd) +{ + /* must be called under down_write(xattr_sem) */ + BUG_ON(xd->refcnt); + + unload_xattr_datum(c, xd); + set_xattr_datum_dead(xd); + spin_lock(&c->erase_completion_lock); + list_add(&xd->xindex, &c->xattr_dead_list); + spin_unlock(&c->erase_completion_lock); + JFFS2_NOTICE("xdatum(xid=%u) was removed without delete marker. " + "An orphan xdatum may be detected on next mounting.\n", xd->xid); +} + +static void delete_xattr_datum(struct jffs2_sb_info *c, struct jffs2_xattr_datum *xd) +{ + /* must be called under jffs2_reserve_space() and down_write(xattr_sem) */ + int rc; + BUG_ON(xd->refcnt); + + unload_xattr_datum(c, xd); + set_xattr_datum_dead(xd); + rc = save_xattr_datum(c, xd); + if (rc) { + JFFS2_NOTICE("xdatum(xid=%u) was removed without delete marker. " + "An orphan xdatum may be detected on next mounting.\n", + xd->xid); + } + spin_lock(&c->erase_completion_lock); + list_add(&xd->xindex, &c->xattr_dead_list); + spin_unlock(&c->erase_completion_lock); +} + /* -------- xref related functions ------------------ * verify_xattr_ref(c, ref) * is used to load xref information from medium. Because summary data does not @@ -450,25 +462,29 @@ static struct jffs2_xattr_datum *create_xattr_datum(struct jffs2_sb_info *c, static int verify_xattr_ref(struct jffs2_sb_info *c, struct jffs2_xattr_ref *ref) { struct jffs2_eraseblock *jeb; + struct jffs2_raw_node_ref *raw; struct jffs2_raw_xref rr; size_t readlen; - uint32_t crc, totlen; + uint32_t crc, offset, totlen; int rc; - BUG_ON(ref_flags(ref->node) != REF_UNCHECKED); + spin_lock(&c->erase_completion_lock); + if (ref_flags(ref->node) != REF_UNCHECKED) + goto complete; + offset = ref_offset(ref->node); + spin_unlock(&c->erase_completion_lock); - rc = jffs2_flash_read(c, ref_offset(ref->node), sizeof(rr), &readlen, (char *)&rr); + rc = jffs2_flash_read(c, offset, sizeof(rr), &readlen, (char *)&rr); if (rc || sizeof(rr) != readlen) { JFFS2_WARNING("jffs2_flash_read()=%d, req=%zu, read=%zu, at %#08x\n", - rc, sizeof(rr), readlen, ref_offset(ref->node)); + rc, sizeof(rr), readlen, offset); return rc ? rc : -EIO; } /* obsolete node */ crc = crc32(0, &rr, sizeof(rr) - 4); if (crc != je32_to_cpu(rr.node_crc)) { - if (je32_to_cpu(rr.node_crc) != 0xffffffff) - JFFS2_ERROR("node CRC failed at %#08x, read=%#08x, calc=%#08x\n", - ref_offset(ref->node), je32_to_cpu(rr.node_crc), crc); + JFFS2_ERROR("node CRC failed at %#08x, read=%#08x, calc=%#08x\n", + offset, je32_to_cpu(rr.node_crc), crc); return EIO; } if (je16_to_cpu(rr.magic) != JFFS2_MAGIC_BITMASK @@ -476,22 +492,28 @@ static int verify_xattr_ref(struct jffs2_sb_info *c, struct jffs2_xattr_ref *ref || je32_to_cpu(rr.totlen) != PAD(sizeof(rr))) { JFFS2_ERROR("inconsistent xref at %#08x, magic=%#04x/%#04x, " "nodetype=%#04x/%#04x, totlen=%u/%zu\n", - ref_offset(ref->node), je16_to_cpu(rr.magic), JFFS2_MAGIC_BITMASK, + offset, je16_to_cpu(rr.magic), JFFS2_MAGIC_BITMASK, je16_to_cpu(rr.nodetype), JFFS2_NODETYPE_XREF, je32_to_cpu(rr.totlen), PAD(sizeof(rr))); return EIO; } ref->ino = je32_to_cpu(rr.ino); ref->xid = je32_to_cpu(rr.xid); - - /* fixup superblock/eraseblock info */ - jeb = &c->blocks[ref_offset(ref->node) / c->sector_size]; - totlen = PAD(sizeof(rr)); + ref->xseqno = je32_to_cpu(rr.xseqno); + if (ref->xseqno > c->highest_xseqno) + c->highest_xseqno = (ref->xseqno & ~XREF_DELETE_MARKER); spin_lock(&c->erase_completion_lock); - c->unchecked_size -= totlen; c->used_size += totlen; - jeb->unchecked_size -= totlen; jeb->used_size += totlen; - ref->node->flash_offset = ref_offset(ref->node) | REF_PRISTINE; + complete: + for (raw=ref->node; raw != (void *)ref; raw=raw->next_in_ino) { + jeb = &c->blocks[ref_offset(raw) / c->sector_size]; + totlen = PAD(ref_totlen(c, jeb, raw)); + if (ref_flags(raw) == REF_UNCHECKED) { + c->unchecked_size -= totlen; c->used_size += totlen; + jeb->unchecked_size -= totlen; jeb->used_size += totlen; + } + raw->flash_offset = ref_offset(raw) | ((ref->node==raw) ? REF_PRISTINE : REF_NORMAL); + } spin_unlock(&c->erase_completion_lock); dbg_xattr("success on verifying xref (ino=%u, xid=%u) at %#08x\n", @@ -499,58 +521,12 @@ static int verify_xattr_ref(struct jffs2_sb_info *c, struct jffs2_xattr_ref *ref return 0; } -static void delete_xattr_ref_node(struct jffs2_sb_info *c, struct jffs2_xattr_ref *ref) -{ - struct jffs2_raw_xref rr; - size_t length; - int rc; - - if (jffs2_sum_active()) { - memset(&rr, 0xff, sizeof(rr)); - rc = jffs2_flash_read(c, ref_offset(ref->node), - sizeof(struct jffs2_unknown_node), - &length, (char *)&rr); - if (rc || length != sizeof(struct jffs2_unknown_node)) { - JFFS2_ERROR("jffs2_flash_read()=%d, req=%zu, read=%zu at %#08x\n", - rc, sizeof(struct jffs2_unknown_node), - length, ref_offset(ref->node)); - } - rc = jffs2_flash_write(c, ref_offset(ref->node), sizeof(rr), - &length, (char *)&rr); - if (rc || length != sizeof(struct jffs2_raw_xref)) { - JFFS2_ERROR("jffs2_flash_write()=%d, req=%zu, wrote=%zu at %#08x\n", - rc, sizeof(rr), length, ref_offset(ref->node)); - } - } - spin_lock(&c->erase_completion_lock); - ref->node->next_in_ino = NULL; - spin_unlock(&c->erase_completion_lock); - jffs2_mark_node_obsolete(c, ref->node); - ref->node = NULL; -} - -static void delete_xattr_ref(struct jffs2_sb_info *c, struct jffs2_xattr_ref *ref) -{ - /* must be called under down_write(xattr_sem) */ - struct jffs2_xattr_datum *xd; - - BUG_ON(!ref->node); - delete_xattr_ref_node(c, ref); - - xd = ref->xd; - xd->refcnt--; - if (!xd->refcnt) - delete_xattr_datum(c, xd); - jffs2_free_xattr_ref(ref); -} - static int save_xattr_ref(struct jffs2_sb_info *c, struct jffs2_xattr_ref *ref) { /* must be called under down_write(xattr_sem) */ - struct jffs2_raw_node_ref *raw; struct jffs2_raw_xref rr; size_t length; - uint32_t phys_ofs = write_ofs(c); + uint32_t xseqno, phys_ofs = write_ofs(c); int ret; rr.magic = cpu_to_je16(JFFS2_MAGIC_BITMASK); @@ -558,8 +534,16 @@ static int save_xattr_ref(struct jffs2_sb_info *c, struct jffs2_xattr_ref *ref) rr.totlen = cpu_to_je32(PAD(sizeof(rr))); rr.hdr_crc = cpu_to_je32(crc32(0, &rr, sizeof(struct jffs2_unknown_node) - 4)); - rr.ino = cpu_to_je32(ref->ic->ino); - rr.xid = cpu_to_je32(ref->xd->xid); + xseqno = (c->highest_xseqno += 2); + if (is_xattr_ref_dead(ref)) { + xseqno |= XREF_DELETE_MARKER; + rr.ino = cpu_to_je32(ref->ino); + rr.xid = cpu_to_je32(ref->xid); + } else { + rr.ino = cpu_to_je32(ref->ic->ino); + rr.xid = cpu_to_je32(ref->xd->xid); + } + rr.xseqno = cpu_to_je32(xseqno); rr.node_crc = cpu_to_je32(crc32(0, &rr, sizeof(rr) - 4)); ret = jffs2_flash_write(c, phys_ofs, sizeof(rr), &length, (char *)&rr); @@ -572,12 +556,9 @@ static int save_xattr_ref(struct jffs2_sb_info *c, struct jffs2_xattr_ref *ref) return ret; } - - raw = jffs2_add_physical_node_ref(c, phys_ofs | REF_PRISTINE, PAD(sizeof(rr)), NULL); - /* FIXME */ raw->next_in_ino = (void *)ref; - if (ref->node) - delete_xattr_ref_node(c, ref); - ref->node = raw; + /* success */ + ref->xseqno = xseqno; + jffs2_add_physical_node_ref(c, phys_ofs | REF_PRISTINE, PAD(sizeof(rr)), (void *)ref); dbg_xattr("success on saving xref (ino=%u, xid=%u)\n", ref->ic->ino, ref->xd->xid); @@ -610,22 +591,116 @@ static struct jffs2_xattr_ref *create_xattr_ref(struct jffs2_sb_info *c, struct return ref; /* success */ } +static void delete_xattr_ref_delay(struct jffs2_sb_info *c, struct jffs2_xattr_ref *ref) +{ + /* must be called under down_write(xattr_sem) */ + struct jffs2_xattr_datum *xd; + + set_xattr_ref_dead(ref); + xd = ref->xd; + ref->ino = ref->ic->ino; + ref->xid = ref->xd->xid; + spin_lock(&c->erase_completion_lock); + ref->next = c->xref_dead_list; + c->xref_dead_list = ref; + spin_unlock(&c->erase_completion_lock); + + JFFS2_NOTICE("xref(ino=%u, xid=%u) was removed without delete marker. " + "An orphan xref may be detected on next mounting.\n", + ref->ino, ref->xid); + + if (!--xd->refcnt) + delete_xattr_datum_delay(c, xd); +} + +static int delete_xattr_ref(struct jffs2_sb_info *c, struct jffs2_xattr_ref *ref, int enforce) +{ + /* must be called under jffs2_reserve_space() and down_write(xattr_sem) */ + struct jffs2_inode_cache *ic; + struct jffs2_xattr_datum *xd; + uint32_t length; + int rc; + + set_xattr_ref_dead(ref); + ic = ref->ic; + xd = ref->xd; + ref->ino = ic->ino; + ref->xid = xd->xid; + rc = save_xattr_ref(c, ref); + if (rc) { + if (!enforce) { + clr_xattr_ref_dead(ref); + ref->ic = ic; + ref->xd = xd; + return rc; + } + JFFS2_WARNING("could not write delete marker of xref(ino=%u, xid=%u). " + "An orphan xref may be detected on next mounting.\n", + ref->ic->ino, ref->xd->xid); + } + spin_lock(&c->erase_completion_lock); + ref->next = c->xref_dead_list; + c->xref_dead_list = ref; + spin_unlock(&c->erase_completion_lock); + + xd->refcnt--; + if (xd->refcnt) + return 0; + + /* delete xdatum */ + unload_xattr_datum(c, xd); + up_write(&c->xattr_sem); + jffs2_complete_reservation(c); + + rc = jffs2_reserve_space(c, PAD(sizeof(struct jffs2_raw_xattr)), &length, + ALLOC_DELETION, JFFS2_SUMMARY_XATTR_SIZE); + if (rc) { + down(&c->alloc_sem); + down_write(&c->xattr_sem); + delete_xattr_datum_delay(c, xd); + } else { + down_write(&c->xattr_sem); + delete_xattr_datum(c, xd); + } + return 0; +} + void jffs2_xattr_delete_inode(struct jffs2_sb_info *c, struct jffs2_inode_cache *ic) { /* It's called from jffs2_clear_inode() on inode removing. When an inode with XATTR is removed, those XATTRs must be removed. */ - struct jffs2_xattr_ref *ref, *_ref; + struct jffs2_xattr_ref *ref; + uint32_t length; + int rc, retry; if (!ic || ic->nlink > 0) return; + down_read(&c->xattr_sem); + if (!ic->xref) { + up_read(&c->xattr_sem); + return; + } + up_read(&c->xattr_sem); + retry: + rc = jffs2_reserve_space(c, PAD(sizeof(struct jffs2_raw_xref)), &length, + ALLOC_DELETION, JFFS2_SUMMARY_XREF_SIZE); down_write(&c->xattr_sem); - for (ref = ic->xref; ref; ref = _ref) { - _ref = ref->next; - delete_xattr_ref(c, ref); + if (ic->xref) { + ref = ic->xref; + ic->xref = ref->next; + if (rc) { + delete_xattr_ref_delay(c, ref); + } else { + delete_xattr_ref(c, ref, 1); + } } - ic->xref = NULL; + retry = ic->xref ? 1 : 0; up_write(&c->xattr_sem); + if (!rc) + jffs2_complete_reservation(c); + if (retry) + goto retry; } void jffs2_xattr_free_inode(struct jffs2_sb_info *c, struct jffs2_inode_cache *ic) @@ -655,7 +730,7 @@ static int check_xattr_ref_inode(struct jffs2_sb_info *c, struct jffs2_inode_cac * duplicate name/value pairs. If duplicate name/value pair would be found, * one will be removed. */ - struct jffs2_xattr_ref *ref, *cmp, **pref; + struct jffs2_xattr_ref *ref, *cmp, **pref, **pcmp; int rc = 0; if (likely(ic->flags & INO_FLAGS_XATTR_CHECKED)) @@ -668,27 +743,32 @@ static int check_xattr_ref_inode(struct jffs2_sb_info *c, struct jffs2_inode_cac rc = load_xattr_datum(c, ref->xd); if (unlikely(rc > 0)) { *pref = ref->next; - delete_xattr_ref(c, ref); + delete_xattr_ref_delay(c, ref); goto retry; } else if (unlikely(rc < 0)) goto out; } - for (cmp=ref->next, pref=&ref->next; cmp; pref=&cmp->next, cmp=cmp->next) { + for (cmp=ref->next, pcmp=&ref->next; cmp; pcmp=&cmp->next, cmp=cmp->next) { if (!cmp->xd->xname) { ref->xd->flags |= JFFS2_XFLAGS_BIND; rc = load_xattr_datum(c, cmp->xd); ref->xd->flags &= ~JFFS2_XFLAGS_BIND; if (unlikely(rc > 0)) { - *pref = cmp->next; - delete_xattr_ref(c, cmp); + *pcmp = cmp->next; + delete_xattr_ref_delay(c, cmp); goto retry; } else if (unlikely(rc < 0)) goto out; } if (ref->xd->xprefix == cmp->xd->xprefix && !strcmp(ref->xd->xname, cmp->xd->xname)) { - *pref = cmp->next; - delete_xattr_ref(c, cmp); + if (ref->xseqno > cmp->xseqno) { + *pcmp = cmp->next; + delete_xattr_ref_delay(c, cmp); + } else { + *pref = ref->next; + delete_xattr_ref_delay(c, ref); + } goto retry; } } @@ -719,9 +799,13 @@ void jffs2_init_xattr_subsystem(struct jffs2_sb_info *c) for (i=0; i < XATTRINDEX_HASHSIZE; i++) INIT_LIST_HEAD(&c->xattrindex[i]); INIT_LIST_HEAD(&c->xattr_unchecked); + INIT_LIST_HEAD(&c->xattr_dead_list); + c->xref_dead_list = NULL; c->xref_temp = NULL; init_rwsem(&c->xattr_sem); + c->highest_xid = 0; + c->highest_xseqno = 0; c->xdatum_mem_usage = 0; c->xdatum_mem_threshold = 32 * 1024; /* Default 32KB */ } @@ -751,7 +835,11 @@ void jffs2_clear_xattr_subsystem(struct jffs2_sb_info *c) _ref = ref->next; jffs2_free_xattr_ref(ref); } - c->xref_temp = NULL; + + for (ref=c->xref_dead_list; ref; ref = _ref) { + _ref = ref->next; + jffs2_free_xattr_ref(ref); + } for (i=0; i < XATTRINDEX_HASHSIZE; i++) { list_for_each_entry_safe(xd, _xd, &c->xattrindex[i], xindex) { @@ -761,64 +849,120 @@ void jffs2_clear_xattr_subsystem(struct jffs2_sb_info *c) jffs2_free_xattr_datum(xd); } } + + list_for_each_entry_safe(xd, _xd, &c->xattr_dead_list, xindex) { + list_del(&xd->xindex); + jffs2_free_xattr_datum(xd); + } } +#define XREF_TMPHASH_SIZE (128) void jffs2_build_xattr_subsystem(struct jffs2_sb_info *c) { struct jffs2_xattr_ref *ref, *_ref; + struct jffs2_xattr_ref *xref_tmphash[XREF_TMPHASH_SIZE]; struct jffs2_xattr_datum *xd, *_xd; struct jffs2_inode_cache *ic; - int i, xdatum_count =0, xdatum_unchecked_count = 0, xref_count = 0; + struct jffs2_raw_node_ref *raw; + int i, xdatum_count = 0, xdatum_unchecked_count = 0, xref_count = 0; BUG_ON(!(c->flags & JFFS2_SB_FLAG_BUILDING)); + /* Phase.1 : Drop dead xdatum */ + for (i=0; i < XATTRINDEX_HASHSIZE; i++) { + list_for_each_entry_safe(xd, _xd, &c->xattrindex[i], xindex) { + BUG_ON(xd->node == (void *)xd); + if (is_xattr_datum_dead(xd)) { + list_del_init(&xd->xindex); + list_add(&xd->xindex, &c->xattr_unchecked); + } + } + } - /* Phase.1 */ + /* Phase.2 : Merge same xref */ + for (i=0; i < XREF_TMPHASH_SIZE; i++) + xref_tmphash[i] = NULL; for (ref=c->xref_temp; ref; ref=_ref) { + struct jffs2_xattr_ref *tmp; + _ref = ref->next; - /* checking REF_UNCHECKED nodes */ if (ref_flags(ref->node) != REF_PRISTINE) { if (verify_xattr_ref(c, ref)) { - delete_xattr_ref_node(c, ref); + BUG_ON(ref->node->next_in_ino != (void *)ref); + ref->node->next_in_ino = NULL; + jffs2_mark_node_obsolete(c, ref->node); jffs2_free_xattr_ref(ref); continue; } } - /* At this point, ref->xid and ref->ino contain XID and inode number. - ref->xd and ref->ic are not valid yet. */ - xd = jffs2_find_xattr_datum(c, ref->xid); - ic = jffs2_get_ino_cache(c, ref->ino); - if (!xd || !ic) { - if (ref_flags(ref->node) != REF_UNCHECKED) - JFFS2_WARNING("xref(ino=%u, xid=%u) is orphan. \n", - ref->ino, ref->xid); - delete_xattr_ref_node(c, ref); + + i = (ref->ino ^ ref->xid) % XREF_TMPHASH_SIZE; + for (tmp=xref_tmphash[i]; tmp; tmp=tmp->next) { + if (tmp->ino == ref->ino && tmp->xid == ref->xid) + break; + } + if (tmp) { + raw = ref->node; + if (ref->xseqno > tmp->xseqno) { + tmp->xseqno = ref->xseqno; + raw->next_in_ino = tmp->node; + tmp->node = raw; + } else { + raw->next_in_ino = tmp->node->next_in_ino; + tmp->node->next_in_ino = raw; + } jffs2_free_xattr_ref(ref); continue; + } else { + ref->next = xref_tmphash[i]; + xref_tmphash[i] = ref; } - ref->xd = xd; - ref->ic = ic; - xd->refcnt++; - ref->next = ic->xref; - ic->xref = ref; - xref_count++; } c->xref_temp = NULL; - /* After this, ref->xid/ino are NEVER used. */ - /* Phase.2 */ + /* Phase.3 : Bind xref with inode_cache and xattr_datum */ + for (i=0; i < XREF_TMPHASH_SIZE; i++) { + for (ref=xref_tmphash[i]; ref; ref=_ref) { + _ref = ref->next; + if (is_xattr_ref_dead(ref)) { + ref->next = c->xref_dead_list; + c->xref_dead_list = ref; + continue; + } + /* At this point, ref->xid and ref->ino contain XID and inode number. + ref->xd and ref->ic are not valid yet. */ + xd = jffs2_find_xattr_datum(c, ref->xid); + ic = jffs2_get_ino_cache(c, ref->ino); + if (!xd || !ic) { + JFFS2_WARNING("xref(ino=%u, xid=%u, xseqno=%u) is orphan. \n", + ref->ino, ref->xid, ref->xseqno); + set_xattr_ref_dead(ref); + ref->next = c->xref_dead_list; + c->xref_dead_list = ref; + continue; + } + ref->xd = xd; + ref->ic = ic; + xd->refcnt++; + ref->next = ic->xref; + ic->xref = ref; + xref_count++; + } + } + + /* Phase.4 : Link unchecked xdatum to xattr_unchecked list */ for (i=0; i < XATTRINDEX_HASHSIZE; i++) { list_for_each_entry_safe(xd, _xd, &c->xattrindex[i], xindex) { list_del_init(&xd->xindex); if (!xd->refcnt) { - if (ref_flags(xd->node) != REF_UNCHECKED) - JFFS2_WARNING("orphan xdatum(xid=%u, version=%u) at %#08x\n", - xd->xid, xd->version, ref_offset(xd->node)); - delete_xattr_datum(c, xd); + JFFS2_WARNING("orphan xdatum(xid=%u, version=%u)\n", + xd->xid, xd->version); + set_xattr_datum_dead(xd); + list_add(&xd->xindex, &c->xattr_unchecked); continue; } - if (ref_flags(xd->node) != REF_PRISTINE) { - dbg_xattr("unchecked xdatum(xid=%u) at %#08x\n", - xd->xid, ref_offset(xd->node)); + if (is_xattr_datum_unchecked(c, xd)) { + dbg_xattr("unchecked xdatum(xid=%u, version=%u)\n", + xd->xid, xd->version); list_add(&xd->xindex, &c->xattr_unchecked); xdatum_unchecked_count++; } @@ -833,28 +977,18 @@ void jffs2_build_xattr_subsystem(struct jffs2_sb_info *c) struct jffs2_xattr_datum *jffs2_setup_xattr_datum(struct jffs2_sb_info *c, uint32_t xid, uint32_t version) { - struct jffs2_xattr_datum *xd, *_xd; + struct jffs2_xattr_datum *xd; - _xd = jffs2_find_xattr_datum(c, xid); - if (_xd) { - dbg_xattr("duplicate xdatum (xid=%u, version=%u/%u) at %#08x\n", - xid, version, _xd->version, ref_offset(_xd->node)); - if (version < _xd->version) - return ERR_PTR(-EEXIST); - } - xd = jffs2_alloc_xattr_datum(); - if (!xd) - return ERR_PTR(-ENOMEM); - xd->xid = xid; - xd->version = version; - if (xd->xid > c->highest_xid) - c->highest_xid = xd->xid; - list_add_tail(&xd->xindex, &c->xattrindex[xid % XATTRINDEX_HASHSIZE]); - - if (_xd) { - list_del_init(&_xd->xindex); - delete_xattr_datum_node(c, _xd); - jffs2_free_xattr_datum(_xd); + xd = jffs2_find_xattr_datum(c, xid); + if (!xd) { + xd = jffs2_alloc_xattr_datum(); + if (!xd) + return ERR_PTR(-ENOMEM); + xd->xid = xid; + xd->version = version; + if (xd->xid > c->highest_xid) + c->highest_xid = xd->xid; + list_add_tail(&xd->xindex, &c->xattrindex[xid % XATTRINDEX_HASHSIZE]); } return xd; } @@ -945,7 +1079,7 @@ ssize_t jffs2_listxattr(struct dentry *dentry, char *buffer, size_t size) rc = load_xattr_datum(c, xd); if (unlikely(rc > 0)) { *pref = ref->next; - delete_xattr_ref(c, ref); + delete_xattr_ref_delay(c, ref); goto retry; } else if (unlikely(rc < 0)) goto out; @@ -1006,7 +1140,7 @@ int do_jffs2_getxattr(struct inode *inode, int xprefix, const char *xname, rc = load_xattr_datum(c, xd); if (unlikely(rc > 0)) { *pref = ref->next; - delete_xattr_ref(c, ref); + delete_xattr_ref_delay(c, ref); goto retry; } else if (unlikely(rc < 0)) { goto out; @@ -1069,7 +1203,7 @@ int do_jffs2_setxattr(struct inode *inode, int xprefix, const char *xname, rc = load_xattr_datum(c, xd); if (unlikely(rc > 0)) { *pref = ref->next; - delete_xattr_ref(c, ref); + delete_xattr_ref_delay(c, ref); goto retry; } else if (unlikely(rc < 0)) goto out; @@ -1081,8 +1215,7 @@ int do_jffs2_setxattr(struct inode *inode, int xprefix, const char *xname, } if (!buffer) { *pref = ref->next; - delete_xattr_ref(c, ref); - rc = 0; + rc = delete_xattr_ref(c, ref, 0); goto out; } goto found; @@ -1094,7 +1227,7 @@ int do_jffs2_setxattr(struct inode *inode, int xprefix, const char *xname, goto out; } if (!buffer) { - rc = -EINVAL; + rc = -ENODATA; goto out; } found: @@ -1110,16 +1243,15 @@ int do_jffs2_setxattr(struct inode *inode, int xprefix, const char *xname, request = PAD(sizeof(struct jffs2_raw_xref)); rc = jffs2_reserve_space(c, request, &length, ALLOC_NORMAL, JFFS2_SUMMARY_XREF_SIZE); + down_write(&c->xattr_sem); if (rc) { JFFS2_WARNING("jffs2_reserve_space()=%d, request=%u\n", rc, request); - down_write(&c->xattr_sem); xd->refcnt--; if (!xd->refcnt) - delete_xattr_datum(c, xd); + delete_xattr_datum_delay(c, xd); up_write(&c->xattr_sem); return rc; } - down_write(&c->xattr_sem); if (ref) *pref = ref->next; newref = create_xattr_ref(c, ic, xd); @@ -1131,9 +1263,21 @@ int do_jffs2_setxattr(struct inode *inode, int xprefix, const char *xname, rc = PTR_ERR(newref); xd->refcnt--; if (!xd->refcnt) - delete_xattr_datum(c, xd); + delete_xattr_datum_delay(c, xd); } else if (ref) { - delete_xattr_ref(c, ref); + up_write(&c->xattr_sem); + jffs2_complete_reservation(c); + + rc = jffs2_reserve_space(c, request, &length, + ALLOC_DELETION, JFFS2_SUMMARY_XREF_SIZE); + down_write(&c->xattr_sem); + if (rc) { + JFFS2_WARNING("jffs2_reserve_space()=%d, request=%u\n", rc, request); + delete_xattr_ref_delay(c, ref); + up_write(&c->xattr_sem); + return 0; + } + delete_xattr_ref(c, ref, 1); } out: up_write(&c->xattr_sem); @@ -1142,38 +1286,37 @@ int do_jffs2_setxattr(struct inode *inode, int xprefix, const char *xname, } /* -------- garbage collector functions ------------- - * jffs2_garbage_collect_xattr_datum(c, xd) + * jffs2_garbage_collect_xattr_datum(c, xd, raw) * is used to move xdatum into new node. - * jffs2_garbage_collect_xattr_ref(c, ref) + * jffs2_garbage_collect_xattr_ref(c, ref, raw) * is used to move xref into new node. * jffs2_verify_xattr(c) * is used to call do_verify_xattr_datum() before garbage collecting. * -------------------------------------------------- */ -int jffs2_garbage_collect_xattr_datum(struct jffs2_sb_info *c, struct jffs2_xattr_datum *xd) +int jffs2_garbage_collect_xattr_datum(struct jffs2_sb_info *c, struct jffs2_xattr_datum *xd, + struct jffs2_raw_node_ref *raw) { uint32_t totlen, length, old_ofs; - int rc = -EINVAL; + int rc = 0; down_write(&c->xattr_sem); - BUG_ON(!xd->node); + if (xd->node != raw) + goto out; + if (is_xattr_datum_dead(xd) && (raw->next_in_ino == (void *)xd)) + goto out; old_ofs = ref_offset(xd->node); totlen = ref_totlen(c, c->gcblock, xd->node); - if (totlen < sizeof(struct jffs2_raw_xattr)) - goto out; - if (!xd->xname) { + if (!is_xattr_datum_dead(xd)) { rc = load_xattr_datum(c, xd); - if (unlikely(rc > 0)) { - delete_xattr_datum_node(c, xd); - rc = 0; - goto out; - } else if (unlikely(rc < 0)) + if (unlikely(rc < 0)) goto out; } + rc = jffs2_reserve_space_gc(c, totlen, &length, JFFS2_SUMMARY_XATTR_SIZE); - if (rc || length < totlen) { - JFFS2_WARNING("jffs2_reserve_space()=%d, request=%u\n", rc, totlen); + if (rc) { + JFFS2_WARNING("jffs2_reserve_space_gc()=%d, request=%u\n", rc, totlen); rc = rc ? rc : -EBADFD; goto out; } @@ -1182,27 +1325,33 @@ int jffs2_garbage_collect_xattr_datum(struct jffs2_sb_info *c, struct jffs2_xatt dbg_xattr("xdatum (xid=%u, version=%u) GC'ed from %#08x to %08x\n", xd->xid, xd->version, old_ofs, ref_offset(xd->node)); out: + if (!rc) + jffs2_mark_node_obsolete(c, raw); up_write(&c->xattr_sem); return rc; } -int jffs2_garbage_collect_xattr_ref(struct jffs2_sb_info *c, struct jffs2_xattr_ref *ref) +int jffs2_garbage_collect_xattr_ref(struct jffs2_sb_info *c, struct jffs2_xattr_ref *ref, + struct jffs2_raw_node_ref *raw) { uint32_t totlen, length, old_ofs; - int rc = -EINVAL; + int rc = 0; down_write(&c->xattr_sem); BUG_ON(!ref->node); + if (ref->node != raw) + goto out; + if (is_xattr_ref_dead(ref) && (raw->next_in_ino == (void *)ref)) + goto out; + old_ofs = ref_offset(ref->node); totlen = ref_totlen(c, c->gcblock, ref->node); - if (totlen != sizeof(struct jffs2_raw_xref)) - goto out; rc = jffs2_reserve_space_gc(c, totlen, &length, JFFS2_SUMMARY_XREF_SIZE); - if (rc || length < totlen) { - JFFS2_WARNING("%s: jffs2_reserve_space() = %d, request = %u\n", + if (rc) { + JFFS2_WARNING("%s: jffs2_reserve_space_gc() = %d, request = %u\n", __FUNCTION__, rc, totlen); rc = rc ? rc : -EBADFD; goto out; @@ -1212,6 +1361,8 @@ int jffs2_garbage_collect_xattr_ref(struct jffs2_sb_info *c, struct jffs2_xattr_ dbg_xattr("xref (ino=%u, xid=%u) GC'ed from %#08x to %08x\n", ref->ic->ino, ref->xd->xid, old_ofs, ref_offset(ref->node)); out: + if (!rc) + jffs2_mark_node_obsolete(c, raw); up_write(&c->xattr_sem); return rc; } @@ -1219,20 +1370,59 @@ int jffs2_garbage_collect_xattr_ref(struct jffs2_sb_info *c, struct jffs2_xattr_ int jffs2_verify_xattr(struct jffs2_sb_info *c) { struct jffs2_xattr_datum *xd, *_xd; + struct jffs2_eraseblock *jeb; + struct jffs2_raw_node_ref *raw; + uint32_t totlen; int rc; down_write(&c->xattr_sem); list_for_each_entry_safe(xd, _xd, &c->xattr_unchecked, xindex) { rc = do_verify_xattr_datum(c, xd); - if (rc == 0) { - list_del_init(&xd->xindex); - break; - } else if (rc > 0) { - list_del_init(&xd->xindex); - delete_xattr_datum_node(c, xd); + if (rc < 0) + continue; + list_del_init(&xd->xindex); + spin_lock(&c->erase_completion_lock); + for (raw=xd->node; raw != (void *)xd; raw=raw->next_in_ino) { + if (ref_flags(raw) != REF_UNCHECKED) + continue; + jeb = &c->blocks[ref_offset(raw) / c->sector_size]; + totlen = PAD(ref_totlen(c, jeb, raw)); + c->unchecked_size -= totlen; c->used_size += totlen; + jeb->unchecked_size -= totlen; jeb->used_size += totlen; + raw->flash_offset = ref_offset(raw) + | ((xd->node == (void *)raw) ? REF_PRISTINE : REF_NORMAL); } + if (is_xattr_datum_dead(xd)) + list_add(&xd->xindex, &c->xattr_dead_list); + spin_unlock(&c->erase_completion_lock); } up_write(&c->xattr_sem); - return list_empty(&c->xattr_unchecked) ? 1 : 0; } + +void jffs2_release_xattr_datum(struct jffs2_sb_info *c, struct jffs2_xattr_datum *xd) +{ + /* must be called under spin_lock(&c->erase_completion_lock) */ + if (xd->node != (void *)xd) + return; + + list_del(&xd->xindex); + jffs2_free_xattr_datum(xd); +} + +void jffs2_release_xattr_ref(struct jffs2_sb_info *c, struct jffs2_xattr_ref *ref) +{ + /* must be called under spin_lock(&c->erase_completion_lock) */ + struct jffs2_xattr_ref *tmp, **ptmp; + + if (ref->node != (void *)ref) + return; + + for (tmp=c->xref_dead_list, ptmp=&c->xref_dead_list; tmp; ptmp=&tmp->next, tmp=tmp->next) { + if (ref == tmp) { + *ptmp = tmp->next; + jffs2_free_xattr_ref(ref); + break; + } + } +} diff --git a/fs/jffs2/xattr.h b/fs/jffs2/xattr.h index 2c199856c58..06ab7b88021 100644 --- a/fs/jffs2/xattr.h +++ b/fs/jffs2/xattr.h @@ -16,6 +16,7 @@ #define JFFS2_XFLAGS_HOT (0x01) /* This datum is HOT */ #define JFFS2_XFLAGS_BIND (0x02) /* This datum is not reclaimed */ +#define JFFS2_XFLAGS_INVALID (0x80) /* This datum contains crc error */ struct jffs2_xattr_datum { @@ -23,7 +24,7 @@ struct jffs2_xattr_datum struct jffs2_raw_node_ref *node; uint8_t class; uint8_t flags; - uint16_t xprefix; /* see JFFS2_XATTR_PREFIX_* */ + uint16_t xprefix; /* see JFFS2_XATTR_PREFIX_* */ struct list_head xindex; /* chained from c->xattrindex[n] */ uint32_t refcnt; /* # of xattr_ref refers this */ @@ -47,6 +48,7 @@ struct jffs2_xattr_ref uint8_t flags; /* Currently unused */ u16 unused; + uint32_t xseqno; union { struct jffs2_inode_cache *ic; /* reference to jffs2_inode_cache */ uint32_t ino; /* only used in scanning/building */ @@ -58,6 +60,34 @@ struct jffs2_xattr_ref struct jffs2_xattr_ref *next; /* chained from ic->xref_list */ }; +#define XDATUM_DELETE_MARKER (0xffffffff) +#define XREF_DELETE_MARKER (0x00000001) +static inline int is_xattr_datum_dead(struct jffs2_xattr_datum *xd) +{ + return (xd->version == XDATUM_DELETE_MARKER); +} + +static inline void set_xattr_datum_dead(struct jffs2_xattr_datum *xd) +{ + xd->version = XDATUM_DELETE_MARKER; +} + +static inline int is_xattr_ref_dead(struct jffs2_xattr_ref *ref) +{ + return ((ref->xseqno & XREF_DELETE_MARKER) != 0); +} + +static inline void set_xattr_ref_dead(struct jffs2_xattr_ref *ref) +{ + ref->xseqno |= XREF_DELETE_MARKER; +} + +static inline void clr_xattr_ref_dead(struct jffs2_xattr_ref *ref) +{ + ref->xseqno &= ~XREF_DELETE_MARKER; +} + + #ifdef CONFIG_JFFS2_FS_XATTR extern void jffs2_init_xattr_subsystem(struct jffs2_sb_info *c); @@ -70,9 +100,13 @@ extern struct jffs2_xattr_datum *jffs2_setup_xattr_datum(struct jffs2_sb_info *c extern void jffs2_xattr_delete_inode(struct jffs2_sb_info *c, struct jffs2_inode_cache *ic); extern void jffs2_xattr_free_inode(struct jffs2_sb_info *c, struct jffs2_inode_cache *ic); -extern int jffs2_garbage_collect_xattr_datum(struct jffs2_sb_info *c, struct jffs2_xattr_datum *xd); -extern int jffs2_garbage_collect_xattr_ref(struct jffs2_sb_info *c, struct jffs2_xattr_ref *ref); +extern int jffs2_garbage_collect_xattr_datum(struct jffs2_sb_info *c, struct jffs2_xattr_datum *xd, + struct jffs2_raw_node_ref *raw); +extern int jffs2_garbage_collect_xattr_ref(struct jffs2_sb_info *c, struct jffs2_xattr_ref *ref, + struct jffs2_raw_node_ref *raw); extern int jffs2_verify_xattr(struct jffs2_sb_info *c); +extern void jffs2_release_xattr_datum(struct jffs2_sb_info *c, struct jffs2_xattr_datum *xd); +extern void jffs2_release_xattr_ref(struct jffs2_sb_info *c, struct jffs2_xattr_ref *ref); extern int do_jffs2_getxattr(struct inode *inode, int xprefix, const char *xname, char *buffer, size_t size); diff --git a/include/linux/jffs2.h b/include/linux/jffs2.h index c6f70660b37..c9c760700bc 100644 --- a/include/linux/jffs2.h +++ b/include/linux/jffs2.h @@ -186,6 +186,7 @@ struct jffs2_raw_xref jint32_t hdr_crc; jint32_t ino; /* inode number */ jint32_t xid; /* XATTR identifier number */ + jint32_t xseqno; /* xref sequencial number */ jint32_t node_crc; } __attribute__((packed)); -- cgit v1.2.3-70-g09d2 From cf7c712c11fb881842534efe98a07f36f1c86c65 Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Mon, 12 Jun 2006 15:49:31 -0700 Subject: [PATCH] 64bit resource: introduce resource_size_t for the start and end of struct resource But do not change it from what it currently is (unsigned long) Based on a patch series originally from Vivek Goyal Cc: Vivek Goyal Signed-off-by: Andrew Morton Signed-off-by: Greg Kroah-Hartman --- include/linux/ioport.h | 4 +++- include/linux/types.h | 2 ++ 2 files changed, 5 insertions(+), 1 deletion(-) (limited to 'include') diff --git a/include/linux/ioport.h b/include/linux/ioport.h index cd6bd001ba4..535bd958589 100644 --- a/include/linux/ioport.h +++ b/include/linux/ioport.h @@ -9,13 +9,15 @@ #define _LINUX_IOPORT_H #include +#include /* * Resources are tree-like, allowing * nesting etc.. */ struct resource { + resource_size_t start; + resource_size_t end; const char *name; - unsigned long start, end; unsigned long flags; struct resource *parent, *sibling, *child; }; diff --git a/include/linux/types.h b/include/linux/types.h index a5e46e783ff..a021e157733 100644 --- a/include/linux/types.h +++ b/include/linux/types.h @@ -177,6 +177,8 @@ typedef __u64 __bitwise __be64; #ifdef __KERNEL__ typedef unsigned __bitwise__ gfp_t; + +typedef unsigned long resource_size_t; #endif struct ustat { -- cgit v1.2.3-70-g09d2 From d75fc8bbccf7c019994bcfd6255d5b56335ed21d Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Mon, 12 Jun 2006 16:09:23 -0700 Subject: [PATCH] 64bit resource: change resource core to use resource_size_t Based on a patch series originally from Vivek Goyal Cc: Vivek Goyal Signed-off-by: Andrew Morton Signed-off-by: Greg Kroah-Hartman --- arch/sparc/kernel/ioport.c | 4 ++-- include/linux/ioport.h | 23 +++++++++++++---------- kernel/resource.c | 34 ++++++++++++++++++---------------- 3 files changed, 33 insertions(+), 28 deletions(-) (limited to 'include') diff --git a/arch/sparc/kernel/ioport.c b/arch/sparc/kernel/ioport.c index 99d716b9507..79d177149fd 100644 --- a/arch/sparc/kernel/ioport.c +++ b/arch/sparc/kernel/ioport.c @@ -208,7 +208,7 @@ _sparc_ioremap(struct resource *res, u32 bus, u32 pa, int sz) pa &= PAGE_MASK; sparc_mapiorange(bus, pa, res->start, res->end - res->start + 1); - return (void __iomem *) (res->start + offset); + return (void __iomem *)(unsigned long)(res->start + offset); } /* @@ -325,7 +325,7 @@ void *sbus_alloc_consistent(struct sbus_dev *sdev, long len, u32 *dma_addrp) res->name = sdev->prom_name; } - return (void *)res->start; + return (void *)(unsigned long)res->start; err_noiommu: release_resource(res); diff --git a/include/linux/ioport.h b/include/linux/ioport.h index 535bd958589..d4895236b7e 100644 --- a/include/linux/ioport.h +++ b/include/linux/ioport.h @@ -98,31 +98,34 @@ extern struct resource * ____request_resource(struct resource *root, struct reso extern int release_resource(struct resource *new); extern __deprecated_for_modules int insert_resource(struct resource *parent, struct resource *new); extern int allocate_resource(struct resource *root, struct resource *new, - unsigned long size, - unsigned long min, unsigned long max, - unsigned long align, + resource_size_t size, resource_size_t min, + resource_size_t max, resource_size_t align, void (*alignf)(void *, struct resource *, - unsigned long, unsigned long), + resource_size_t, resource_size_t), void *alignf_data); -int adjust_resource(struct resource *res, unsigned long start, - unsigned long size); +int adjust_resource(struct resource *res, resource_size_t start, + resource_size_t size); /* Convenience shorthand with allocation */ #define request_region(start,n,name) __request_region(&ioport_resource, (start), (n), (name)) #define request_mem_region(start,n,name) __request_region(&iomem_resource, (start), (n), (name)) #define rename_region(region, newname) do { (region)->name = (newname); } while (0) -extern struct resource * __request_region(struct resource *, unsigned long start, unsigned long n, const char *name); +extern struct resource * __request_region(struct resource *, + resource_size_t start, + resource_size_t n, const char *name); /* Compatibility cruft */ #define release_region(start,n) __release_region(&ioport_resource, (start), (n)) #define check_mem_region(start,n) __check_region(&iomem_resource, (start), (n)) #define release_mem_region(start,n) __release_region(&iomem_resource, (start), (n)) -extern int __check_region(struct resource *, unsigned long, unsigned long); -extern void __release_region(struct resource *, unsigned long, unsigned long); +extern int __check_region(struct resource *, resource_size_t, resource_size_t); +extern void __release_region(struct resource *, resource_size_t, + resource_size_t); -static inline int __deprecated check_region(unsigned long s, unsigned long n) +static inline int __deprecated check_region(resource_size_t s, + resource_size_t n) { return __check_region(&ioport_resource, s, n); } diff --git a/kernel/resource.c b/kernel/resource.c index ea5f7811a40..54835c02ab3 100644 --- a/kernel/resource.c +++ b/kernel/resource.c @@ -151,8 +151,8 @@ __initcall(ioresources_init); /* Return the conflict entry if you can't request it */ static struct resource * __request_resource(struct resource *root, struct resource *new) { - unsigned long start = new->start; - unsigned long end = new->end; + resource_size_t start = new->start; + resource_size_t end = new->end; struct resource *tmp, **p; if (end < start) @@ -236,11 +236,10 @@ EXPORT_SYMBOL(release_resource); * Find empty slot in the resource tree given range and alignment. */ static int find_resource(struct resource *root, struct resource *new, - unsigned long size, - unsigned long min, unsigned long max, - unsigned long align, + resource_size_t size, resource_size_t min, + resource_size_t max, resource_size_t align, void (*alignf)(void *, struct resource *, - unsigned long, unsigned long), + resource_size_t, resource_size_t), void *alignf_data) { struct resource *this = root->child; @@ -282,11 +281,10 @@ static int find_resource(struct resource *root, struct resource *new, * Allocate empty slot in the resource tree given range and alignment. */ int allocate_resource(struct resource *root, struct resource *new, - unsigned long size, - unsigned long min, unsigned long max, - unsigned long align, + resource_size_t size, resource_size_t min, + resource_size_t max, resource_size_t align, void (*alignf)(void *, struct resource *, - unsigned long, unsigned long), + resource_size_t, resource_size_t), void *alignf_data) { int err; @@ -378,10 +376,10 @@ EXPORT_SYMBOL(insert_resource); * arguments. Returns -EBUSY if it can't fit. Existing children of * the resource are assumed to be immutable. */ -int adjust_resource(struct resource *res, unsigned long start, unsigned long size) +int adjust_resource(struct resource *res, resource_size_t start, resource_size_t size) { struct resource *tmp, *parent = res->parent; - unsigned long end = start + size - 1; + resource_size_t end = start + size - 1; int result = -EBUSY; write_lock(&resource_lock); @@ -428,7 +426,9 @@ EXPORT_SYMBOL(adjust_resource); * * Release-region releases a matching busy region. */ -struct resource * __request_region(struct resource *parent, unsigned long start, unsigned long n, const char *name) +struct resource * __request_region(struct resource *parent, + resource_size_t start, resource_size_t n, + const char *name) { struct resource *res = kzalloc(sizeof(*res), GFP_KERNEL); @@ -464,7 +464,8 @@ struct resource * __request_region(struct resource *parent, unsigned long start, EXPORT_SYMBOL(__request_region); -int __check_region(struct resource *parent, unsigned long start, unsigned long n) +int __check_region(struct resource *parent, resource_size_t start, + resource_size_t n) { struct resource * res; @@ -479,10 +480,11 @@ int __check_region(struct resource *parent, unsigned long start, unsigned long n EXPORT_SYMBOL(__check_region); -void __release_region(struct resource *parent, unsigned long start, unsigned long n) +void __release_region(struct resource *parent, resource_size_t start, + resource_size_t n) { struct resource **p; - unsigned long end; + resource_size_t end; p = &parent->child; end = start + n - 1; -- cgit v1.2.3-70-g09d2 From e31dd6e4520439ceae4753f32dd2da2c345e929a Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Mon, 12 Jun 2006 17:06:02 -0700 Subject: [PATCH] 64bit resource: change pci core and arch code to use resource_size_t Based on a patch series originally from Vivek Goyal Cc: Vivek Goyal Signed-off-by: Andrew Morton Signed-off-by: Greg Kroah-Hartman --- arch/alpha/kernel/pci.c | 4 ++-- arch/arm/kernel/bios32.c | 6 +++--- arch/cris/arch-v32/drivers/pci/bios.c | 4 ++-- arch/frv/mb93090-mb00/pci-frv.c | 4 ++-- arch/i386/pci/i386.c | 4 ++-- arch/ia64/pci/pci.c | 2 +- arch/m68knommu/kernel/comempci.c | 3 ++- arch/mips/pci/pci.c | 4 ++-- arch/mips/pmc-sierra/yosemite/ht.c | 4 ++-- arch/parisc/kernel/pci.c | 2 +- arch/powerpc/kernel/pci_32.c | 10 +++++----- arch/powerpc/kernel/pci_64.c | 4 ++-- arch/ppc/kernel/pci.c | 12 ++++++------ arch/sh/boards/mpc1211/pci.c | 4 ++-- arch/sh/boards/overdrive/galileo.c | 2 +- arch/sh/drivers/pci/pci.c | 6 +++--- arch/sh64/kernel/pcibios.c | 4 ++-- arch/sparc/kernel/pcic.c | 2 +- arch/sparc64/kernel/pci.c | 2 +- arch/v850/kernel/rte_mb_a_pci.c | 2 +- arch/xtensa/kernel/pci.c | 6 +++--- drivers/pci/bus.c | 10 +++++----- drivers/pci/pci-sysfs.c | 4 ++-- drivers/pci/pci.h | 6 +++--- drivers/pci/proc.c | 4 ++-- drivers/pci/rom.c | 10 +++++----- drivers/pci/setup-res.c | 6 +++--- include/asm-arm/mach/pci.h | 2 +- include/asm-powerpc/pci.h | 2 +- include/asm-ppc/pci.h | 2 +- include/linux/pci.h | 13 +++++++------ 31 files changed, 76 insertions(+), 74 deletions(-) (limited to 'include') diff --git a/arch/alpha/kernel/pci.c b/arch/alpha/kernel/pci.c index 2a8b364c822..4ea6711e55a 100644 --- a/arch/alpha/kernel/pci.c +++ b/arch/alpha/kernel/pci.c @@ -124,12 +124,12 @@ DECLARE_PCI_FIXUP_FINAL(PCI_ANY_ID, PCI_ANY_ID, pcibios_fixup_final); void pcibios_align_resource(void *data, struct resource *res, - unsigned long size, unsigned long align) + resource_size_t size, resource_size_t align) { struct pci_dev *dev = data; struct pci_controller *hose = dev->sysdata; unsigned long alignto; - unsigned long start = res->start; + resource_size_t start = res->start; if (res->flags & IORESOURCE_IO) { /* Make sure we start at our min on all hoses */ diff --git a/arch/arm/kernel/bios32.c b/arch/arm/kernel/bios32.c index 302fc140154..45da06fc1ba 100644 --- a/arch/arm/kernel/bios32.c +++ b/arch/arm/kernel/bios32.c @@ -304,7 +304,7 @@ static inline int pdev_bad_for_parity(struct pci_dev *dev) static void __devinit pdev_fixup_device_resources(struct pci_sys_data *root, struct pci_dev *dev) { - unsigned long offset; + resource_size_t offset; int i; for (i = 0; i < PCI_NUM_RESOURCES; i++) { @@ -634,9 +634,9 @@ char * __init pcibios_setup(char *str) * which might be mirrored at 0x0100-0x03ff.. */ void pcibios_align_resource(void *data, struct resource *res, - unsigned long size, unsigned long align) + resource_size_t size, resource_size_t align) { - unsigned long start = res->start; + resource_size_t start = res->start; if (res->flags & IORESOURCE_IO && start & 0x300) start = (start + 0x3ff) & ~0x3ff; diff --git a/arch/cris/arch-v32/drivers/pci/bios.c b/arch/cris/arch-v32/drivers/pci/bios.c index 1e9d062103a..a2b9c60c277 100644 --- a/arch/cris/arch-v32/drivers/pci/bios.c +++ b/arch/cris/arch-v32/drivers/pci/bios.c @@ -43,10 +43,10 @@ int pci_mmap_page_range(struct pci_dev *dev, struct vm_area_struct *vma, void pcibios_align_resource(void *data, struct resource *res, - unsigned long size, unsigned long align) + resource_size_t size, resource_size_t align) { if (res->flags & IORESOURCE_IO) { - unsigned long start = res->start; + resource_size_t start = res->start; if (start & 0x300) { start = (start + 0x3ff) & ~0x3ff; diff --git a/arch/frv/mb93090-mb00/pci-frv.c b/arch/frv/mb93090-mb00/pci-frv.c index 0a26bf6f1cd..4f165c93be4 100644 --- a/arch/frv/mb93090-mb00/pci-frv.c +++ b/arch/frv/mb93090-mb00/pci-frv.c @@ -64,10 +64,10 @@ pcibios_update_resource(struct pci_dev *dev, struct resource *root, */ void pcibios_align_resource(void *data, struct resource *res, - unsigned long size, unsigned long align) + resource_size_t size, resource_size_t align) { if (res->flags & IORESOURCE_IO) { - unsigned long start = res->start; + resource_size_t start = res->start; if (start & 0x300) { start = (start + 0x3ff) & ~0x3ff; diff --git a/arch/i386/pci/i386.c b/arch/i386/pci/i386.c index a151f7a99f5..10154a2cac6 100644 --- a/arch/i386/pci/i386.c +++ b/arch/i386/pci/i386.c @@ -48,10 +48,10 @@ */ void pcibios_align_resource(void *data, struct resource *res, - unsigned long size, unsigned long align) + resource_size_t size, resource_size_t align) { if (res->flags & IORESOURCE_IO) { - unsigned long start = res->start; + resource_size_t start = res->start; if (start & 0x300) { start = (start + 0x3ff) & ~0x3ff; diff --git a/arch/ia64/pci/pci.c b/arch/ia64/pci/pci.c index 77375a55da3..5bef0e3603f 100644 --- a/arch/ia64/pci/pci.c +++ b/arch/ia64/pci/pci.c @@ -568,7 +568,7 @@ pcibios_disable_device (struct pci_dev *dev) void pcibios_align_resource (void *data, struct resource *res, - unsigned long size, unsigned long align) + resource_size_t size, resource_size_t align) { } diff --git a/arch/m68knommu/kernel/comempci.c b/arch/m68knommu/kernel/comempci.c index 8670938f110..db7a0c1ceba 100644 --- a/arch/m68knommu/kernel/comempci.c +++ b/arch/m68knommu/kernel/comempci.c @@ -357,7 +357,8 @@ void pcibios_fixup_bus(struct pci_bus *b) /*****************************************************************************/ -void pcibios_align_resource(void *data, struct resource *res, unsigned long size, unsigned long align) +void pcibios_align_resource(void *data, struct resource *res, + resource_size_t size, resource_size_t align) { } diff --git a/arch/mips/pci/pci.c b/arch/mips/pci/pci.c index 4dfce154d4a..ba66f8c9bd4 100644 --- a/arch/mips/pci/pci.c +++ b/arch/mips/pci/pci.c @@ -51,11 +51,11 @@ unsigned long PCIBIOS_MIN_MEM = 0; */ void pcibios_align_resource(void *data, struct resource *res, - unsigned long size, unsigned long align) + resource_size_t size, resource_size_t align) { struct pci_dev *dev = data; struct pci_controller *hose = dev->sysdata; - unsigned long start = res->start; + resource_size_t start = res->start; if (res->flags & IORESOURCE_IO) { /* Make sure we start at our min on all hoses */ diff --git a/arch/mips/pmc-sierra/yosemite/ht.c b/arch/mips/pmc-sierra/yosemite/ht.c index 54b65a80abf..fb523ebcafa 100644 --- a/arch/mips/pmc-sierra/yosemite/ht.c +++ b/arch/mips/pmc-sierra/yosemite/ht.c @@ -383,12 +383,12 @@ void pcibios_update_resource(struct pci_dev *dev, struct resource *root, void pcibios_align_resource(void *data, struct resource *res, - unsigned long size, unsigned long align) + resource_size_t size, resource_size_t align) { struct pci_dev *dev = data; if (res->flags & IORESOURCE_IO) { - unsigned long start = res->start; + resource_size_t start = res->start; /* We need to avoid collisions with `mirrored' VGA ports and other strange ISA hardware, so we always want the diff --git a/arch/parisc/kernel/pci.c b/arch/parisc/kernel/pci.c index 79c7db2705f..7d6967ee367 100644 --- a/arch/parisc/kernel/pci.c +++ b/arch/parisc/kernel/pci.c @@ -289,7 +289,7 @@ EXPORT_SYMBOL(pcibios_bus_to_resource); * than res->start. */ void pcibios_align_resource(void *data, struct resource *res, - unsigned long size, unsigned long alignment) + resource_size_t size, resource_size_t alignment) { unsigned long mask, align; diff --git a/arch/powerpc/kernel/pci_32.c b/arch/powerpc/kernel/pci_32.c index d9e2506db5f..8474355a1a4 100644 --- a/arch/powerpc/kernel/pci_32.c +++ b/arch/powerpc/kernel/pci_32.c @@ -173,18 +173,18 @@ EXPORT_SYMBOL(pcibios_bus_to_resource); * but we want to try to avoid allocating at 0x2900-0x2bff * which might have be mirrored at 0x0100-0x03ff.. */ -void pcibios_align_resource(void *data, struct resource *res, unsigned long size, - unsigned long align) +void pcibios_align_resource(void *data, struct resource *res, + resource_size_t size, resource_size_t align) { struct pci_dev *dev = data; if (res->flags & IORESOURCE_IO) { - unsigned long start = res->start; + resource_size_t start = res->start; if (size > 0x100) { printk(KERN_ERR "PCI: I/O Region %s/%d too large" " (%lld bytes)\n", pci_name(dev), - dev->resource - res, size); + dev->resource - res, (unsigned long long)size); } if (start & 0x300) { @@ -1756,7 +1756,7 @@ long sys_pciconfig_iobase(long which, unsigned long bus, unsigned long devfn) void pci_resource_to_user(const struct pci_dev *dev, int bar, const struct resource *rsrc, - u64 *start, u64 *end) + resource_size_t *start, resource_size_t *end) { struct pci_controller *hose = pci_bus_to_hose(dev->bus->number); unsigned long offset = 0; diff --git a/arch/powerpc/kernel/pci_64.c b/arch/powerpc/kernel/pci_64.c index 247937dd8b7..286aa52aae3 100644 --- a/arch/powerpc/kernel/pci_64.c +++ b/arch/powerpc/kernel/pci_64.c @@ -138,11 +138,11 @@ EXPORT_SYMBOL(pcibios_bus_to_resource); * which might have be mirrored at 0x0100-0x03ff.. */ void pcibios_align_resource(void *data, struct resource *res, - unsigned long size, unsigned long align) + resource_size_t size, resource_size_t align) { struct pci_dev *dev = data; struct pci_controller *hose = pci_bus_to_host(dev->bus); - unsigned long start = res->start; + resource_size_t start = res->start; unsigned long alignto; if (res->flags & IORESOURCE_IO) { diff --git a/arch/ppc/kernel/pci.c b/arch/ppc/kernel/pci.c index 8544e100d71..242bb052be6 100644 --- a/arch/ppc/kernel/pci.c +++ b/arch/ppc/kernel/pci.c @@ -171,13 +171,13 @@ EXPORT_SYMBOL(pcibios_bus_to_resource); * but we want to try to avoid allocating at 0x2900-0x2bff * which might have be mirrored at 0x0100-0x03ff.. */ -void pcibios_align_resource(void *data, struct resource *res, unsigned long size, - unsigned long align) +void pcibios_align_resource(void *data, struct resource *res, + resource_size_t size, resource_size_t align) { struct pci_dev *dev = data; if (res->flags & IORESOURCE_IO) { - unsigned long start = res->start; + resource_size_t start = res->start; if (size > 0x100) { printk(KERN_ERR "PCI: I/O Region %s/%d too large" @@ -960,8 +960,8 @@ static pgprot_t __pci_mmap_set_pgprot(struct pci_dev *dev, struct resource *rp, else prot |= _PAGE_GUARDED; - printk("PCI map for %s:%llx, prot: %llx\n", pci_name(dev), rp->start, - prot); + printk("PCI map for %s:%llx, prot: %lx\n", pci_name(dev), + (unsigned long long)rp->start, prot); return __pgprot(prot); } @@ -1130,7 +1130,7 @@ long sys_pciconfig_iobase(long which, unsigned long bus, unsigned long devfn) void pci_resource_to_user(const struct pci_dev *dev, int bar, const struct resource *rsrc, - u64 *start, u64 *end) + resource_size_t *start, resource_size_t *end) { struct pci_controller *hose = pci_bus_to_hose(dev->bus->number); unsigned long offset = 0; diff --git a/arch/sh/boards/mpc1211/pci.c b/arch/sh/boards/mpc1211/pci.c index ba3a6543975..9f7ccd33ffb 100644 --- a/arch/sh/boards/mpc1211/pci.c +++ b/arch/sh/boards/mpc1211/pci.c @@ -273,9 +273,9 @@ void __init pcibios_fixup_irqs(void) } void pcibios_align_resource(void *data, struct resource *res, - unsigned long size, unsigned long align) + resource_size_t size, resource_size_t align) { - unsigned long start = res->start; + resource_size_t start = res->start; if (res->flags & IORESOURCE_IO) { if (start >= 0x10000UL) { diff --git a/arch/sh/boards/overdrive/galileo.c b/arch/sh/boards/overdrive/galileo.c index 276fa11ee4c..b055809d2ac 100644 --- a/arch/sh/boards/overdrive/galileo.c +++ b/arch/sh/boards/overdrive/galileo.c @@ -536,7 +536,7 @@ void __init pcibios_fixup_bus(struct pci_bus *bus) } void pcibios_align_resource(void *data, struct resource *res, - unsigned long size) + resource_size_t size) { } diff --git a/arch/sh/drivers/pci/pci.c b/arch/sh/drivers/pci/pci.c index c1669905abe..3d546ba329c 100644 --- a/arch/sh/drivers/pci/pci.c +++ b/arch/sh/drivers/pci/pci.c @@ -75,7 +75,7 @@ pcibios_update_resource(struct pci_dev *dev, struct resource *root, } void pcibios_align_resource(void *data, struct resource *res, - unsigned long size, unsigned long align) + resource_size_t size, resource_size_t align) __attribute__ ((weak)); /* @@ -85,10 +85,10 @@ void pcibios_align_resource(void *data, struct resource *res, * modulo 0x400. */ void pcibios_align_resource(void *data, struct resource *res, - unsigned long size, unsigned long align) + resource_size_t size, resource_size_t align) { if (res->flags & IORESOURCE_IO) { - unsigned long start = res->start; + resource_size_t start = res->start; if (start & 0x300) { start = (start + 0x3ff) & ~0x3ff; diff --git a/arch/sh64/kernel/pcibios.c b/arch/sh64/kernel/pcibios.c index 50c61dcb9fa..945920bc24d 100644 --- a/arch/sh64/kernel/pcibios.c +++ b/arch/sh64/kernel/pcibios.c @@ -69,10 +69,10 @@ pcibios_update_resource(struct pci_dev *dev, struct resource *root, * modulo 0x400. */ void pcibios_align_resource(void *data, struct resource *res, - unsigned long size, unsigned long align) + resource_size_t size, resource_size_t align) { if (res->flags & IORESOURCE_IO) { - unsigned long start = res->start; + resource_size_t start = res->start; if (start & 0x300) { start = (start + 0x3ff) & ~0x3ff; diff --git a/arch/sparc/kernel/pcic.c b/arch/sparc/kernel/pcic.c index bcfdddd0418..5df3ebdc0ab 100644 --- a/arch/sparc/kernel/pcic.c +++ b/arch/sparc/kernel/pcic.c @@ -860,7 +860,7 @@ char * __init pcibios_setup(char *str) } void pcibios_align_resource(void *data, struct resource *res, - unsigned long size, unsigned long align) + resource_size_t size, resource_size_t align) { } diff --git a/arch/sparc64/kernel/pci.c b/arch/sparc64/kernel/pci.c index 6c9e3e94aba..20ca9ec8fd3 100644 --- a/arch/sparc64/kernel/pci.c +++ b/arch/sparc64/kernel/pci.c @@ -357,7 +357,7 @@ void pcibios_update_irq(struct pci_dev *pdev, int irq) } void pcibios_align_resource(void *data, struct resource *res, - unsigned long size, unsigned long align) + resource_size_t size, resource_size_t align) { } diff --git a/arch/v850/kernel/rte_mb_a_pci.c b/arch/v850/kernel/rte_mb_a_pci.c index ffbb6d073bf..3a7c5c9c3ac 100644 --- a/arch/v850/kernel/rte_mb_a_pci.c +++ b/arch/v850/kernel/rte_mb_a_pci.c @@ -329,7 +329,7 @@ void pcibios_fixup_bus(struct pci_bus *b) void pcibios_align_resource (void *data, struct resource *res, - unsigned long size, unsigned long align) + resource_size_t size, resource_size_t align) { } diff --git a/arch/xtensa/kernel/pci.c b/arch/xtensa/kernel/pci.c index c6f471b9eaa..eda029fc897 100644 --- a/arch/xtensa/kernel/pci.c +++ b/arch/xtensa/kernel/pci.c @@ -71,13 +71,13 @@ static int pci_bus_count; * which might have be mirrored at 0x0100-0x03ff.. */ void -pcibios_align_resource(void *data, struct resource *res, unsigned long size, - unsigned long align) +pcibios_align_resource(void *data, struct resource *res, resource_size_t size, + resource_size_t align) { struct pci_dev *dev = data; if (res->flags & IORESOURCE_IO) { - unsigned long start = res->start; + resource_size_t start = res->start; if (size > 0x100) { printk(KERN_ERR "PCI: I/O Region %s/%d too large" diff --git a/drivers/pci/bus.c b/drivers/pci/bus.c index 72309268202..5f7db9d2436 100644 --- a/drivers/pci/bus.c +++ b/drivers/pci/bus.c @@ -34,11 +34,11 @@ */ int pci_bus_alloc_resource(struct pci_bus *bus, struct resource *res, - unsigned long size, unsigned long align, unsigned long min, - unsigned int type_mask, - void (*alignf)(void *, struct resource *, - unsigned long, unsigned long), - void *alignf_data) + resource_size_t size, resource_size_t align, + resource_size_t min, unsigned int type_mask, + void (*alignf)(void *, struct resource *, resource_size_t, + resource_size_t), + void *alignf_data) { int i, ret = -ENOMEM; diff --git a/drivers/pci/pci-sysfs.c b/drivers/pci/pci-sysfs.c index bc405c035ce..606f9b6f70e 100644 --- a/drivers/pci/pci-sysfs.c +++ b/drivers/pci/pci-sysfs.c @@ -87,7 +87,7 @@ resource_show(struct device * dev, struct device_attribute *attr, char * buf) char * str = buf; int i; int max = 7; - u64 start, end; + resource_size_t start, end; if (pci_dev->subordinate) max = DEVICE_COUNT_RESOURCE; @@ -365,7 +365,7 @@ pci_mmap_resource(struct kobject *kobj, struct bin_attribute *attr, struct device, kobj)); struct resource *res = (struct resource *)attr->private; enum pci_mmap_state mmap_type; - u64 start, end; + resource_size_t start, end; int i; for (i = 0; i < PCI_ROM_RESOURCE; i++) diff --git a/drivers/pci/pci.h b/drivers/pci/pci.h index 29bdeca031a..9cc842b666e 100644 --- a/drivers/pci/pci.h +++ b/drivers/pci/pci.h @@ -6,10 +6,10 @@ extern int pci_create_sysfs_dev_files(struct pci_dev *pdev); extern void pci_remove_sysfs_dev_files(struct pci_dev *pdev); extern void pci_cleanup_rom(struct pci_dev *dev); extern int pci_bus_alloc_resource(struct pci_bus *bus, struct resource *res, - unsigned long size, unsigned long align, - unsigned long min, unsigned int type_mask, + resource_size_t size, resource_size_t align, + resource_size_t min, unsigned int type_mask, void (*alignf)(void *, struct resource *, - unsigned long, unsigned long), + resource_size_t, resource_size_t), void *alignf_data); /* Firmware callbacks */ extern int (*platform_pci_choose_state)(struct pci_dev *dev, pm_message_t state); diff --git a/drivers/pci/proc.c b/drivers/pci/proc.c index 20dfd77bd8c..99cf3337976 100644 --- a/drivers/pci/proc.c +++ b/drivers/pci/proc.c @@ -350,14 +350,14 @@ static int show_device(struct seq_file *m, void *v) dev->irq); /* Here should be 7 and not PCI_NUM_RESOURCES as we need to preserve compatibility */ for (i=0; i<7; i++) { - u64 start, end; + resource_size_t start, end; pci_resource_to_user(dev, i, &dev->resource[i], &start, &end); seq_printf(m, "\t%16llx", (unsigned long long)(start | (dev->resource[i].flags & PCI_REGION_FLAG_MASK))); } for (i=0; i<7; i++) { - u64 start, end; + resource_size_t start, end; pci_resource_to_user(dev, i, &dev->resource[i], &start, &end); seq_printf(m, "\t%16llx", dev->resource[i].start < dev->resource[i].end ? diff --git a/drivers/pci/rom.c b/drivers/pci/rom.c index 598a115cd00..cbb69cf4131 100644 --- a/drivers/pci/rom.c +++ b/drivers/pci/rom.c @@ -80,8 +80,8 @@ void __iomem *pci_map_rom(struct pci_dev *pdev, size_t *size) } else { if (res->flags & IORESOURCE_ROM_COPY) { *size = pci_resource_len(pdev, PCI_ROM_RESOURCE); - return (void __iomem *)pci_resource_start(pdev, - PCI_ROM_RESOURCE); + return (void __iomem *)(unsigned long) + pci_resource_start(pdev, PCI_ROM_RESOURCE); } else { /* assign the ROM an address if it doesn't have one */ if (res->parent == NULL && @@ -170,11 +170,11 @@ void __iomem *pci_map_rom_copy(struct pci_dev *pdev, size_t *size) return rom; res->end = res->start + *size; - memcpy_fromio((void*)res->start, rom, *size); + memcpy_fromio((void*)(unsigned long)res->start, rom, *size); pci_unmap_rom(pdev, rom); res->flags |= IORESOURCE_ROM_COPY; - return (void __iomem *)res->start; + return (void __iomem *)(unsigned long)res->start; } /** @@ -227,7 +227,7 @@ void pci_cleanup_rom(struct pci_dev *pdev) { struct resource *res = &pdev->resource[PCI_ROM_RESOURCE]; if (res->flags & IORESOURCE_ROM_COPY) { - kfree((void*)res->start); + kfree((void*)(unsigned long)res->start); res->flags &= ~IORESOURCE_ROM_COPY; res->start = 0; res->end = 0; diff --git a/drivers/pci/setup-res.c b/drivers/pci/setup-res.c index f5ff0d3ba62..ab78e4bbdd8 100644 --- a/drivers/pci/setup-res.c +++ b/drivers/pci/setup-res.c @@ -121,7 +121,7 @@ int pci_assign_resource(struct pci_dev *dev, int resno) { struct pci_bus *bus = dev->bus; struct resource *res = dev->resource + resno; - unsigned long size, min, align; + resource_size_t size, min, align; int ret; size = res->end - res->start + 1; @@ -209,7 +209,7 @@ pdev_sort_resources(struct pci_dev *dev, struct resource_list *head) for (i = 0; i < PCI_NUM_RESOURCES; i++) { struct resource *r; struct resource_list *list, *tmp; - unsigned long r_align; + resource_size_t r_align; r = &dev->resource[i]; r_align = r->end - r->start; @@ -225,7 +225,7 @@ pdev_sort_resources(struct pci_dev *dev, struct resource_list *head) } r_align = (i < PCI_BRIDGE_RESOURCES) ? r_align + 1 : r->start; for (list = head; ; list = list->next) { - unsigned long align = 0; + resource_size_t align = 0; struct resource_list *ln = list->next; int idx; diff --git a/include/asm-arm/mach/pci.h b/include/asm-arm/mach/pci.h index 25d540ed007..923e0ca6620 100644 --- a/include/asm-arm/mach/pci.h +++ b/include/asm-arm/mach/pci.h @@ -28,7 +28,7 @@ struct hw_pci { struct pci_sys_data { struct list_head node; int busnr; /* primary bus number */ - unsigned long mem_offset; /* bus->cpu memory mapping offset */ + u64 mem_offset; /* bus->cpu memory mapping offset */ unsigned long io_offset; /* bus->cpu IO mapping offset */ struct pci_bus *bus; /* PCI bus */ struct resource *resource[3]; /* Primary PCI bus resources */ diff --git a/include/asm-powerpc/pci.h b/include/asm-powerpc/pci.h index 5d2c9e6c4be..46afd29b904 100644 --- a/include/asm-powerpc/pci.h +++ b/include/asm-powerpc/pci.h @@ -242,7 +242,7 @@ extern pgprot_t pci_phys_mem_access_prot(struct file *file, #define HAVE_ARCH_PCI_RESOURCE_TO_USER extern void pci_resource_to_user(const struct pci_dev *dev, int bar, const struct resource *rsrc, - u64 *start, u64 *end); + resource_size_t *start, resource_size_t *end); #endif /* CONFIG_PPC_MULTIPLATFORM || CONFIG_PPC32 */ #endif /* __KERNEL__ */ diff --git a/include/asm-ppc/pci.h b/include/asm-ppc/pci.h index 61434edbad7..11ffaaa5da1 100644 --- a/include/asm-ppc/pci.h +++ b/include/asm-ppc/pci.h @@ -133,7 +133,7 @@ extern pgprot_t pci_phys_mem_access_prot(struct file *file, #define HAVE_ARCH_PCI_RESOURCE_TO_USER extern void pci_resource_to_user(const struct pci_dev *dev, int bar, const struct resource *rsrc, - u64 *start, u64 *end); + resource_size_t *start, resource_size_t *end); #endif /* __KERNEL__ */ diff --git a/include/linux/pci.h b/include/linux/pci.h index 62a8c22f5f6..983fca251b2 100644 --- a/include/linux/pci.h +++ b/include/linux/pci.h @@ -404,8 +404,8 @@ int pcibios_enable_device(struct pci_dev *, int mask); char *pcibios_setup (char *str); /* Used only when drivers/pci/setup.c is used */ -void pcibios_align_resource(void *, struct resource *, - unsigned long, unsigned long); +void pcibios_align_resource(void *, struct resource *, resource_size_t, + resource_size_t); void pcibios_update_irq(struct pci_dev *, int irq); /* Generic PCI functions used internally */ @@ -532,10 +532,10 @@ void pci_release_region(struct pci_dev *, int); /* drivers/pci/bus.c */ int pci_bus_alloc_resource(struct pci_bus *bus, struct resource *res, - unsigned long size, unsigned long align, - unsigned long min, unsigned int type_mask, + resource_size_t size, resource_size_t align, + resource_size_t min, unsigned int type_mask, void (*alignf)(void *, struct resource *, - unsigned long, unsigned long), + resource_size_t, resource_size_t), void *alignf_data); void pci_enable_bridges(struct pci_bus *bus); @@ -730,7 +730,8 @@ static inline char *pci_name(struct pci_dev *pdev) */ #ifndef HAVE_ARCH_PCI_RESOURCE_TO_USER static inline void pci_resource_to_user(const struct pci_dev *dev, int bar, - const struct resource *rsrc, u64 *start, u64 *end) + const struct resource *rsrc, resource_size_t *start, + resource_size_t *end) { *start = rsrc->start; *end = rsrc->end; -- cgit v1.2.3-70-g09d2 From b60ba8343b78b182c03cf239d4342785376c1ad1 Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Mon, 12 Jun 2006 17:07:07 -0700 Subject: [PATCH] 64bit resource: change pnp core to use resource_size_t Based on a patch series originally from Vivek Goyal Cc: Vivek Goyal Signed-off-by: Andrew Morton Signed-off-by: Greg Kroah-Hartman --- drivers/pnp/interface.c | 8 ++++---- drivers/pnp/manager.c | 15 ++++++++++----- drivers/pnp/resource.c | 8 ++++---- include/linux/pnp.h | 7 +++++-- 4 files changed, 23 insertions(+), 15 deletions(-) (limited to 'include') diff --git a/drivers/pnp/interface.c b/drivers/pnp/interface.c index a2d8ce7fef9..3163e3d73da 100644 --- a/drivers/pnp/interface.c +++ b/drivers/pnp/interface.c @@ -264,7 +264,7 @@ static ssize_t pnp_show_current_resources(struct device *dmdev, struct device_at if (pnp_port_flags(dev, i) & IORESOURCE_DISABLED) pnp_printf(buffer," disabled\n"); else - pnp_printf(buffer," 0x%lx-0x%lx\n", + pnp_printf(buffer," 0x%llx-0x%llx\n", pnp_port_start(dev, i), pnp_port_end(dev, i)); } @@ -275,7 +275,7 @@ static ssize_t pnp_show_current_resources(struct device *dmdev, struct device_at if (pnp_mem_flags(dev, i) & IORESOURCE_DISABLED) pnp_printf(buffer," disabled\n"); else - pnp_printf(buffer," 0x%lx-0x%lx\n", + pnp_printf(buffer," 0x%llx-0x%llx\n", pnp_mem_start(dev, i), pnp_mem_end(dev, i)); } @@ -286,7 +286,7 @@ static ssize_t pnp_show_current_resources(struct device *dmdev, struct device_at if (pnp_irq_flags(dev, i) & IORESOURCE_DISABLED) pnp_printf(buffer," disabled\n"); else - pnp_printf(buffer," %ld\n", + pnp_printf(buffer," %lld\n", pnp_irq(dev, i)); } } @@ -296,7 +296,7 @@ static ssize_t pnp_show_current_resources(struct device *dmdev, struct device_at if (pnp_dma_flags(dev, i) & IORESOURCE_DISABLED) pnp_printf(buffer," disabled\n"); else - pnp_printf(buffer," %ld\n", + pnp_printf(buffer," %lld\n", pnp_dma(dev, i)); } } diff --git a/drivers/pnp/manager.c b/drivers/pnp/manager.c index 6fff109bdab..1d7a5b87f4c 100644 --- a/drivers/pnp/manager.c +++ b/drivers/pnp/manager.c @@ -20,7 +20,8 @@ DECLARE_MUTEX(pnp_res_mutex); static int pnp_assign_port(struct pnp_dev *dev, struct pnp_port *rule, int idx) { - unsigned long *start, *end, *flags; + resource_size_t *start, *end; + unsigned long *flags; if (!dev || !rule) return -EINVAL; @@ -63,7 +64,8 @@ static int pnp_assign_port(struct pnp_dev *dev, struct pnp_port *rule, int idx) static int pnp_assign_mem(struct pnp_dev *dev, struct pnp_mem *rule, int idx) { - unsigned long *start, *end, *flags; + resource_size_t *start, *end; + unsigned long *flags; if (!dev || !rule) return -EINVAL; @@ -116,7 +118,8 @@ static int pnp_assign_mem(struct pnp_dev *dev, struct pnp_mem *rule, int idx) static int pnp_assign_irq(struct pnp_dev * dev, struct pnp_irq *rule, int idx) { - unsigned long *start, *end, *flags; + resource_size_t *start, *end; + unsigned long *flags; int i; /* IRQ priority: this table is good for i386 */ @@ -168,7 +171,8 @@ static int pnp_assign_irq(struct pnp_dev * dev, struct pnp_irq *rule, int idx) static int pnp_assign_dma(struct pnp_dev *dev, struct pnp_dma *rule, int idx) { - unsigned long *start, *end, *flags; + resource_size_t *start, *end; + unsigned long *flags; int i; /* DMA priority: this table is good for i386 */ @@ -582,7 +586,8 @@ int pnp_disable_dev(struct pnp_dev *dev) * @size: size of region * */ -void pnp_resource_change(struct resource *resource, unsigned long start, unsigned long size) +void pnp_resource_change(struct resource *resource, resource_size_t start, + resource_size_t size) { if (resource == NULL) return; diff --git a/drivers/pnp/resource.c b/drivers/pnp/resource.c index 6ded527169f..7bb892f58cc 100644 --- a/drivers/pnp/resource.c +++ b/drivers/pnp/resource.c @@ -241,7 +241,7 @@ int pnp_check_port(struct pnp_dev * dev, int idx) { int tmp; struct pnp_dev *tdev; - unsigned long *port, *end, *tport, *tend; + resource_size_t *port, *end, *tport, *tend; port = &dev->res.port_resource[idx].start; end = &dev->res.port_resource[idx].end; @@ -297,7 +297,7 @@ int pnp_check_mem(struct pnp_dev * dev, int idx) { int tmp; struct pnp_dev *tdev; - unsigned long *addr, *end, *taddr, *tend; + resource_size_t *addr, *end, *taddr, *tend; addr = &dev->res.mem_resource[idx].start; end = &dev->res.mem_resource[idx].end; @@ -358,7 +358,7 @@ int pnp_check_irq(struct pnp_dev * dev, int idx) { int tmp; struct pnp_dev *tdev; - unsigned long * irq = &dev->res.irq_resource[idx].start; + resource_size_t * irq = &dev->res.irq_resource[idx].start; /* if the resource doesn't exist, don't complain about it */ if (cannot_compare(dev->res.irq_resource[idx].flags)) @@ -423,7 +423,7 @@ int pnp_check_dma(struct pnp_dev * dev, int idx) #ifndef CONFIG_IA64 int tmp; struct pnp_dev *tdev; - unsigned long * dma = &dev->res.dma_resource[idx].start; + resource_size_t * dma = &dev->res.dma_resource[idx].start; /* if the resource doesn't exist, don't complain about it */ if (cannot_compare(dev->res.dma_resource[idx].flags)) diff --git a/include/linux/pnp.h b/include/linux/pnp.h index 93b0959eb40..ab8a8dd8d64 100644 --- a/include/linux/pnp.h +++ b/include/linux/pnp.h @@ -389,7 +389,8 @@ int pnp_start_dev(struct pnp_dev *dev); int pnp_stop_dev(struct pnp_dev *dev); int pnp_activate_dev(struct pnp_dev *dev); int pnp_disable_dev(struct pnp_dev *dev); -void pnp_resource_change(struct resource *resource, unsigned long start, unsigned long size); +void pnp_resource_change(struct resource *resource, resource_size_t start, + resource_size_t size); /* protocol helpers */ int pnp_is_active(struct pnp_dev * dev); @@ -434,7 +435,9 @@ static inline int pnp_start_dev(struct pnp_dev *dev) { return -ENODEV; } static inline int pnp_stop_dev(struct pnp_dev *dev) { return -ENODEV; } static inline int pnp_activate_dev(struct pnp_dev *dev) { return -ENODEV; } static inline int pnp_disable_dev(struct pnp_dev *dev) { return -ENODEV; } -static inline void pnp_resource_change(struct resource *resource, unsigned long start, unsigned long size) { } +static inline void pnp_resource_change(struct resource *resource, + resource_size_t start, + resource_size_t size) { } /* protocol helpers */ static inline int pnp_is_active(struct pnp_dev * dev) { return 0; } -- cgit v1.2.3-70-g09d2 From 6550e07f41ce8473ed684dac54fbfbd42183ffda Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Mon, 12 Jun 2006 17:11:31 -0700 Subject: [PATCH] 64bit Resource: finally enable 64bit resource sizes Introduce the Kconfig entry and actually switch to a 64bit value, if wanted, for resource_size_t. Based on a patch series originally from Vivek Goyal Cc: Vivek Goyal Signed-off-by: Andrew Morton Signed-off-by: Greg Kroah-Hartman --- arch/i386/Kconfig | 1 + include/linux/types.h | 7 ++++++- kernel/resource.c | 8 +++----- mm/Kconfig | 6 ++++++ 4 files changed, 16 insertions(+), 6 deletions(-) (limited to 'include') diff --git a/arch/i386/Kconfig b/arch/i386/Kconfig index 47c08bcd9b2..7e46ad7a7b6 100644 --- a/arch/i386/Kconfig +++ b/arch/i386/Kconfig @@ -529,6 +529,7 @@ config X86_PAE bool depends on HIGHMEM64G default y + select RESOURCES_64BIT # Common NUMA Features config NUMA diff --git a/include/linux/types.h b/include/linux/types.h index a021e157733..3f235660a3c 100644 --- a/include/linux/types.h +++ b/include/linux/types.h @@ -178,9 +178,14 @@ typedef __u64 __bitwise __be64; #ifdef __KERNEL__ typedef unsigned __bitwise__ gfp_t; -typedef unsigned long resource_size_t; +#ifdef CONFIG_RESOURCES_64BIT +typedef u64 resource_size_t; +#else +typedef u32 resource_size_t; #endif +#endif /* __KERNEL__ */ + struct ustat { __kernel_daddr_t f_tfree; __kernel_ino_t f_tinode; diff --git a/kernel/resource.c b/kernel/resource.c index 54835c02ab3..cc73029088a 100644 --- a/kernel/resource.c +++ b/kernel/resource.c @@ -23,20 +23,18 @@ struct resource ioport_resource = { .name = "PCI IO", - .start = 0x0000, + .start = 0, .end = IO_SPACE_LIMIT, .flags = IORESOURCE_IO, }; - EXPORT_SYMBOL(ioport_resource); struct resource iomem_resource = { .name = "PCI mem", - .start = 0UL, - .end = ~0UL, + .start = 0, + .end = -1, .flags = IORESOURCE_MEM, }; - EXPORT_SYMBOL(iomem_resource); static DEFINE_RWLOCK(resource_lock); diff --git a/mm/Kconfig b/mm/Kconfig index 66e65ab3942..e3644b0062b 100644 --- a/mm/Kconfig +++ b/mm/Kconfig @@ -145,3 +145,9 @@ config MIGRATION while the virtual addresses are not changed. This is useful for example on NUMA systems to put pages nearer to the processors accessing the page. + +config RESOURCES_64BIT + bool "64 bit Memory and IO resources (EXPERIMENTAL)" if (!64BIT && EXPERIMENTAL) + default 64BIT + help + This option allows memory and IO resources to be 64 bit. -- cgit v1.2.3-70-g09d2 From 64f495323c9a902b3e59fe0a588585102bb3b13e Mon Sep 17 00:00:00 2001 From: Kyle McMartin Date: Sat, 22 Apr 2006 00:48:22 -0600 Subject: [PARISC] Ensure all ldcw uses are ldcw,co on pa2.0 ldcw,co should always be used on pa2.0, otherwise the strict cache width alignment requirement is not relaxed. Signed-off-by: Kyle McMartin --- arch/parisc/kernel/entry.S | 6 +++--- arch/parisc/kernel/syscall.S | 2 +- include/asm-parisc/assembly.h | 2 ++ include/asm-parisc/system.h | 26 ++++++++++++++------------ 4 files changed, 20 insertions(+), 16 deletions(-) (limited to 'include') diff --git a/arch/parisc/kernel/entry.S b/arch/parisc/kernel/entry.S index d9e53cf0372..630730c32a5 100644 --- a/arch/parisc/kernel/entry.S +++ b/arch/parisc/kernel/entry.S @@ -1638,7 +1638,7 @@ dbit_trap_20w: load32 PA(pa_dbit_lock),t0 dbit_spin_20w: - ldcw 0(t0),t1 + LDCW 0(t0),t1 cmpib,= 0,t1,dbit_spin_20w nop @@ -1674,7 +1674,7 @@ dbit_trap_11: load32 PA(pa_dbit_lock),t0 dbit_spin_11: - ldcw 0(t0),t1 + LDCW 0(t0),t1 cmpib,= 0,t1,dbit_spin_11 nop @@ -1714,7 +1714,7 @@ dbit_trap_20: load32 PA(pa_dbit_lock),t0 dbit_spin_20: - ldcw 0(t0),t1 + LDCW 0(t0),t1 cmpib,= 0,t1,dbit_spin_20 nop diff --git a/arch/parisc/kernel/syscall.S b/arch/parisc/kernel/syscall.S index 479d9a017cd..a028c990cbf 100644 --- a/arch/parisc/kernel/syscall.S +++ b/arch/parisc/kernel/syscall.S @@ -541,7 +541,7 @@ cas_nocontend: # endif /* ENABLE_LWS_DEBUG */ - ldcw 0(%sr2,%r20), %r28 /* Try to acquire the lock */ + LDCW 0(%sr2,%r20), %r28 /* Try to acquire the lock */ cmpb,<>,n %r0, %r28, cas_action /* Did we get it? */ cas_wouldblock: ldo 2(%r0), %r28 /* 2nd case */ diff --git a/include/asm-parisc/assembly.h b/include/asm-parisc/assembly.h index 3ce3440d1b0..1a7bfe699e0 100644 --- a/include/asm-parisc/assembly.h +++ b/include/asm-parisc/assembly.h @@ -48,6 +48,7 @@ #define CALLEE_SAVE_FRAME_SIZE (CALLEE_REG_FRAME_SIZE + CALLEE_FLOAT_FRAME_SIZE) #ifdef CONFIG_PA20 +#define LDCW ldcw,co #define BL b,l # ifdef CONFIG_64BIT # define LEVEL 2.0w @@ -55,6 +56,7 @@ # define LEVEL 2.0 # endif #else +#define LDCW ldcw #define BL bl #define LEVEL 1.1 #endif diff --git a/include/asm-parisc/system.h b/include/asm-parisc/system.h index 863876134b2..5fe2d2329ab 100644 --- a/include/asm-parisc/system.h +++ b/include/asm-parisc/system.h @@ -155,13 +155,14 @@ static inline void set_eiem(unsigned long val) type and dynamically select the 16-byte aligned int from the array for the semaphore. */ -#define __PA_LDCW_ALIGNMENT 16 -#define __ldcw_align(a) ({ \ - unsigned long __ret = (unsigned long) &(a)->lock[0]; \ - __ret = (__ret + __PA_LDCW_ALIGNMENT - 1) & ~(__PA_LDCW_ALIGNMENT - 1); \ - (volatile unsigned int *) __ret; \ +#define __PA_LDCW_ALIGNMENT 16 +#define __ldcw_align(a) ({ \ + unsigned long __ret = (unsigned long) &(a)->lock[0]; \ + __ret = (__ret + __PA_LDCW_ALIGNMENT - 1) \ + & ~(__PA_LDCW_ALIGNMENT - 1); \ + (volatile unsigned int *) __ret; \ }) -#define LDCW "ldcw" +#define __LDCW "ldcw" #else /*CONFIG_PA20*/ /* From: "Jim Hull" @@ -171,17 +172,18 @@ static inline void set_eiem(unsigned long val) they only require "natural" alignment (4-byte for ldcw, 8-byte for ldcd). */ -#define __PA_LDCW_ALIGNMENT 4 +#define __PA_LDCW_ALIGNMENT 4 #define __ldcw_align(a) ((volatile unsigned int *)a) -#define LDCW "ldcw,co" +#define __LDCW "ldcw,co" #endif /*!CONFIG_PA20*/ /* LDCW, the only atomic read-write operation PA-RISC has. *sigh*. */ -#define __ldcw(a) ({ \ - unsigned __ret; \ - __asm__ __volatile__(LDCW " 0(%1),%0" : "=r" (__ret) : "r" (a)); \ - __ret; \ +#define __ldcw(a) ({ \ + unsigned __ret; \ + __asm__ __volatile__(__LDCW " 0(%1),%0" \ + : "=r" (__ret) : "r" (a)); \ + __ret; \ }) #ifdef CONFIG_SMP -- cgit v1.2.3-70-g09d2 From c8224e0074f1dce12e95e53ca469f6fe49cc9101 Mon Sep 17 00:00:00 2001 From: Kyle McMartin Date: Fri, 21 Apr 2006 02:20:37 +0000 Subject: [PARISC] Whitespace clean up in asm/pgtable.h Signed-off-by: Kyle McMartin --- include/asm-parisc/pgtable.h | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'include') diff --git a/include/asm-parisc/pgtable.h b/include/asm-parisc/pgtable.h index b6bcc672ba8..5066c54dae0 100644 --- a/include/asm-parisc/pgtable.h +++ b/include/asm-parisc/pgtable.h @@ -506,13 +506,13 @@ static inline void ptep_set_wrprotect(struct mm_struct *mm, unsigned long addr, /* TLB page size encoding - see table 3-1 in parisc20.pdf */ #define _PAGE_SIZE_ENCODING_4K 0 -#define _PAGE_SIZE_ENCODING_16K 1 -#define _PAGE_SIZE_ENCODING_64K 2 +#define _PAGE_SIZE_ENCODING_16K 1 +#define _PAGE_SIZE_ENCODING_64K 2 #define _PAGE_SIZE_ENCODING_256K 3 #define _PAGE_SIZE_ENCODING_1M 4 #define _PAGE_SIZE_ENCODING_4M 5 -#define _PAGE_SIZE_ENCODING_16M 6 -#define _PAGE_SIZE_ENCODING_64M 7 +#define _PAGE_SIZE_ENCODING_16M 6 +#define _PAGE_SIZE_ENCODING_64M 7 #if defined(CONFIG_PARISC_PAGE_SIZE_4KB) # define _PAGE_SIZE_ENCODING_DEFAULT _PAGE_SIZE_ENCODING_4K -- cgit v1.2.3-70-g09d2 From 3fd3a74f45c935f7d6d5c2fb48f06324b18826b7 Mon Sep 17 00:00:00 2001 From: Carlos O'Donell Date: Sat, 22 Apr 2006 14:47:21 -0600 Subject: [PARISC] Use FIXUP_BRANCH_CLOBBER to asm clobber list Joel Soete noticed correctly that the fixup's clobbers must be listed as the ASM clobbers. FIXUP_BRANCH in unaligned.c has a new macro which lists all the clobbers in the fixup, we use this throughout the file to simplify the process of listing clobbers in the future. A missing "r1" clobber is added to our uaccess.h for the 64-bit __put_kernel_asm. Interestingly this is a pretty serious bug since gcc generates pretty good use of r1 as a temporary and the uses of __put_kernel_asm are varied and dangerous if r1 is scratched during an invalid write. Signed-off-by: Joel Soete Signed-off-by: Carlos O'Donell Signed-off-by: Kyle McMartin --- arch/parisc/kernel/unaligned.c | 18 ++++++++++-------- include/asm-parisc/uaccess.h | 9 +++++++-- 2 files changed, 17 insertions(+), 10 deletions(-) (limited to 'include') diff --git a/arch/parisc/kernel/unaligned.c b/arch/parisc/kernel/unaligned.c index de0a1b21cb4..92328fbddb3 100644 --- a/arch/parisc/kernel/unaligned.c +++ b/arch/parisc/kernel/unaligned.c @@ -43,6 +43,8 @@ "\tldil L%%" #lbl ", %%r1\n" \ "\tldo R%%" #lbl "(%%r1), %%r1\n" \ "\tbv,n %%r0(%%r1)\n" +/* If you use FIXUP_BRANCH, then you must list this clobber */ +#define FIXUP_BRANCH_CLOBBER "r1" /* 1111 1100 0000 0000 0001 0011 1100 0000 */ #define OPCODE1(a,b,c) ((a)<<26|(b)<<12|(c)<<6) @@ -157,7 +159,7 @@ static int emulate_ldh(struct pt_regs *regs, int toreg) " .previous\n" : "=r" (val), "=r" (ret) : "0" (val), "r" (saddr), "r" (regs->isr) - : "r20" ); + : "r20", FIXUP_BRANCH_CLOBBER ); DPRINTF("val = 0x" RFMT "\n", val); @@ -202,7 +204,7 @@ static int emulate_ldw(struct pt_regs *regs, int toreg, int flop) " .previous\n" : "=r" (val), "=r" (ret) : "0" (val), "r" (saddr), "r" (regs->isr) - : "r19", "r20" ); + : "r19", "r20", FIXUP_BRANCH_CLOBBER ); DPRINTF("val = 0x" RFMT "\n", val); @@ -253,7 +255,7 @@ static int emulate_ldd(struct pt_regs *regs, int toreg, int flop) " .previous\n" : "=r" (val), "=r" (ret) : "0" (val), "r" (saddr), "r" (regs->isr) - : "r19", "r20" ); + : "r19", "r20", FIXUP_BRANCH_CLOBBER ); #else { unsigned long valh=0,vall=0; @@ -287,7 +289,7 @@ static int emulate_ldd(struct pt_regs *regs, int toreg, int flop) " .previous\n" : "=r" (valh), "=r" (vall), "=r" (ret) : "0" (valh), "1" (vall), "r" (saddr), "r" (regs->isr) - : "r19", "r20" ); + : "r19", "r20", FIXUP_BRANCH_CLOBBER ); val=((__u64)valh<<32)|(__u64)vall; } #endif @@ -335,7 +337,7 @@ static int emulate_sth(struct pt_regs *regs, int frreg) " .previous\n" : "=r" (ret) : "r" (val), "r" (regs->ior), "r" (regs->isr) - : "r19" ); + : "r19", FIXUP_BRANCH_CLOBBER ); return ret; } @@ -389,7 +391,7 @@ static int emulate_stw(struct pt_regs *regs, int frreg, int flop) " .previous\n" : "=r" (ret) : "r" (val), "r" (regs->ior), "r" (regs->isr) - : "r19", "r20", "r21", "r22", "r1" ); + : "r19", "r20", "r21", "r22", "r1", FIXUP_BRANCH_CLOBBER ); return 0; } @@ -450,7 +452,7 @@ static int emulate_std(struct pt_regs *regs, int frreg, int flop) " .previous\n" : "=r" (ret) : "r" (val), "r" (regs->ior), "r" (regs->isr) - : "r19", "r20", "r21", "r22", "r1" ); + : "r19", "r20", "r21", "r22", "r1", FIXUP_BRANCH_CLOBBER ); #else { unsigned long valh=(val>>32),vall=(val&0xffffffffl); @@ -495,7 +497,7 @@ static int emulate_std(struct pt_regs *regs, int frreg, int flop) " .previous\n" : "=r" (ret) : "r" (valh), "r" (vall), "r" (regs->ior), "r" (regs->isr) - : "r19", "r20", "r21", "r1" ); + : "r19", "r20", "r21", "r1", FIXUP_BRANCH_CLOBBER ); } #endif diff --git a/include/asm-parisc/uaccess.h b/include/asm-parisc/uaccess.h index f6c417c8c48..d973e8b3466 100644 --- a/include/asm-parisc/uaccess.h +++ b/include/asm-parisc/uaccess.h @@ -172,7 +172,11 @@ struct exception_data { /* * The "__put_user/kernel_asm()" macros tell gcc they read from memory * instead of writing. This is because they do not write to any memory - * gcc knows about, so there are no aliasing issues. + * gcc knows about, so there are no aliasing issues. These macros must + * also be aware that "fixup_put_user_skip_[12]" are executed in the + * context of the fault, and any registers used there must be listed + * as clobbers. In this case only "r1" is used by the current routines. + * r8/r9 are already listed as err/val. */ #ifdef __LP64__ @@ -183,7 +187,8 @@ struct exception_data { "\t.dword\t1b,fixup_put_user_skip_1\n" \ "\t.previous" \ : "=r"(__pu_err) \ - : "r"(ptr), "r"(x), "0"(__pu_err)) + : "r"(ptr), "r"(x), "0"(__pu_err) \ + : "r1") #define __put_user_asm(stx,x,ptr) \ __asm__ __volatile__ ( \ -- cgit v1.2.3-70-g09d2 From 1b6d1b5e073ad9313ad64da49dff649f5c06e546 Mon Sep 17 00:00:00 2001 From: Kyle McMartin Date: Fri, 19 May 2006 20:18:17 +0000 Subject: [PARISC] Use MFIA in current_text_addr on pa2.0 processors Signed-off-by: Kyle McMartin --- include/asm-parisc/processor.h | 14 ++++++-------- 1 file changed, 6 insertions(+), 8 deletions(-) (limited to 'include') diff --git a/include/asm-parisc/processor.h b/include/asm-parisc/processor.h index ca49dc91f4f..b73626f040d 100644 --- a/include/asm-parisc/processor.h +++ b/include/asm-parisc/processor.h @@ -26,14 +26,12 @@ * Default implementation of macro that returns current * instruction pointer ("program counter"). */ - -/* We cannot use MFIA as it was added for PA2.0 - prumpf - - At one point there were no "0f/0b" type local symbols in gas for - PA-RISC. This is no longer true, but this still seems like the - nicest way to implement this. */ - -#define current_text_addr() ({ void *pc; __asm__("\n\tblr 0,%0\n\tnop":"=r" (pc)); pc; }) +#ifdef CONFIG_PA20 +#define current_ia(x) __asm__("mfia %0" : "=r"(x)) +#else /* mfia added in pa2.0 */ +#define current_ia(x) __asm__("blr 0,%0\n\tnop" : "=r"(x)) +#endif +#define current_text_addr() ({ void *pc; current_ia(pc); pc; }) #define TASK_SIZE (current->thread.task_size) #define TASK_UNMAPPED_BASE (current->thread.map_base) -- cgit v1.2.3-70-g09d2 From c95f2e5f2f6f61d734a025414c9eb81872a5c831 Mon Sep 17 00:00:00 2001 From: Carlos O'Donell Date: Wed, 21 Jun 2006 03:43:52 +0000 Subject: [PARISC] Remove unconditional #define PIC in syscall macros Signed-off-by: Carlos O'Donell Signed-off-by: Kyle McMartin --- include/asm-parisc/unistd.h | 5 ----- 1 file changed, 5 deletions(-) (limited to 'include') diff --git a/include/asm-parisc/unistd.h b/include/asm-parisc/unistd.h index 12b867238a4..27bcfad1c3e 100644 --- a/include/asm-parisc/unistd.h +++ b/include/asm-parisc/unistd.h @@ -797,11 +797,6 @@ #define SYS_ify(syscall_name) __NR_##syscall_name -/* Assume all syscalls are done from PIC code just to be - * safe. The worst case scenario is that you lose a register - * and save/restore r19 across the syscall. */ -#define PIC - #ifndef ASM_LINE_SEP # define ASM_LINE_SEP ; #endif -- cgit v1.2.3-70-g09d2 From 8ffaeaf42e91930888df09d696a8a6ebe056d0e0 Mon Sep 17 00:00:00 2001 From: Thibaut Varene Date: Wed, 3 May 2006 17:27:35 -0600 Subject: [PARISC] PDC_CHASSIS is implemented on all machines This patch removes a limitation of the original code, so that CHASSIS codes can be sent to all machines. On machines with a LCD panel, this code displays "INI" during bootup, "RUN" when the system is booted and running, "FLT" when a panic occurs, etc. This part of the code can be enabled/disabled through CONFIG_PDC_CHASSIS This patch also adds minimalistic support for Chassis warnings, through a proc entry '/proc/chassis', which will reflect the warnings status (PSU or fans failure when they happen, NVRAM battery level and temperature thresholds overflows). This part of the code can be enabled/disabled through CONFIG_PDC_CHASSIS_WARN Signed-off-by: Thibaut VARENE Signed-off-by: Kyle McMartin --- arch/parisc/kernel/firmware.c | 22 +++++++++--- arch/parisc/kernel/pdc_chassis.c | 73 ++++++++++++++++++++++++++++++++++------ drivers/parisc/Kconfig | 33 ++++++++++++++---- include/asm-parisc/pdc.h | 1 + 4 files changed, 107 insertions(+), 22 deletions(-) (limited to 'include') diff --git a/arch/parisc/kernel/firmware.c b/arch/parisc/kernel/firmware.c index 2dc06b8e181..0596f27340c 100644 --- a/arch/parisc/kernel/firmware.c +++ b/arch/parisc/kernel/firmware.c @@ -11,7 +11,7 @@ * Copyright 1999 The Puffin Group, (Alex deVries, David Kennedy) * Copyright 2003 Grant Grundler * Copyright 2003,2004 Ryan Bradetich - * Copyright 2004 Thibaut VARENE + * Copyright 2004,2006 Thibaut VARENE * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License as published by @@ -252,10 +252,8 @@ int pdc_pat_chassis_send_log(unsigned long state, unsigned long data) #endif /** - * pdc_chassis_disp - Updates display + * pdc_chassis_disp - Updates chassis code * @retval: -1 on error, 0 on success - * - * Works on old PDC only (E class, others?) */ int pdc_chassis_disp(unsigned long disp) { @@ -268,6 +266,22 @@ int pdc_chassis_disp(unsigned long disp) return retval; } +/** + * pdc_chassis_warn - Fetches chassis warnings + * @retval: -1 on error, 0 on success + */ +int pdc_chassis_warn(unsigned long *warn) +{ + int retval = 0; + + spin_lock_irq(&pdc_lock); + retval = mem_pdc_call(PDC_CHASSIS, PDC_CHASSIS_WARN, __pa(pdc_result)); + *warn = pdc_result[0]; + spin_unlock_irq(&pdc_lock); + + return retval; +} + /** * pdc_coproc_cfg - To identify coprocessors attached to the processor. * @pdc_coproc_info: Return buffer address. diff --git a/arch/parisc/kernel/pdc_chassis.c b/arch/parisc/kernel/pdc_chassis.c index a45e2e2ffd9..51e86c0365c 100644 --- a/arch/parisc/kernel/pdc_chassis.c +++ b/arch/parisc/kernel/pdc_chassis.c @@ -1,8 +1,8 @@ /* - * interfaces to log Chassis Codes via PDC (firmware) + * interfaces to Chassis Codes via PDC (firmware) * * Copyright (C) 2002 Laurent Canet - * Copyright (C) 2002-2004 Thibaut VARENE + * Copyright (C) 2002-2006 Thibaut VARENE * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License, version 2, as @@ -16,6 +16,9 @@ * You should have received a copy of the GNU General Public License * along with this program; if not, write to the Free Software * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA + * + * TODO: poll chassis warns, trigger (configurable) machine shutdown when + * needed. */ #undef PDC_CHASSIS_DEBUG @@ -30,6 +33,7 @@ #include #include #include +#include #include #include @@ -38,7 +42,6 @@ #ifdef CONFIG_PDC_CHASSIS -static int pdc_chassis_old __read_mostly = 0; static unsigned int pdc_chassis_enabled __read_mostly = 1; @@ -64,7 +67,7 @@ __setup("pdcchassis=", pdc_chassis_setup); * Currently, only E class and A180 are known to work with this. * Inspired by Christoph Plattner */ - +#if 0 static void __init pdc_chassis_checkold(void) { switch(CPU_HVERSION) { @@ -73,7 +76,6 @@ static void __init pdc_chassis_checkold(void) case 0x482: /* E45 */ case 0x483: /* E55 */ case 0x516: /* A180 */ - pdc_chassis_old = 1; break; default: @@ -81,7 +83,7 @@ static void __init pdc_chassis_checkold(void) } DPRINTK(KERN_DEBUG "%s: pdc_chassis_checkold(); pdc_chassis_old = %d\n", __FILE__, pdc_chassis_old); } - +#endif /** * pdc_chassis_panic_event() - Called by the panic handler. @@ -136,14 +138,13 @@ void __init parisc_pdc_chassis_init(void) DPRINTK(KERN_DEBUG "%s: parisc_pdc_chassis_init()\n", __FILE__); /* Let see if we have something to handle... */ - /* Check for PDC_PAT or old LED Panel */ - pdc_chassis_checkold(); + /* Check for PDC_PAT */ if (is_pdc_pat()) { printk(KERN_INFO "Enabling PDC_PAT chassis codes support.\n"); handle = 1; } - else if (unlikely(pdc_chassis_old)) { - printk(KERN_INFO "Enabling old style chassis LED panel support.\n"); + else { + printk(KERN_INFO "Enabling regular chassis codes support.\n"); handle = 1; } @@ -215,9 +216,12 @@ int pdc_chassis_send_status(int message) } } else retval = -1; #else - if (unlikely(pdc_chassis_old)) { + if (1) { switch (message) { case PDC_CHASSIS_DIRECT_BSTART: + retval = pdc_chassis_disp(PDC_CHASSIS_DISP_DATA(OSTAT_INIT)); + break; + case PDC_CHASSIS_DIRECT_BCOMPLETE: retval = pdc_chassis_disp(PDC_CHASSIS_DISP_DATA(OSTAT_RUN)); break; @@ -244,3 +248,50 @@ int pdc_chassis_send_status(int message) #endif /* CONFIG_PDC_CHASSIS */ return retval; } + +#ifdef CONFIG_PDC_CHASSIS_WARN +#ifdef CONFIG_PROC_FS +static int pdc_chassis_warn_pread(char *page, char **start, off_t off, + int count, int *eof, void *data) +{ + char *out = page; + int len, ret; + unsigned long warn; + u32 warnreg; + + ret = pdc_chassis_warn(&warn); + if (ret != PDC_OK) + return -EIO; + + warnreg = (warn & 0xFFFFFFFF); + + if ((warnreg >> 24) & 0xFF) + out += sprintf(out, "Chassis component failure! (eg fan or PSU): 0x%.2x\n", ((warnreg >> 24) & 0xFF)); + + out += sprintf(out, "Battery: %s\n", (warnreg & 0x04) ? "Low!" : "OK"); + out += sprintf(out, "Temp low: %s\n", (warnreg & 0x02) ? "Exceeded!" : "OK"); + out += sprintf(out, "Temp mid: %s\n", (warnreg & 0x01) ? "Exceeded!" : "OK"); + + len = out - page - off; + if (len < count) { + *eof = 1; + if (len <= 0) return 0; + } else { + len = count; + } + *start = page + off; + return len; +} + +static int __init pdc_chassis_create_procfs(void) +{ + printk(KERN_INFO "Enabling PDC chassis warnings support.\n"); + create_proc_read_entry("chassis", 0400, NULL, pdc_chassis_warn_pread, + NULL); + return 0; +} + +__initcall(pdc_chassis_create_procfs); + +#endif /* CONFIG_PROC_FS */ +#endif /* CONFIG_PDC_CHASSIS_WARN */ diff --git a/drivers/parisc/Kconfig b/drivers/parisc/Kconfig index 3f5de867acd..1d3b84b4af3 100644 --- a/drivers/parisc/Kconfig +++ b/drivers/parisc/Kconfig @@ -140,18 +140,37 @@ config CHASSIS_LCD_LED If unsure, say Y. config PDC_CHASSIS - bool "PDC chassis State Panel support" + bool "PDC chassis state codes support" default y help - Say Y here if you want to enable support for the LED State front - panel as found on E class, and support for the GSP Virtual Front - Panel (LED State and message logging) as found on high end - servers such as A, L and N-class. - - This has nothing to do with Chassis LCD and LED support. + Say Y here if you want to enable support for Chassis codes. + That includes support for LED State front panel as found on E + class, and support for the GSP Virtual Front Panel (LED State and + message logging) as found on high end servers such as A, L and + N-class. + This driver will also display progress messages on LCD display, + such as "INI", "RUN" and "FLT", and might thus clobber messages + shown by the LED/LCD driver. + This driver updates the state panel (LED and/or LCD) upon system + state change (eg: boot, shutdown or panic). If unsure, say Y. + +config PDC_CHASSIS_WARN + bool "PDC chassis warnings support" + depends on PROC_FS + default y + help + Say Y here if you want to enable support for Chassis warnings. + This will add a proc entry '/proc/chassis' giving some information + about the overall health state of the system. + This includes NVRAM battery level, overtemp or failures such as + fans or power units. + + If unsure, say Y. + + config PDC_STABLE tristate "PDC Stable Storage support" depends on SYSFS diff --git a/include/asm-parisc/pdc.h b/include/asm-parisc/pdc.h index 08364f957e7..7d8a71a2854 100644 --- a/include/asm-parisc/pdc.h +++ b/include/asm-parisc/pdc.h @@ -719,6 +719,7 @@ void setup_pdc(void); /* in inventory.c */ int pdc_add_valid(unsigned long address); int pdc_chassis_info(struct pdc_chassis_info *chassis_info, void *led_info, unsigned long len); int pdc_chassis_disp(unsigned long disp); +int pdc_chassis_warn(unsigned long *warn); int pdc_coproc_cfg(struct pdc_coproc_cfg *pdc_coproc_info); int pdc_iodc_read(unsigned long *actcnt, unsigned long hpa, unsigned int index, void *iodc_data, unsigned int iodc_data_size); -- cgit v1.2.3-70-g09d2 From a3ea84fabacd9cc5bcc3fda67c35e692ca10dc8c Mon Sep 17 00:00:00 2001 From: Kyle McMartin Date: Fri, 16 Jun 2006 19:10:02 +0000 Subject: [PARISC] Add is_compat_task() helper ... And convert signal.c and ptrace.c to use it instead of open coded equivalents. Signed-off-by: Kyle McMartin --- arch/parisc/kernel/ptrace.c | 8 ++++---- arch/parisc/kernel/signal.c | 20 ++++++++++---------- include/asm-parisc/compat.h | 11 +++++++++++ 3 files changed, 25 insertions(+), 14 deletions(-) (limited to 'include') diff --git a/arch/parisc/kernel/ptrace.c b/arch/parisc/kernel/ptrace.c index 413292f1a4a..3f28de97455 100644 --- a/arch/parisc/kernel/ptrace.c +++ b/arch/parisc/kernel/ptrace.c @@ -91,7 +91,7 @@ long arch_ptrace(struct task_struct *child, long request, long addr, long data) int copied; #ifdef __LP64__ - if (personality(child->personality) == PER_LINUX32) { + if (__is_compat_task(child)) { unsigned int tmp; addr &= 0xffffffffL; @@ -123,7 +123,7 @@ long arch_ptrace(struct task_struct *child, long request, long addr, long data) case PTRACE_POKEDATA: ret = 0; #ifdef __LP64__ - if (personality(child->personality) == PER_LINUX32) { + if (__is_compat_task(child)) { unsigned int tmp = (unsigned int)data; DBG("sys_ptrace(POKE%s, %d, %lx, %lx)\n", request == PTRACE_POKETEXT ? "TEXT" : "DATA", @@ -146,7 +146,7 @@ long arch_ptrace(struct task_struct *child, long request, long addr, long data) case PTRACE_PEEKUSR: { ret = -EIO; #ifdef __LP64__ - if (personality(child->personality) == PER_LINUX32) { + if (__is_compat_task(child)) { unsigned int tmp; if (addr & (sizeof(int)-1)) @@ -205,7 +205,7 @@ long arch_ptrace(struct task_struct *child, long request, long addr, long data) goto out_tsk; } #ifdef __LP64__ - if (personality(child->personality) == PER_LINUX32) { + if (__is_compat_task(child)) { if (addr & (sizeof(int)-1)) goto out_tsk; if ((addr = translate_usr_offset(addr)) < 0) diff --git a/arch/parisc/kernel/signal.c b/arch/parisc/kernel/signal.c index cc38edfd90c..bb83880c5ee 100644 --- a/arch/parisc/kernel/signal.c +++ b/arch/parisc/kernel/signal.c @@ -76,7 +76,7 @@ sys_rt_sigsuspend(sigset_t __user *unewset, size_t sigsetsize, struct pt_regs *r #ifdef __LP64__ compat_sigset_t newset32; - if(personality(current->personality) == PER_LINUX32){ + if (is_compat_task()) { /* XXX: Don't preclude handling different sized sigset_t's. */ if (sigsetsize != sizeof(compat_sigset_t)) return -EINVAL; @@ -153,7 +153,7 @@ sys_rt_sigreturn(struct pt_regs *regs, int in_syscall) compat_sigset_t compat_set; struct compat_rt_sigframe __user * compat_frame; - if(personality(current->personality) == PER_LINUX32) + if (is_compat_task()) sigframe_size = PARISC_RT_SIGFRAME_SIZE32; #endif @@ -166,7 +166,7 @@ sys_rt_sigreturn(struct pt_regs *regs, int in_syscall) #ifdef __LP64__ compat_frame = (struct compat_rt_sigframe __user *)frame; - if(personality(current->personality) == PER_LINUX32){ + if (is_compat_task()) { DBG(2,"sys_rt_sigreturn: ELF32 process.\n"); if (__copy_from_user(&compat_set, &compat_frame->uc.uc_sigmask, sizeof(compat_set))) goto give_sigsegv; @@ -186,7 +186,7 @@ sys_rt_sigreturn(struct pt_regs *regs, int in_syscall) /* Good thing we saved the old gr[30], eh? */ #ifdef __LP64__ - if(personality(current->personality) == PER_LINUX32){ + if (is_compat_task()) { DBG(1,"sys_rt_sigreturn: compat_frame->uc.uc_mcontext 0x%p\n", &compat_frame->uc.uc_mcontext); // FIXME: Load upper half from register file @@ -315,7 +315,7 @@ setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info, compat_frame = (struct compat_rt_sigframe __user *)frame; - if(personality(current->personality) == PER_LINUX32) { + if (is_compat_task()) { DBG(1,"setup_rt_frame: frame->info = 0x%p\n", &compat_frame->info); err |= copy_siginfo_to_user32(&compat_frame->info, info); DBG(1,"SETUP_RT_FRAME: 1\n"); @@ -392,7 +392,7 @@ setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info, haddr = A(ka->sa.sa_handler); /* The sa_handler may be a pointer to a function descriptor */ #ifdef __LP64__ - if(personality(current->personality) == PER_LINUX32) { + if (is_compat_task()) { #endif if (haddr & PA_PLABEL_FDESC) { Elf32_Fdesc fdesc; @@ -427,19 +427,19 @@ setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info, */ sigframe_size = PARISC_RT_SIGFRAME_SIZE; #ifdef __LP64__ - if(personality(current->personality) == PER_LINUX32) + if (is_compat_task()) sigframe_size = PARISC_RT_SIGFRAME_SIZE32; #endif if (in_syscall) { regs->gr[31] = haddr; #ifdef __LP64__ - if(personality(current->personality) == PER_LINUX) + if (personality(current->personality) == PER_LINUX) sigframe_size |= 1; #endif } else { unsigned long psw = USER_PSW; #ifdef __LP64__ - if(personality(current->personality) == PER_LINUX) + if (personality(current->personality) == PER_LINUX) psw |= PSW_W; #endif @@ -464,7 +464,7 @@ setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info, regs->gr[26] = sig; /* signal number */ #ifdef __LP64__ - if(personality(current->personality) == PER_LINUX32){ + if (is_compat_task()) { regs->gr[25] = A(&compat_frame->info); /* siginfo pointer */ regs->gr[24] = A(&compat_frame->uc); /* ucontext pointer */ } else diff --git a/include/asm-parisc/compat.h b/include/asm-parisc/compat.h index 289624d8b2d..71b4eeea205 100644 --- a/include/asm-parisc/compat.h +++ b/include/asm-parisc/compat.h @@ -5,6 +5,7 @@ */ #include #include +#include #define COMPAT_USER_HZ 100 @@ -149,4 +150,14 @@ static __inline__ void __user *compat_alloc_user_space(long len) return (void __user *)regs->gr[30]; } +static inline int __is_compat_task(struct task_struct *t) +{ + return personality(t->personality) == PER_LINUX32; +} + +static inline int is_compat_task(void) +{ + return __is_compat_task(current); +} + #endif /* _ASM_PARISC_COMPAT_H */ -- cgit v1.2.3-70-g09d2 From e5a2e7fdb53028ce187c0ce0ae0d45ca7546fd5e Mon Sep 17 00:00:00 2001 From: Kyle McMartin Date: Wed, 14 Jun 2006 20:26:25 +0000 Subject: [PARISC] Match show_cache_info with reality show_cache_info and struct pdc_cache_cf were out of sync with published documentation. Fix the reporting of cache associativity and update the pdc_cache_cf bitfields to match documentation. Signed-off-by: Kyle McMartin --- arch/parisc/kernel/cache.c | 18 ++++++++++-------- include/asm-parisc/pdc.h | 4 ++-- 2 files changed, 12 insertions(+), 10 deletions(-) (limited to 'include') diff --git a/arch/parisc/kernel/cache.c b/arch/parisc/kernel/cache.c index c057ad7605b..851519cbad6 100644 --- a/arch/parisc/kernel/cache.c +++ b/arch/parisc/kernel/cache.c @@ -97,15 +97,17 @@ update_mmu_cache(struct vm_area_struct *vma, unsigned long address, pte_t pte) void show_cache_info(struct seq_file *m) { + char buf[32]; + seq_printf(m, "I-cache\t\t: %ld KB\n", cache_info.ic_size/1024 ); - seq_printf(m, "D-cache\t\t: %ld KB (%s%s, %d-way associative)\n", + if (cache_info.dc_loop == 1) + snprintf(buf, 32, "%lu-way associative", cache_info.dc_loop); + seq_printf(m, "D-cache\t\t: %ld KB (%s%s, %s)\n", cache_info.dc_size/1024, (cache_info.dc_conf.cc_wt ? "WT":"WB"), (cache_info.dc_conf.cc_sh ? ", shared I/D":""), - (cache_info.dc_conf.cc_assoc) - ); - + ((cache_info.dc_loop == 1) ? "direct mapped" : buf)); seq_printf(m, "ITLB entries\t: %ld\n" "DTLB entries\t: %ld%s\n", cache_info.it_size, cache_info.dt_size, @@ -158,11 +160,11 @@ parisc_cache_init(void) cache_info.dc_conf.cc_block, cache_info.dc_conf.cc_line, cache_info.dc_conf.cc_shift); - printk(" wt %d sh %d cst %d assoc %d\n", + printk(" wt %d sh %d cst %d hv %d\n", cache_info.dc_conf.cc_wt, cache_info.dc_conf.cc_sh, cache_info.dc_conf.cc_cst, - cache_info.dc_conf.cc_assoc); + cache_info.dc_conf.cc_hv); printk("IC base 0x%lx stride 0x%lx count 0x%lx loop 0x%lx\n", cache_info.ic_base, @@ -176,11 +178,11 @@ parisc_cache_init(void) cache_info.ic_conf.cc_block, cache_info.ic_conf.cc_line, cache_info.ic_conf.cc_shift); - printk(" wt %d sh %d cst %d assoc %d\n", + printk(" wt %d sh %d cst %d hv %d\n", cache_info.ic_conf.cc_wt, cache_info.ic_conf.cc_sh, cache_info.ic_conf.cc_cst, - cache_info.ic_conf.cc_assoc); + cache_info.ic_conf.cc_hv); printk("D-TLB conf: sh %d page %d cst %d aid %d pad1 %d \n", cache_info.dt_conf.tc_sh, diff --git a/include/asm-parisc/pdc.h b/include/asm-parisc/pdc.h index 7d8a71a2854..592f13bcc98 100644 --- a/include/asm-parisc/pdc.h +++ b/include/asm-parisc/pdc.h @@ -352,8 +352,8 @@ struct pdc_cache_cf { /* for PDC_CACHE (I/D-caches) */ cc_wt : 1, /* 0 = WT-Dcache, 1 = WB-Dcache */ cc_sh : 2, /* 0 = separate I/D-cache, else shared I/D-cache */ cc_cst : 3, /* 0 = incoherent D-cache, 1=coherent D-cache */ - cc_pad1 : 5, /* reserved */ - cc_assoc: 8; /* associativity of I/D-cache */ + cc_pad1 : 10, /* reserved */ + cc_hv : 3; /* hversion dependent */ }; struct pdc_tlb_cf { /* for PDC_CACHE (I/D-TLB's) */ -- cgit v1.2.3-70-g09d2 From a9d2d386c42b816a81eb1d02dfb7b502ba8cea82 Mon Sep 17 00:00:00 2001 From: Kyle McMartin Date: Fri, 16 Jun 2006 18:20:00 -0400 Subject: [PARISC] Ensure Space ID hashing is turned off Check PDC_CACHE to see if spaceid hashing is turned on, and fail to boot if that is the case. However, some old machines do not implement the PDC_CACHE_RET_SPID firmware call, so continue to boot if the call fails because of PDC_BAD_OPTION (but fail in all other error returns). Signed-off-by: Kyle McMartin --- arch/parisc/kernel/cache.c | 10 +++++++++- arch/parisc/kernel/firmware.c | 20 ++++++++++++++++++++ include/asm-parisc/pdc.h | 1 + 3 files changed, 30 insertions(+), 1 deletion(-) (limited to 'include') diff --git a/arch/parisc/kernel/cache.c b/arch/parisc/kernel/cache.c index 851519cbad6..bc7c4a4e26a 100644 --- a/arch/parisc/kernel/cache.c +++ b/arch/parisc/kernel/cache.c @@ -236,7 +236,8 @@ parisc_cache_init(void) void disable_sr_hashing(void) { - int srhash_type; + int srhash_type, retval; + unsigned long space_bits; switch (boot_cpu_data.cpu_type) { case pcx: /* We shouldn't get this far. setup.c should prevent it. */ @@ -262,6 +263,13 @@ void disable_sr_hashing(void) } disable_sr_hashing_asm(srhash_type); + + retval = pdc_spaceid_bits(&space_bits); + /* If this procedure isn't implemented, don't panic. */ + if (retval < 0 && retval != PDC_BAD_OPTION) + panic("pdc_spaceid_bits call failed.\n"); + if (space_bits != 0) + panic("SpaceID hashing is still on!\n"); } void flush_dcache_page(struct page *page) diff --git a/arch/parisc/kernel/firmware.c b/arch/parisc/kernel/firmware.c index 0596f27340c..c80c277454f 100644 --- a/arch/parisc/kernel/firmware.c +++ b/arch/parisc/kernel/firmware.c @@ -512,6 +512,26 @@ int pdc_cache_info(struct pdc_cache_info *cache_info) return retval; } +/** + * pdc_spaceid_bits - Return whether Space ID hashing is turned on. + * @space_bits: Should be 0, if not, bad mojo! + * + * Returns information about Space ID hashing. + */ +int pdc_spaceid_bits(unsigned long *space_bits) +{ + int retval; + + spin_lock_irq(&pdc_lock); + pdc_result[0] = 0; + retval = mem_pdc_call(PDC_CACHE, PDC_CACHE_RET_SPID, __pa(pdc_result), 0); + convert_to_wide(pdc_result); + *space_bits = pdc_result[0]; + spin_unlock_irq(&pdc_lock); + + return retval; +} + #ifndef CONFIG_PA20 /** * pdc_btlb_info - Return block TLB information. diff --git a/include/asm-parisc/pdc.h b/include/asm-parisc/pdc.h index 592f13bcc98..3d3446ec207 100644 --- a/include/asm-parisc/pdc.h +++ b/include/asm-parisc/pdc.h @@ -733,6 +733,7 @@ int pdc_model_cpuid(unsigned long *cpu_id); int pdc_model_versions(unsigned long *versions, int id); int pdc_model_capabilities(unsigned long *capabilities); int pdc_cache_info(struct pdc_cache_info *cache); +int pdc_spaceid_bits(unsigned long *space_bits); #ifndef CONFIG_PA20 int pdc_btlb_info(struct pdc_btlb_info *btlb); int pdc_mem_map_hpa(struct pdc_memory_map *r_addr, struct pdc_module_path *mod_path); -- cgit v1.2.3-70-g09d2 From ec1fdc24c2ae012b078ba0187ceef208e08a3aec Mon Sep 17 00:00:00 2001 From: Kyle McMartin Date: Wed, 21 Jun 2006 19:27:29 +0000 Subject: [PARISC] OS_ID_LINUX == 0x0006 We were assigned an OS_ID of 0x0006. Consistently use OS_ID_LINUX instead of using the magic number. Also update the OS_ID_ defines in asm/pdc.h to reflect this. Signed-off-by: Kyle McMartin --- arch/parisc/kernel/firmware.c | 4 +++- arch/parisc/kernel/setup.c | 2 +- drivers/parisc/pdc_stable.c | 4 ++-- include/asm-parisc/pdc.h | 3 +-- 4 files changed, 7 insertions(+), 6 deletions(-) (limited to 'include') diff --git a/arch/parisc/kernel/firmware.c b/arch/parisc/kernel/firmware.c index c80c277454f..4398d2a9578 100644 --- a/arch/parisc/kernel/firmware.c +++ b/arch/parisc/kernel/firmware.c @@ -407,7 +407,9 @@ int pdc_model_info(struct pdc_model *model) * pdc_model_sysmodel - Get the system model name. * @name: A char array of at least 81 characters. * - * Get system model name from PDC ROM (e.g. 9000/715 or 9000/778/B160L) + * Get system model name from PDC ROM (e.g. 9000/715 or 9000/778/B160L). + * Using OS_ID_HPUX will return the equivalent of the 'modelname' command + * on HP/UX. */ int pdc_model_sysmodel(char *name) { diff --git a/arch/parisc/kernel/setup.c b/arch/parisc/kernel/setup.c index 8471486a756..278f4b9f6a3 100644 --- a/arch/parisc/kernel/setup.c +++ b/arch/parisc/kernel/setup.c @@ -303,7 +303,7 @@ extern void eisa_init(void); static int __init parisc_init(void) { - u32 osid = (0x0006 << 16); + u32 osid = (OS_ID_LINUX << 16); parisc_proc_mkdir(); parisc_init_resources(); diff --git a/drivers/parisc/pdc_stable.c b/drivers/parisc/pdc_stable.c index a1094210cd1..97b34585804 100644 --- a/drivers/parisc/pdc_stable.c +++ b/drivers/parisc/pdc_stable.c @@ -884,7 +884,7 @@ pdcs_osdep1_write(struct subsystem *entry, const char *buf, size_t count) if (!entry || !buf || !count) return -EINVAL; - if (unlikely(pdcs_osid != 0x0006)) + if (unlikely(pdcs_osid != OS_ID_LINUX)) return -EPERM; if (count > 16) @@ -926,7 +926,7 @@ pdcs_osdep2_write(struct subsystem *entry, const char *buf, size_t count) if (unlikely(pdcs_size <= 224)) return -ENOSYS; - if (unlikely(pdcs_osid != 0x0006)) + if (unlikely(pdcs_osid != OS_ID_LINUX)) return -EPERM; size = pdcs_size - 224; diff --git a/include/asm-parisc/pdc.h b/include/asm-parisc/pdc.h index 3d3446ec207..33bff615d6e 100644 --- a/include/asm-parisc/pdc.h +++ b/include/asm-parisc/pdc.h @@ -278,12 +278,11 @@ typedef struct { /* constants for OS (NVM...) */ #define OS_ID_NONE 0 /* Undefined OS ID */ #define OS_ID_HPUX 1 /* HP-UX OS */ -#define OS_ID_LINUX OS_ID_HPUX /* just use the same value as hpux */ #define OS_ID_MPEXL 2 /* MPE XL OS */ #define OS_ID_OSF 3 /* OSF OS */ #define OS_ID_HPRT 4 /* HP-RT OS */ #define OS_ID_NOVEL 5 /* NOVELL OS */ -#define OS_ID_NT 6 /* NT OS */ +#define OS_ID_LINUX 6 /* Linux */ /* constants for PDC_CHASSIS */ -- cgit v1.2.3-70-g09d2 From 67a061a191017f984a1ef0ff73ae988b9b15f6d3 Mon Sep 17 00:00:00 2001 From: Kyle McMartin Date: Sun, 25 Jun 2006 16:58:57 +0000 Subject: [PARISC] Add os_id_to_string helper Add a helper to asm/pdc.h to translate OS_ID values to strings and use it in the pdc_stable driver. Signed-off-by: Kyle McMartin --- drivers/parisc/pdc_stable.c | 14 ++------------ include/asm-parisc/pdc.h | 13 +++++++++++++ 2 files changed, 15 insertions(+), 12 deletions(-) (limited to 'include') diff --git a/drivers/parisc/pdc_stable.c b/drivers/parisc/pdc_stable.c index 97b34585804..ea1b7a63598 100644 --- a/drivers/parisc/pdc_stable.c +++ b/drivers/parisc/pdc_stable.c @@ -622,22 +622,12 @@ static ssize_t pdcs_osid_read(struct subsystem *entry, char *buf) { char *out = buf; - char *tmpstr = NULL; if (!entry || !buf) return -EINVAL; - switch (pdcs_osid) { - case 0x0000: tmpstr = "No OS"; break; - case 0x0001: tmpstr = "HP-UX"; break; - case 0x0002: tmpstr = "MPE-iX"; break; - case 0x0003: tmpstr = "OSF"; break; - case 0x0004: tmpstr = "HP-RT"; break; - case 0x0005: tmpstr = "Novell Netware"; break; - case 0x0006: tmpstr = "Linux"; break; - default: tmpstr = "Unknown"; break; - } - out += sprintf(out, "%s dependent data (0x%.4x)\n", tmpstr, pdcs_osid); + out += sprintf(out, "%s dependent data (0x%.4x)\n", + os_id_to_string(pdcs_osid), pdcs_osid); return out - buf; } diff --git a/include/asm-parisc/pdc.h b/include/asm-parisc/pdc.h index 33bff615d6e..7a21f106ade 100644 --- a/include/asm-parisc/pdc.h +++ b/include/asm-parisc/pdc.h @@ -284,6 +284,19 @@ typedef struct { #define OS_ID_NOVEL 5 /* NOVELL OS */ #define OS_ID_LINUX 6 /* Linux */ +static inline char * os_id_to_string(u16 os_id) { + switch(os_id) { + case OS_ID_NONE: return "No OS"; + case OS_ID_HPUX: return "HP-UX"; + case OS_ID_MPEXL: return "MPE-iX"; + case OS_ID_OSF: return "OSF"; + case OS_ID_HPRT: return "HP-RT"; + case OS_ID_NOVEL: return "Novell Netware"; + case OS_ID_LINUX: return "Linux"; + default: return "Unknown"; + } +} + /* constants for PDC_CHASSIS */ #define OSTAT_OFF 0 -- cgit v1.2.3-70-g09d2 From c1a7a755be26f68d7f21ee769584149a96185ea8 Mon Sep 17 00:00:00 2001 From: Kyle McMartin Date: Mon, 26 Jun 2006 11:28:09 -0400 Subject: [PARISC] Move os_id_to_string() inside #ifndef __ASSEMBLY__ Assembly files should be protected from os_id_to_string()... Pass me a brown paper bag, please. Signed-off-by: Kyle McMartin --- include/asm-parisc/pdc.h | 25 ++++++++++++------------- 1 file changed, 12 insertions(+), 13 deletions(-) (limited to 'include') diff --git a/include/asm-parisc/pdc.h b/include/asm-parisc/pdc.h index 7a21f106ade..c9b2e35326e 100644 --- a/include/asm-parisc/pdc.h +++ b/include/asm-parisc/pdc.h @@ -284,19 +284,6 @@ typedef struct { #define OS_ID_NOVEL 5 /* NOVELL OS */ #define OS_ID_LINUX 6 /* Linux */ -static inline char * os_id_to_string(u16 os_id) { - switch(os_id) { - case OS_ID_NONE: return "No OS"; - case OS_ID_HPUX: return "HP-UX"; - case OS_ID_MPEXL: return "MPE-iX"; - case OS_ID_OSF: return "OSF"; - case OS_ID_HPRT: return "HP-RT"; - case OS_ID_NOVEL: return "Novell Netware"; - case OS_ID_LINUX: return "Linux"; - default: return "Unknown"; - } -} - /* constants for PDC_CHASSIS */ #define OSTAT_OFF 0 @@ -789,6 +776,18 @@ int pdc_sti_call(unsigned long func, unsigned long flags, extern void pdc_init(void); +static inline char * os_id_to_string(u16 os_id) { + switch(os_id) { + case OS_ID_NONE: return "No OS"; + case OS_ID_HPUX: return "HP-UX"; + case OS_ID_MPEXL: return "MPE-iX"; + case OS_ID_OSF: return "OSF"; + case OS_ID_HPRT: return "HP-RT"; + case OS_ID_NOVEL: return "Novell Netware"; + case OS_ID_LINUX: return "Linux"; + default: return "Unknown"; + } +} #endif /* __ASSEMBLY__ */ #endif /* _PARISC_PDC_H */ -- cgit v1.2.3-70-g09d2 From bc02af93dd2bbddce1b55e0a493f833a1b7cf140 Mon Sep 17 00:00:00 2001 From: Yasunori Goto Date: Tue, 27 Jun 2006 02:53:30 -0700 Subject: [PATCH] pgdat allocation for new node add (specify node id) Change the name of old add_memory() to arch_add_memory. And use node id to get pgdat for the node at NODE_DATA(). Note: Powerpc's old add_memory() is defined as __devinit. However, add_memory() is usually called only after bootup. I suppose it may be redundant. But, I'm not well known about powerpc. So, I keep it. (But, __meminit is better at least.) Signed-off-by: Yasunori Goto Cc: Dave Hansen Cc: "Brown, Len" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/i386/mm/init.c | 2 +- arch/ia64/mm/init.c | 5 ++-- arch/powerpc/mm/mem.c | 11 +++++-- arch/x86_64/mm/init.c | 66 ++++++++++++++++++++++++------------------ drivers/acpi/Kconfig | 2 +- drivers/acpi/acpi_memhotplug.c | 3 +- drivers/base/memory.c | 4 ++- include/linux/memory_hotplug.h | 13 ++++++++- mm/memory_hotplug.c | 11 +++++++ 9 files changed, 78 insertions(+), 39 deletions(-) (limited to 'include') diff --git a/arch/i386/mm/init.c b/arch/i386/mm/init.c index bf19513f0ce..46859253179 100644 --- a/arch/i386/mm/init.c +++ b/arch/i386/mm/init.c @@ -654,7 +654,7 @@ void __init mem_init(void) */ #ifdef CONFIG_MEMORY_HOTPLUG #ifndef CONFIG_NEED_MULTIPLE_NODES -int add_memory(u64 start, u64 size) +int arch_add_memory(int nid, u64 start, u64 size) { struct pglist_data *pgdata = &contig_page_data; struct zone *zone = pgdata->node_zones + MAX_NR_ZONES-1; diff --git a/arch/ia64/mm/init.c b/arch/ia64/mm/init.c index 11f08001f8c..38306e98f04 100644 --- a/arch/ia64/mm/init.c +++ b/arch/ia64/mm/init.c @@ -652,7 +652,7 @@ void online_page(struct page *page) num_physpages++; } -int add_memory(u64 start, u64 size) +int arch_add_memory(int nid, u64 start, u64 size) { pg_data_t *pgdat; struct zone *zone; @@ -660,7 +660,7 @@ int add_memory(u64 start, u64 size) unsigned long nr_pages = size >> PAGE_SHIFT; int ret; - pgdat = NODE_DATA(0); + pgdat = NODE_DATA(nid); zone = pgdat->node_zones + ZONE_NORMAL; ret = __add_pages(zone, start_pfn, nr_pages); @@ -671,7 +671,6 @@ int add_memory(u64 start, u64 size) return ret; } -EXPORT_SYMBOL_GPL(add_memory); int remove_memory(u64 start, u64 size) { diff --git a/arch/powerpc/mm/mem.c b/arch/powerpc/mm/mem.c index 69f3b9a20be..089d939a0b3 100644 --- a/arch/powerpc/mm/mem.c +++ b/arch/powerpc/mm/mem.c @@ -114,15 +114,20 @@ void online_page(struct page *page) num_physpages++; } -int __devinit add_memory(u64 start, u64 size) +#ifdef CONFIG_NUMA +int memory_add_physaddr_to_nid(u64 start) +{ + return hot_add_scn_to_nid(start); +} +#endif + +int __devinit arch_add_memory(int nid, u64 start, u64 size) { struct pglist_data *pgdata; struct zone *zone; - int nid; unsigned long start_pfn = start >> PAGE_SHIFT; unsigned long nr_pages = size >> PAGE_SHIFT; - nid = hot_add_scn_to_nid(start); pgdata = NODE_DATA(nid); start = (unsigned long)__va(start); diff --git a/arch/x86_64/mm/init.c b/arch/x86_64/mm/init.c index 02add1d1dfa..51fbf3eddf1 100644 --- a/arch/x86_64/mm/init.c +++ b/arch/x86_64/mm/init.c @@ -506,8 +506,6 @@ void __init clear_kernel_mapping(unsigned long address, unsigned long size) /* * Memory hotplug specific functions */ -#if defined(CONFIG_ACPI_HOTPLUG_MEMORY) || defined(CONFIG_ACPI_HOTPLUG_MEMORY_MODULE) - void online_page(struct page *page) { ClearPageReserved(page); @@ -517,31 +515,17 @@ void online_page(struct page *page) num_physpages++; } -#ifndef CONFIG_MEMORY_HOTPLUG +#ifdef CONFIG_MEMORY_HOTPLUG /* - * Memory Hotadd without sparsemem. The mem_maps have been allocated in advance, - * just online the pages. + * XXX: memory_add_physaddr_to_nid() is to find node id from physical address + * via probe interface of sysfs. If acpi notifies hot-add event, then it + * can tell node id by searching dsdt. But, probe interface doesn't have + * node id. So, return 0 as node id at this time. */ -int __add_pages(struct zone *z, unsigned long start_pfn, unsigned long nr_pages) +#ifdef CONFIG_NUMA +int memory_add_physaddr_to_nid(u64 start) { - int err = -EIO; - unsigned long pfn; - unsigned long total = 0, mem = 0; - for (pfn = start_pfn; pfn < start_pfn + nr_pages; pfn++) { - if (pfn_valid(pfn)) { - online_page(pfn_to_page(pfn)); - err = 0; - mem++; - } - total++; - } - if (!err) { - z->spanned_pages += total; - z->present_pages += mem; - z->zone_pgdat->node_spanned_pages += total; - z->zone_pgdat->node_present_pages += mem; - } - return err; + return 0; } #endif @@ -549,9 +533,9 @@ int __add_pages(struct zone *z, unsigned long start_pfn, unsigned long nr_pages) * Memory is added always to NORMAL zone. This means you will never get * additional DMA/DMA32 memory. */ -int add_memory(u64 start, u64 size) +int arch_add_memory(int nid, u64 start, u64 size) { - struct pglist_data *pgdat = NODE_DATA(0); + struct pglist_data *pgdat = NODE_DATA(nid); struct zone *zone = pgdat->node_zones + MAX_NR_ZONES-2; unsigned long start_pfn = start >> PAGE_SHIFT; unsigned long nr_pages = size >> PAGE_SHIFT; @@ -568,7 +552,7 @@ error: printk("%s: Problem encountered in __add_pages!\n", __func__); return ret; } -EXPORT_SYMBOL_GPL(add_memory); +EXPORT_SYMBOL_GPL(arch_add_memory); int remove_memory(u64 start, u64 size) { @@ -576,7 +560,33 @@ int remove_memory(u64 start, u64 size) } EXPORT_SYMBOL_GPL(remove_memory); -#endif +#else /* CONFIG_MEMORY_HOTPLUG */ +/* + * Memory Hotadd without sparsemem. The mem_maps have been allocated in advance, + * just online the pages. + */ +int __add_pages(struct zone *z, unsigned long start_pfn, unsigned long nr_pages) +{ + int err = -EIO; + unsigned long pfn; + unsigned long total = 0, mem = 0; + for (pfn = start_pfn; pfn < start_pfn + nr_pages; pfn++) { + if (pfn_valid(pfn)) { + online_page(pfn_to_page(pfn)); + err = 0; + mem++; + } + total++; + } + if (!err) { + z->spanned_pages += total; + z->present_pages += mem; + z->zone_pgdat->node_spanned_pages += total; + z->zone_pgdat->node_present_pages += mem; + } + return err; +} +#endif /* CONFIG_MEMORY_HOTPLUG */ static struct kcore_list kcore_mem, kcore_vmalloc, kcore_kernel, kcore_modules, kcore_vsyscall; diff --git a/drivers/acpi/Kconfig b/drivers/acpi/Kconfig index 94b8d820c51..610d2cc02cf 100644 --- a/drivers/acpi/Kconfig +++ b/drivers/acpi/Kconfig @@ -328,7 +328,7 @@ config ACPI_CONTAINER config ACPI_HOTPLUG_MEMORY tristate "Memory Hotplug" depends on ACPI - depends on MEMORY_HOTPLUG || X86_64 + depends on MEMORY_HOTPLUG default n help This driver adds supports for ACPI Memory Hotplug. This driver diff --git a/drivers/acpi/acpi_memhotplug.c b/drivers/acpi/acpi_memhotplug.c index 5652569b376..0424326eae1 100644 --- a/drivers/acpi/acpi_memhotplug.c +++ b/drivers/acpi/acpi_memhotplug.c @@ -215,6 +215,7 @@ static int acpi_memory_enable_device(struct acpi_memory_device *mem_device) { int result, num_enabled = 0; struct acpi_memory_info *info; + int node = 0; ACPI_FUNCTION_TRACE("acpi_memory_enable_device"); @@ -245,7 +246,7 @@ static int acpi_memory_enable_device(struct acpi_memory_device *mem_device) continue; } - result = add_memory(info->start_addr, info->length); + result = add_memory(node, info->start_addr, info->length); if (result) continue; info->enabled = 1; diff --git a/drivers/base/memory.c b/drivers/base/memory.c index dd547af4681..c6b7d9c4b65 100644 --- a/drivers/base/memory.c +++ b/drivers/base/memory.c @@ -306,11 +306,13 @@ static ssize_t memory_probe_store(struct class *class, const char *buf, size_t count) { u64 phys_addr; + int nid; int ret; phys_addr = simple_strtoull(buf, NULL, 0); - ret = add_memory(phys_addr, PAGES_PER_SECTION << PAGE_SHIFT); + nid = memory_add_physaddr_to_nid(phys_addr); + ret = add_memory(nid, phys_addr, PAGES_PER_SECTION << PAGE_SHIFT); if (ret) count = ret; diff --git a/include/linux/memory_hotplug.h b/include/linux/memory_hotplug.h index 91120638617..29c1472efad 100644 --- a/include/linux/memory_hotplug.h +++ b/include/linux/memory_hotplug.h @@ -63,6 +63,16 @@ extern int online_pages(unsigned long, unsigned long); /* reasonably generic interface to expand the physical pages in a zone */ extern int __add_pages(struct zone *zone, unsigned long start_pfn, unsigned long nr_pages); + +#ifdef CONFIG_NUMA +extern int memory_add_physaddr_to_nid(u64 start); +#else +static inline int memory_add_physaddr_to_nid(u64 start) +{ + return 0; +} +#endif + #else /* ! CONFIG_MEMORY_HOTPLUG */ /* * Stub functions for when hotplug is off @@ -99,7 +109,8 @@ static inline int __remove_pages(struct zone *zone, unsigned long start_pfn, return -ENOSYS; } -extern int add_memory(u64 start, u64 size); +extern int add_memory(int nid, u64 start, u64 size); +extern int arch_add_memory(int nid, u64 start, u64 size); extern int remove_memory(u64 start, u64 size); #endif /* __LINUX_MEMORY_HOTPLUG_H */ diff --git a/mm/memory_hotplug.c b/mm/memory_hotplug.c index 841a077d5ae..6cdeabe9f6d 100644 --- a/mm/memory_hotplug.c +++ b/mm/memory_hotplug.c @@ -163,3 +163,14 @@ int online_pages(unsigned long pfn, unsigned long nr_pages) vm_total_pages = nr_free_pagecache_pages(); return 0; } + +int add_memory(int nid, u64 start, u64 size) +{ + int ret; + + /* call arch's memory hotadd */ + ret = arch_add_memory(nid, start, size); + + return ret; +} +EXPORT_SYMBOL_GPL(add_memory); -- cgit v1.2.3-70-g09d2 From 1e3590e2e4a38e8390fdac5bda23330bf2801838 Mon Sep 17 00:00:00 2001 From: Yasunori Goto Date: Tue, 27 Jun 2006 02:53:31 -0700 Subject: [PATCH] pgdat allocation for new node add (get node id by acpi) This is to find node id from acpi's handle of memory_device in DSDT. _PXM for the new node can be found by acpi_get_pxm() by using new memory's handle. So, node id can be found by pxm_to_nid_map[]. This patch becomes simpler than v2 of node hot-add patch. Because old add_memory() function doesn't have node id parameter. So, kernel must find its handle by physical address via DSDT again. But, v3 just give node id to add_memory() now. Signed-off-by: Yasunori Goto Cc: Dave Hansen Cc: "Brown, Len" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/acpi/acpi_memhotplug.c | 3 ++- drivers/acpi/numa.c | 15 ++++++++++++++- include/linux/acpi.h | 6 ++++++ 3 files changed, 22 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/drivers/acpi/acpi_memhotplug.c b/drivers/acpi/acpi_memhotplug.c index 0424326eae1..1012284ff4f 100644 --- a/drivers/acpi/acpi_memhotplug.c +++ b/drivers/acpi/acpi_memhotplug.c @@ -215,7 +215,7 @@ static int acpi_memory_enable_device(struct acpi_memory_device *mem_device) { int result, num_enabled = 0; struct acpi_memory_info *info; - int node = 0; + int node; ACPI_FUNCTION_TRACE("acpi_memory_enable_device"); @@ -228,6 +228,7 @@ static int acpi_memory_enable_device(struct acpi_memory_device *mem_device) return result; } + node = acpi_get_node(mem_device->handle); /* * Tell the VM there is more memory here... * Note: Assume that this function returns zero on success diff --git a/drivers/acpi/numa.c b/drivers/acpi/numa.c index e2c1a16078c..13d6d5bdea2 100644 --- a/drivers/acpi/numa.c +++ b/drivers/acpi/numa.c @@ -254,5 +254,18 @@ int acpi_get_pxm(acpi_handle h) } while (ACPI_SUCCESS(status)); return -1; } - EXPORT_SYMBOL(acpi_get_pxm); + +int acpi_get_node(acpi_handle *handle) +{ + int pxm, node = -1; + + ACPI_FUNCTION_TRACE("acpi_get_node"); + + pxm = acpi_get_pxm(handle); + if (pxm >= 0) + node = acpi_map_pxm_to_node(pxm); + + return_VALUE(node); +} +EXPORT_SYMBOL(acpi_get_node); diff --git a/include/linux/acpi.h b/include/linux/acpi.h index 90d6df1551e..88b5dfd8ee1 100644 --- a/include/linux/acpi.h +++ b/include/linux/acpi.h @@ -528,12 +528,18 @@ static inline void acpi_set_cstate_limit(unsigned int new_limit) { return; } #ifdef CONFIG_ACPI_NUMA int acpi_get_pxm(acpi_handle handle); +int acpi_get_node(acpi_handle *handle); #else static inline int acpi_get_pxm(acpi_handle handle) { return 0; } +static inline int acpi_get_node(acpi_handle *handle) +{ + return 0; +} #endif +extern int acpi_paddr_to_node(u64 start_addr, u64 size); extern int pnpacpi_disabled; -- cgit v1.2.3-70-g09d2 From 306d6cbe86e2e6603ac3162e1294d5c75cfdeca6 Mon Sep 17 00:00:00 2001 From: Yasunori Goto Date: Tue, 27 Jun 2006 02:53:32 -0700 Subject: [PATCH] pgdat allocation for new node add (generic alloc node_data) For node hotplug, basically we have to allocate new pgdat. But, there are several types of implementations of pgdat. 1. Allocate only pgdat. This style allocate only pgdat area. And its address is recorded in node_data[]. It is most popular style. 2. Static array of pgdat In this case, all of pgdats are static array. Some archs use this style. 3. Allocate not only pgdat, but also per node data. To increase performance, each node has copy of some data as a per node data. So, this area must be allocated too. Ia64 is this style. Ia64 has the copies of node_data[] array on each per node data to increase performance. In this series of patches, treat (1) as generic arch. generic archs can use generic function. (2) and (3) should have its own if necessary. This patch defines pgdat allocator. Updating NODE_DATA() macro function is in other patch. Signed-off-by: Yasonori Goto Signed-off-by: KAMEZAWA Hiroyuki Cc: Dave Hansen Cc: "Brown, Len" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/memory_hotplug.h | 55 ++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 55 insertions(+) (limited to 'include') diff --git a/include/linux/memory_hotplug.h b/include/linux/memory_hotplug.h index 29c1472efad..c6fd2c0323f 100644 --- a/include/linux/memory_hotplug.h +++ b/include/linux/memory_hotplug.h @@ -73,6 +73,61 @@ static inline int memory_add_physaddr_to_nid(u64 start) } #endif +#ifdef CONFIG_HAVE_ARCH_NODEDATA_EXTENSION +/* + * For supporting node-hotadd, we have to allocate a new pgdat. + * + * If an arch has generic style NODE_DATA(), + * node_data[nid] = kzalloc() works well. But it depends on the architecture. + * + * In general, generic_alloc_nodedata() is used. + * Now, arch_free_nodedata() is just defined for error path of node_hot_add. + * + */ +static inline pg_data_t *arch_alloc_nodedata(int nid) +{ + return NULL; +} +static inline void arch_free_nodedata(pg_data_t *pgdat) +{ +} + +#else /* CONFIG_HAVE_ARCH_NODEDATA_EXTENSION */ + +#define arch_alloc_nodedata(nid) generic_alloc_nodedata(nid) +#define arch_free_nodedata(pgdat) generic_free_nodedata(pgdat) + +#ifdef CONFIG_NUMA +/* + * If ARCH_HAS_NODEDATA_EXTENSION=n, this func is used to allocate pgdat. + * XXX: kmalloc_node() can't work well to get new node's memory at this time. + * Because, pgdat for the new node is not allocated/initialized yet itself. + * To use new node's memory, more consideration will be necessary. + */ +#define generic_alloc_nodedata(nid) \ +({ \ + kzalloc(sizeof(pg_data_t), GFP_KERNEL); \ +}) +/* + * This definition is just for error path in node hotadd. + * For node hotremove, we have to replace this. + */ +#define generic_free_nodedata(pgdat) kfree(pgdat) + +#else /* !CONFIG_NUMA */ + +/* never called */ +static inline pg_data_t *generic_alloc_nodedata(int nid) +{ + BUG(); + return NULL; +} +static inline void generic_free_nodedata(pg_data_t *pgdat) +{ +} +#endif /* CONFIG_NUMA */ +#endif /* CONFIG_HAVE_ARCH_NODEDATA_EXTENSION */ + #else /* ! CONFIG_MEMORY_HOTPLUG */ /* * Stub functions for when hotplug is off -- cgit v1.2.3-70-g09d2 From 10ad400b49aca15ecf83b0fde7e35e4064b15c85 Mon Sep 17 00:00:00 2001 From: Yasunori Goto Date: Tue, 27 Jun 2006 02:53:33 -0700 Subject: [PATCH] pgdat allocation for new node add (refresh node_data[]) Refresh NODE_DATA() for generic archs. In this case, NODE_DATA(nid) == node_data[nid]. node_data[] is array of address of pgdat. So, refresh is quite simple. Signed-off-by: Yasunori Goto Signed-off-by: KAMEZAWA Hiroyuki Cc: Dave Hansen Cc: "Brown, Len" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/ia64/Kconfig | 4 ++++ include/linux/memory_hotplug.h | 12 ++++++++++++ 2 files changed, 16 insertions(+) (limited to 'include') diff --git a/arch/ia64/Kconfig b/arch/ia64/Kconfig index 18318749884..a56df7bf022 100644 --- a/arch/ia64/Kconfig +++ b/arch/ia64/Kconfig @@ -374,6 +374,10 @@ config HAVE_ARCH_EARLY_PFN_TO_NID def_bool y depends on NEED_MULTIPLE_NODES +config HAVE_ARCH_NODEDATA_EXTENSION + def_bool y + depends on NUMA + config IA32_SUPPORT bool "Support for Linux/x86 binaries" help diff --git a/include/linux/memory_hotplug.h b/include/linux/memory_hotplug.h index c6fd2c0323f..569b1f6c27d 100644 --- a/include/linux/memory_hotplug.h +++ b/include/linux/memory_hotplug.h @@ -91,6 +91,9 @@ static inline pg_data_t *arch_alloc_nodedata(int nid) static inline void arch_free_nodedata(pg_data_t *pgdat) { } +static inline void arch_refresh_nodedata(int nid, pg_data_t *pgdat) +{ +} #else /* CONFIG_HAVE_ARCH_NODEDATA_EXTENSION */ @@ -114,6 +117,12 @@ static inline void arch_free_nodedata(pg_data_t *pgdat) */ #define generic_free_nodedata(pgdat) kfree(pgdat) +extern pg_data_t *node_data[]; +static inline void arch_refresh_nodedata(int nid, pg_data_t *pgdat) +{ + node_data[nid] = pgdat; +} + #else /* !CONFIG_NUMA */ /* never called */ @@ -125,6 +134,9 @@ static inline pg_data_t *generic_alloc_nodedata(int nid) static inline void generic_free_nodedata(pg_data_t *pgdat) { } +static inline void arch_refresh_nodedata(int nid, pg_data_t *pgdat) +{ +} #endif /* CONFIG_NUMA */ #endif /* CONFIG_HAVE_ARCH_NODEDATA_EXTENSION */ -- cgit v1.2.3-70-g09d2 From 3218ae14b1e3ee2ab81df30ed690c8e864d23316 Mon Sep 17 00:00:00 2001 From: Yasunori Goto Date: Tue, 27 Jun 2006 02:53:33 -0700 Subject: [PATCH] pgdat allocation for new node add (export kswapd start func) When node is hot-added, kswapd for the node should start. This export kswapd start function as kswapd_run() to use at add_memory(). [akpm@osdl.org: daemonize() isn't needed when using the kthread API] Signed-off-by: Yasunori Goto Signed-off-by: KAMEZAWA Hiroyuki Cc: Dave Hansen Cc: "Brown, Len" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/swap.h | 2 ++ mm/vmscan.c | 37 ++++++++++++++++++++++++++----------- 2 files changed, 28 insertions(+), 11 deletions(-) (limited to 'include') diff --git a/include/linux/swap.h b/include/linux/swap.h index dc3f3aa0c83..c41e2d6d1ac 100644 --- a/include/linux/swap.h +++ b/include/linux/swap.h @@ -199,6 +199,8 @@ static inline int zone_reclaim(struct zone *z, gfp_t mask, unsigned int order) } #endif +extern int kswapd_run(int nid); + #ifdef CONFIG_MMU /* linux/mm/shmem.c */ extern int shmem_unuse(swp_entry_t entry, struct page *page); diff --git a/mm/vmscan.c b/mm/vmscan.c index 72babac71de..f03da33d914 100644 --- a/mm/vmscan.c +++ b/mm/vmscan.c @@ -34,6 +34,7 @@ #include #include #include +#include #include #include @@ -1223,7 +1224,6 @@ static int kswapd(void *p) }; cpumask_t cpumask; - daemonize("kswapd%d", pgdat->node_id); cpumask = node_to_cpumask(pgdat->node_id); if (!cpus_empty(cpumask)) set_cpus_allowed(tsk, cpumask); @@ -1468,20 +1468,35 @@ static int cpu_callback(struct notifier_block *nfb, } #endif /* CONFIG_HOTPLUG_CPU */ +/* + * This kswapd start function will be called by init and node-hot-add. + * On node-hot-add, kswapd will moved to proper cpus if cpus are hot-added. + */ +int kswapd_run(int nid) +{ + pg_data_t *pgdat = NODE_DATA(nid); + int ret = 0; + + if (pgdat->kswapd) + return 0; + + pgdat->kswapd = kthread_run(kswapd, pgdat, "kswapd%d", nid); + if (IS_ERR(pgdat->kswapd)) { + /* failure at boot is fatal */ + BUG_ON(system_state == SYSTEM_BOOTING); + printk("Failed to start kswapd on node %d\n",nid); + ret = -1; + } + return ret; +} + static int __init kswapd_init(void) { - pg_data_t *pgdat; + int nid; swap_setup(); - for_each_online_pgdat(pgdat) { - pid_t pid; - - pid = kernel_thread(kswapd, pgdat, CLONE_KERNEL); - BUG_ON(pid < 0); - read_lock(&tasklist_lock); - pgdat->kswapd = find_task_by_pid(pid); - read_unlock(&tasklist_lock); - } + for_each_online_node(nid) + kswapd_run(nid); hotcpu_notifier(cpu_callback, 0); return 0; } -- cgit v1.2.3-70-g09d2 From 2842f11419704f8707fffc82e10d2263427fc130 Mon Sep 17 00:00:00 2001 From: KAMEZAWA Hiroyuki Date: Tue, 27 Jun 2006 02:53:36 -0700 Subject: [PATCH] catch valid mem range at onlining memory This patch allows hot-add memory which is not aligned to section. Now, hot-added memory has to be aligned to section size. Considering big section sized archs, this is not useful. When hot-added memory is registerd as iomem resoruce by iomem resource patch, we can make use of that information to detect valid memory range. Note: With this, not-aligned memory can be registerd. To allow hot-add memory with holes, we have to do more work around add_memory(). (It doesn't allows add memory to already existing mem section.) Signed-off-by: KAMEZAWA Hiroyuki Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/ioport.h | 3 +++ kernel/resource.c | 38 ++++++++++++++++++++++++++++++++++++++ mm/memory_hotplug.c | 28 ++++++++++++++++++++++++---- 3 files changed, 65 insertions(+), 4 deletions(-) (limited to 'include') diff --git a/include/linux/ioport.h b/include/linux/ioport.h index cd6bd001ba4..edfc733b157 100644 --- a/include/linux/ioport.h +++ b/include/linux/ioport.h @@ -105,6 +105,9 @@ extern int allocate_resource(struct resource *root, struct resource *new, int adjust_resource(struct resource *res, unsigned long start, unsigned long size); +/* get registered SYSTEM_RAM resources in specified area */ +extern int find_next_system_ram(struct resource *res); + /* Convenience shorthand with allocation */ #define request_region(start,n,name) __request_region(&ioport_resource, (start), (n), (name)) #define request_mem_region(start,n,name) __request_region(&iomem_resource, (start), (n), (name)) diff --git a/kernel/resource.c b/kernel/resource.c index e3080fcc66a..2404f9b0bc4 100644 --- a/kernel/resource.c +++ b/kernel/resource.c @@ -232,6 +232,44 @@ int release_resource(struct resource *old) EXPORT_SYMBOL(release_resource); +#ifdef CONFIG_MEMORY_HOTPLUG +/* + * Finds the lowest memory reosurce exists within [res->start.res->end) + * the caller must specify res->start, res->end, res->flags. + * If found, returns 0, res is overwritten, if not found, returns -1. + */ +int find_next_system_ram(struct resource *res) +{ + resource_size_t start, end; + struct resource *p; + + BUG_ON(!res); + + start = res->start; + end = res->end; + + read_lock(&resource_lock); + for (p = iomem_resource.child; p ; p = p->sibling) { + /* system ram is just marked as IORESOURCE_MEM */ + if (p->flags != res->flags) + continue; + if (p->start > end) { + p = NULL; + break; + } + if (p->start >= start) + break; + } + read_unlock(&resource_lock); + if (!p) + return -1; + /* copy data */ + res->start = p->start; + res->end = p->end; + return 0; +} +#endif + /* * Find empty slot in the resource tree given range and alignment. */ diff --git a/mm/memory_hotplug.c b/mm/memory_hotplug.c index 0b11a854344..f13783e81eb 100644 --- a/mm/memory_hotplug.c +++ b/mm/memory_hotplug.c @@ -127,6 +127,9 @@ int online_pages(unsigned long pfn, unsigned long nr_pages) unsigned long i; unsigned long flags; unsigned long onlined_pages = 0; + struct resource res; + u64 section_end; + unsigned long start_pfn; struct zone *zone; int need_zonelists_rebuild = 0; @@ -149,10 +152,27 @@ int online_pages(unsigned long pfn, unsigned long nr_pages) if (!populated_zone(zone)) need_zonelists_rebuild = 1; - for (i = 0; i < nr_pages; i++) { - struct page *page = pfn_to_page(pfn + i); - online_page(page); - onlined_pages++; + res.start = (u64)pfn << PAGE_SHIFT; + res.end = res.start + ((u64)nr_pages << PAGE_SHIFT) - 1; + res.flags = IORESOURCE_MEM; /* we just need system ram */ + section_end = res.end; + + while (find_next_system_ram(&res) >= 0) { + start_pfn = (unsigned long)(res.start >> PAGE_SHIFT); + nr_pages = (unsigned long) + ((res.end + 1 - res.start) >> PAGE_SHIFT); + + if (PageReserved(pfn_to_page(start_pfn))) { + /* this region's page is not onlined now */ + for (i = 0; i < nr_pages; i++) { + struct page *page = pfn_to_page(start_pfn + i); + online_page(page); + onlined_pages++; + } + } + + res.start = res.end + 1; + res.end = section_end; } zone->present_pages += onlined_pages; zone->zone_pgdat->node_present_pages += onlined_pages; -- cgit v1.2.3-70-g09d2 From 0fc44159bfcb5b0afa178f9c3f50db23aebc76ff Mon Sep 17 00:00:00 2001 From: Yasunori Goto Date: Tue, 27 Jun 2006 02:53:38 -0700 Subject: [PATCH] Register sysfs file for hotplugged new node When new node becomes enable by hot-add, new sysfs file must be created for new node. So, if new node is enabled by add_memory(), register_one_node() is called to create it. In addition, I386's arch_register_node() and a part of register_nodes() of powerpc are consolidated to register_one_node() as a generic_code(). This is tested by Tiger4(IPF) with node hot-plug emulation. Signed-off-by: Keiichiro Tokunaga Signed-off-by: Yasunori Goto Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/i386/kernel/topology.c | 9 +++------ arch/ia64/kernel/topology.c | 15 +++------------ arch/powerpc/kernel/sysfs.c | 15 ++------------- drivers/base/node.c | 25 +++++++++++++++++++++++++ include/asm-i386/cpu.h | 2 -- include/asm-i386/node.h | 29 ----------------------------- include/linux/node.h | 4 ++++ mm/memory_hotplug.c | 12 +++++++++++- 8 files changed, 48 insertions(+), 63 deletions(-) delete mode 100644 include/asm-i386/node.h (limited to 'include') diff --git a/arch/i386/kernel/topology.c b/arch/i386/kernel/topology.c index 296355292c7..1eecc2e1bd4 100644 --- a/arch/i386/kernel/topology.c +++ b/arch/i386/kernel/topology.c @@ -38,7 +38,7 @@ int arch_register_cpu(int num){ #ifdef CONFIG_NUMA int node = cpu_to_node(num); if (node_online(node)) - parent = &node_devices[node].node; + parent = &node_devices[parent_node(node)]; #endif /* CONFIG_NUMA */ /* @@ -61,7 +61,7 @@ void arch_unregister_cpu(int num) { #ifdef CONFIG_NUMA int node = cpu_to_node(num); if (node_online(node)) - parent = &node_devices[node].node; + parent = &node_devices[parent_node(node)]; #endif /* CONFIG_NUMA */ return unregister_cpu(&cpu_devices[num].cpu, parent); @@ -74,16 +74,13 @@ EXPORT_SYMBOL(arch_unregister_cpu); #ifdef CONFIG_NUMA #include -#include - -struct i386_node node_devices[MAX_NUMNODES]; static int __init topology_init(void) { int i; for_each_online_node(i) - arch_register_node(i); + register_one_node(i); for_each_present_cpu(i) arch_register_cpu(i); diff --git a/arch/ia64/kernel/topology.c b/arch/ia64/kernel/topology.c index 879edb51d1e..42cb05bdc68 100644 --- a/arch/ia64/kernel/topology.c +++ b/arch/ia64/kernel/topology.c @@ -26,9 +26,6 @@ #include #include -#ifdef CONFIG_NUMA -static struct node *sysfs_nodes; -#endif static struct ia64_cpu *sysfs_cpus; int arch_register_cpu(int num) @@ -36,7 +33,7 @@ int arch_register_cpu(int num) struct node *parent = NULL; #ifdef CONFIG_NUMA - parent = &sysfs_nodes[cpu_to_node(num)]; + parent = &node_devices[cpu_to_node(num)]; #endif /* CONFIG_NUMA */ #if defined (CONFIG_ACPI) && defined (CONFIG_HOTPLUG_CPU) @@ -59,7 +56,7 @@ void arch_unregister_cpu(int num) #ifdef CONFIG_NUMA int node = cpu_to_node(num); - parent = &sysfs_nodes[node]; + parent = &node_devices[node]; #endif /* CONFIG_NUMA */ return unregister_cpu(&sysfs_cpus[num].cpu, parent); @@ -74,17 +71,11 @@ static int __init topology_init(void) int i, err = 0; #ifdef CONFIG_NUMA - sysfs_nodes = kzalloc(sizeof(struct node) * MAX_NUMNODES, GFP_KERNEL); - if (!sysfs_nodes) { - err = -ENOMEM; - goto out; - } - /* * MCD - Do we want to register all ONLINE nodes, or all POSSIBLE nodes? */ for_each_online_node(i) { - if ((err = register_node(&sysfs_nodes[i], i, 0))) + if ((err = register_one_node(i))) goto out; } #endif diff --git a/arch/powerpc/kernel/sysfs.c b/arch/powerpc/kernel/sysfs.c index 5bc2585c803..338491d1604 100644 --- a/arch/powerpc/kernel/sysfs.c +++ b/arch/powerpc/kernel/sysfs.c @@ -304,23 +304,12 @@ static struct notifier_block sysfs_cpu_nb = { /* NUMA stuff */ #ifdef CONFIG_NUMA -static struct node node_devices[MAX_NUMNODES]; - static void register_nodes(void) { int i; - for (i = 0; i < MAX_NUMNODES; i++) { - if (node_online(i)) { - int p_node = parent_node(i); - struct node *parent = NULL; - - if (p_node != i) - parent = &node_devices[p_node]; - - register_node(&node_devices[i], i, parent); - } - } + for (i = 0; i < MAX_NUMNODES; i++) + register_one_node(i); } int sysfs_add_device_to_node(struct sys_device *dev, int nid) diff --git a/drivers/base/node.c b/drivers/base/node.c index c80c3aeed00..cbd0f62b487 100644 --- a/drivers/base/node.c +++ b/drivers/base/node.c @@ -190,6 +190,31 @@ void unregister_node(struct node *node) sysdev_unregister(&node->sysdev); } +struct node node_devices[MAX_NUMNODES]; + +int register_one_node(int nid) +{ + int error = 0; + + if (node_online(nid)) { + int p_node = parent_node(nid); + struct node *parent = NULL; + + if (p_node != nid) + parent = &node_devices[p_node]; + + error = register_node(&node_devices[nid], nid, parent); + } + + return error; + +} + +void unregister_one_node(int nid) +{ + unregister_node(&node_devices[nid]); +} + static int __init register_node_type(void) { return sysdev_class_register(&node_class); diff --git a/include/asm-i386/cpu.h b/include/asm-i386/cpu.h index e7252c216ca..b1bc7b1b64b 100644 --- a/include/asm-i386/cpu.h +++ b/include/asm-i386/cpu.h @@ -7,8 +7,6 @@ #include #include -#include - struct i386_cpu { struct cpu cpu; }; diff --git a/include/asm-i386/node.h b/include/asm-i386/node.h deleted file mode 100644 index e13c6ffa72a..00000000000 --- a/include/asm-i386/node.h +++ /dev/null @@ -1,29 +0,0 @@ -#ifndef _ASM_I386_NODE_H_ -#define _ASM_I386_NODE_H_ - -#include -#include -#include -#include -#include - -struct i386_node { - struct node node; -}; -extern struct i386_node node_devices[MAX_NUMNODES]; - -static inline int arch_register_node(int num){ - int p_node; - struct node *parent = NULL; - - if (!node_online(num)) - return 0; - p_node = parent_node(num); - - if (p_node != num) - parent = &node_devices[p_node].node; - - return register_node(&node_devices[num].node, num, parent); -} - -#endif /* _ASM_I386_NODE_H_ */ diff --git a/include/linux/node.h b/include/linux/node.h index 254dc3de650..1e5347527fa 100644 --- a/include/linux/node.h +++ b/include/linux/node.h @@ -26,8 +26,12 @@ struct node { struct sys_device sysdev; }; +extern struct node node_devices[]; + extern int register_node(struct node *, int, struct node *); extern void unregister_node(struct node *node); +extern int register_one_node(int nid); +extern void unregister_one_node(int nid); #define to_node(sys_device) container_of(sys_device, struct node, sysdev) diff --git a/mm/memory_hotplug.c b/mm/memory_hotplug.c index f13783e81eb..ea4038838b0 100644 --- a/mm/memory_hotplug.c +++ b/mm/memory_hotplug.c @@ -256,9 +256,19 @@ int add_memory(int nid, u64 start, u64 size) if (ret < 0) goto error; - /* we online node here. we have no error path from here. */ + /* we online node here. we can't roll back from here. */ node_set_online(nid); + if (new_pgdat) { + ret = register_one_node(nid); + /* + * If sysfs file of new node can't create, cpu on the node + * can't be hot-added. There is no rollback way now. + * So, check by BUG_ON() to catch it reluctantly.. + */ + BUG_ON(ret); + } + /* register this memory as resource */ register_memory_resource(start, size); -- cgit v1.2.3-70-g09d2 From 7049027c6f0098eb6b23b8f6ca65a905541faf81 Mon Sep 17 00:00:00 2001 From: Yasunori Goto Date: Tue, 27 Jun 2006 02:53:39 -0700 Subject: [PATCH] pgdat allocation and update for ia64 of memory hotplug: update pgdat address array This is to refresh node_data[] array for ia64. As I mentioned previous patches, ia64 has copies of information of pgdat address array on each node as per node data. At v2 of node_add, this function used stop_machine_run() to update them. (I wished that they were copied safety as much as possible.) But, in this patch, this arrays are just copied simply, and set node_online_map bit after completion of pgdat initialization. So, kernel must touch NODE_DATA() macro after checking node_online_map(). (Current code has already done it.) This is more simple way for just hot-add..... Note : It will be problem when hot-remove will occur, because, even if online_map bit is set, kernel may touch NODE_DATA() due to race condition. :-( Signed-off-by: Yasunori Goto Cc: "Luck, Tony" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/ia64/mm/discontig.c | 24 +++++++++++++++++++----- include/asm-ia64/nodedata.h | 12 ++++++++++++ include/linux/memory_hotplug.h | 4 +--- 3 files changed, 32 insertions(+), 8 deletions(-) (limited to 'include') diff --git a/arch/ia64/mm/discontig.c b/arch/ia64/mm/discontig.c index 83153ac1879..9153465d7fc 100644 --- a/arch/ia64/mm/discontig.c +++ b/arch/ia64/mm/discontig.c @@ -308,6 +308,17 @@ static void __init reserve_pernode_space(void) } } +static void __meminit scatter_node_data(void) +{ + pg_data_t **dst; + int node; + + for_each_online_node(node) { + dst = LOCAL_DATA_ADDR(pgdat_list[node])->pg_data_ptrs; + memcpy(dst, pgdat_list, sizeof(pgdat_list)); + } +} + /** * initialize_pernode_data - fixup per-cpu & per-node pointers * @@ -320,11 +331,8 @@ static void __init initialize_pernode_data(void) { int cpu, node; - /* Copy the pg_data_t list to each node and init the node field */ - for_each_online_node(node) { - memcpy(mem_data[node].node_data->pg_data_ptrs, pgdat_list, - sizeof(pgdat_list)); - } + scatter_node_data(); + #ifdef CONFIG_SMP /* Set the node_data pointer for each per-cpu struct */ for (cpu = 0; cpu < NR_CPUS; cpu++) { @@ -783,3 +791,9 @@ void __init paging_init(void) zero_page_memmap_ptr = virt_to_page(ia64_imva(empty_zero_page)); } + +void arch_refresh_nodedata(int update_node, pg_data_t *update_pgdat) +{ + pgdat_list[update_node] = update_pgdat; + scatter_node_data(); +} diff --git a/include/asm-ia64/nodedata.h b/include/asm-ia64/nodedata.h index a140310bf84..2fb337b0e9b 100644 --- a/include/asm-ia64/nodedata.h +++ b/include/asm-ia64/nodedata.h @@ -46,6 +46,18 @@ struct ia64_node_data { */ #define NODE_DATA(nid) (local_node_data->pg_data_ptrs[nid]) +/* + * LOCAL_DATA_ADDR - This is to calculate the address of other node's + * "local_node_data" at hot-plug phase. The local_node_data + * is pointed by per_cpu_page. Kernel usually use it for + * just executing cpu. However, when new node is hot-added, + * the addresses of local data for other nodes are necessary + * to update all of them. + */ +#define LOCAL_DATA_ADDR(pgdat) \ + ((struct ia64_node_data *)((u64)(pgdat) + \ + L1_CACHE_ALIGN(sizeof(struct pglist_data)))) + #endif /* CONFIG_NUMA */ #endif /* _ASM_IA64_NODEDATA_H */ diff --git a/include/linux/memory_hotplug.h b/include/linux/memory_hotplug.h index 569b1f6c27d..9b6260007e5 100644 --- a/include/linux/memory_hotplug.h +++ b/include/linux/memory_hotplug.h @@ -91,9 +91,7 @@ static inline pg_data_t *arch_alloc_nodedata(int nid) static inline void arch_free_nodedata(pg_data_t *pgdat) { } -static inline void arch_refresh_nodedata(int nid, pg_data_t *pgdat) -{ -} +extern void arch_refresh_nodedata(int nid, pg_data_t *pgdat); #else /* CONFIG_HAVE_ARCH_NODEDATA_EXTENSION */ -- cgit v1.2.3-70-g09d2 From dd0932d9d4301bd58a4d5a634a3a8298c4fc5e24 Mon Sep 17 00:00:00 2001 From: Yasunori Goto Date: Tue, 27 Jun 2006 02:53:40 -0700 Subject: [PATCH] pgdat allocation and update for ia64 of memory hotplug: allocate pgdat and per node data This is a patch to allocate pgdat and per node data area for ia64. The size for them can be calculated by compute_pernodesize(). Signed-off-by: Yasunori Goto Cc: "Luck, Tony" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/ia64/mm/discontig.c | 16 ++++++++++++++-- include/linux/memory_hotplug.h | 9 ++------- 2 files changed, 16 insertions(+), 9 deletions(-) (limited to 'include') diff --git a/arch/ia64/mm/discontig.c b/arch/ia64/mm/discontig.c index 9153465d7fc..525b082eb66 100644 --- a/arch/ia64/mm/discontig.c +++ b/arch/ia64/mm/discontig.c @@ -100,7 +100,7 @@ static int __init build_node_maps(unsigned long start, unsigned long len, * acpi_boot_init() (which builds the node_to_cpu_mask array) hasn't been * called yet. Note that node 0 will also count all non-existent cpus. */ -static int __init early_nr_cpus_node(int node) +static int __meminit early_nr_cpus_node(int node) { int cpu, n = 0; @@ -115,7 +115,7 @@ static int __init early_nr_cpus_node(int node) * compute_pernodesize - compute size of pernode data * @node: the node id. */ -static unsigned long __init compute_pernodesize(int node) +static unsigned long __meminit compute_pernodesize(int node) { unsigned long pernodesize = 0, cpus; @@ -792,6 +792,18 @@ void __init paging_init(void) zero_page_memmap_ptr = virt_to_page(ia64_imva(empty_zero_page)); } +pg_data_t *arch_alloc_nodedata(int nid) +{ + unsigned long size = compute_pernodesize(nid); + + return kzalloc(size, GFP_KERNEL); +} + +void arch_free_nodedata(pg_data_t *pgdat) +{ + kfree(pgdat); +} + void arch_refresh_nodedata(int update_node, pg_data_t *update_pgdat) { pgdat_list[update_node] = update_pgdat; diff --git a/include/linux/memory_hotplug.h b/include/linux/memory_hotplug.h index 9b6260007e5..218501cfaeb 100644 --- a/include/linux/memory_hotplug.h +++ b/include/linux/memory_hotplug.h @@ -84,13 +84,8 @@ static inline int memory_add_physaddr_to_nid(u64 start) * Now, arch_free_nodedata() is just defined for error path of node_hot_add. * */ -static inline pg_data_t *arch_alloc_nodedata(int nid) -{ - return NULL; -} -static inline void arch_free_nodedata(pg_data_t *pgdat) -{ -} +extern pg_data_t *arch_alloc_nodedata(int nid); +extern void arch_free_nodedata(pg_data_t *pgdat); extern void arch_refresh_nodedata(int nid, pg_data_t *pgdat); #else /* CONFIG_HAVE_ARCH_NODEDATA_EXTENSION */ -- cgit v1.2.3-70-g09d2 From 76b67ed9dce69a6a329cdd66f94af1787f417b62 Mon Sep 17 00:00:00 2001 From: KAMEZAWA Hiroyuki Date: Tue, 27 Jun 2006 02:53:41 -0700 Subject: [PATCH] node hotplug: register cpu: remove node struct With Goto-san's patch, we can add new pgdat/node at runtime. I'm now considering node-hot-add with cpu + memory on ACPI. I found acpi container, which describes node, could evaluate cpu before memory. This means cpu-hot-add occurs before memory hot add. In most part, cpu-hot-add doesn't depend on node hot add. But register_cpu(), which creates symbolic link from node to cpu, requires that node should be onlined before register_cpu(). When a node is onlined, its pgdat should be there. This patch-set holds off creating symbolic link from node to cpu until node is onlined. This removes node arguments from register_cpu(). Now, register_cpu() requires 'struct node' as its argument. But the array of struct node is now unified in driver/base/node.c now (By Goto's node hotplug patch). We can get struct node in generic way. So, this argument is not necessary now. This patch also guarantees add cpu under node only when node is onlined. It is necessary for node-hot-add vs. cpu-hot-add patch following this. Moreover, register_cpu calculates cpu->node_id by cpu_to_node() without regard to its 'struct node *root' argument. This patch removes it. Also modify callers of register_cpu()/unregister_cpu, whose args are changed by register-cpu-remove-node-struct patch. [Brice.Goglin@ens-lyon.org: fix it] Signed-off-by: KAMEZAWA Hiroyuki Cc: Yasunori Goto Cc: Ashok Raj Cc: Dave Hansen Signed-off-by: Brice Goglin Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/alpha/kernel/setup.c | 2 +- arch/arm/kernel/setup.c | 2 +- arch/i386/kernel/topology.c | 23 ++++------------------- arch/ia64/kernel/topology.c | 17 ++--------------- arch/m32r/kernel/setup.c | 2 +- arch/mips/kernel/smp.c | 2 +- arch/parisc/kernel/topology.c | 3 +-- arch/powerpc/kernel/setup_32.c | 2 +- arch/powerpc/kernel/sysfs.c | 12 +----------- arch/ppc/kernel/setup.c | 2 +- arch/s390/kernel/smp.c | 2 +- arch/sh/kernel/setup.c | 2 +- arch/sh64/kernel/setup.c | 2 +- arch/sparc64/kernel/setup.c | 2 +- drivers/base/cpu.c | 18 ++++++++---------- drivers/base/node.c | 36 ++++++++++++++++++++++++++++++++++++ include/linux/cpu.h | 4 ++-- include/linux/node.h | 13 +++++++++++++ 18 files changed, 77 insertions(+), 69 deletions(-) (limited to 'include') diff --git a/arch/alpha/kernel/setup.c b/arch/alpha/kernel/setup.c index 558b8336855..254c507a608 100644 --- a/arch/alpha/kernel/setup.c +++ b/arch/alpha/kernel/setup.c @@ -481,7 +481,7 @@ register_cpus(void) struct cpu *p = kzalloc(sizeof(*p), GFP_KERNEL); if (!p) return -ENOMEM; - register_cpu(p, i, NULL); + register_cpu(p, i); } return 0; } diff --git a/arch/arm/kernel/setup.c b/arch/arm/kernel/setup.c index 9fc9af88c60..093ccba0503 100644 --- a/arch/arm/kernel/setup.c +++ b/arch/arm/kernel/setup.c @@ -808,7 +808,7 @@ static int __init topology_init(void) int cpu; for_each_possible_cpu(cpu) - register_cpu(&per_cpu(cpu_data, cpu).cpu, cpu, NULL); + register_cpu(&per_cpu(cpu_data, cpu).cpu, cpu); return 0; } diff --git a/arch/i386/kernel/topology.c b/arch/i386/kernel/topology.c index 1eecc2e1bd4..e2e281d4bcc 100644 --- a/arch/i386/kernel/topology.c +++ b/arch/i386/kernel/topology.c @@ -32,15 +32,8 @@ static struct i386_cpu cpu_devices[NR_CPUS]; -int arch_register_cpu(int num){ - struct node *parent = NULL; - -#ifdef CONFIG_NUMA - int node = cpu_to_node(num); - if (node_online(node)) - parent = &node_devices[parent_node(node)]; -#endif /* CONFIG_NUMA */ - +int arch_register_cpu(int num) +{ /* * CPU0 cannot be offlined due to several * restrictions and assumptions in kernel. This basically @@ -50,21 +43,13 @@ int arch_register_cpu(int num){ if (!num) cpu_devices[num].cpu.no_control = 1; - return register_cpu(&cpu_devices[num].cpu, num, parent); + return register_cpu(&cpu_devices[num].cpu, num); } #ifdef CONFIG_HOTPLUG_CPU void arch_unregister_cpu(int num) { - struct node *parent = NULL; - -#ifdef CONFIG_NUMA - int node = cpu_to_node(num); - if (node_online(node)) - parent = &node_devices[parent_node(node)]; -#endif /* CONFIG_NUMA */ - - return unregister_cpu(&cpu_devices[num].cpu, parent); + return unregister_cpu(&cpu_devices[num].cpu); } EXPORT_SYMBOL(arch_register_cpu); EXPORT_SYMBOL(arch_unregister_cpu); diff --git a/arch/ia64/kernel/topology.c b/arch/ia64/kernel/topology.c index 42cb05bdc68..5737c9a061e 100644 --- a/arch/ia64/kernel/topology.c +++ b/arch/ia64/kernel/topology.c @@ -30,12 +30,6 @@ static struct ia64_cpu *sysfs_cpus; int arch_register_cpu(int num) { - struct node *parent = NULL; - -#ifdef CONFIG_NUMA - parent = &node_devices[cpu_to_node(num)]; -#endif /* CONFIG_NUMA */ - #if defined (CONFIG_ACPI) && defined (CONFIG_HOTPLUG_CPU) /* * If CPEI cannot be re-targetted, and this is @@ -45,21 +39,14 @@ int arch_register_cpu(int num) sysfs_cpus[num].cpu.no_control = 1; #endif - return register_cpu(&sysfs_cpus[num].cpu, num, parent); + return register_cpu(&sysfs_cpus[num].cpu, num); } #ifdef CONFIG_HOTPLUG_CPU void arch_unregister_cpu(int num) { - struct node *parent = NULL; - -#ifdef CONFIG_NUMA - int node = cpu_to_node(num); - parent = &node_devices[node]; -#endif /* CONFIG_NUMA */ - - return unregister_cpu(&sysfs_cpus[num].cpu, parent); + return unregister_cpu(&sysfs_cpus[num].cpu); } EXPORT_SYMBOL(arch_register_cpu); EXPORT_SYMBOL(arch_unregister_cpu); diff --git a/arch/m32r/kernel/setup.c b/arch/m32r/kernel/setup.c index 3cd3c2988a4..1ff483c8a4c 100644 --- a/arch/m32r/kernel/setup.c +++ b/arch/m32r/kernel/setup.c @@ -275,7 +275,7 @@ static int __init topology_init(void) int i; for_each_present_cpu(i) - register_cpu(&cpu_devices[i], i, NULL); + register_cpu(&cpu_devices[i], i); return 0; } diff --git a/arch/mips/kernel/smp.c b/arch/mips/kernel/smp.c index 298f82fe844..9096a5ea422 100644 --- a/arch/mips/kernel/smp.c +++ b/arch/mips/kernel/smp.c @@ -446,7 +446,7 @@ static int __init topology_init(void) int ret; for_each_present_cpu(cpu) { - ret = register_cpu(&per_cpu(cpu_devices, cpu), cpu, NULL); + ret = register_cpu(&per_cpu(cpu_devices, cpu), cpu); if (ret) printk(KERN_WARNING "topology_init: register_cpu %d " "failed (%d)\n", cpu, ret); diff --git a/arch/parisc/kernel/topology.c b/arch/parisc/kernel/topology.c index 3ba040050e4..068b20d822e 100644 --- a/arch/parisc/kernel/topology.c +++ b/arch/parisc/kernel/topology.c @@ -26,11 +26,10 @@ static struct cpu cpu_devices[NR_CPUS] __read_mostly; static int __init topology_init(void) { - struct node *parent = NULL; int num; for_each_present_cpu(num) { - register_cpu(&cpu_devices[num], num, parent); + register_cpu(&cpu_devices[num], num); } return 0; } diff --git a/arch/powerpc/kernel/setup_32.c b/arch/powerpc/kernel/setup_32.c index e5a44812441..0932a62a1c9 100644 --- a/arch/powerpc/kernel/setup_32.c +++ b/arch/powerpc/kernel/setup_32.c @@ -215,7 +215,7 @@ int __init ppc_init(void) /* register CPU devices */ for_each_possible_cpu(i) - register_cpu(&cpu_devices[i], i, NULL); + register_cpu(&cpu_devices[i], i); /* call platform init */ if (ppc_md.init != NULL) { diff --git a/arch/powerpc/kernel/sysfs.c b/arch/powerpc/kernel/sysfs.c index 338491d1604..412ad00e222 100644 --- a/arch/powerpc/kernel/sysfs.c +++ b/arch/powerpc/kernel/sysfs.c @@ -348,23 +348,13 @@ static SYSDEV_ATTR(physical_id, 0444, show_physical_id, NULL); static int __init topology_init(void) { int cpu; - struct node *parent = NULL; register_nodes(); - register_cpu_notifier(&sysfs_cpu_nb); for_each_possible_cpu(cpu) { struct cpu *c = &per_cpu(cpu_devices, cpu); -#ifdef CONFIG_NUMA - /* The node to which a cpu belongs can't be known - * until the cpu is made present. - */ - parent = NULL; - if (cpu_present(cpu)) - parent = &node_devices[cpu_to_node(cpu)]; -#endif /* * For now, we just see if the system supports making * the RTAS calls for CPU hotplug. But, there may be a @@ -376,7 +366,7 @@ static int __init topology_init(void) c->no_control = 1; if (cpu_online(cpu) || (c->no_control == 0)) { - register_cpu(c, cpu, parent); + register_cpu(c, cpu); sysdev_create_file(&c->sysdev, &attr_physical_id); } diff --git a/arch/ppc/kernel/setup.c b/arch/ppc/kernel/setup.c index 1f79e84ab46..4b4607d89bf 100644 --- a/arch/ppc/kernel/setup.c +++ b/arch/ppc/kernel/setup.c @@ -475,7 +475,7 @@ int __init ppc_init(void) /* register CPU devices */ for_each_possible_cpu(i) - register_cpu(&cpu_devices[i], i, NULL); + register_cpu(&cpu_devices[i], i); /* call platform init */ if (ppc_md.init != NULL) { diff --git a/arch/s390/kernel/smp.c b/arch/s390/kernel/smp.c index 343120c9223..8e03219eea7 100644 --- a/arch/s390/kernel/smp.c +++ b/arch/s390/kernel/smp.c @@ -869,7 +869,7 @@ static int __init topology_init(void) int ret; for_each_possible_cpu(cpu) { - ret = register_cpu(&per_cpu(cpu_devices, cpu), cpu, NULL); + ret = register_cpu(&per_cpu(cpu_devices, cpu), cpu); if (ret) printk(KERN_WARNING "topology_init: register_cpu %d " "failed (%d)\n", cpu, ret); diff --git a/arch/sh/kernel/setup.c b/arch/sh/kernel/setup.c index bb229ef030f..9af22116c9a 100644 --- a/arch/sh/kernel/setup.c +++ b/arch/sh/kernel/setup.c @@ -402,7 +402,7 @@ static int __init topology_init(void) int cpu_id; for_each_possible_cpu(cpu_id) - register_cpu(&cpu[cpu_id], cpu_id, NULL); + register_cpu(&cpu[cpu_id], cpu_id); return 0; } diff --git a/arch/sh64/kernel/setup.c b/arch/sh64/kernel/setup.c index d2711c9c9d1..da98d8dbcf9 100644 --- a/arch/sh64/kernel/setup.c +++ b/arch/sh64/kernel/setup.c @@ -309,7 +309,7 @@ static struct cpu cpu[1]; static int __init topology_init(void) { - return register_cpu(cpu, 0, NULL); + return register_cpu(cpu, 0); } subsys_initcall(topology_init); diff --git a/arch/sparc64/kernel/setup.c b/arch/sparc64/kernel/setup.c index a6a7d816834..116d9632002 100644 --- a/arch/sparc64/kernel/setup.c +++ b/arch/sparc64/kernel/setup.c @@ -537,7 +537,7 @@ static int __init topology_init(void) for_each_possible_cpu(i) { struct cpu *p = kzalloc(sizeof(*p), GFP_KERNEL); if (p) { - register_cpu(p, i, NULL); + register_cpu(p, i); err = 0; } } diff --git a/drivers/base/cpu.c b/drivers/base/cpu.c index dd712b24ec9..3972d8ac978 100644 --- a/drivers/base/cpu.c +++ b/drivers/base/cpu.c @@ -8,6 +8,7 @@ #include #include #include +#include #include "base.h" @@ -57,13 +58,12 @@ static void __devinit register_cpu_control(struct cpu *cpu) { sysdev_create_file(&cpu->sysdev, &attr_online); } -void unregister_cpu(struct cpu *cpu, struct node *root) +void unregister_cpu(struct cpu *cpu) { int logical_cpu = cpu->sysdev.id; - if (root) - sysfs_remove_link(&root->sysdev.kobj, - kobject_name(&cpu->sysdev.kobj)); + unregister_cpu_under_node(logical_cpu, cpu_to_node(logical_cpu)); + sysdev_remove_file(&cpu->sysdev, &attr_online); sysdev_unregister(&cpu->sysdev); @@ -109,23 +109,21 @@ static SYSDEV_ATTR(crash_notes, 0400, show_crash_notes, NULL); * * Initialize and register the CPU device. */ -int __devinit register_cpu(struct cpu *cpu, int num, struct node *root) +int __devinit register_cpu(struct cpu *cpu, int num) { int error; - cpu->node_id = cpu_to_node(num); cpu->sysdev.id = num; cpu->sysdev.cls = &cpu_sysdev_class; error = sysdev_register(&cpu->sysdev); - if (!error && root) - error = sysfs_create_link(&root->sysdev.kobj, - &cpu->sysdev.kobj, - kobject_name(&cpu->sysdev.kobj)); + if (!error && !cpu->no_control) register_cpu_control(cpu); if (!error) cpu_sys_devices[num] = &cpu->sysdev; + if (!error) + register_cpu_under_node(num, cpu_to_node(num)); #ifdef CONFIG_KEXEC if (!error) diff --git a/drivers/base/node.c b/drivers/base/node.c index cbd0f62b487..eae2bdc183b 100644 --- a/drivers/base/node.c +++ b/drivers/base/node.c @@ -11,6 +11,7 @@ #include #include #include +#include static struct sysdev_class node_class = { set_kset_name("node"), @@ -192,9 +193,38 @@ void unregister_node(struct node *node) struct node node_devices[MAX_NUMNODES]; +/* + * register cpu under node + */ +int register_cpu_under_node(unsigned int cpu, unsigned int nid) +{ + if (node_online(nid)) { + struct sys_device *obj = get_cpu_sysdev(cpu); + if (!obj) + return 0; + return sysfs_create_link(&node_devices[nid].sysdev.kobj, + &obj->kobj, + kobject_name(&obj->kobj)); + } + + return 0; +} + +int unregister_cpu_under_node(unsigned int cpu, unsigned int nid) +{ + if (node_online(nid)) { + struct sys_device *obj = get_cpu_sysdev(cpu); + if (obj) + sysfs_remove_link(&node_devices[nid].sysdev.kobj, + kobject_name(&obj->kobj)); + } + return 0; +} + int register_one_node(int nid) { int error = 0; + int cpu; if (node_online(nid)) { int p_node = parent_node(nid); @@ -204,6 +234,12 @@ int register_one_node(int nid) parent = &node_devices[p_node]; error = register_node(&node_devices[nid], nid, parent); + + /* link cpu under this node */ + for_each_present_cpu(cpu) { + if (cpu_to_node(cpu) == nid) + register_cpu_under_node(cpu, nid); + } } return error; diff --git a/include/linux/cpu.h b/include/linux/cpu.h index 08d50c53aab..b23bf1c8add 100644 --- a/include/linux/cpu.h +++ b/include/linux/cpu.h @@ -31,10 +31,10 @@ struct cpu { struct sys_device sysdev; }; -extern int register_cpu(struct cpu *, int, struct node *); +extern int register_cpu(struct cpu *cpu, int num); extern struct sys_device *get_cpu_sysdev(unsigned cpu); #ifdef CONFIG_HOTPLUG_CPU -extern void unregister_cpu(struct cpu *, struct node *); +extern void unregister_cpu(struct cpu *cpu); #endif struct notifier_block; diff --git a/include/linux/node.h b/include/linux/node.h index 1e5347527fa..81dcec84cd8 100644 --- a/include/linux/node.h +++ b/include/linux/node.h @@ -32,6 +32,19 @@ extern int register_node(struct node *, int, struct node *); extern void unregister_node(struct node *node); extern int register_one_node(int nid); extern void unregister_one_node(int nid); +#ifdef CONFIG_NUMA +extern int register_cpu_under_node(unsigned int cpu, unsigned int nid); +extern int unregister_cpu_under_node(unsigned int cpu, unsigned int nid); +#else +static inline int register_cpu_under_node(unsigned int cpu, unsigned int nid) +{ + return 0; +} +static inline int unregister_cpu_under_node(unsigned int cpu, unsigned int nid) +{ + return 0; +} +#endif #define to_node(sys_device) container_of(sys_device, struct node, sysdev) -- cgit v1.2.3-70-g09d2 From 19eadf98c8167eac843580683317b99572e2abf0 Mon Sep 17 00:00:00 2001 From: Rusty Russell Date: Tue, 27 Jun 2006 02:53:44 -0700 Subject: [PATCH] x86: increase interrupt vector range Remove the limit of 256 interrupt vectors by changing the value stored in orig_{e,r}ax to be the complemented interrupt vector. The orig_{e,r}ax needs to be < 0 to allow the signal code to distinguish between return from interrupt and return from syscall. With this change applied, NR_IRQS can be > 256. Xen extends the IRQ numbering space to include room for dynamically allocated virtual interrupts (in the range 256-511), which requires a more permissive interface to do_IRQ. Signed-off-by: Ian Pratt Signed-off-by: Christian Limpach Signed-off-by: Chris Wright Signed-off-by: Rusty Russell Cc: "Protasevich, Natalie" Cc: Andi Kleen Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/i386/kernel/entry.S | 4 ++-- arch/i386/kernel/irq.c | 4 ++-- arch/x86_64/kernel/entry.S | 2 +- arch/x86_64/kernel/irq.c | 4 ++-- arch/x86_64/kernel/smp.c | 4 ++-- include/asm-x86_64/hw_irq.h | 2 +- 6 files changed, 10 insertions(+), 10 deletions(-) (limited to 'include') diff --git a/arch/i386/kernel/entry.S b/arch/i386/kernel/entry.S index e6e4506e749..8713e0248a0 100644 --- a/arch/i386/kernel/entry.S +++ b/arch/i386/kernel/entry.S @@ -515,7 +515,7 @@ ENTRY(irq_entries_start) .if vector CFI_ADJUST_CFA_OFFSET -4 .endif -1: pushl $vector-256 +1: pushl $~(vector) CFI_ADJUST_CFA_OFFSET 4 jmp common_interrupt .data @@ -535,7 +535,7 @@ common_interrupt: #define BUILD_INTERRUPT(name, nr) \ ENTRY(name) \ RING0_INT_FRAME; \ - pushl $nr-256; \ + pushl $~(nr); \ CFI_ADJUST_CFA_OFFSET 4; \ SAVE_ALL; \ movl %esp,%eax; \ diff --git a/arch/i386/kernel/irq.c b/arch/i386/kernel/irq.c index 061533e0cb5..586b15f6741 100644 --- a/arch/i386/kernel/irq.c +++ b/arch/i386/kernel/irq.c @@ -53,8 +53,8 @@ static union irq_ctx *softirq_ctx[NR_CPUS] __read_mostly; */ fastcall unsigned int do_IRQ(struct pt_regs *regs) { - /* high bits used in ret_from_ code */ - int irq = regs->orig_eax & 0xff; + /* high bit used in ret_from_ code */ + int irq = ~regs->orig_eax; #ifdef CONFIG_4KSTACKS union irq_ctx *curctx, *irqctx; u32 *isp; diff --git a/arch/x86_64/kernel/entry.S b/arch/x86_64/kernel/entry.S index 7290e72b9a3..22cac4487b5 100644 --- a/arch/x86_64/kernel/entry.S +++ b/arch/x86_64/kernel/entry.S @@ -588,7 +588,7 @@ END(common_interrupt) */ .macro apicinterrupt num,func INTR_FRAME - pushq $\num-256 + pushq $~(\num) CFI_ADJUST_CFA_OFFSET 8 interrupt \func jmp ret_from_intr diff --git a/arch/x86_64/kernel/irq.c b/arch/x86_64/kernel/irq.c index 59518d4d435..3be0a7e4bf0 100644 --- a/arch/x86_64/kernel/irq.c +++ b/arch/x86_64/kernel/irq.c @@ -115,8 +115,8 @@ skip: */ asmlinkage unsigned int do_IRQ(struct pt_regs *regs) { - /* high bits used in ret_from_ code */ - unsigned irq = regs->orig_rax & 0xff; + /* high bit used in ret_from_ code */ + unsigned irq = ~regs->orig_rax; exit_idle(); irq_enter(); diff --git a/arch/x86_64/kernel/smp.c b/arch/x86_64/kernel/smp.c index acee4bc3f6f..5a1c0a3bf87 100644 --- a/arch/x86_64/kernel/smp.c +++ b/arch/x86_64/kernel/smp.c @@ -135,10 +135,10 @@ asmlinkage void smp_invalidate_interrupt(struct pt_regs *regs) cpu = smp_processor_id(); /* - * orig_rax contains the interrupt vector - 256. + * orig_rax contains the negated interrupt vector. * Use that to determine where the sender put the data. */ - sender = regs->orig_rax + 256 - INVALIDATE_TLB_VECTOR_START; + sender = ~regs->orig_rax - INVALIDATE_TLB_VECTOR_START; f = &per_cpu(flush_state, sender); if (!cpu_isset(cpu, f->flush_cpumask)) diff --git a/include/asm-x86_64/hw_irq.h b/include/asm-x86_64/hw_irq.h index 1b2ac55d320..93187746278 100644 --- a/include/asm-x86_64/hw_irq.h +++ b/include/asm-x86_64/hw_irq.h @@ -124,7 +124,7 @@ asmlinkage void IRQ_NAME(nr); \ __asm__( \ "\n.p2align\n" \ "IRQ" #nr "_interrupt:\n\t" \ - "push $" #nr "-256 ; " \ + "push $~(" #nr ") ; " \ "jmp common_interrupt"); #if defined(CONFIG_X86_IO_APIC) -- cgit v1.2.3-70-g09d2 From 4b89aff930d632be10d557d08d1b60dee7163dbe Mon Sep 17 00:00:00 2001 From: Rohit Seth Date: Tue, 27 Jun 2006 02:53:46 -0700 Subject: [PATCH] i386: move phys_proc_id and cpu_core_id to cpuinfo_x86 Move the phys_core_id and cpu_core_id to cpuinfo_x86 structure. Similar patch for x86_64 is already accepted by Andi earlier this week. [akpm@osdl.org: fix warning] Signed-off-by: Rohit Seth Cc: Andi Kleen Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/i386/kernel/cpu/amd.c | 6 +++--- arch/i386/kernel/cpu/common.c | 15 +++++++-------- arch/i386/kernel/cpu/proc.c | 4 ++-- arch/i386/kernel/smpboot.c | 16 +++++----------- include/asm-i386/processor.h | 8 +++++--- include/asm-i386/topology.h | 6 ++---- 6 files changed, 24 insertions(+), 31 deletions(-) (limited to 'include') diff --git a/arch/i386/kernel/cpu/amd.c b/arch/i386/kernel/cpu/amd.c index fd0457c9c82..e6a2d6b80cd 100644 --- a/arch/i386/kernel/cpu/amd.c +++ b/arch/i386/kernel/cpu/amd.c @@ -235,10 +235,10 @@ static void __init init_amd(struct cpuinfo_x86 *c) while ((1 << bits) < c->x86_max_cores) bits++; } - cpu_core_id[cpu] = phys_proc_id[cpu] & ((1<>= bits; + c->cpu_core_id = c->phys_proc_id & ((1<phys_proc_id >>= bits; printk(KERN_INFO "CPU %d(%d) -> Core %d\n", - cpu, c->x86_max_cores, cpu_core_id[cpu]); + cpu, c->x86_max_cores, c->cpu_core_id); } #endif diff --git a/arch/i386/kernel/cpu/common.c b/arch/i386/kernel/cpu/common.c index 640364d9b66..2fa401f19f3 100644 --- a/arch/i386/kernel/cpu/common.c +++ b/arch/i386/kernel/cpu/common.c @@ -319,7 +319,7 @@ void __cpuinit generic_identify(struct cpuinfo_x86 * c) early_intel_workaround(c); #ifdef CONFIG_X86_HT - phys_proc_id[smp_processor_id()] = (cpuid_ebx(1) >> 24) & 0xff; + c->phys_proc_id = (cpuid_ebx(1) >> 24) & 0xff; #endif } @@ -477,11 +477,9 @@ void __cpuinit detect_ht(struct cpuinfo_x86 *c) { u32 eax, ebx, ecx, edx; int index_msb, core_bits; - int cpu = smp_processor_id(); cpuid(1, &eax, &ebx, &ecx, &edx); - if (!cpu_has(c, X86_FEATURE_HT) || cpu_has(c, X86_FEATURE_CMP_LEGACY)) return; @@ -492,16 +490,17 @@ void __cpuinit detect_ht(struct cpuinfo_x86 *c) } else if (smp_num_siblings > 1 ) { if (smp_num_siblings > NR_CPUS) { - printk(KERN_WARNING "CPU: Unsupported number of the siblings %d", smp_num_siblings); + printk(KERN_WARNING "CPU: Unsupported number of the " + "siblings %d", smp_num_siblings); smp_num_siblings = 1; return; } index_msb = get_count_order(smp_num_siblings); - phys_proc_id[cpu] = phys_pkg_id((ebx >> 24) & 0xFF, index_msb); + c->phys_proc_id = phys_pkg_id((ebx >> 24) & 0xFF, index_msb); printk(KERN_INFO "CPU: Physical Processor ID: %d\n", - phys_proc_id[cpu]); + c->phys_proc_id); smp_num_siblings = smp_num_siblings / c->x86_max_cores; @@ -509,12 +508,12 @@ void __cpuinit detect_ht(struct cpuinfo_x86 *c) core_bits = get_count_order(c->x86_max_cores); - cpu_core_id[cpu] = phys_pkg_id((ebx >> 24) & 0xFF, index_msb) & + c->cpu_core_id = phys_pkg_id((ebx >> 24) & 0xFF, index_msb) & ((1 << core_bits) - 1); if (c->x86_max_cores > 1) printk(KERN_INFO "CPU: Processor Core ID: %d\n", - cpu_core_id[cpu]); + c->cpu_core_id); } } #endif diff --git a/arch/i386/kernel/cpu/proc.c b/arch/i386/kernel/cpu/proc.c index 4b03f9f5832..f54a15268ed 100644 --- a/arch/i386/kernel/cpu/proc.c +++ b/arch/i386/kernel/cpu/proc.c @@ -109,9 +109,9 @@ static int show_cpuinfo(struct seq_file *m, void *v) seq_printf(m, "cache size\t: %d KB\n", c->x86_cache_size); #ifdef CONFIG_X86_HT if (c->x86_max_cores * smp_num_siblings > 1) { - seq_printf(m, "physical id\t: %d\n", phys_proc_id[n]); + seq_printf(m, "physical id\t: %d\n", c->phys_proc_id); seq_printf(m, "siblings\t: %d\n", cpus_weight(cpu_core_map[n])); - seq_printf(m, "core id\t\t: %d\n", cpu_core_id[n]); + seq_printf(m, "core id\t\t: %d\n", c->cpu_core_id); seq_printf(m, "cpu cores\t: %d\n", c->booted_cores); } #endif diff --git a/arch/i386/kernel/smpboot.c b/arch/i386/kernel/smpboot.c index 9466a3c9ff0..ab5275beddf 100644 --- a/arch/i386/kernel/smpboot.c +++ b/arch/i386/kernel/smpboot.c @@ -67,12 +67,6 @@ int smp_num_siblings = 1; EXPORT_SYMBOL(smp_num_siblings); #endif -/* Package ID of each logical CPU */ -int phys_proc_id[NR_CPUS] __read_mostly = {[0 ... NR_CPUS-1] = BAD_APICID}; - -/* Core ID of each logical CPU */ -int cpu_core_id[NR_CPUS] __read_mostly = {[0 ... NR_CPUS-1] = BAD_APICID}; - /* Last level cache ID of each logical CPU */ int cpu_llc_id[NR_CPUS] __cpuinitdata = {[0 ... NR_CPUS-1] = BAD_APICID}; @@ -473,8 +467,8 @@ set_cpu_sibling_map(int cpu) if (smp_num_siblings > 1) { for_each_cpu_mask(i, cpu_sibling_setup_map) { - if (phys_proc_id[cpu] == phys_proc_id[i] && - cpu_core_id[cpu] == cpu_core_id[i]) { + if (c[cpu].phys_proc_id == c[i].phys_proc_id && + c[cpu].cpu_core_id == c[i].cpu_core_id) { cpu_set(i, cpu_sibling_map[cpu]); cpu_set(cpu, cpu_sibling_map[i]); cpu_set(i, cpu_core_map[cpu]); @@ -501,7 +495,7 @@ set_cpu_sibling_map(int cpu) cpu_set(i, c[cpu].llc_shared_map); cpu_set(cpu, c[i].llc_shared_map); } - if (phys_proc_id[cpu] == phys_proc_id[i]) { + if (c[cpu].phys_proc_id == c[i].phys_proc_id) { cpu_set(i, cpu_core_map[cpu]); cpu_set(cpu, cpu_core_map[i]); /* @@ -1353,8 +1347,8 @@ remove_siblinginfo(int cpu) cpu_clear(cpu, cpu_sibling_map[sibling]); cpus_clear(cpu_sibling_map[cpu]); cpus_clear(cpu_core_map[cpu]); - phys_proc_id[cpu] = BAD_APICID; - cpu_core_id[cpu] = BAD_APICID; + c[cpu].phys_proc_id = 0; + c[cpu].cpu_core_id = 0; cpu_clear(cpu, cpu_sibling_setup_map); } diff --git a/include/asm-i386/processor.h b/include/asm-i386/processor.h index 55ea992da32..b32346d62e1 100644 --- a/include/asm-i386/processor.h +++ b/include/asm-i386/processor.h @@ -71,8 +71,12 @@ struct cpuinfo_x86 { cpumask_t llc_shared_map; /* cpus sharing the last level cache */ #endif unsigned char x86_max_cores; /* cpuid returned max cores value */ - unsigned char booted_cores; /* number of cores as seen by OS */ unsigned char apicid; +#ifdef CONFIG_SMP + unsigned char booted_cores; /* number of cores as seen by OS */ + __u8 phys_proc_id; /* Physical processor id. */ + __u8 cpu_core_id; /* Core id */ +#endif } __attribute__((__aligned__(SMP_CACHE_BYTES))); #define X86_VENDOR_INTEL 0 @@ -104,8 +108,6 @@ extern struct cpuinfo_x86 cpu_data[]; #define current_cpu_data boot_cpu_data #endif -extern int phys_proc_id[NR_CPUS]; -extern int cpu_core_id[NR_CPUS]; extern int cpu_llc_id[NR_CPUS]; extern char ignore_fpu_irq; diff --git a/include/asm-i386/topology.h b/include/asm-i386/topology.h index b94e5eeef91..aa4185ee81f 100644 --- a/include/asm-i386/topology.h +++ b/include/asm-i386/topology.h @@ -28,10 +28,8 @@ #define _ASM_I386_TOPOLOGY_H #ifdef CONFIG_X86_HT -#define topology_physical_package_id(cpu) \ - (phys_proc_id[cpu] == BAD_APICID ? -1 : phys_proc_id[cpu]) -#define topology_core_id(cpu) \ - (cpu_core_id[cpu] == BAD_APICID ? 0 : cpu_core_id[cpu]) +#define topology_physical_package_id(cpu) (cpu_data[cpu].phys_proc_id) +#define topology_core_id(cpu) (cpu_data[cpu].cpu_core_id) #define topology_core_siblings(cpu) (cpu_core_map[cpu]) #define topology_thread_siblings(cpu) (cpu_sibling_map[cpu]) #endif -- cgit v1.2.3-70-g09d2 From c723e084606ca1c81e91b80b2c0c44bde7bbc4df Mon Sep 17 00:00:00 2001 From: Chuck Ebbert <76306.1226@compuserve.com> Date: Tue, 27 Jun 2006 02:53:47 -0700 Subject: [PATCH] i386: use C code for current_thread_info() Using C code for current_thread_info() lets the compiler optimize it. With gcc 4.0.2, kernel is smaller: text data bss dec hex filename 3645212 555556 312024 4512792 44dc18 2.6.17-rc6-nb-post/vmlinux 3647276 555556 312024 4514856 44e428 2.6.17-rc6-nb/vmlinux ------- -2064 Signed-off-by: Chuck Ebbert <76306.1226@compuserve.com> Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/asm-i386/thread_info.h | 10 ++++------ 1 file changed, 4 insertions(+), 6 deletions(-) (limited to 'include') diff --git a/include/asm-i386/thread_info.h b/include/asm-i386/thread_info.h index fdbc7f422ea..ff1e2b1a7c8 100644 --- a/include/asm-i386/thread_info.h +++ b/include/asm-i386/thread_info.h @@ -83,17 +83,15 @@ struct thread_info { #define init_stack (init_thread_union.stack) +/* how to get the current stack pointer from C */ +register unsigned long current_stack_pointer asm("esp") __attribute_used__; + /* how to get the thread information struct from C */ static inline struct thread_info *current_thread_info(void) { - struct thread_info *ti; - __asm__("andl %%esp,%0; ":"=r" (ti) : "0" (~(THREAD_SIZE - 1))); - return ti; + return (struct thread_info *)(current_stack_pointer & ~(THREAD_SIZE - 1)); } -/* how to get the current stack pointer from C */ -register unsigned long current_stack_pointer asm("esp") __attribute_used__; - /* thread information allocation */ #ifdef CONFIG_DEBUG_STACK_USAGE #define alloc_thread_info(tsk) \ -- cgit v1.2.3-70-g09d2 From e6e5494cb23d1933735ee47cc674ffe1c4afed6f Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Tue, 27 Jun 2006 02:53:50 -0700 Subject: [PATCH] vdso: randomize the i386 vDSO by moving it into a vma Move the i386 VDSO down into a vma and thus randomize it. Besides the security implications, this feature also helps debuggers, which can COW a vma-backed VDSO just like a normal DSO and can thus do single-stepping and other debugging features. It's good for hypervisors (Xen, VMWare) too, which typically live in the same high-mapped address space as the VDSO, hence whenever the VDSO is used, they get lots of guest pagefaults and have to fix such guest accesses up - which slows things down instead of speeding things up (the primary purpose of the VDSO). There's a new CONFIG_COMPAT_VDSO (default=y) option, which provides support for older glibcs that still rely on a prelinked high-mapped VDSO. Newer distributions (using glibc 2.3.3 or later) can turn this option off. Turning it off is also recommended for security reasons: attackers cannot use the predictable high-mapped VDSO page as syscall trampoline anymore. There is a new vdso=[0|1] boot option as well, and a runtime /proc/sys/vm/vdso_enabled sysctl switch, that allows the VDSO to be turned on/off. (This version of the VDSO-randomization patch also has working ELF coredumping, the previous patch crashed in the coredumping code.) This code is a combined work of the exec-shield VDSO randomization code and Gerd Hoffmann's hypervisor-centric VDSO patch. Rusty Russell started this patch and i completed it. [akpm@osdl.org: cleanups] [akpm@osdl.org: compile fix] [akpm@osdl.org: compile fix 2] [akpm@osdl.org: compile fix 3] [akpm@osdl.org: revernt MAXMEM change] Signed-off-by: Ingo Molnar Signed-off-by: Arjan van de Ven Cc: Gerd Hoffmann Cc: Rusty Russell Cc: Zachary Amsden Cc: Andi Kleen Cc: Jan Beulich Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- Documentation/kernel-parameters.txt | 4 ++ arch/i386/Kconfig | 11 +++ arch/i386/kernel/asm-offsets.c | 4 +- arch/i386/kernel/entry.S | 7 +- arch/i386/kernel/signal.c | 4 +- arch/i386/kernel/sysenter.c | 128 +++++++++++++++++++++++++++++++++-- arch/i386/kernel/vsyscall-sysenter.S | 4 +- arch/i386/kernel/vsyscall.lds.S | 4 +- fs/proc/task_mmu.c | 30 ++++---- include/asm-i386/elf.h | 53 +++++++++++---- include/asm-i386/fixmap.h | 10 +-- include/asm-i386/mmu.h | 1 + include/asm-i386/page.h | 3 + include/asm-i386/thread_info.h | 1 + include/asm-i386/unwind.h | 4 +- include/linux/mm.h | 2 + include/linux/sysctl.h | 1 + kernel/sysctl.c | 12 ++++ 18 files changed, 235 insertions(+), 48 deletions(-) (limited to 'include') diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt index 2e352a605fc..0d189c93eea 100644 --- a/Documentation/kernel-parameters.txt +++ b/Documentation/kernel-parameters.txt @@ -1669,6 +1669,10 @@ running once the system is up. usbhid.mousepoll= [USBHID] The interval which mice are to be polled at. + vdso= [IA-32] + vdso=1: enable VDSO (default) + vdso=0: disable VDSO mapping + video= [FB] Frame buffer configuration See Documentation/fb/modedb.txt. diff --git a/arch/i386/Kconfig b/arch/i386/Kconfig index 6662f8c4479..3bb221db164 100644 --- a/arch/i386/Kconfig +++ b/arch/i386/Kconfig @@ -780,6 +780,17 @@ config HOTPLUG_CPU enable suspend on SMP systems. CPUs can be controlled through /sys/devices/system/cpu. +config COMPAT_VDSO + bool "Compat VDSO support" + default y + help + Map the VDSO to the predictable old-style address too. + ---help--- + Say N here if you are running a sufficiently recent glibc + version (2.3.3 or later), to remove the high-mapped + VDSO mapping and to exclusively use the randomized VDSO. + + If unsure, say Y. endmenu diff --git a/arch/i386/kernel/asm-offsets.c b/arch/i386/kernel/asm-offsets.c index 1c3a809e642..c80271f8f08 100644 --- a/arch/i386/kernel/asm-offsets.c +++ b/arch/i386/kernel/asm-offsets.c @@ -14,6 +14,7 @@ #include #include #include +#include #define DEFINE(sym, val) \ asm volatile("\n->" #sym " %0 " #val : : "i" (val)) @@ -54,6 +55,7 @@ void foo(void) OFFSET(TI_preempt_count, thread_info, preempt_count); OFFSET(TI_addr_limit, thread_info, addr_limit); OFFSET(TI_restart_block, thread_info, restart_block); + OFFSET(TI_sysenter_return, thread_info, sysenter_return); BLANK(); OFFSET(EXEC_DOMAIN_handler, exec_domain, handler); @@ -69,7 +71,7 @@ void foo(void) sizeof(struct tss_struct)); DEFINE(PAGE_SIZE_asm, PAGE_SIZE); - DEFINE(VSYSCALL_BASE, __fix_to_virt(FIX_VSYSCALL)); + DEFINE(VDSO_PRELINK, VDSO_PRELINK); OFFSET(crypto_tfm_ctx_offset, crypto_tfm, __crt_ctx); } diff --git a/arch/i386/kernel/entry.S b/arch/i386/kernel/entry.S index e8d2630fd19..fbdb933251b 100644 --- a/arch/i386/kernel/entry.S +++ b/arch/i386/kernel/entry.S @@ -270,7 +270,12 @@ sysenter_past_esp: pushl $(__USER_CS) CFI_ADJUST_CFA_OFFSET 4 /*CFI_REL_OFFSET cs, 0*/ - pushl $SYSENTER_RETURN + /* + * Push current_thread_info()->sysenter_return to the stack. + * A tiny bit of offset fixup is necessary - 4*4 means the 4 words + * pushed above; +8 corresponds to copy_thread's esp0 setting. + */ + pushl (TI_sysenter_return-THREAD_SIZE+8+4*4)(%esp) CFI_ADJUST_CFA_OFFSET 4 CFI_REL_OFFSET eip, 0 diff --git a/arch/i386/kernel/signal.c b/arch/i386/kernel/signal.c index 5c352c3a9e7..43002cfb40c 100644 --- a/arch/i386/kernel/signal.c +++ b/arch/i386/kernel/signal.c @@ -351,7 +351,7 @@ static int setup_frame(int sig, struct k_sigaction *ka, goto give_sigsegv; } - restorer = &__kernel_sigreturn; + restorer = (void *)VDSO_SYM(&__kernel_sigreturn); if (ka->sa.sa_flags & SA_RESTORER) restorer = ka->sa.sa_restorer; @@ -447,7 +447,7 @@ static int setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info, goto give_sigsegv; /* Set up to return from userspace. */ - restorer = &__kernel_rt_sigreturn; + restorer = (void *)VDSO_SYM(&__kernel_rt_sigreturn); if (ka->sa.sa_flags & SA_RESTORER) restorer = ka->sa.sa_restorer; err |= __put_user(restorer, &frame->pretcode); diff --git a/arch/i386/kernel/sysenter.c b/arch/i386/kernel/sysenter.c index 0bada1870bd..c60419dee01 100644 --- a/arch/i386/kernel/sysenter.c +++ b/arch/i386/kernel/sysenter.c @@ -2,6 +2,8 @@ * linux/arch/i386/kernel/sysenter.c * * (C) Copyright 2002 Linus Torvalds + * Portions based on the vdso-randomization code from exec-shield: + * Copyright(C) 2005-2006, Red Hat, Inc., Ingo Molnar * * This file contains the needed initializations to support sysenter. */ @@ -13,12 +15,31 @@ #include #include #include +#include +#include #include #include #include #include +/* + * Should the kernel map a VDSO page into processes and pass its + * address down to glibc upon exec()? + */ +unsigned int __read_mostly vdso_enabled = 1; + +EXPORT_SYMBOL_GPL(vdso_enabled); + +static int __init vdso_setup(char *s) +{ + vdso_enabled = simple_strtoul(s, NULL, 0); + + return 1; +} + +__setup("vdso=", vdso_setup); + extern asmlinkage void sysenter_entry(void); void enable_sep_cpu(void) @@ -45,23 +66,122 @@ void enable_sep_cpu(void) */ extern const char vsyscall_int80_start, vsyscall_int80_end; extern const char vsyscall_sysenter_start, vsyscall_sysenter_end; +static void *syscall_page; int __init sysenter_setup(void) { - void *page = (void *)get_zeroed_page(GFP_ATOMIC); + syscall_page = (void *)get_zeroed_page(GFP_ATOMIC); - __set_fixmap(FIX_VSYSCALL, __pa(page), PAGE_READONLY_EXEC); +#ifdef CONFIG_COMPAT_VDSO + __set_fixmap(FIX_VDSO, __pa(syscall_page), PAGE_READONLY); + printk("Compat vDSO mapped to %08lx.\n", __fix_to_virt(FIX_VDSO)); +#else + /* + * In the non-compat case the ELF coredumping code needs the fixmap: + */ + __set_fixmap(FIX_VDSO, __pa(syscall_page), PAGE_KERNEL_RO); +#endif if (!boot_cpu_has(X86_FEATURE_SEP)) { - memcpy(page, + memcpy(syscall_page, &vsyscall_int80_start, &vsyscall_int80_end - &vsyscall_int80_start); return 0; } - memcpy(page, + memcpy(syscall_page, &vsyscall_sysenter_start, &vsyscall_sysenter_end - &vsyscall_sysenter_start); return 0; } + +static struct page *syscall_nopage(struct vm_area_struct *vma, + unsigned long adr, int *type) +{ + struct page *p = virt_to_page(adr - vma->vm_start + syscall_page); + get_page(p); + return p; +} + +/* Prevent VMA merging */ +static void syscall_vma_close(struct vm_area_struct *vma) +{ +} + +static struct vm_operations_struct syscall_vm_ops = { + .close = syscall_vma_close, + .nopage = syscall_nopage, +}; + +/* Defined in vsyscall-sysenter.S */ +extern void SYSENTER_RETURN; + +/* Setup a VMA at program startup for the vsyscall page */ +int arch_setup_additional_pages(struct linux_binprm *bprm, int exstack) +{ + struct vm_area_struct *vma; + struct mm_struct *mm = current->mm; + unsigned long addr; + int ret; + + down_write(&mm->mmap_sem); + addr = get_unmapped_area(NULL, 0, PAGE_SIZE, 0, 0); + if (IS_ERR_VALUE(addr)) { + ret = addr; + goto up_fail; + } + + vma = kmem_cache_zalloc(vm_area_cachep, SLAB_KERNEL); + if (!vma) { + ret = -ENOMEM; + goto up_fail; + } + + vma->vm_start = addr; + vma->vm_end = addr + PAGE_SIZE; + /* MAYWRITE to allow gdb to COW and set breakpoints */ + vma->vm_flags = VM_READ|VM_EXEC|VM_MAYREAD|VM_MAYEXEC|VM_MAYWRITE; + vma->vm_flags |= mm->def_flags; + vma->vm_page_prot = protection_map[vma->vm_flags & 7]; + vma->vm_ops = &syscall_vm_ops; + vma->vm_mm = mm; + + ret = insert_vm_struct(mm, vma); + if (ret) + goto free_vma; + + current->mm->context.vdso = (void *)addr; + current_thread_info()->sysenter_return = + (void *)VDSO_SYM(&SYSENTER_RETURN); + mm->total_vm++; +up_fail: + up_write(&mm->mmap_sem); + return ret; + +free_vma: + kmem_cache_free(vm_area_cachep, vma); + return ret; +} + +const char *arch_vma_name(struct vm_area_struct *vma) +{ + if (vma->vm_mm && vma->vm_start == (long)vma->vm_mm->context.vdso) + return "[vdso]"; + return NULL; +} + +struct vm_area_struct *get_gate_vma(struct task_struct *tsk) +{ + return NULL; +} + +int in_gate_area(struct task_struct *task, unsigned long addr) +{ + return 0; +} + +int in_gate_area_no_task(unsigned long addr) +{ + return 0; +} diff --git a/arch/i386/kernel/vsyscall-sysenter.S b/arch/i386/kernel/vsyscall-sysenter.S index 3b62baa6a37..1a36d26e15e 100644 --- a/arch/i386/kernel/vsyscall-sysenter.S +++ b/arch/i386/kernel/vsyscall-sysenter.S @@ -42,10 +42,10 @@ __kernel_vsyscall: /* 7: align return point with nop's to make disassembly easier */ .space 7,0x90 - /* 14: System call restart point is here! (SYSENTER_RETURN - 2) */ + /* 14: System call restart point is here! (SYSENTER_RETURN-2) */ jmp .Lenter_kernel /* 16: System call normal return point is here! */ - .globl SYSENTER_RETURN /* Symbol used by entry.S. */ + .globl SYSENTER_RETURN /* Symbol used by sysenter.c */ SYSENTER_RETURN: pop %ebp .Lpop_ebp: diff --git a/arch/i386/kernel/vsyscall.lds.S b/arch/i386/kernel/vsyscall.lds.S index 98699ca6e52..e26975fc68b 100644 --- a/arch/i386/kernel/vsyscall.lds.S +++ b/arch/i386/kernel/vsyscall.lds.S @@ -7,7 +7,7 @@ SECTIONS { - . = VSYSCALL_BASE + SIZEOF_HEADERS; + . = VDSO_PRELINK + SIZEOF_HEADERS; .hash : { *(.hash) } :text .dynsym : { *(.dynsym) } @@ -20,7 +20,7 @@ SECTIONS For the layouts to match, we need to skip more than enough space for the dynamic symbol table et al. If this amount is insufficient, ld -shared will barf. Just increase it here. */ - . = VSYSCALL_BASE + 0x400; + . = VDSO_PRELINK + 0x400; .text : { *(.text) } :text =0x90909090 .note : { *(.note.*) } :text :note diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c index 0137ec4c136..0a163a4f776 100644 --- a/fs/proc/task_mmu.c +++ b/fs/proc/task_mmu.c @@ -122,6 +122,11 @@ struct mem_size_stats unsigned long private_dirty; }; +__attribute__((weak)) const char *arch_vma_name(struct vm_area_struct *vma) +{ + return NULL; +} + static int show_map_internal(struct seq_file *m, void *v, struct mem_size_stats *mss) { struct proc_maps_private *priv = m->private; @@ -158,22 +163,23 @@ static int show_map_internal(struct seq_file *m, void *v, struct mem_size_stats pad_len_spaces(m, len); seq_path(m, file->f_vfsmnt, file->f_dentry, "\n"); } else { - if (mm) { - if (vma->vm_start <= mm->start_brk && + const char *name = arch_vma_name(vma); + if (!name) { + if (mm) { + if (vma->vm_start <= mm->start_brk && vma->vm_end >= mm->brk) { - pad_len_spaces(m, len); - seq_puts(m, "[heap]"); - } else { - if (vma->vm_start <= mm->start_stack && - vma->vm_end >= mm->start_stack) { - - pad_len_spaces(m, len); - seq_puts(m, "[stack]"); + name = "[heap]"; + } else if (vma->vm_start <= mm->start_stack && + vma->vm_end >= mm->start_stack) { + name = "[stack]"; } + } else { + name = "[vdso]"; } - } else { + } + if (name) { pad_len_spaces(m, len); - seq_puts(m, "[vdso]"); + seq_puts(m, name); } } seq_putc(m, '\n'); diff --git a/include/asm-i386/elf.h b/include/asm-i386/elf.h index 4153d80e4d2..1eac92cb5b1 100644 --- a/include/asm-i386/elf.h +++ b/include/asm-i386/elf.h @@ -10,6 +10,7 @@ #include #include /* for savesegment */ #include +#include #include @@ -129,15 +130,41 @@ extern int dump_task_extended_fpu (struct task_struct *, struct user_fxsr_struct #define ELF_CORE_COPY_FPREGS(tsk, elf_fpregs) dump_task_fpu(tsk, elf_fpregs) #define ELF_CORE_COPY_XFPREGS(tsk, elf_xfpregs) dump_task_extended_fpu(tsk, elf_xfpregs) -#define VSYSCALL_BASE (__fix_to_virt(FIX_VSYSCALL)) -#define VSYSCALL_EHDR ((const struct elfhdr *) VSYSCALL_BASE) -#define VSYSCALL_ENTRY ((unsigned long) &__kernel_vsyscall) +#define VDSO_HIGH_BASE (__fix_to_virt(FIX_VDSO)) +#define VDSO_BASE ((unsigned long)current->mm->context.vdso) + +#ifdef CONFIG_COMPAT_VDSO +# define VDSO_COMPAT_BASE VDSO_HIGH_BASE +# define VDSO_PRELINK VDSO_HIGH_BASE +#else +# define VDSO_COMPAT_BASE VDSO_BASE +# define VDSO_PRELINK 0 +#endif + +#define VDSO_COMPAT_SYM(x) \ + (VDSO_COMPAT_BASE + (unsigned long)(x) - VDSO_PRELINK) + +#define VDSO_SYM(x) \ + (VDSO_BASE + (unsigned long)(x) - VDSO_PRELINK) + +#define VDSO_HIGH_EHDR ((const struct elfhdr *) VDSO_HIGH_BASE) +#define VDSO_EHDR ((const struct elfhdr *) VDSO_COMPAT_BASE) + extern void __kernel_vsyscall; +#define VDSO_ENTRY VDSO_SYM(&__kernel_vsyscall) + +#define ARCH_HAS_SETUP_ADDITIONAL_PAGES +struct linux_binprm; +extern int arch_setup_additional_pages(struct linux_binprm *bprm, + int executable_stack); + +extern unsigned int vdso_enabled; + #define ARCH_DLINFO \ -do { \ - NEW_AUX_ENT(AT_SYSINFO, VSYSCALL_ENTRY); \ - NEW_AUX_ENT(AT_SYSINFO_EHDR, VSYSCALL_BASE); \ +do if (vdso_enabled) { \ + NEW_AUX_ENT(AT_SYSINFO, VDSO_ENTRY); \ + NEW_AUX_ENT(AT_SYSINFO_EHDR, VDSO_COMPAT_BASE); \ } while (0) /* @@ -148,15 +175,15 @@ do { \ * Dumping its extra ELF program headers includes all the other information * a debugger needs to easily find how the vsyscall DSO was being used. */ -#define ELF_CORE_EXTRA_PHDRS (VSYSCALL_EHDR->e_phnum) +#define ELF_CORE_EXTRA_PHDRS (VDSO_HIGH_EHDR->e_phnum) #define ELF_CORE_WRITE_EXTRA_PHDRS \ do { \ const struct elf_phdr *const vsyscall_phdrs = \ - (const struct elf_phdr *) (VSYSCALL_BASE \ - + VSYSCALL_EHDR->e_phoff); \ + (const struct elf_phdr *) (VDSO_HIGH_BASE \ + + VDSO_HIGH_EHDR->e_phoff); \ int i; \ Elf32_Off ofs = 0; \ - for (i = 0; i < VSYSCALL_EHDR->e_phnum; ++i) { \ + for (i = 0; i < VDSO_HIGH_EHDR->e_phnum; ++i) { \ struct elf_phdr phdr = vsyscall_phdrs[i]; \ if (phdr.p_type == PT_LOAD) { \ BUG_ON(ofs != 0); \ @@ -174,10 +201,10 @@ do { \ #define ELF_CORE_WRITE_EXTRA_DATA \ do { \ const struct elf_phdr *const vsyscall_phdrs = \ - (const struct elf_phdr *) (VSYSCALL_BASE \ - + VSYSCALL_EHDR->e_phoff); \ + (const struct elf_phdr *) (VDSO_HIGH_BASE \ + + VDSO_HIGH_EHDR->e_phoff); \ int i; \ - for (i = 0; i < VSYSCALL_EHDR->e_phnum; ++i) { \ + for (i = 0; i < VDSO_HIGH_EHDR->e_phnum; ++i) { \ if (vsyscall_phdrs[i].p_type == PT_LOAD) \ DUMP_WRITE((void *) vsyscall_phdrs[i].p_vaddr, \ PAGE_ALIGN(vsyscall_phdrs[i].p_memsz)); \ diff --git a/include/asm-i386/fixmap.h b/include/asm-i386/fixmap.h index f7e068f4d2f..a48cc3f7ccc 100644 --- a/include/asm-i386/fixmap.h +++ b/include/asm-i386/fixmap.h @@ -51,7 +51,7 @@ */ enum fixed_addresses { FIX_HOLE, - FIX_VSYSCALL, + FIX_VDSO, #ifdef CONFIG_X86_LOCAL_APIC FIX_APIC_BASE, /* local (CPU) APIC) -- required for SMP or not */ #endif @@ -115,14 +115,6 @@ extern void __set_fixmap (enum fixed_addresses idx, #define __fix_to_virt(x) (FIXADDR_TOP - ((x) << PAGE_SHIFT)) #define __virt_to_fix(x) ((FIXADDR_TOP - ((x)&PAGE_MASK)) >> PAGE_SHIFT) -/* - * This is the range that is readable by user mode, and things - * acting like user mode such as get_user_pages. - */ -#define FIXADDR_USER_START (__fix_to_virt(FIX_VSYSCALL)) -#define FIXADDR_USER_END (FIXADDR_USER_START + PAGE_SIZE) - - extern void __this_fixmap_does_not_exist(void); /* diff --git a/include/asm-i386/mmu.h b/include/asm-i386/mmu.h index f431a0b86d4..8358dd3df7a 100644 --- a/include/asm-i386/mmu.h +++ b/include/asm-i386/mmu.h @@ -12,6 +12,7 @@ typedef struct { int size; struct semaphore sem; void *ldt; + void *vdso; } mm_context_t; #endif diff --git a/include/asm-i386/page.h b/include/asm-i386/page.h index e3a552fa553..f5bf544c729 100644 --- a/include/asm-i386/page.h +++ b/include/asm-i386/page.h @@ -96,6 +96,8 @@ typedef struct { unsigned long pgprot; } pgprot_t; #ifndef __ASSEMBLY__ +struct vm_area_struct; + /* * This much address space is reserved for vmalloc() and iomap() * as well as fixmap mappings. @@ -139,6 +141,7 @@ extern int page_is_ram(unsigned long pagenr); #include #include +#define __HAVE_ARCH_GATE_AREA 1 #endif /* __KERNEL__ */ #endif /* _I386_PAGE_H */ diff --git a/include/asm-i386/thread_info.h b/include/asm-i386/thread_info.h index ff1e2b1a7c8..2833fa2c0dd 100644 --- a/include/asm-i386/thread_info.h +++ b/include/asm-i386/thread_info.h @@ -37,6 +37,7 @@ struct thread_info { 0-0xBFFFFFFF for user-thead 0-0xFFFFFFFF for kernel-thread */ + void *sysenter_return; struct restart_block restart_block; unsigned long previous_esp; /* ESP of the previous stack in case diff --git a/include/asm-i386/unwind.h b/include/asm-i386/unwind.h index d480f2e3821..69f0f1df672 100644 --- a/include/asm-i386/unwind.h +++ b/include/asm-i386/unwind.h @@ -78,8 +78,8 @@ static inline int arch_unw_user_mode(const struct unwind_frame_info *info) return user_mode_vm(&info->regs); #else return info->regs.eip < PAGE_OFFSET - || (info->regs.eip >= __fix_to_virt(FIX_VSYSCALL) - && info->regs.eip < __fix_to_virt(FIX_VSYSCALL) + PAGE_SIZE) + || (info->regs.eip >= __fix_to_virt(FIX_VDSO) + && info->regs.eip < __fix_to_virt(FIX_VDSO) + PAGE_SIZE) || info->regs.esp < PAGE_OFFSET; #endif } diff --git a/include/linux/mm.h b/include/linux/mm.h index a929ea197e4..ff1fa87df8d 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -1065,5 +1065,7 @@ void drop_slab(void); extern int randomize_va_space; #endif +const char *arch_vma_name(struct vm_area_struct *vma); + #endif /* __KERNEL__ */ #endif /* _LINUX_MM_H */ diff --git a/include/linux/sysctl.h b/include/linux/sysctl.h index 349ef908a22..bee12a7a057 100644 --- a/include/linux/sysctl.h +++ b/include/linux/sysctl.h @@ -189,6 +189,7 @@ enum VM_ZONE_RECLAIM_MODE=31, /* reclaim local zone memory before going off node */ VM_ZONE_RECLAIM_INTERVAL=32, /* time period to wait after reclaim failure */ VM_PANIC_ON_OOM=33, /* panic at out-of-memory */ + VM_VDSO_ENABLED=34, /* map VDSO into new processes? */ }; diff --git a/kernel/sysctl.c b/kernel/sysctl.c index f1a4eb1a655..f54afed8426 100644 --- a/kernel/sysctl.c +++ b/kernel/sysctl.c @@ -927,6 +927,18 @@ static ctl_table vm_table[] = { .proc_handler = &proc_dointvec_jiffies, .strategy = &sysctl_jiffies, }, +#endif +#ifdef CONFIG_X86_32 + { + .ctl_name = VM_VDSO_ENABLED, + .procname = "vdso_enabled", + .data = &vdso_enabled, + .maxlen = sizeof(vdso_enabled), + .mode = 0644, + .proc_handler = &proc_dointvec, + .strategy = &sysctl_intvec, + .extra1 = &zero, + }, #endif { .ctl_name = 0 } }; -- cgit v1.2.3-70-g09d2 From c9cf55285e87ac423c45d9efca750d3f50234d10 Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Tue, 27 Jun 2006 02:53:52 -0700 Subject: [PATCH] add poison.h and patch primary users Localize poison values into one header file for better documentation and easier/quicker debugging and so that the same values won't be used for multiple purposes. Use these constants in core arch., mm, driver, and fs code. Signed-off-by: Randy Dunlap Acked-by: Matt Mackall Cc: Paul Mackerras Cc: Benjamin Herrenschmidt Cc: "David S. Miller" Cc: Andi Kleen Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/i386/mm/init.c | 3 ++- arch/powerpc/mm/init_64.c | 3 ++- arch/sparc64/mm/init.c | 3 ++- arch/x86_64/mm/init.c | 7 +++++-- drivers/base/dmapool.c | 3 +-- fs/jbd/journal.c | 3 ++- include/linux/list.h | 9 +-------- include/linux/poison.h | 45 +++++++++++++++++++++++++++++++++++++++++++++ mm/slab.c | 12 +----------- 9 files changed, 61 insertions(+), 27 deletions(-) create mode 100644 include/linux/poison.h (limited to 'include') diff --git a/arch/i386/mm/init.c b/arch/i386/mm/init.c index 46859253179..f84b16e007f 100644 --- a/arch/i386/mm/init.c +++ b/arch/i386/mm/init.c @@ -23,6 +23,7 @@ #include #include #include +#include #include #include #include @@ -753,7 +754,7 @@ void free_init_pages(char *what, unsigned long begin, unsigned long end) for (addr = begin; addr < end; addr += PAGE_SIZE) { ClearPageReserved(virt_to_page(addr)); init_page_count(virt_to_page(addr)); - memset((void *)addr, 0xcc, PAGE_SIZE); + memset((void *)addr, POISON_FREE_INITMEM, PAGE_SIZE); free_page(addr); totalram_pages++; } diff --git a/arch/powerpc/mm/init_64.c b/arch/powerpc/mm/init_64.c index 9e30f968c18..d454caada26 100644 --- a/arch/powerpc/mm/init_64.c +++ b/arch/powerpc/mm/init_64.c @@ -41,6 +41,7 @@ #include #include #include +#include #include #include @@ -90,7 +91,7 @@ void free_initmem(void) addr = (unsigned long)__init_begin; for (; addr < (unsigned long)__init_end; addr += PAGE_SIZE) { - memset((void *)addr, 0xcc, PAGE_SIZE); + memset((void *)addr, POISON_FREE_INITMEM, PAGE_SIZE); ClearPageReserved(virt_to_page(addr)); init_page_count(virt_to_page(addr)); free_page(addr); diff --git a/arch/sparc64/mm/init.c b/arch/sparc64/mm/init.c index 5c2bcf354ce..cb75a27adb5 100644 --- a/arch/sparc64/mm/init.c +++ b/arch/sparc64/mm/init.c @@ -18,6 +18,7 @@ #include #include #include +#include #include #include #include @@ -1520,7 +1521,7 @@ void free_initmem(void) page = (addr + ((unsigned long) __va(kern_base)) - ((unsigned long) KERNBASE)); - memset((void *)addr, 0xcc, PAGE_SIZE); + memset((void *)addr, POISON_FREE_INITMEM, PAGE_SIZE); p = virt_to_page(page); ClearPageReserved(p); diff --git a/arch/x86_64/mm/init.c b/arch/x86_64/mm/init.c index 51fbf3eddf1..95bd232ff0c 100644 --- a/arch/x86_64/mm/init.c +++ b/arch/x86_64/mm/init.c @@ -23,6 +23,7 @@ #include #include #include +#include #include #include #include @@ -660,7 +661,8 @@ void free_init_pages(char *what, unsigned long begin, unsigned long end) for (addr = begin; addr < end; addr += PAGE_SIZE) { ClearPageReserved(virt_to_page(addr)); init_page_count(virt_to_page(addr)); - memset((void *)(addr & ~(PAGE_SIZE-1)), 0xcc, PAGE_SIZE); + memset((void *)(addr & ~(PAGE_SIZE-1)), + POISON_FREE_INITMEM, PAGE_SIZE); free_page(addr); totalram_pages++; } @@ -668,7 +670,8 @@ void free_init_pages(char *what, unsigned long begin, unsigned long end) void free_initmem(void) { - memset(__initdata_begin, 0xba, __initdata_end - __initdata_begin); + memset(__initdata_begin, POISON_FREE_INITDATA, + __initdata_end - __initdata_begin); free_init_pages("unused kernel memory", (unsigned long)(&__init_begin), (unsigned long)(&__init_end)); diff --git a/drivers/base/dmapool.c b/drivers/base/dmapool.c index e2f64f91ed0..33c5cce1560 100644 --- a/drivers/base/dmapool.c +++ b/drivers/base/dmapool.c @@ -7,6 +7,7 @@ #include #include #include +#include /* * Pool allocator ... wraps the dma_alloc_coherent page allocator, so @@ -35,8 +36,6 @@ struct dma_page { /* cacheable header for 'allocation' bytes */ }; #define POOL_TIMEOUT_JIFFIES ((100 /* msec */ * HZ) / 1000) -#define POOL_POISON_FREED 0xa7 /* !inuse */ -#define POOL_POISON_ALLOCATED 0xa9 /* !initted */ static DECLARE_MUTEX (pools_lock); diff --git a/fs/jbd/journal.c b/fs/jbd/journal.c index 7f96b5cb678..8c9b28dff11 100644 --- a/fs/jbd/journal.c +++ b/fs/jbd/journal.c @@ -34,6 +34,7 @@ #include #include #include +#include #include #include @@ -1675,7 +1676,7 @@ static void journal_free_journal_head(struct journal_head *jh) { #ifdef CONFIG_JBD_DEBUG atomic_dec(&nr_journal_heads); - memset(jh, 0x5b, sizeof(*jh)); + memset(jh, JBD_POISON_FREE, sizeof(*jh)); #endif kmem_cache_free(journal_head_cache, jh); } diff --git a/include/linux/list.h b/include/linux/list.h index 37ca31b21bb..6b74adf5297 100644 --- a/include/linux/list.h +++ b/include/linux/list.h @@ -4,17 +4,10 @@ #ifdef __KERNEL__ #include +#include #include #include -/* - * These are non-NULL pointers that will result in page faults - * under normal circumstances, used to verify that nobody uses - * non-initialized list entries. - */ -#define LIST_POISON1 ((void *) 0x00100100) -#define LIST_POISON2 ((void *) 0x00200200) - /* * Simple doubly linked list implementation. * diff --git a/include/linux/poison.h b/include/linux/poison.h new file mode 100644 index 00000000000..d536de7dd82 --- /dev/null +++ b/include/linux/poison.h @@ -0,0 +1,45 @@ +#ifndef _LINUX_POISON_H +#define _LINUX_POISON_H + +/********** include/linux/list.h **********/ +/* + * These are non-NULL pointers that will result in page faults + * under normal circumstances, used to verify that nobody uses + * non-initialized list entries. + */ +#define LIST_POISON1 ((void *) 0x00100100) +#define LIST_POISON2 ((void *) 0x00200200) + +/********** mm/slab.c **********/ +/* + * Magic nums for obj red zoning. + * Placed in the first word before and the first word after an obj. + */ +#define RED_INACTIVE 0x5A2CF071UL /* when obj is inactive */ +#define RED_ACTIVE 0x170FC2A5UL /* when obj is active */ + +/* ...and for poisoning */ +#define POISON_INUSE 0x5a /* for use-uninitialised poisoning */ +#define POISON_FREE 0x6b /* for use-after-free poisoning */ +#define POISON_END 0xa5 /* end-byte of poisoning */ + +/********** arch/$ARCH/mm/init.c **********/ +#define POISON_FREE_INITMEM 0xcc + +/********** arch/x86_64/mm/init.c **********/ +#define POISON_FREE_INITDATA 0xba + +/********** arch/ia64/hp/common/sba_iommu.c **********/ +/* + * arch/ia64/hp/common/sba_iommu.c uses a 16-byte poison string with a + * value of "SBAIOMMU POISON\0" for spill-over poisoning. + */ + +/********** fs/jbd/journal.c **********/ +#define JBD_POISON_FREE 0x5b + +/********** drivers/base/dmapool.c **********/ +#define POOL_POISON_FREED 0xa7 /* !inuse */ +#define POOL_POISON_ALLOCATED 0xa9 /* !initted */ + +#endif diff --git a/mm/slab.c b/mm/slab.c index 98ac20bc0de..47982c2d9f3 100644 --- a/mm/slab.c +++ b/mm/slab.c @@ -89,6 +89,7 @@ #include #include #include +#include #include #include #include @@ -492,17 +493,6 @@ struct kmem_cache { #endif #if DEBUG -/* - * Magic nums for obj red zoning. - * Placed in the first word before and the first word after an obj. - */ -#define RED_INACTIVE 0x5A2CF071UL /* when obj is inactive */ -#define RED_ACTIVE 0x170FC2A5UL /* when obj is active */ - -/* ...and for poisoning */ -#define POISON_INUSE 0x5a /* for use-uninitialised poisoning */ -#define POISON_FREE 0x6b /* for use-after-free poisoning */ -#define POISON_END 0xa5 /* end-byte of poisoning */ /* * memory layout of objects: -- cgit v1.2.3-70-g09d2 From b3c681e09193559ba15f6c9562bd37045f120a96 Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Tue, 27 Jun 2006 02:53:53 -0700 Subject: [PATCH] update two drivers for poison.h Update two drivers to use poison.h. Signed-off-by: Randy Dunlap Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/atm/firestream.c | 3 ++- include/linux/poison.h | 6 ++++++ sound/oss/via82cxxx_audio.c | 5 +++-- 3 files changed, 11 insertions(+), 3 deletions(-) (limited to 'include') diff --git a/drivers/atm/firestream.c b/drivers/atm/firestream.c index f2eeaf9dc56..1bca86edf57 100644 --- a/drivers/atm/firestream.c +++ b/drivers/atm/firestream.c @@ -33,6 +33,7 @@ #include #include #include +#include #include #include #include @@ -754,7 +755,7 @@ static void process_txdone_queue (struct fs_dev *dev, struct queue *q) fs_kfree_skb (skb); fs_dprintk (FS_DEBUG_ALLOC, "Free trans-d: %p\n", td); - memset (td, 0x12, sizeof (struct FS_BPENTRY)); + memset (td, ATM_POISON_FREE, sizeof(struct FS_BPENTRY)); kfree (td); break; default: diff --git a/include/linux/poison.h b/include/linux/poison.h index d536de7dd82..4109f37b7b6 100644 --- a/include/linux/poison.h +++ b/include/linux/poison.h @@ -42,4 +42,10 @@ #define POOL_POISON_FREED 0xa7 /* !inuse */ #define POOL_POISON_ALLOCATED 0xa9 /* !initted */ +/********** drivers/atm/ **********/ +#define ATM_POISON_FREE 0x12 + +/********** sound/oss/ **********/ +#define OSS_POISON_FREE 0xAB + #endif diff --git a/sound/oss/via82cxxx_audio.c b/sound/oss/via82cxxx_audio.c index 1a921ee71ab..2343dedd44a 100644 --- a/sound/oss/via82cxxx_audio.c +++ b/sound/oss/via82cxxx_audio.c @@ -24,6 +24,7 @@ #include #include #include +#include #include #include #include @@ -3522,7 +3523,7 @@ err_out_have_mixer: err_out_kfree: #ifndef VIA_NDEBUG - memset (card, 0xAB, sizeof (*card)); /* poison memory */ + memset (card, OSS_POISON_FREE, sizeof (*card)); /* poison memory */ #endif kfree (card); @@ -3559,7 +3560,7 @@ static void __devexit via_remove_one (struct pci_dev *pdev) via_ac97_cleanup (card); #ifndef VIA_NDEBUG - memset (card, 0xAB, sizeof (*card)); /* poison memory */ + memset (card, OSS_POISON_FREE, sizeof (*card)); /* poison memory */ #endif kfree (card); -- cgit v1.2.3-70-g09d2 From a7807a32bbb027ab9955b96734fdc7f1e6497a9f Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Tue, 27 Jun 2006 02:53:54 -0700 Subject: [PATCH] poison: add & use more constants Add more poison values to include/linux/poison.h. It's not clear to me whether some others should be added or not, so I haven't added any of these: ./include/linux/libata.h:#define ATA_TAG_POISON 0xfafbfcfdU ./arch/ppc/8260_io/fcc_enet.c:1918: memset((char *)(&(immap->im_dprambase[(mem_addr+64)])), 0x88, 32); ./drivers/usb/mon/mon_text.c:429: memset(mem, 0xe5, sizeof(struct mon_event_text)); ./drivers/char/ftape/lowlevel/ftape-ctl.c:738: memset(ft_buffer[i]->address, 0xAA, FT_BUFF_SIZE); ./drivers/block/sx8.c:/* 0xf is just arbitrary, non-zero noise; this is sorta like poisoning */ Signed-off-by: Randy Dunlap Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/poison.h | 7 +++++++ kernel/mutex-debug.c | 5 +++-- security/keys/key.c | 3 ++- 3 files changed, 12 insertions(+), 3 deletions(-) (limited to 'include') diff --git a/include/linux/poison.h b/include/linux/poison.h index 4109f37b7b6..a5347c02432 100644 --- a/include/linux/poison.h +++ b/include/linux/poison.h @@ -45,6 +45,13 @@ /********** drivers/atm/ **********/ #define ATM_POISON_FREE 0x12 +/********** kernel/mutexes **********/ +#define MUTEX_DEBUG_INIT 0x11 +#define MUTEX_DEBUG_FREE 0x22 + +/********** security/ **********/ +#define KEY_DESTROY 0xbd + /********** sound/oss/ **********/ #define OSS_POISON_FREE 0xAB diff --git a/kernel/mutex-debug.c b/kernel/mutex-debug.c index 036b6285b15..e38e4bac97c 100644 --- a/kernel/mutex-debug.c +++ b/kernel/mutex-debug.c @@ -16,6 +16,7 @@ #include #include #include +#include #include #include #include @@ -381,7 +382,7 @@ void debug_mutex_set_owner(struct mutex *lock, void debug_mutex_init_waiter(struct mutex_waiter *waiter) { - memset(waiter, 0x11, sizeof(*waiter)); + memset(waiter, MUTEX_DEBUG_INIT, sizeof(*waiter)); waiter->magic = waiter; INIT_LIST_HEAD(&waiter->list); } @@ -397,7 +398,7 @@ void debug_mutex_wake_waiter(struct mutex *lock, struct mutex_waiter *waiter) void debug_mutex_free_waiter(struct mutex_waiter *waiter) { DEBUG_WARN_ON(!list_empty(&waiter->list)); - memset(waiter, 0x22, sizeof(*waiter)); + memset(waiter, MUTEX_DEBUG_FREE, sizeof(*waiter)); } void debug_mutex_add_waiter(struct mutex *lock, struct mutex_waiter *waiter, diff --git a/security/keys/key.c b/security/keys/key.c index 43295ca37b5..80de8c3e9cc 100644 --- a/security/keys/key.c +++ b/security/keys/key.c @@ -11,6 +11,7 @@ #include #include +#include #include #include #include @@ -988,7 +989,7 @@ void unregister_key_type(struct key_type *ktype) if (key->type == ktype) { if (ktype->destroy) ktype->destroy(key); - memset(&key->payload, 0xbd, sizeof(key->payload)); + memset(&key->payload, KEY_DESTROY, sizeof(key->payload)); } } -- cgit v1.2.3-70-g09d2 From b6cd0b772dcc5dc9b4c03d53946474dee399fa72 Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Tue, 27 Jun 2006 02:53:54 -0700 Subject: [PATCH] fs/buffer.c: cleanups - add a proper prototype for the following global function: - buffer_init() - make the following needlessly global function static: - end_buffer_async_write() Signed-off-by: Adrian Bunk Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/buffer.c | 3 +-- include/linux/buffer_head.h | 2 +- init/main.c | 2 +- 3 files changed, 3 insertions(+), 4 deletions(-) (limited to 'include') diff --git a/fs/buffer.c b/fs/buffer.c index 373bb6292bd..f23bb647db4 100644 --- a/fs/buffer.c +++ b/fs/buffer.c @@ -564,7 +564,7 @@ still_busy: * Completion handler for block_write_full_page() - pages which are unlocked * during I/O, and which have PageWriteback cleared upon I/O completion. */ -void end_buffer_async_write(struct buffer_head *bh, int uptodate) +static void end_buffer_async_write(struct buffer_head *bh, int uptodate) { char b[BDEVNAME_SIZE]; unsigned long flags; @@ -3166,7 +3166,6 @@ EXPORT_SYMBOL(block_sync_page); EXPORT_SYMBOL(block_truncate_page); EXPORT_SYMBOL(block_write_full_page); EXPORT_SYMBOL(cont_prepare_write); -EXPORT_SYMBOL(end_buffer_async_write); EXPORT_SYMBOL(end_buffer_read_sync); EXPORT_SYMBOL(end_buffer_write_sync); EXPORT_SYMBOL(file_fsync); diff --git a/include/linux/buffer_head.h b/include/linux/buffer_head.h index fb7e9b7ccbe..737e407d0cd 100644 --- a/include/linux/buffer_head.h +++ b/include/linux/buffer_head.h @@ -149,7 +149,6 @@ void create_empty_buffers(struct page *, unsigned long, unsigned long b_state); void end_buffer_read_sync(struct buffer_head *bh, int uptodate); void end_buffer_write_sync(struct buffer_head *bh, int uptodate); -void end_buffer_async_write(struct buffer_head *bh, int uptodate); /* Things to do with buffers at mapping->private_list */ void mark_buffer_dirty_inode(struct buffer_head *bh, struct inode *inode); @@ -214,6 +213,7 @@ int nobh_truncate_page(struct address_space *, loff_t); int nobh_writepage(struct page *page, get_block_t *get_block, struct writeback_control *wbc); +void buffer_init(void); /* * inline definitions diff --git a/init/main.c b/init/main.c index 80af1a52485..0d57f6ccb63 100644 --- a/init/main.c +++ b/init/main.c @@ -48,6 +48,7 @@ #include #include #include +#include #include #include @@ -80,7 +81,6 @@ extern void mca_init(void); extern void sbus_init(void); extern void sysctl_init(void); extern void signals_init(void); -extern void buffer_init(void); extern void pidhash_init(void); extern void pidmap_init(void); extern void prio_tree_init(void); -- cgit v1.2.3-70-g09d2 From 34af946a22724c4e2b204957f2b24b22a0fb121c Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Tue, 27 Jun 2006 02:53:55 -0700 Subject: [PATCH] spin/rwlock init cleanups locking init cleanups: - convert " = SPIN_LOCK_UNLOCKED" to spin_lock_init() or DEFINE_SPINLOCK() - convert rwlocks in a similar manner this patch was generated automatically. Motivation: - cleanliness - lockdep needs control of lock initialization, which the open-coded variants do not give - it's also useful for -rt and for lock debugging in general Signed-off-by: Ingo Molnar Signed-off-by: Arjan van de Ven Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/ia64/sn/kernel/irq.c | 2 +- arch/mips/kernel/smtc.c | 4 ++-- arch/powerpc/platforms/cell/spufs/switch.c | 2 +- arch/powerpc/platforms/powermac/pfunc_core.c | 2 +- arch/powerpc/platforms/pseries/eeh_event.c | 2 +- arch/powerpc/sysdev/mmio_nvram.c | 2 +- arch/xtensa/kernel/time.c | 2 +- arch/xtensa/kernel/traps.c | 2 +- drivers/char/drm/drm_memory_debug.h | 2 +- drivers/char/drm/via_dmablit.c | 2 +- drivers/char/epca.c | 2 +- drivers/char/moxa.c | 2 +- drivers/char/specialix.c | 2 +- drivers/char/sx.c | 2 +- drivers/isdn/gigaset/common.c | 2 +- drivers/leds/led-core.c | 2 +- drivers/leds/led-triggers.c | 2 +- drivers/misc/ibmasm/module.c | 2 +- drivers/pcmcia/m8xx_pcmcia.c | 4 ++-- drivers/rapidio/rio-access.c | 4 ++-- drivers/rtc/rtc-sa1100.c | 2 +- drivers/rtc/rtc-vr41xx.c | 2 +- drivers/s390/block/dasd_eer.c | 2 +- drivers/scsi/libata-core.c | 2 +- drivers/sn/ioc3.c | 2 +- drivers/video/backlight/hp680_bl.c | 2 +- fs/nfsd/nfs4state.c | 2 +- fs/ocfs2/cluster/heartbeat.c | 2 +- fs/ocfs2/cluster/tcp.c | 2 +- fs/ocfs2/dlm/dlmdomain.c | 2 +- fs/ocfs2/dlm/dlmlock.c | 2 +- fs/ocfs2/dlm/dlmrecovery.c | 4 ++-- fs/ocfs2/dlmglue.c | 2 +- fs/ocfs2/journal.c | 2 +- include/asm-alpha/core_t2.h | 2 +- kernel/audit.c | 2 +- mm/sparse.c | 2 +- net/ipv6/route.c | 2 +- net/sunrpc/auth_gss/gss_krb5_seal.c | 2 +- net/tipc/bcast.c | 4 ++-- net/tipc/bearer.c | 2 +- net/tipc/config.c | 2 +- net/tipc/dbg.c | 2 +- net/tipc/handler.c | 2 +- net/tipc/name_table.c | 4 ++-- net/tipc/net.c | 2 +- net/tipc/node.c | 2 +- net/tipc/port.c | 4 ++-- net/tipc/ref.c | 4 ++-- net/tipc/subscr.c | 2 +- net/tipc/user_reg.c | 2 +- 51 files changed, 59 insertions(+), 59 deletions(-) (limited to 'include') diff --git a/arch/ia64/sn/kernel/irq.c b/arch/ia64/sn/kernel/irq.c index dc8e2b69671..677c6c0fd66 100644 --- a/arch/ia64/sn/kernel/irq.c +++ b/arch/ia64/sn/kernel/irq.c @@ -27,7 +27,7 @@ static void unregister_intr_pda(struct sn_irq_info *sn_irq_info); int sn_force_interrupt_flag = 1; extern int sn_ioif_inited; struct list_head **sn_irq_lh; -static spinlock_t sn_irq_info_lock = SPIN_LOCK_UNLOCKED; /* non-IRQ lock */ +static DEFINE_SPINLOCK(sn_irq_info_lock); /* non-IRQ lock */ u64 sn_intr_alloc(nasid_t local_nasid, int local_widget, struct sn_irq_info *sn_irq_info, diff --git a/arch/mips/kernel/smtc.c b/arch/mips/kernel/smtc.c index 2e8e52c135e..70cf09afdf5 100644 --- a/arch/mips/kernel/smtc.c +++ b/arch/mips/kernel/smtc.c @@ -367,7 +367,7 @@ void mipsmt_prepare_cpus(void) dvpe(); dmt(); - freeIPIq.lock = SPIN_LOCK_UNLOCKED; + spin_lock_init(&freeIPIq.lock); /* * We probably don't have as many VPEs as we do SMP "CPUs", @@ -375,7 +375,7 @@ void mipsmt_prepare_cpus(void) */ for (i=0; ilscsa = lscsa; - csa->register_lock = SPIN_LOCK_UNLOCKED; + spin_lock_init(&csa->register_lock); /* Set LS pages reserved to allow for user-space mapping. */ for (p = lscsa->ls; p < lscsa->ls + LS_SIZE; p += PAGE_SIZE) diff --git a/arch/powerpc/platforms/powermac/pfunc_core.c b/arch/powerpc/platforms/powermac/pfunc_core.c index 047f954a89e..93e7505debc 100644 --- a/arch/powerpc/platforms/powermac/pfunc_core.c +++ b/arch/powerpc/platforms/powermac/pfunc_core.c @@ -546,7 +546,7 @@ struct pmf_device { }; static LIST_HEAD(pmf_devices); -static spinlock_t pmf_lock = SPIN_LOCK_UNLOCKED; +static DEFINE_SPINLOCK(pmf_lock); static DEFINE_MUTEX(pmf_irq_mutex); static void pmf_release_device(struct kref *kref) diff --git a/arch/powerpc/platforms/pseries/eeh_event.c b/arch/powerpc/platforms/pseries/eeh_event.c index 8f2d12935b9..45ccc687e57 100644 --- a/arch/powerpc/platforms/pseries/eeh_event.c +++ b/arch/powerpc/platforms/pseries/eeh_event.c @@ -35,7 +35,7 @@ */ /* EEH event workqueue setup. */ -static spinlock_t eeh_eventlist_lock = SPIN_LOCK_UNLOCKED; +static DEFINE_SPINLOCK(eeh_eventlist_lock); LIST_HEAD(eeh_eventlist); static void eeh_thread_launcher(void *); DECLARE_WORK(eeh_event_wq, eeh_thread_launcher, NULL); diff --git a/arch/powerpc/sysdev/mmio_nvram.c b/arch/powerpc/sysdev/mmio_nvram.c index 74e0d31a355..615350d46b5 100644 --- a/arch/powerpc/sysdev/mmio_nvram.c +++ b/arch/powerpc/sysdev/mmio_nvram.c @@ -32,7 +32,7 @@ static void __iomem *mmio_nvram_start; static long mmio_nvram_len; -static spinlock_t mmio_nvram_lock = SPIN_LOCK_UNLOCKED; +static DEFINE_SPINLOCK(mmio_nvram_lock); static ssize_t mmio_nvram_read(char *buf, size_t count, loff_t *index) { diff --git a/arch/xtensa/kernel/time.c b/arch/xtensa/kernel/time.c index 937d81f62f4..fe14909f45e 100644 --- a/arch/xtensa/kernel/time.c +++ b/arch/xtensa/kernel/time.c @@ -29,7 +29,7 @@ extern volatile unsigned long wall_jiffies; -spinlock_t rtc_lock = SPIN_LOCK_UNLOCKED; +DEFINE_SPINLOCK(rtc_lock); EXPORT_SYMBOL(rtc_lock); diff --git a/arch/xtensa/kernel/traps.c b/arch/xtensa/kernel/traps.c index 225d64d73f0..27e409089a7 100644 --- a/arch/xtensa/kernel/traps.c +++ b/arch/xtensa/kernel/traps.c @@ -461,7 +461,7 @@ void show_code(unsigned int *pc) } } -spinlock_t die_lock = SPIN_LOCK_UNLOCKED; +DEFINE_SPINLOCK(die_lock); void die(const char * str, struct pt_regs * regs, long err) { diff --git a/drivers/char/drm/drm_memory_debug.h b/drivers/char/drm/drm_memory_debug.h index 6543b9a14c4..d117cc99719 100644 --- a/drivers/char/drm/drm_memory_debug.h +++ b/drivers/char/drm/drm_memory_debug.h @@ -43,7 +43,7 @@ typedef struct drm_mem_stats { unsigned long bytes_freed; } drm_mem_stats_t; -static spinlock_t drm_mem_lock = SPIN_LOCK_UNLOCKED; +static DEFINE_SPINLOCK(drm_mem_lock); static unsigned long drm_ram_available = 0; /* In pages */ static unsigned long drm_ram_used = 0; static drm_mem_stats_t drm_mem_stats[] = diff --git a/drivers/char/drm/via_dmablit.c b/drivers/char/drm/via_dmablit.c index b7f17457b42..78a81a4a99c 100644 --- a/drivers/char/drm/via_dmablit.c +++ b/drivers/char/drm/via_dmablit.c @@ -557,7 +557,7 @@ via_init_dmablit(drm_device_t *dev) blitq->num_outstanding = 0; blitq->is_active = 0; blitq->aborting = 0; - blitq->blit_lock = SPIN_LOCK_UNLOCKED; + spin_lock_init(&blitq->blit_lock); for (j=0; jblit_queue + j); } diff --git a/drivers/char/epca.c b/drivers/char/epca.c index 9cad8501d62..dc0602ae850 100644 --- a/drivers/char/epca.c +++ b/drivers/char/epca.c @@ -80,7 +80,7 @@ static int invalid_lilo_config; /* The ISA boards do window flipping into the same spaces so its only sane with a single lock. It's still pretty efficient */ -static spinlock_t epca_lock = SPIN_LOCK_UNLOCKED; +static DEFINE_SPINLOCK(epca_lock); /* ----------------------------------------------------------------------- MAXBOARDS is typically 12, but ISA and EISA cards are restricted to diff --git a/drivers/char/moxa.c b/drivers/char/moxa.c index f43c2e04ead..01247cccb89 100644 --- a/drivers/char/moxa.c +++ b/drivers/char/moxa.c @@ -301,7 +301,7 @@ static struct tty_operations moxa_ops = { .tiocmset = moxa_tiocmset, }; -static spinlock_t moxa_lock = SPIN_LOCK_UNLOCKED; +static DEFINE_SPINLOCK(moxa_lock); #ifdef CONFIG_PCI static int moxa_get_PCI_conf(struct pci_dev *p, int board_type, moxa_board_conf * board) diff --git a/drivers/char/specialix.c b/drivers/char/specialix.c index 1b5330299e3..d2d6b01dcd0 100644 --- a/drivers/char/specialix.c +++ b/drivers/char/specialix.c @@ -2477,7 +2477,7 @@ static int __init specialix_init(void) #endif for (i = 0; i < SX_NBOARD; i++) - sx_board[i].lock = SPIN_LOCK_UNLOCKED; + spin_lock_init(&sx_board[i].lock); if (sx_init_drivers()) { func_exit(); diff --git a/drivers/char/sx.c b/drivers/char/sx.c index 3b474723027..76b9107f7f8 100644 --- a/drivers/char/sx.c +++ b/drivers/char/sx.c @@ -2320,7 +2320,7 @@ static int sx_init_portstructs (int nboards, int nports) #ifdef NEW_WRITE_LOCKING port->gs.port_write_mutex = MUTEX; #endif - port->gs.driver_lock = SPIN_LOCK_UNLOCKED; + spin_lock_init(&port->gs.driver_lock); /* * Initializing wait queue */ diff --git a/drivers/isdn/gigaset/common.c b/drivers/isdn/gigaset/common.c index acb7e265678..2a56bf33a67 100644 --- a/drivers/isdn/gigaset/common.c +++ b/drivers/isdn/gigaset/common.c @@ -981,7 +981,7 @@ exit: EXPORT_SYMBOL_GPL(gigaset_stop); static LIST_HEAD(drivers); -static spinlock_t driver_lock = SPIN_LOCK_UNLOCKED; +static DEFINE_SPINLOCK(driver_lock); struct cardstate *gigaset_get_cs_by_id(int id) { diff --git a/drivers/leds/led-core.c b/drivers/leds/led-core.c index fe6541326c7..9b015f9af35 100644 --- a/drivers/leds/led-core.c +++ b/drivers/leds/led-core.c @@ -18,7 +18,7 @@ #include #include "leds.h" -rwlock_t leds_list_lock = RW_LOCK_UNLOCKED; +DEFINE_RWLOCK(leds_list_lock); LIST_HEAD(leds_list); EXPORT_SYMBOL_GPL(leds_list); diff --git a/drivers/leds/led-triggers.c b/drivers/leds/led-triggers.c index 5e2cd8be119..1b1ce652396 100644 --- a/drivers/leds/led-triggers.c +++ b/drivers/leds/led-triggers.c @@ -26,7 +26,7 @@ /* * Nests outside led_cdev->trigger_lock */ -static rwlock_t triggers_list_lock = RW_LOCK_UNLOCKED; +static DEFINE_RWLOCK(triggers_list_lock); static LIST_HEAD(trigger_list); ssize_t led_trigger_store(struct class_device *dev, const char *buf, diff --git a/drivers/misc/ibmasm/module.c b/drivers/misc/ibmasm/module.c index 1fdf03fd2da..9706cc19134 100644 --- a/drivers/misc/ibmasm/module.c +++ b/drivers/misc/ibmasm/module.c @@ -85,7 +85,7 @@ static int __devinit ibmasm_init_one(struct pci_dev *pdev, const struct pci_devi } memset(sp, 0, sizeof(struct service_processor)); - sp->lock = SPIN_LOCK_UNLOCKED; + spin_lock_init(&sp->lock); INIT_LIST_HEAD(&sp->command_queue); pci_set_drvdata(pdev, (void *)sp); diff --git a/drivers/pcmcia/m8xx_pcmcia.c b/drivers/pcmcia/m8xx_pcmcia.c index 0e07d953511..d0f68ab8f04 100644 --- a/drivers/pcmcia/m8xx_pcmcia.c +++ b/drivers/pcmcia/m8xx_pcmcia.c @@ -157,7 +157,7 @@ MODULE_LICENSE("Dual MPL/GPL"); static int pcmcia_schlvl = PCMCIA_SCHLVL; -static spinlock_t events_lock = SPIN_LOCK_UNLOCKED; +static DEFINE_SPINLOCK(events_lock); #define PCMCIA_SOCKET_KEY_5V 1 @@ -644,7 +644,7 @@ static struct platform_device m8xx_device = { }; static u32 pending_events[PCMCIA_SOCKETS_NO]; -static spinlock_t pending_event_lock = SPIN_LOCK_UNLOCKED; +static DEFINE_SPINLOCK(pending_event_lock); static irqreturn_t m8xx_interrupt(int irq, void *dev, struct pt_regs *regs) { diff --git a/drivers/rapidio/rio-access.c b/drivers/rapidio/rio-access.c index b9fab2ae3a3..8b56bbdd011 100644 --- a/drivers/rapidio/rio-access.c +++ b/drivers/rapidio/rio-access.c @@ -17,8 +17,8 @@ * These interrupt-safe spinlocks protect all accesses to RIO * configuration space and doorbell access. */ -static spinlock_t rio_config_lock = SPIN_LOCK_UNLOCKED; -static spinlock_t rio_doorbell_lock = SPIN_LOCK_UNLOCKED; +static DEFINE_SPINLOCK(rio_config_lock); +static DEFINE_SPINLOCK(rio_doorbell_lock); /* * Wrappers for all RIO configuration access functions. They just check diff --git a/drivers/rtc/rtc-sa1100.c b/drivers/rtc/rtc-sa1100.c index ab486fbc828..9cd1cb304bb 100644 --- a/drivers/rtc/rtc-sa1100.c +++ b/drivers/rtc/rtc-sa1100.c @@ -45,7 +45,7 @@ static unsigned long rtc_freq = 1024; static struct rtc_time rtc_alarm; -static spinlock_t sa1100_rtc_lock = SPIN_LOCK_UNLOCKED; +static DEFINE_SPINLOCK(sa1100_rtc_lock); static int rtc_update_alarm(struct rtc_time *alrm) { diff --git a/drivers/rtc/rtc-vr41xx.c b/drivers/rtc/rtc-vr41xx.c index 33e029207e2..4b9291dd444 100644 --- a/drivers/rtc/rtc-vr41xx.c +++ b/drivers/rtc/rtc-vr41xx.c @@ -93,7 +93,7 @@ static void __iomem *rtc2_base; static unsigned long epoch = 1970; /* Jan 1 1970 00:00:00 */ -static spinlock_t rtc_lock = SPIN_LOCK_UNLOCKED; +static DEFINE_SPINLOCK(rtc_lock); static char rtc_name[] = "RTC"; static unsigned long periodic_frequency; static unsigned long periodic_count; diff --git a/drivers/s390/block/dasd_eer.c b/drivers/s390/block/dasd_eer.c index 2d946b6ca07..2d8af709947 100644 --- a/drivers/s390/block/dasd_eer.c +++ b/drivers/s390/block/dasd_eer.c @@ -89,7 +89,7 @@ struct eerbuffer { }; static LIST_HEAD(bufferlist); -static spinlock_t bufferlock = SPIN_LOCK_UNLOCKED; +static DEFINE_SPINLOCK(bufferlock); static DECLARE_WAIT_QUEUE_HEAD(dasd_eer_read_wait_queue); /* diff --git a/drivers/scsi/libata-core.c b/drivers/scsi/libata-core.c index 6c66877be2b..855ce9a9d94 100644 --- a/drivers/scsi/libata-core.c +++ b/drivers/scsi/libata-core.c @@ -5733,7 +5733,7 @@ module_init(ata_init); module_exit(ata_exit); static unsigned long ratelimit_time; -static spinlock_t ata_ratelimit_lock = SPIN_LOCK_UNLOCKED; +static DEFINE_SPINLOCK(ata_ratelimit_lock); int ata_ratelimit(void) { diff --git a/drivers/sn/ioc3.c b/drivers/sn/ioc3.c index 501316b198e..ed946311d3a 100644 --- a/drivers/sn/ioc3.c +++ b/drivers/sn/ioc3.c @@ -26,7 +26,7 @@ static DECLARE_RWSEM(ioc3_devices_rwsem); static struct ioc3_submodule *ioc3_submodules[IOC3_MAX_SUBMODULES]; static struct ioc3_submodule *ioc3_ethernet; -static rwlock_t ioc3_submodules_lock = RW_LOCK_UNLOCKED; +static DEFINE_RWLOCK(ioc3_submodules_lock); /* NIC probing code */ diff --git a/drivers/video/backlight/hp680_bl.c b/drivers/video/backlight/hp680_bl.c index a71e984c93d..ffc72ae3ada 100644 --- a/drivers/video/backlight/hp680_bl.c +++ b/drivers/video/backlight/hp680_bl.c @@ -27,7 +27,7 @@ static int hp680bl_suspended; static int current_intensity = 0; -static spinlock_t bl_lock = SPIN_LOCK_UNLOCKED; +static DEFINE_SPINLOCK(bl_lock); static struct backlight_device *hp680_backlight_device; static void hp680bl_send_intensity(struct backlight_device *bd) diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c index 1630b5670dc..7c7d01672d3 100644 --- a/fs/nfsd/nfs4state.c +++ b/fs/nfsd/nfs4state.c @@ -123,7 +123,7 @@ static void release_stateid(struct nfs4_stateid *stp, int flags); */ /* recall_lock protects the del_recall_lru */ -static spinlock_t recall_lock = SPIN_LOCK_UNLOCKED; +static DEFINE_SPINLOCK(recall_lock); static struct list_head del_recall_lru; static void diff --git a/fs/ocfs2/cluster/heartbeat.c b/fs/ocfs2/cluster/heartbeat.c index 21f38accd03..1d26cfcd9f8 100644 --- a/fs/ocfs2/cluster/heartbeat.c +++ b/fs/ocfs2/cluster/heartbeat.c @@ -54,7 +54,7 @@ static DECLARE_RWSEM(o2hb_callback_sem); * multiple hb threads are watching multiple regions. A node is live * whenever any of the threads sees activity from the node in its region. */ -static spinlock_t o2hb_live_lock = SPIN_LOCK_UNLOCKED; +static DEFINE_SPINLOCK(o2hb_live_lock); static struct list_head o2hb_live_slots[O2NM_MAX_NODES]; static unsigned long o2hb_live_node_bitmap[BITS_TO_LONGS(O2NM_MAX_NODES)]; static LIST_HEAD(o2hb_node_events); diff --git a/fs/ocfs2/cluster/tcp.c b/fs/ocfs2/cluster/tcp.c index 0f60cc0d398..1591eb37a72 100644 --- a/fs/ocfs2/cluster/tcp.c +++ b/fs/ocfs2/cluster/tcp.c @@ -108,7 +108,7 @@ ##args); \ } while (0) -static rwlock_t o2net_handler_lock = RW_LOCK_UNLOCKED; +static DEFINE_RWLOCK(o2net_handler_lock); static struct rb_root o2net_handler_tree = RB_ROOT; static struct o2net_node o2net_nodes[O2NM_MAX_NODES]; diff --git a/fs/ocfs2/dlm/dlmdomain.c b/fs/ocfs2/dlm/dlmdomain.c index ba27c5c5e95..b8c23f7ba67 100644 --- a/fs/ocfs2/dlm/dlmdomain.c +++ b/fs/ocfs2/dlm/dlmdomain.c @@ -88,7 +88,7 @@ out_free: * */ -spinlock_t dlm_domain_lock = SPIN_LOCK_UNLOCKED; +DEFINE_SPINLOCK(dlm_domain_lock); LIST_HEAD(dlm_domains); static DECLARE_WAIT_QUEUE_HEAD(dlm_domain_events); diff --git a/fs/ocfs2/dlm/dlmlock.c b/fs/ocfs2/dlm/dlmlock.c index d6f89577e25..5ca57ec650c 100644 --- a/fs/ocfs2/dlm/dlmlock.c +++ b/fs/ocfs2/dlm/dlmlock.c @@ -53,7 +53,7 @@ #define MLOG_MASK_PREFIX ML_DLM #include "cluster/masklog.h" -static spinlock_t dlm_cookie_lock = SPIN_LOCK_UNLOCKED; +static DEFINE_SPINLOCK(dlm_cookie_lock); static u64 dlm_next_cookie = 1; static enum dlm_status dlm_send_remote_lock_request(struct dlm_ctxt *dlm, diff --git a/fs/ocfs2/dlm/dlmrecovery.c b/fs/ocfs2/dlm/dlmrecovery.c index da399013516..29b2845f370 100644 --- a/fs/ocfs2/dlm/dlmrecovery.c +++ b/fs/ocfs2/dlm/dlmrecovery.c @@ -98,8 +98,8 @@ static void dlm_mig_lockres_worker(struct dlm_work_item *item, void *data); static u64 dlm_get_next_mig_cookie(void); -static spinlock_t dlm_reco_state_lock = SPIN_LOCK_UNLOCKED; -static spinlock_t dlm_mig_cookie_lock = SPIN_LOCK_UNLOCKED; +static DEFINE_SPINLOCK(dlm_reco_state_lock); +static DEFINE_SPINLOCK(dlm_mig_cookie_lock); static u64 dlm_mig_cookie = 1; static u64 dlm_get_next_mig_cookie(void) diff --git a/fs/ocfs2/dlmglue.c b/fs/ocfs2/dlmglue.c index 64cd52860c8..4acd37286bd 100644 --- a/fs/ocfs2/dlmglue.c +++ b/fs/ocfs2/dlmglue.c @@ -242,7 +242,7 @@ static void ocfs2_build_lock_name(enum ocfs2_lock_type type, mlog_exit_void(); } -static spinlock_t ocfs2_dlm_tracking_lock = SPIN_LOCK_UNLOCKED; +static DEFINE_SPINLOCK(ocfs2_dlm_tracking_lock); static void ocfs2_add_lockres_tracking(struct ocfs2_lock_res *res, struct ocfs2_dlm_debug *dlm_debug) diff --git a/fs/ocfs2/journal.c b/fs/ocfs2/journal.c index 3fe8781c22c..910a601b2e9 100644 --- a/fs/ocfs2/journal.c +++ b/fs/ocfs2/journal.c @@ -49,7 +49,7 @@ #include "buffer_head_io.h" -spinlock_t trans_inc_lock = SPIN_LOCK_UNLOCKED; +DEFINE_SPINLOCK(trans_inc_lock); static int ocfs2_force_read_journal(struct inode *inode); static int ocfs2_recover_node(struct ocfs2_super *osb, diff --git a/include/asm-alpha/core_t2.h b/include/asm-alpha/core_t2.h index dba70c62a16..457c34b6eb0 100644 --- a/include/asm-alpha/core_t2.h +++ b/include/asm-alpha/core_t2.h @@ -435,7 +435,7 @@ static inline void t2_outl(u32 b, unsigned long addr) set_hae(msb); \ } -static spinlock_t t2_hae_lock = SPIN_LOCK_UNLOCKED; +static DEFINE_SPINLOCK(t2_hae_lock); __EXTERN_INLINE u8 t2_readb(const volatile void __iomem *xaddr) { diff --git a/kernel/audit.c b/kernel/audit.c index 7dfac7031bd..82443fb433e 100644 --- a/kernel/audit.c +++ b/kernel/audit.c @@ -818,7 +818,7 @@ err: */ unsigned int audit_serial(void) { - static spinlock_t serial_lock = SPIN_LOCK_UNLOCKED; + static DEFINE_SPINLOCK(serial_lock); static unsigned int serial = 0; unsigned long flags; diff --git a/mm/sparse.c b/mm/sparse.c index e0a3fe48aa3..c7a2b3a0e46 100644 --- a/mm/sparse.c +++ b/mm/sparse.c @@ -45,7 +45,7 @@ static struct mem_section *sparse_index_alloc(int nid) static int sparse_index_init(unsigned long section_nr, int nid) { - static spinlock_t index_init_lock = SPIN_LOCK_UNLOCKED; + static DEFINE_SPINLOCK(index_init_lock); unsigned long root = SECTION_NR_TO_ROOT(section_nr); struct mem_section *section; int ret = 0; diff --git a/net/ipv6/route.c b/net/ipv6/route.c index 8a777932786..e728980160d 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -349,7 +349,7 @@ static struct rt6_info *rt6_select(struct rt6_info **head, int oif, (strict & RT6_SELECT_F_REACHABLE) && last && last != rt0) { /* no entries matched; do round-robin */ - static spinlock_t lock = SPIN_LOCK_UNLOCKED; + static DEFINE_SPINLOCK(lock); spin_lock(&lock); *head = rt0->u.next; rt0->u.next = last->u.next; diff --git a/net/sunrpc/auth_gss/gss_krb5_seal.c b/net/sunrpc/auth_gss/gss_krb5_seal.c index f43311221a7..2f312164d6d 100644 --- a/net/sunrpc/auth_gss/gss_krb5_seal.c +++ b/net/sunrpc/auth_gss/gss_krb5_seal.c @@ -70,7 +70,7 @@ # define RPCDBG_FACILITY RPCDBG_AUTH #endif -spinlock_t krb5_seq_lock = SPIN_LOCK_UNLOCKED; +DEFINE_SPINLOCK(krb5_seq_lock); u32 gss_get_mic_kerberos(struct gss_ctx *gss_ctx, struct xdr_buf *text, diff --git a/net/tipc/bcast.c b/net/tipc/bcast.c index 54128040a12..1bb75703f38 100644 --- a/net/tipc/bcast.c +++ b/net/tipc/bcast.c @@ -117,7 +117,7 @@ struct bclink { static struct bcbearer *bcbearer = NULL; static struct bclink *bclink = NULL; static struct link *bcl = NULL; -static spinlock_t bc_lock = SPIN_LOCK_UNLOCKED; +static DEFINE_SPINLOCK(bc_lock); char tipc_bclink_name[] = "multicast-link"; @@ -796,7 +796,7 @@ int tipc_bclink_init(void) memset(bclink, 0, sizeof(struct bclink)); INIT_LIST_HEAD(&bcl->waiting_ports); bcl->next_out_no = 1; - bclink->node.lock = SPIN_LOCK_UNLOCKED; + spin_lock_init(&bclink->node.lock); bcl->owner = &bclink->node; bcl->max_pkt = MAX_PKT_DEFAULT_MCAST; tipc_link_set_queue_limits(bcl, BCLINK_WIN_DEFAULT); diff --git a/net/tipc/bearer.c b/net/tipc/bearer.c index 4fa24b5e891..7ef17a449cf 100644 --- a/net/tipc/bearer.c +++ b/net/tipc/bearer.c @@ -566,7 +566,7 @@ restart: b_ptr->link_req = tipc_disc_init_link_req(b_ptr, &m_ptr->bcast_addr, bcast_scope, 2); } - b_ptr->publ.lock = SPIN_LOCK_UNLOCKED; + spin_lock_init(&b_ptr->publ.lock); write_unlock_bh(&tipc_net_lock); info("Enabled bearer <%s>, discovery domain %s, priority %u\n", name, addr_string_fill(addr_string, bcast_scope), priority); diff --git a/net/tipc/config.c b/net/tipc/config.c index 3ec502fac8c..285e1bc2d88 100644 --- a/net/tipc/config.c +++ b/net/tipc/config.c @@ -63,7 +63,7 @@ struct manager { static struct manager mng = { 0}; -static spinlock_t config_lock = SPIN_LOCK_UNLOCKED; +static DEFINE_SPINLOCK(config_lock); static const void *req_tlv_area; /* request message TLV area */ static int req_tlv_space; /* request message TLV area size */ diff --git a/net/tipc/dbg.c b/net/tipc/dbg.c index 26ef95d5fe3..55130655e1e 100644 --- a/net/tipc/dbg.c +++ b/net/tipc/dbg.c @@ -41,7 +41,7 @@ #define MAX_STRING 512 static char print_string[MAX_STRING]; -static spinlock_t print_lock = SPIN_LOCK_UNLOCKED; +static DEFINE_SPINLOCK(print_lock); static struct print_buf cons_buf = { NULL, 0, NULL, NULL }; struct print_buf *TIPC_CONS = &cons_buf; diff --git a/net/tipc/handler.c b/net/tipc/handler.c index 966f70a1b60..ae6ddf00a1a 100644 --- a/net/tipc/handler.c +++ b/net/tipc/handler.c @@ -44,7 +44,7 @@ struct queue_item { static kmem_cache_t *tipc_queue_item_cache; static struct list_head signal_queue_head; -static spinlock_t qitem_lock = SPIN_LOCK_UNLOCKED; +static DEFINE_SPINLOCK(qitem_lock); static int handler_enabled = 0; static void process_signal_queue(unsigned long dummy); diff --git a/net/tipc/name_table.c b/net/tipc/name_table.c index 38571306aba..a6926ff07bc 100644 --- a/net/tipc/name_table.c +++ b/net/tipc/name_table.c @@ -101,7 +101,7 @@ struct name_table { static struct name_table table = { NULL } ; static atomic_t rsv_publ_ok = ATOMIC_INIT(0); -rwlock_t tipc_nametbl_lock = RW_LOCK_UNLOCKED; +DEFINE_RWLOCK(tipc_nametbl_lock); static int hash(int x) @@ -172,7 +172,7 @@ static struct name_seq *tipc_nameseq_create(u32 type, struct hlist_head *seq_hea } memset(nseq, 0, sizeof(*nseq)); - nseq->lock = SPIN_LOCK_UNLOCKED; + spin_lock_init(&nseq->lock); nseq->type = type; nseq->sseqs = sseq; dbg("tipc_nameseq_create(): nseq = %p, type %u, ssseqs %p, ff: %u\n", diff --git a/net/tipc/net.c b/net/tipc/net.c index f7c8223ddf7..e5a359ab493 100644 --- a/net/tipc/net.c +++ b/net/tipc/net.c @@ -115,7 +115,7 @@ * - A local spin_lock protecting the queue of subscriber events. */ -rwlock_t tipc_net_lock = RW_LOCK_UNLOCKED; +DEFINE_RWLOCK(tipc_net_lock); struct network tipc_net = { NULL }; struct node *tipc_net_select_remote_node(u32 addr, u32 ref) diff --git a/net/tipc/node.c b/net/tipc/node.c index ce9678efa98..861322b935d 100644 --- a/net/tipc/node.c +++ b/net/tipc/node.c @@ -77,7 +77,7 @@ struct node *tipc_node_create(u32 addr) memset(n_ptr, 0, sizeof(*n_ptr)); n_ptr->addr = addr; - n_ptr->lock = SPIN_LOCK_UNLOCKED; + spin_lock_init(&n_ptr->lock); INIT_LIST_HEAD(&n_ptr->nsub); n_ptr->owner = c_ptr; tipc_cltr_attach_node(c_ptr, n_ptr); diff --git a/net/tipc/port.c b/net/tipc/port.c index 47d97404e3e..3251c8d8e53 100644 --- a/net/tipc/port.c +++ b/net/tipc/port.c @@ -57,8 +57,8 @@ static struct sk_buff *msg_queue_head = NULL; static struct sk_buff *msg_queue_tail = NULL; -spinlock_t tipc_port_list_lock = SPIN_LOCK_UNLOCKED; -static spinlock_t queue_lock = SPIN_LOCK_UNLOCKED; +DEFINE_SPINLOCK(tipc_port_list_lock); +static DEFINE_SPINLOCK(queue_lock); static LIST_HEAD(ports); static void port_handle_node_down(unsigned long ref); diff --git a/net/tipc/ref.c b/net/tipc/ref.c index d2f0cce10e2..596d3c8ff75 100644 --- a/net/tipc/ref.c +++ b/net/tipc/ref.c @@ -63,7 +63,7 @@ struct ref_table tipc_ref_table = { NULL }; -static rwlock_t ref_table_lock = RW_LOCK_UNLOCKED; +static DEFINE_RWLOCK(ref_table_lock); /** * tipc_ref_table_init - create reference table for objects @@ -87,7 +87,7 @@ int tipc_ref_table_init(u32 requested_size, u32 start) index_mask = sz - 1; for (i = sz - 1; i >= 0; i--) { table[i].object = NULL; - table[i].lock = SPIN_LOCK_UNLOCKED; + spin_lock_init(&table[i].lock); table[i].data.next_plus_upper = (start & ~index_mask) + i - 1; } tipc_ref_table.entries = table; diff --git a/net/tipc/subscr.c b/net/tipc/subscr.c index fc171875660..e19b4bcd67e 100644 --- a/net/tipc/subscr.c +++ b/net/tipc/subscr.c @@ -457,7 +457,7 @@ int tipc_subscr_start(void) int res = -1; memset(&topsrv, 0, sizeof (topsrv)); - topsrv.lock = SPIN_LOCK_UNLOCKED; + spin_lock_init(&topsrv.lock); INIT_LIST_HEAD(&topsrv.subscriber_list); spin_lock_bh(&topsrv.lock); diff --git a/net/tipc/user_reg.c b/net/tipc/user_reg.c index 3f3f933976e..1e3ae57c722 100644 --- a/net/tipc/user_reg.c +++ b/net/tipc/user_reg.c @@ -67,7 +67,7 @@ struct tipc_user { static struct tipc_user *users = NULL; static u32 next_free_user = MAX_USERID + 1; -static spinlock_t reg_lock = SPIN_LOCK_UNLOCKED; +static DEFINE_SPINLOCK(reg_lock); /** * reg_init - create TIPC user registry (but don't activate it) -- cgit v1.2.3-70-g09d2 From 1c0f16e5cdff59f3b132a1b0c0d44a941f8813d2 Mon Sep 17 00:00:00 2001 From: David Woodhouse Date: Tue, 27 Jun 2006 02:53:56 -0700 Subject: [PATCH] Remove gratuitous inclusion of from We include config.h on the compiler command line. There's no need for it to be included again. Signed-off-by: David Woodhouse Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/dmaengine.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include') diff --git a/include/linux/dmaengine.h b/include/linux/dmaengine.h index 78b236ca04f..272010a6078 100644 --- a/include/linux/dmaengine.h +++ b/include/linux/dmaengine.h @@ -20,7 +20,7 @@ */ #ifndef DMAENGINE_H #define DMAENGINE_H -#include + #ifdef CONFIG_DMA_ENGINE #include -- cgit v1.2.3-70-g09d2 From c32e066057fe0914da262c94e52cefb142f965b4 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Tue, 27 Jun 2006 02:54:04 -0700 Subject: [PATCH] rcutorture: add call_rcu_bh() operations Add operations for the call_rcu_bh() variant of RCU. Also add an rcu_batches_completed_bh() function, which is needed by rcutorture. Signed-off-by: Paul E. McKenney Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/rcupdate.h | 1 + kernel/rcupdate.c | 10 ++++++++++ kernel/rcutorture.c | 40 ++++++++++++++++++++++++++++++++++++++-- 3 files changed, 49 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/include/linux/rcupdate.h b/include/linux/rcupdate.h index 6312758393b..48dfe00070c 100644 --- a/include/linux/rcupdate.h +++ b/include/linux/rcupdate.h @@ -258,6 +258,7 @@ extern void rcu_init(void); extern void rcu_check_callbacks(int cpu, int user); extern void rcu_restart_cpu(int cpu); extern long rcu_batches_completed(void); +extern long rcu_batches_completed_bh(void); /* Exported interfaces */ extern void FASTCALL(call_rcu(struct rcu_head *head, diff --git a/kernel/rcupdate.c b/kernel/rcupdate.c index 20e9710fc21..c0e1cb95dd4 100644 --- a/kernel/rcupdate.c +++ b/kernel/rcupdate.c @@ -182,6 +182,15 @@ long rcu_batches_completed(void) return rcu_ctrlblk.completed; } +/* + * Return the number of RCU batches processed thus far. Useful + * for debug and statistics. + */ +long rcu_batches_completed_bh(void) +{ + return rcu_bh_ctrlblk.completed; +} + static void rcu_barrier_callback(struct rcu_head *notused) { if (atomic_dec_and_test(&rcu_barrier_cpu_count)) @@ -619,6 +628,7 @@ module_param(qlowmark, int, 0); module_param(rsinterval, int, 0); #endif EXPORT_SYMBOL_GPL(rcu_batches_completed); +EXPORT_SYMBOL_GPL(rcu_batches_completed_bh); EXPORT_SYMBOL_GPL(call_rcu); EXPORT_SYMBOL_GPL(call_rcu_bh); EXPORT_SYMBOL_GPL(synchronize_rcu); diff --git a/kernel/rcutorture.c b/kernel/rcutorture.c index c96b5edd6ed..4d1c3d24712 100644 --- a/kernel/rcutorture.c +++ b/kernel/rcutorture.c @@ -66,7 +66,7 @@ MODULE_PARM_DESC(test_no_idle_hz, "Test support for tickless idle CPUs"); module_param(shuffle_interval, int, 0); MODULE_PARM_DESC(shuffle_interval, "Number of seconds between shuffles"); module_param(torture_type, charp, 0); -MODULE_PARM_DESC(torture_type, "Type of RCU to torture (rcu)"); +MODULE_PARM_DESC(torture_type, "Type of RCU to torture (rcu, rcu_bh)"); #define TORTURE_FLAG "-torture:" #define PRINTK_STRING(s) \ @@ -246,8 +246,44 @@ static struct rcu_torture_ops rcu_ops = { .name = "rcu" }; +/* + * Definitions for rcu_bh torture testing. + */ + +static int rcu_bh_torture_read_lock(void) +{ + rcu_read_lock_bh(); + return 0; +} + +static void rcu_bh_torture_read_unlock(int idx) +{ + rcu_read_unlock_bh(); +} + +static int rcu_bh_torture_completed(void) +{ + return rcu_batches_completed_bh(); +} + +static void rcu_bh_torture_deferred_free(struct rcu_torture *p) +{ + call_rcu_bh(&p->rtort_rcu, rcu_torture_cb); +} + +static struct rcu_torture_ops rcu_bh_ops = { + .init = NULL, + .cleanup = NULL, + .readlock = rcu_bh_torture_read_lock, + .readunlock = rcu_bh_torture_read_unlock, + .completed = rcu_bh_torture_completed, + .deferredfree = rcu_bh_torture_deferred_free, + .stats = NULL, + .name = "rcu_bh" +}; + static struct rcu_torture_ops *torture_ops[] = - { &rcu_ops, NULL }; + { &rcu_ops, &rcu_bh_ops, NULL }; /* * RCU torture writer kthread. Repeatedly substitutes a new structure -- cgit v1.2.3-70-g09d2 From 65edc68c345cbe21d0b0375c3452a3ed5e322868 Mon Sep 17 00:00:00 2001 From: Chandra Seetharaman Date: Tue, 27 Jun 2006 02:54:08 -0700 Subject: [PATCH] cpu hotplug: make [un]register_cpu_notifier init time only CPUs come online only at init time (unless CONFIG_HOTPLUG_CPU is defined). So, cpu_notifier functionality need to be available only at init time. This patch makes register_cpu_notifier() available only at init time, unless CONFIG_HOTPLUG_CPU is defined. This patch exports register_cpu_notifier() and unregister_cpu_notifier() only if CONFIG_HOTPLUG_CPU is defined. Signed-off-by: Chandra Seetharaman Cc: Ashok Raj Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/cpufreq/cpufreq.c | 8 +++++--- drivers/cpufreq/cpufreq_stats.c | 4 ++-- include/linux/cpu.h | 6 ++++++ kernel/cpu.c | 8 +++++--- 4 files changed, 18 insertions(+), 8 deletions(-) (limited to 'include') diff --git a/drivers/cpufreq/cpufreq.c b/drivers/cpufreq/cpufreq.c index 486ef666470..3533e26f837 100644 --- a/drivers/cpufreq/cpufreq.c +++ b/drivers/cpufreq/cpufreq.c @@ -1497,7 +1497,8 @@ int cpufreq_update_policy(unsigned int cpu) } EXPORT_SYMBOL(cpufreq_update_policy); -static int __cpuinit cpufreq_cpu_callback(struct notifier_block *nfb, +#ifdef CONFIG_HOTPLUG_CPU +static int cpufreq_cpu_callback(struct notifier_block *nfb, unsigned long action, void *hcpu) { unsigned int cpu = (unsigned long)hcpu; @@ -1536,6 +1537,7 @@ static struct notifier_block cpufreq_cpu_notifier = { .notifier_call = cpufreq_cpu_callback, }; +#endif /* CONFIG_HOTPLUG_CPU */ /********************************************************************* * REGISTER / UNREGISTER CPUFREQ DRIVER * @@ -1596,7 +1598,7 @@ int cpufreq_register_driver(struct cpufreq_driver *driver_data) } if (!ret) { - register_cpu_notifier(&cpufreq_cpu_notifier); + register_hotcpu_notifier(&cpufreq_cpu_notifier); dprintk("driver %s up and running\n", driver_data->name); cpufreq_debug_enable_ratelimit(); } @@ -1628,7 +1630,7 @@ int cpufreq_unregister_driver(struct cpufreq_driver *driver) dprintk("unregistering driver %s\n", driver->name); sysdev_driver_unregister(&cpu_sysdev_class, &cpufreq_sysdev_driver); - unregister_cpu_notifier(&cpufreq_cpu_notifier); + unregister_hotcpu_notifier(&cpufreq_cpu_notifier); spin_lock_irqsave(&cpufreq_driver_lock, flags); cpufreq_driver = NULL; diff --git a/drivers/cpufreq/cpufreq_stats.c b/drivers/cpufreq/cpufreq_stats.c index c576c0b3f45..145061b8472 100644 --- a/drivers/cpufreq/cpufreq_stats.c +++ b/drivers/cpufreq/cpufreq_stats.c @@ -350,7 +350,7 @@ __init cpufreq_stats_init(void) return ret; } - register_cpu_notifier(&cpufreq_stat_cpu_notifier); + register_hotcpu_notifier(&cpufreq_stat_cpu_notifier); lock_cpu_hotplug(); for_each_online_cpu(cpu) { cpufreq_stat_cpu_callback(&cpufreq_stat_cpu_notifier, CPU_ONLINE, @@ -368,7 +368,7 @@ __exit cpufreq_stats_exit(void) CPUFREQ_POLICY_NOTIFIER); cpufreq_unregister_notifier(¬ifier_trans_block, CPUFREQ_TRANSITION_NOTIFIER); - unregister_cpu_notifier(&cpufreq_stat_cpu_notifier); + unregister_hotcpu_notifier(&cpufreq_stat_cpu_notifier); lock_cpu_hotplug(); for_each_online_cpu(cpu) { cpufreq_stat_cpu_callback(&cpufreq_stat_cpu_notifier, CPU_DEAD, diff --git a/include/linux/cpu.h b/include/linux/cpu.h index b23bf1c8add..cdfe471a70a 100644 --- a/include/linux/cpu.h +++ b/include/linux/cpu.h @@ -41,7 +41,13 @@ struct notifier_block; #ifdef CONFIG_SMP /* Need to know about CPUs going up/down? */ extern int register_cpu_notifier(struct notifier_block *nb); +#ifdef CONFIG_HOTPLUG_CPU extern void unregister_cpu_notifier(struct notifier_block *nb); +#else +static inline void unregister_cpu_notifier(struct notifier_block *nb) +{ +} +#endif extern int current_in_cpu_hotplug(void); int cpu_up(unsigned int cpu); diff --git a/kernel/cpu.c b/kernel/cpu.c index 03dcd981846..70fbf2e8376 100644 --- a/kernel/cpu.c +++ b/kernel/cpu.c @@ -18,7 +18,7 @@ /* This protects CPUs going up and down... */ static DEFINE_MUTEX(cpucontrol); -static BLOCKING_NOTIFIER_HEAD(cpu_chain); +static __cpuinitdata BLOCKING_NOTIFIER_HEAD(cpu_chain); #ifdef CONFIG_HOTPLUG_CPU static struct task_struct *lock_cpu_hotplug_owner; @@ -69,10 +69,13 @@ EXPORT_SYMBOL_GPL(lock_cpu_hotplug_interruptible); #endif /* CONFIG_HOTPLUG_CPU */ /* Need to know about CPUs going up/down? */ -int register_cpu_notifier(struct notifier_block *nb) +int __cpuinit register_cpu_notifier(struct notifier_block *nb) { return blocking_notifier_chain_register(&cpu_chain, nb); } + +#ifdef CONFIG_HOTPLUG_CPU + EXPORT_SYMBOL(register_cpu_notifier); void unregister_cpu_notifier(struct notifier_block *nb) @@ -81,7 +84,6 @@ void unregister_cpu_notifier(struct notifier_block *nb) } EXPORT_SYMBOL(unregister_cpu_notifier); -#ifdef CONFIG_HOTPLUG_CPU static inline void check_for_tasks(int cpu) { struct task_struct *p; -- cgit v1.2.3-70-g09d2 From 39f4885c56073ecafd482d7e10dd7b17900fa312 Mon Sep 17 00:00:00 2001 From: Chandra Seetharaman Date: Tue, 27 Jun 2006 02:54:10 -0700 Subject: [PATCH] cpu hotplug: add hotplug versions of cpu_notifier Define new macros register_hotcpu_notifier() and unregister_hotcpu_notifier() that redefines register_cpu_notifier() and unregister_cpu_notifier() for use only when HOTPLUG_CPU is defined. Signed-off-by: Chandra Seetharaman Cc: Ashok Raj Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/cpu.h | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'include') diff --git a/include/linux/cpu.h b/include/linux/cpu.h index cdfe471a70a..a3caf6866ba 100644 --- a/include/linux/cpu.h +++ b/include/linux/cpu.h @@ -79,6 +79,8 @@ extern int lock_cpu_hotplug_interruptible(void); { .notifier_call = fn, .priority = pri }; \ register_cpu_notifier(&fn##_nb); \ } +#define register_hotcpu_notifier(nb) register_cpu_notifier(nb) +#define unregister_hotcpu_notifier(nb) unregister_cpu_notifier(nb) int cpu_down(unsigned int cpu); #define cpu_is_offline(cpu) unlikely(!cpu_online(cpu)) #else @@ -86,6 +88,8 @@ int cpu_down(unsigned int cpu); #define unlock_cpu_hotplug() do { } while (0) #define lock_cpu_hotplug_interruptible() 0 #define hotcpu_notifier(fn, pri) +#define register_hotcpu_notifier(nb) +#define unregister_hotcpu_notifier(nb) /* CPUs don't go offline once they're online w/o CONFIG_HOTPLUG_CPU */ static inline int cpu_is_offline(int cpu) { return 0; } -- cgit v1.2.3-70-g09d2 From 62c83cde9282a9580994a12b3063e677181b5ebe Mon Sep 17 00:00:00 2001 From: Jim Cromie Date: Tue, 27 Jun 2006 02:54:13 -0700 Subject: [PATCH] chardev: GPIO for SCx200 & PC-8736x: whitespace pre-clean GPIO SUPPORT FOR SCx200 & PC8736x The patch-set reworks the 2.4 vintage scx200_gpio driver for modern 2.6, and refactors GPIO support to reuse it in a new driver for the GPIO on PC-8736x chips. Its handy for the Soekris.com net-4801, which has both chips. These patches have been seen recently on Kernel-Mentors, and then Kernel-Newbies ML, where Jesper Juhl kindly reviewed it. His feedback has been incorporated. Thanks Jesper ! Its also gone to soekris-tech@soekris.com for possible testing by linux folks, I've gotten 1 promise so far. Theyre mostly BSD folk over there, but we'll see.. Device-file & Sysfs The driver preserves the existing device-file interface, including the write/cmd set, but adds v to 'view' the pin-settings & configs by inducing, via gpio_dump(), a dev_info() call. Its a fairly crappy way to get status, but it sticks to the syslog approach, conservatively. Allowing users to voluntarily trigger logging is good, it gives them a familiar way to confirm their app's control & use of the pins, and I've thus reduced the pin-mode-updates from dev_info to dev_dbg. I've recently bolted on a proto sysfs interface for both new drivers. Im not including those patches here; they (the patch + doc-pre-patch) are still quite raw (and unreviewed on KNML), and since they 'invent' a convention for GPIO, a proper vetting is needed. Since this patchset is much bigger than my previous ones, Id like to keep things simpler, and address it 1st, before bolting on more stuff. The driver-split The Geode CPU and the PC-87366 Super-IO chip have GPIO units which share a common pin-architecture (same pin features, with same bits controlling), but with different addressing mechanics and port organizations. The vintage driver expresses the pin capabilities with pin-mode commands [OoPpTt],etc that change the pin configurations, and since the 2 chips share pin-arch, we can reuse the read(), write() commands, once the implementation is suitably adjusted. The patchset adds a vtable: struct nsc_gpio_ops, to abstract the existing gpio operations, then adjusts fileops.write() code to invoke operations via that vtable. Driver specific open()s set private_data to the vtable so its available for use by write(). The vtable gets the gpio_dump() too, since its user-friendly, and (could be construed as) part of the current device-file interface. To support use of dev_dbg() in write() & _dump(), the vtable gets a dev ptr too, set by both scx200 & pc8736x _gpio drivers. heres how the pins are presented in syslog: [ 1890.176223] scx200_gpio.0: io00: 0x0044 TS OD PUE EDGE LO DEBOUNCE [ 1890.287223] scx200_gpio.0: io01: 0x0003 OE PP PUD EDGE LO nsc_gpio.c: new file is new home of several file-ops methods, which are modified to get their vtable from filp->private_data, and use it where needed. scx200_gpio.c: keeps some of its existing gpio routines, but now wires them up via the vtable (they're invoked by nsc_gpio.c:nsc_gpio_write() thru this vtable). A driver-spcific open() initializes filp->private_data with the vtable. Once the split is clean, and the scx200_gpio driver is working, we copy and modify the function and variable names, and rework the access-method bodies for the different addressing scheme. Heres a working overview of the patchset: # series file for GPIO # Spring Cleaning gpio-scx/patch.preclean # scripts/Lindent fixes, editor-ctrl comments # API Modernization gpio-scx/patch.api26 # what I learned from LDD3 gpio-scx/patch.platform-dev-2 # get pdev, support for dev_dbg() gpio-scx/patch.unsigned-minor # fix to match std practice # Debuggability gpio-scx/patch.dump-diet # shrink gpio_dump() gpio-scx/patch.viewpins # add new 'command' to call dump() gpio-scx/patch.init-refactor # pull shadow-register init to sub # Access-Abstraction (add vtable) gpio-scx/patch.access-vtable # introduce nsg_gpio_ops vtable, w dump gpio-scx/patch.vtable-calls # add & use the vtable in scx200_gpio gpio-scx/patch.nscgpio-shell # add empty driver for common-fops # move code under abstraction gpio-scx/patch.migrate-fops # move file-ops methods from scx200_gpio gpio-scx/patch.common-dump # mv scx200.c:scx200_gpio_dump() to nsc_gpio.c gpio-scx/patch.add-pc8736x-gpio # add new driver, like old, w chip adapt # gpio-scx/patch.add-DEBUG # enable all dev_dbg()s # Cleanups # finish printk -> dev_dbg() etc gpio-scx/patch.pdev-pc8736x # new drvr needs pdev too, gpio-scx/patch.devdbg-nscgpio # add device to 'vtable', use in dev_dbg() # gpio-scx/patch.pin-config-view # another 'c' 'command' # gpio-scx/quiet-getset # take out excess dbg stuff (pretty quiet now) gpio-scx/patch.shadow-current # imitate scx200_gpio's shadow regs in pc87* # post KMentors-post patches .. gpio-scx/patch.mutexes # use mutexes for config-locks gpio-scx/patch.viewpins-values # extend dump to obsolete separate 'c' cmd gpio-scx/patch.kconfig # add stuff for kbuild # TBC # combine api26 with pdev, which is just one step. # merge c&v commands to single do-all-fn # delay viewpins, dump-diet should also un-ifdef it too. diff.sys-gpio-rollup-1 This patch: Removed editor format-control comments, and used scripts/Lindent to clean up whitespace, then deleted the bogus chunks :-( Signed-off-by: Jim Cromie Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/i386/kernel/scx200.c | 7 ------- drivers/char/scx200_gpio.c | 26 +++++++++----------------- include/linux/scx200.h | 7 ------- include/linux/scx200_gpio.h | 7 ------- 4 files changed, 9 insertions(+), 38 deletions(-) (limited to 'include') diff --git a/arch/i386/kernel/scx200.c b/arch/i386/kernel/scx200.c index 321f5fd26e7..4d3d8e811fe 100644 --- a/arch/i386/kernel/scx200.c +++ b/arch/i386/kernel/scx200.c @@ -159,10 +159,3 @@ EXPORT_SYMBOL(scx200_gpio_base); EXPORT_SYMBOL(scx200_gpio_shadow); EXPORT_SYMBOL(scx200_gpio_configure); EXPORT_SYMBOL(scx200_cb_base); - -/* - Local variables: - compile-command: "make -k -C ../../.. SUBDIRS=arch/i386/kernel modules" - c-basic-offset: 8 - End: -*/ diff --git a/drivers/char/scx200_gpio.c b/drivers/char/scx200_gpio.c index 664a6e97eb1..dd051ea6755 100644 --- a/drivers/char/scx200_gpio.c +++ b/drivers/char/scx200_gpio.c @@ -1,4 +1,4 @@ -/* linux/drivers/char/scx200_gpio.c +/* linux/drivers/char/scx200_gpio.c National Semiconductor SCx200 GPIO driver. Allows a user space process to play with the GPIO pins. @@ -26,7 +26,7 @@ static int major = 0; /* default to dynamic major */ module_param(major, int, 0); MODULE_PARM_DESC(major, "Major device number"); -static ssize_t scx200_gpio_write(struct file *file, const char __user *data, +static ssize_t scx200_gpio_write(struct file *file, const char __user *data, size_t len, loff_t *ppos) { unsigned m = iminor(file->f_dentry->d_inode); @@ -34,15 +34,14 @@ static ssize_t scx200_gpio_write(struct file *file, const char __user *data, for (i = 0; i < len; ++i) { char c; - if (get_user(c, data+i)) + if (get_user(c, data + i)) return -EFAULT; - switch (c) - { - case '0': - scx200_gpio_set(m, 0); + switch (c) { + case '0': + scx200_gpio_set(m, 0); break; - case '1': - scx200_gpio_set(m, 1); + case '1': + scx200_gpio_set(m, 1); break; case 'O': printk(KERN_INFO NAME ": GPIO%d output enabled\n", m); @@ -83,7 +82,7 @@ static ssize_t scx200_gpio_read(struct file *file, char __user *buf, value = scx200_gpio_get(m); if (put_user(value ? '1' : '0', buf)) return -EFAULT; - + return 1; } @@ -140,10 +139,3 @@ static void __exit scx200_gpio_cleanup(void) module_init(scx200_gpio_init); module_exit(scx200_gpio_cleanup); - -/* - Local variables: - compile-command: "make -k -C ../.. SUBDIRS=drivers/char modules" - c-basic-offset: 8 - End: -*/ diff --git a/include/linux/scx200.h b/include/linux/scx200.h index a22f9e173ad..693c0557e70 100644 --- a/include/linux/scx200.h +++ b/include/linux/scx200.h @@ -49,10 +49,3 @@ extern unsigned scx200_cb_base; #define SCx200_REV 0x3d /* Revision Register */ #define SCx200_CBA 0x3e /* Configuration Base Address Register */ #define SCx200_CBA_SCRATCH 0x64 /* Configuration Base Address Scratchpad */ - -/* - Local variables: - compile-command: "make -C ../.. bzImage modules" - c-basic-offset: 8 - End: -*/ diff --git a/include/linux/scx200_gpio.h b/include/linux/scx200_gpio.h index 30cdd648ba7..83ab8a1521f 100644 --- a/include/linux/scx200_gpio.h +++ b/include/linux/scx200_gpio.h @@ -87,10 +87,3 @@ static inline void scx200_gpio_change(int index) { #undef __SCx200_GPIO_SHADOW #undef __SCx200_GPIO_INDEX #undef __SCx200_GPIO_OUT - -/* - Local variables: - compile-command: "make -C ../.. bzImage modules" - c-basic-offset: 8 - End: -*/ -- cgit v1.2.3-70-g09d2 From 55b8c0455b8aeb80f94183fa3aa42e3fa62b1705 Mon Sep 17 00:00:00 2001 From: Jim Cromie Date: Tue, 27 Jun 2006 02:54:15 -0700 Subject: [PATCH] chardev: GPIO for SCx200 & PC-8736x: device minor numbers are unsigned ints Per kernel headers, device minor numbers are unsigned ints. Do the same in this driver. Signed-off-by: Jim Cromie Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/i386/kernel/scx200.c | 2 +- include/linux/scx200_gpio.h | 14 +++++++------- 2 files changed, 8 insertions(+), 8 deletions(-) (limited to 'include') diff --git a/arch/i386/kernel/scx200.c b/arch/i386/kernel/scx200.c index 4d3d8e811fe..18f895ce248 100644 --- a/arch/i386/kernel/scx200.c +++ b/arch/i386/kernel/scx200.c @@ -87,7 +87,7 @@ static int __devinit scx200_probe(struct pci_dev *pdev, const struct pci_device_ return 0; } -u32 scx200_gpio_configure(int index, u32 mask, u32 bits) +u32 scx200_gpio_configure(unsigned index, u32 mask, u32 bits) { u32 config, new_config; unsigned long flags; diff --git a/include/linux/scx200_gpio.h b/include/linux/scx200_gpio.h index 83ab8a1521f..90dd069cc14 100644 --- a/include/linux/scx200_gpio.h +++ b/include/linux/scx200_gpio.h @@ -1,6 +1,6 @@ #include -u32 scx200_gpio_configure(int index, u32 set, u32 clear); +u32 scx200_gpio_configure(unsigned index, u32 set, u32 clear); extern unsigned scx200_gpio_base; extern long scx200_gpio_shadow[2]; @@ -17,7 +17,7 @@ extern long scx200_gpio_shadow[2]; /* returns the value of the GPIO pin */ -static inline int scx200_gpio_get(int index) { +static inline int scx200_gpio_get(unsigned index) { __SCx200_GPIO_BANK; __SCx200_GPIO_IOADDR + 0x04; __SCx200_GPIO_INDEX; @@ -29,7 +29,7 @@ static inline int scx200_gpio_get(int index) { driven if the GPIO is configured as an output, it might not be the state of the GPIO right now if the GPIO is configured as an input) */ -static inline int scx200_gpio_current(int index) { +static inline int scx200_gpio_current(unsigned index) { __SCx200_GPIO_BANK; __SCx200_GPIO_INDEX; @@ -38,7 +38,7 @@ static inline int scx200_gpio_current(int index) { /* drive the GPIO signal high */ -static inline void scx200_gpio_set_high(int index) { +static inline void scx200_gpio_set_high(unsigned index) { __SCx200_GPIO_BANK; __SCx200_GPIO_IOADDR; __SCx200_GPIO_SHADOW; @@ -49,7 +49,7 @@ static inline void scx200_gpio_set_high(int index) { /* drive the GPIO signal low */ -static inline void scx200_gpio_set_low(int index) { +static inline void scx200_gpio_set_low(unsigned index) { __SCx200_GPIO_BANK; __SCx200_GPIO_IOADDR; __SCx200_GPIO_SHADOW; @@ -60,7 +60,7 @@ static inline void scx200_gpio_set_low(int index) { /* drive the GPIO signal to state */ -static inline void scx200_gpio_set(int index, int state) { +static inline void scx200_gpio_set(unsigned index, int state) { __SCx200_GPIO_BANK; __SCx200_GPIO_IOADDR; __SCx200_GPIO_SHADOW; @@ -73,7 +73,7 @@ static inline void scx200_gpio_set(int index, int state) { } /* toggle the GPIO signal */ -static inline void scx200_gpio_change(int index) { +static inline void scx200_gpio_change(unsigned index) { __SCx200_GPIO_BANK; __SCx200_GPIO_IOADDR; __SCx200_GPIO_SHADOW; -- cgit v1.2.3-70-g09d2 From fe3a168a2ce1c93837cdf8fe27a3e67795155f90 Mon Sep 17 00:00:00 2001 From: Jim Cromie Date: Tue, 27 Jun 2006 02:54:18 -0700 Subject: [PATCH] chardev: GPIO for SCx200 & PC-8736x: add gpio-ops vtable Abstract the gpio operations into a new nsc_gpio_ops vtable. Signed-off-by: Jim Cromie Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/char/scx200_gpio.c | 13 +++++++++++++ include/linux/nsc_gpio.h | 33 +++++++++++++++++++++++++++++++++ 2 files changed, 46 insertions(+) create mode 100644 include/linux/nsc_gpio.h (limited to 'include') diff --git a/drivers/char/scx200_gpio.c b/drivers/char/scx200_gpio.c index a1a56c5c8a8..f9994ed0181 100644 --- a/drivers/char/scx200_gpio.c +++ b/drivers/char/scx200_gpio.c @@ -20,6 +20,7 @@ #include #include +#include #define NAME "scx200_gpio" #define DEVNAME NAME @@ -36,6 +37,18 @@ MODULE_PARM_DESC(major, "Major device number"); extern void scx200_gpio_dump(unsigned index); +struct nsc_gpio_ops scx200_access = { + .owner = THIS_MODULE, + .gpio_config = scx200_gpio_configure, + .gpio_dump = scx200_gpio_dump, + .gpio_get = scx200_gpio_get, + .gpio_set = scx200_gpio_set, + .gpio_set_high = scx200_gpio_set_high, + .gpio_set_low = scx200_gpio_set_low, + .gpio_change = scx200_gpio_change, + .gpio_current = scx200_gpio_current +}; + static ssize_t scx200_gpio_write(struct file *file, const char __user *data, size_t len, loff_t *ppos) { diff --git a/include/linux/nsc_gpio.h b/include/linux/nsc_gpio.h new file mode 100644 index 00000000000..3ad8ae9dcb5 --- /dev/null +++ b/include/linux/nsc_gpio.h @@ -0,0 +1,33 @@ +/** + nsc_gpio.c + + National Semiconductor GPIO common access methods. + + struct nsc_gpio_ops abstracts the low-level access + operations for the GPIO units on 2 NSC chip families; the GEODE + integrated CPU, and the PC-8736[03456] integrated PC-peripheral + chips. + + The GPIO units on these chips have the same pin architecture, but + the access methods differ. Thus, scx200_gpio and pc8736x_gpio + implement their own versions of these routines; and use the common + file-operations routines implemented in nsc_gpio module. + + Copyright (c) 2005 Jim Cromie + + NB: this work was tested on the Geode SC-1100 and PC-87366 chips. + NSC sold the GEODE line to AMD, and the PC-8736x line to Winbond. +*/ + +struct nsc_gpio_ops { + struct module* owner; + u32 (*gpio_config) (unsigned iminor, u32 mask, u32 bits); + void (*gpio_dump) (unsigned iminor); + int (*gpio_get) (unsigned iminor); + void (*gpio_set) (unsigned iminor, int state); + void (*gpio_set_high)(unsigned iminor); + void (*gpio_set_low) (unsigned iminor); + void (*gpio_change) (unsigned iminor); + int (*gpio_current) (unsigned iminor); +}; + -- cgit v1.2.3-70-g09d2 From 1a66fdf083bf2b60c4d12feb970bc7373b59e33a Mon Sep 17 00:00:00 2001 From: Jim Cromie Date: Tue, 27 Jun 2006 02:54:20 -0700 Subject: [PATCH] chardev: GPIO for SCx200 & PC-8736x: migrate file-ops to common module Now that the read(), write() file-ops are dispatching gpio-ops via the vtable, they are generic, and can be moved 'verbatim' to the nsc_gpio common-support module. After the move, various symbols are renamed to update 'scx200_' to 'nsc_', and headers are adjusted accordingly. Signed-off-by: Jim Cromie Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/char/nsc_gpio.c | 97 +++++++++++++++++++++++++++++++++++++++++----- drivers/char/scx200_gpio.c | 88 ++++------------------------------------- include/linux/nsc_gpio.h | 5 +++ 3 files changed, 100 insertions(+), 90 deletions(-) (limited to 'include') diff --git a/drivers/char/nsc_gpio.c b/drivers/char/nsc_gpio.c index 42d95904967..3842c272711 100644 --- a/drivers/char/nsc_gpio.c +++ b/drivers/char/nsc_gpio.c @@ -19,9 +19,89 @@ #define NAME "nsc_gpio" -MODULE_AUTHOR("Jim Cromie "); -MODULE_DESCRIPTION("NatSemi GPIO Common Methods"); -MODULE_LICENSE("GPL"); +ssize_t nsc_gpio_write(struct file *file, const char __user *data, + size_t len, loff_t *ppos) +{ + unsigned m = iminor(file->f_dentry->d_inode); + struct nsc_gpio_ops *amp = file->private_data; + size_t i; + int err = 0; + + for (i = 0; i < len; ++i) { + char c; + if (get_user(c, data + i)) + return -EFAULT; + switch (c) { + case '0': + amp->gpio_set(m, 0); + break; + case '1': + amp->gpio_set(m, 1); + break; + case 'O': + printk(KERN_INFO NAME ": GPIO%d output enabled\n", m); + amp->gpio_config(m, ~1, 1); + break; + case 'o': + printk(KERN_INFO NAME ": GPIO%d output disabled\n", m); + amp->gpio_config(m, ~1, 0); + break; + case 'T': + printk(KERN_INFO NAME ": GPIO%d output is push pull\n", + m); + amp->gpio_config(m, ~2, 2); + break; + case 't': + printk(KERN_INFO NAME ": GPIO%d output is open drain\n", + m); + amp->gpio_config(m, ~2, 0); + break; + case 'P': + printk(KERN_INFO NAME ": GPIO%d pull up enabled\n", m); + amp->gpio_config(m, ~4, 4); + break; + case 'p': + printk(KERN_INFO NAME ": GPIO%d pull up disabled\n", m); + amp->gpio_config(m, ~4, 0); + break; + + case 'v': + /* View Current pin settings */ + amp->gpio_dump(m); + break; + case '\n': + /* end of settings string, do nothing */ + break; + default: + printk(KERN_ERR NAME + ": GPIO-%2d bad setting: chr<0x%2x>\n", m, + (int)c); + err++; + } + } + if (err) + return -EINVAL; /* full string handled, report error */ + + return len; +} + +ssize_t nsc_gpio_read(struct file *file, char __user * buf, + size_t len, loff_t * ppos) +{ + unsigned m = iminor(file->f_dentry->d_inode); + int value; + struct nsc_gpio_ops *amp = file->private_data; + + value = amp->gpio_get(m); + if (put_user(value ? '1' : '0', buf)) + return -EFAULT; + + return 1; +} + +/* common routines for both scx200_gpio and pc87360_gpio */ +EXPORT_SYMBOL(nsc_gpio_write); +EXPORT_SYMBOL(nsc_gpio_read); static int __init nsc_gpio_init(void) { @@ -34,12 +114,9 @@ static void __exit nsc_gpio_cleanup(void) printk(KERN_DEBUG NAME " cleanup\n"); } -/* prepare for - common routines for both scx200_gpio and pc87360_gpio -EXPORT_SYMBOL(scx200_gpio_write); -EXPORT_SYMBOL(scx200_gpio_read); -EXPORT_SYMBOL(scx200_gpio_release); -*/ - module_init(nsc_gpio_init); module_exit(nsc_gpio_cleanup); + +MODULE_AUTHOR("Jim Cromie "); +MODULE_DESCRIPTION("NatSemi GPIO Common Methods"); +MODULE_LICENSE("GPL"); diff --git a/drivers/char/scx200_gpio.c b/drivers/char/scx200_gpio.c index 15dfb95ebc7..eb9a8477759 100644 --- a/drivers/char/scx200_gpio.c +++ b/drivers/char/scx200_gpio.c @@ -37,6 +37,12 @@ MODULE_PARM_DESC(major, "Major device number"); extern void scx200_gpio_dump(unsigned index); +extern ssize_t nsc_gpio_write(struct file *file, const char __user *data, + size_t len, loff_t *ppos); + +extern ssize_t nsc_gpio_read(struct file *file, char __user *buf, + size_t len, loff_t *ppos); + struct nsc_gpio_ops scx200_access = { .owner = THIS_MODULE, .gpio_config = scx200_gpio_configure, @@ -49,84 +55,6 @@ struct nsc_gpio_ops scx200_access = { .gpio_current = scx200_gpio_current }; -static ssize_t scx200_gpio_write(struct file *file, const char __user *data, - size_t len, loff_t *ppos) -{ - unsigned m = iminor(file->f_dentry->d_inode); - struct nsc_gpio_ops *amp = file->private_data; - size_t i; - int err = 0; - - for (i = 0; i < len; ++i) { - char c; - if (get_user(c, data + i)) - return -EFAULT; - switch (c) { - case '0': - amp->gpio_set(m, 0); - break; - case '1': - amp->gpio_set(m, 1); - break; - case 'O': - printk(KERN_INFO NAME ": GPIO%d output enabled\n", m); - amp->gpio_config(m, ~1, 1); - break; - case 'o': - printk(KERN_INFO NAME ": GPIO%d output disabled\n", m); - amp->gpio_config(m, ~1, 0); - break; - case 'T': - printk(KERN_INFO NAME ": GPIO%d output is push pull\n", m); - amp->gpio_config(m, ~2, 2); - break; - case 't': - printk(KERN_INFO NAME ": GPIO%d output is open drain\n", m); - amp->gpio_config(m, ~2, 0); - break; - case 'P': - printk(KERN_INFO NAME ": GPIO%d pull up enabled\n", m); - amp->gpio_config(m, ~4, 4); - break; - case 'p': - printk(KERN_INFO NAME ": GPIO%d pull up disabled\n", m); - amp->gpio_config(m, ~4, 0); - break; - - case 'v': - /* View Current pin settings */ - amp->gpio_dump(m); - break; - case '\n': - /* end of settings string, do nothing */ - break; - default: - printk(KERN_ERR NAME - ": GPIO-%2d bad setting: chr<0x%2x>\n", m, - (int)c); - err++; - } - } - if (err) - return -EINVAL; /* full string handled, report error */ - - return len; -} - -static ssize_t scx200_gpio_read(struct file *file, char __user *buf, - size_t len, loff_t *ppos) -{ - unsigned m = iminor(file->f_dentry->d_inode); - int value; - struct nsc_gpio_ops *amp = file->private_data; - - value = amp->gpio_get(m); - if (put_user(value ? '1' : '0', buf)) - return -EFAULT; - - return 1; -} - static int scx200_gpio_open(struct inode *inode, struct file *file) { unsigned m = iminor(inode); @@ -145,8 +73,8 @@ static int scx200_gpio_release(struct inode *inode, struct file *file) static struct file_operations scx200_gpio_fops = { .owner = THIS_MODULE, - .write = scx200_gpio_write, - .read = scx200_gpio_read, + .write = nsc_gpio_write, + .read = nsc_gpio_read, .open = scx200_gpio_open, .release = scx200_gpio_release, }; diff --git a/include/linux/nsc_gpio.h b/include/linux/nsc_gpio.h index 3ad8ae9dcb5..27bf66f7386 100644 --- a/include/linux/nsc_gpio.h +++ b/include/linux/nsc_gpio.h @@ -31,3 +31,8 @@ struct nsc_gpio_ops { int (*gpio_current) (unsigned iminor); }; +extern ssize_t nsc_gpio_write(struct file *file, const char __user *data, + size_t len, loff_t *ppos); + +extern ssize_t nsc_gpio_read(struct file *file, char __user *buf, + size_t len, loff_t *ppos); -- cgit v1.2.3-70-g09d2 From f31000e573da052b6b8bcc21faff520b4e2eda7a Mon Sep 17 00:00:00 2001 From: Jim Cromie Date: Tue, 27 Jun 2006 02:54:23 -0700 Subject: [PATCH] chardev: GPIO for SCx200 & PC-8736x: use dev_dbg in common module Use of dev_dbg() and friends is considered good practice. dev_dbg() needs a struct device *devp, but nsc_gpio is only a helper module, so it doesnt have/need its own. To provide devp to the user-modules (scx200 & pc8736x _gpio), we add it to the vtable, and set it during init. Also squeeze nsc_gpio_dump()'s format a little. [ 199.259879] pc8736x_gpio.0: io09: 0x0044 TS OD PUE EDGE LO DEBOUNCE Signed-off-by: Jim Cromie Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/char/nsc_gpio.c | 45 +++++++++++++++++++++++++-------------------- drivers/char/pc8736x_gpio.c | 3 +-- drivers/char/scx200_gpio.c | 11 +++-------- include/linux/nsc_gpio.h | 6 +++++- 4 files changed, 34 insertions(+), 31 deletions(-) (limited to 'include') diff --git a/drivers/char/nsc_gpio.c b/drivers/char/nsc_gpio.c index 929d68486c1..d0b5d65a73f 100644 --- a/drivers/char/nsc_gpio.c +++ b/drivers/char/nsc_gpio.c @@ -14,22 +14,27 @@ #include #include #include +#include #include #include #define NAME "nsc_gpio" -void nsc_gpio_dump(unsigned index, u32 config) +void nsc_gpio_dump(struct nsc_gpio_ops *amp, unsigned index) { - printk(KERN_INFO NAME ": GPIO-%02u: 0x%08lx %s %s %s %s %s %s %s\n", - index, (unsigned long)config, - (config & 1) ? "OE" : "TS", /* output-enabled/tristate */ - (config & 2) ? "PP" : "OD", /* push pull / open drain */ - (config & 4) ? "PUE" : "PUD", /* pull up enabled/disabled */ - (config & 8) ? "LOCKED" : "", /* locked / unlocked */ - (config & 16) ? "LEVEL" : "EDGE",/* level/edge input */ - (config & 32) ? "HI" : "LO", /* trigger on rise/fall edge */ - (config & 64) ? "DEBOUNCE" : ""); /* debounce */ + /* retrieve current config w/o changing it */ + u32 config = amp->gpio_config(index, ~0, 0); + + /* user requested via 'v' command, so its INFO */ + dev_info(amp->dev, "io%02u: 0x%04x %s %s %s %s %s %s %s\n", + index, config, + (config & 1) ? "OE" : "TS", /* output-enabled/tristate */ + (config & 2) ? "PP" : "OD", /* push pull / open drain */ + (config & 4) ? "PUE" : "PUD", /* pull up enabled/disabled */ + (config & 8) ? "LOCKED" : "", /* locked / unlocked */ + (config & 16) ? "LEVEL" : "EDGE",/* level/edge input */ + (config & 32) ? "HI" : "LO", /* trigger on rise/fall edge */ + (config & 64) ? "DEBOUNCE" : ""); /* debounce */ } ssize_t nsc_gpio_write(struct file *file, const char __user *data, @@ -37,6 +42,7 @@ ssize_t nsc_gpio_write(struct file *file, const char __user *data, { unsigned m = iminor(file->f_dentry->d_inode); struct nsc_gpio_ops *amp = file->private_data; + struct device *dev = amp->dev; size_t i; int err = 0; @@ -52,42 +58,41 @@ ssize_t nsc_gpio_write(struct file *file, const char __user *data, amp->gpio_set(m, 1); break; case 'O': - printk(KERN_INFO NAME ": GPIO%d output enabled\n", m); + dev_dbg(dev, "GPIO%d output enabled\n", m); amp->gpio_config(m, ~1, 1); break; case 'o': - printk(KERN_INFO NAME ": GPIO%d output disabled\n", m); + dev_dbg(dev, "GPIO%d output disabled\n", m); amp->gpio_config(m, ~1, 0); break; case 'T': - printk(KERN_INFO NAME ": GPIO%d output is push pull\n", + dev_dbg(dev, "GPIO%d output is push pull\n", m); amp->gpio_config(m, ~2, 2); break; case 't': - printk(KERN_INFO NAME ": GPIO%d output is open drain\n", + dev_dbg(dev, "GPIO%d output is open drain\n", m); amp->gpio_config(m, ~2, 0); break; case 'P': - printk(KERN_INFO NAME ": GPIO%d pull up enabled\n", m); + dev_dbg(dev, "GPIO%d pull up enabled\n", m); amp->gpio_config(m, ~4, 4); break; case 'p': - printk(KERN_INFO NAME ": GPIO%d pull up disabled\n", m); + dev_dbg(dev, "GPIO%d pull up disabled\n", m); amp->gpio_config(m, ~4, 0); break; case 'v': /* View Current pin settings */ - amp->gpio_dump(m); + amp->gpio_dump(amp, m); break; case '\n': /* end of settings string, do nothing */ break; default: - printk(KERN_ERR NAME - ": GPIO-%2d bad setting: chr<0x%2x>\n", m, - (int)c); + dev_err(dev, "io%2d bad setting: chr<0x%2x>\n", + m, (int)c); err++; } } diff --git a/drivers/char/pc8736x_gpio.c b/drivers/char/pc8736x_gpio.c index b16fbef816c..48ff1fc8b06 100644 --- a/drivers/char/pc8736x_gpio.c +++ b/drivers/char/pc8736x_gpio.c @@ -207,8 +207,6 @@ static void pc8736x_gpio_change(unsigned index) pc8736x_gpio_set(index, !pc8736x_gpio_get(index)); } -extern void nsc_gpio_dump(unsigned iminor); - static struct nsc_gpio_ops pc8736x_access = { .owner = THIS_MODULE, .gpio_config = pc8736x_gpio_configure, @@ -260,6 +258,7 @@ static int __init pc8736x_gpio_init(void) dev_err(&pdev->dev, "no device found\n"); goto undo_platform_dev_add; } + pc8736x_access.dev = &pdev->dev; /* Verify that chip and it's GPIO unit are both enabled. My BIOS does this, so I take minimum action here diff --git a/drivers/char/scx200_gpio.c b/drivers/char/scx200_gpio.c index 442367b3f5d..5a280a33040 100644 --- a/drivers/char/scx200_gpio.c +++ b/drivers/char/scx200_gpio.c @@ -35,14 +35,6 @@ static int major = 0; /* default to dynamic major */ module_param(major, int, 0); MODULE_PARM_DESC(major, "Major device number"); -extern void nsc_gpio_dump(unsigned index); - -extern ssize_t nsc_gpio_write(struct file *file, const char __user *data, - size_t len, loff_t *ppos); - -extern ssize_t nsc_gpio_read(struct file *file, char __user *buf, - size_t len, loff_t *ppos); - struct nsc_gpio_ops scx200_access = { .owner = THIS_MODULE, .gpio_config = scx200_gpio_configure, @@ -101,6 +93,9 @@ static int __init scx200_gpio_init(void) if (rc) goto undo_malloc; + /* nsc_gpio uses dev_dbg(), so needs this */ + scx200_access.dev = &pdev->dev; + if (major) rc = register_chrdev_region(dev, num_pins, "scx200_gpio"); else { diff --git a/include/linux/nsc_gpio.h b/include/linux/nsc_gpio.h index 27bf66f7386..135742cfada 100644 --- a/include/linux/nsc_gpio.h +++ b/include/linux/nsc_gpio.h @@ -22,13 +22,14 @@ struct nsc_gpio_ops { struct module* owner; u32 (*gpio_config) (unsigned iminor, u32 mask, u32 bits); - void (*gpio_dump) (unsigned iminor); + void (*gpio_dump) (struct nsc_gpio_ops *amp, unsigned iminor); int (*gpio_get) (unsigned iminor); void (*gpio_set) (unsigned iminor, int state); void (*gpio_set_high)(unsigned iminor); void (*gpio_set_low) (unsigned iminor); void (*gpio_change) (unsigned iminor); int (*gpio_current) (unsigned iminor); + struct device* dev; /* for dev_dbg() support, set in init */ }; extern ssize_t nsc_gpio_write(struct file *file, const char __user *data, @@ -36,3 +37,6 @@ extern ssize_t nsc_gpio_write(struct file *file, const char __user *data, extern ssize_t nsc_gpio_read(struct file *file, char __user *buf, size_t len, loff_t *ppos); + +extern void nsc_gpio_dump(struct nsc_gpio_ops *amp, unsigned index); + -- cgit v1.2.3-70-g09d2 From 2dd73a4f09beacadde827a032cf15fd8b1fa3d48 Mon Sep 17 00:00:00 2001 From: Peter Williams Date: Tue, 27 Jun 2006 02:54:34 -0700 Subject: [PATCH] sched: implement smpnice Problem: The introduction of separate run queues per CPU has brought with it "nice" enforcement problems that are best described by a simple example. For the sake of argument suppose that on a single CPU machine with a nice==19 hard spinner and a nice==0 hard spinner running that the nice==0 task gets 95% of the CPU and the nice==19 task gets 5% of the CPU. Now suppose that there is a system with 2 CPUs and 2 nice==19 hard spinners and 2 nice==0 hard spinners running. The user of this system would be entitled to expect that the nice==0 tasks each get 95% of a CPU and the nice==19 tasks only get 5% each. However, whether this expectation is met is pretty much down to luck as there are four equally likely distributions of the tasks to the CPUs that the load balancing code will consider to be balanced with loads of 2.0 for each CPU. Two of these distributions involve one nice==0 and one nice==19 task per CPU and in these circumstances the users expectations will be met. The other two distributions both involve both nice==0 tasks being on one CPU and both nice==19 being on the other CPU and each task will get 50% of a CPU and the user's expectations will not be met. Solution: The solution to this problem that is implemented in the attached patch is to use weighted loads when determining if the system is balanced and, when an imbalance is detected, to move an amount of weighted load between run queues (as opposed to a number of tasks) to restore the balance. Once again, the easiest way to explain why both of these measures are necessary is to use a simple example. Suppose that (in a slight variation of the above example) that we have a two CPU system with 4 nice==0 and 4 nice=19 hard spinning tasks running and that the 4 nice==0 tasks are on one CPU and the 4 nice==19 tasks are on the other CPU. The weighted loads for the two CPUs would be 4.0 and 0.2 respectively and the load balancing code would move 2 tasks resulting in one CPU with a load of 2.0 and the other with load of 2.2. If this was considered to be a big enough imbalance to justify moving a task and that task was moved using the current move_tasks() then it would move the highest priority task that it found and this would result in one CPU with a load of 3.0 and the other with a load of 1.2 which would result in the movement of a task in the opposite direction and so on -- infinite loop. If, on the other hand, an amount of load to be moved is calculated from the imbalance (in this case 0.1) and move_tasks() skips tasks until it find ones whose contributions to the weighted load are less than this amount it would move two of the nice==19 tasks resulting in a system with 2 nice==0 and 2 nice=19 on each CPU with loads of 2.1 for each CPU. One of the advantages of this mechanism is that on a system where all tasks have nice==0 the load balancing calculations would be mathematically identical to the current load balancing code. Notes: struct task_struct: has a new field load_weight which (in a trade off of space for speed) stores the contribution that this task makes to a CPU's weighted load when it is runnable. struct runqueue: has a new field raw_weighted_load which is the sum of the load_weight values for the currently runnable tasks on this run queue. This field always needs to be updated when nr_running is updated so two new inline functions inc_nr_running() and dec_nr_running() have been created to make sure that this happens. This also offers a convenient way to optimize away this part of the smpnice mechanism when CONFIG_SMP is not defined. int try_to_wake_up(): in this function the value SCHED_LOAD_BALANCE is used to represent the load contribution of a single task in various calculations in the code that decides which CPU to put the waking task on. While this would be a valid on a system where the nice values for the runnable tasks were distributed evenly around zero it will lead to anomalous load balancing if the distribution is skewed in either direction. To overcome this problem SCHED_LOAD_SCALE has been replaced by the load_weight for the relevant task or by the average load_weight per task for the queue in question (as appropriate). int move_tasks(): The modifications to this function were complicated by the fact that active_load_balance() uses it to move exactly one task without checking whether an imbalance actually exists. This precluded the simple overloading of max_nr_move with max_load_move and necessitated the addition of the latter as an extra argument to the function. The internal implementation is then modified to move up to max_nr_move tasks and max_load_move of weighted load. This slightly complicates the code where move_tasks() is called and if ever active_load_balance() is changed to not use move_tasks() the implementation of move_tasks() should be simplified accordingly. struct sched_group *find_busiest_group(): Similar to try_to_wake_up(), there are places in this function where SCHED_LOAD_SCALE is used to represent the load contribution of a single task and the same issues are created. A similar solution is adopted except that it is now the average per task contribution to a group's load (as opposed to a run queue) that is required. As this value is not directly available from the group it is calculated on the fly as the queues in the groups are visited when determining the busiest group. A key change to this function is that it is no longer to scale down *imbalance on exit as move_tasks() uses the load in its scaled form. void set_user_nice(): has been modified to update the task's load_weight field when it's nice value and also to ensure that its run queue's raw_weighted_load field is updated if it was runnable. From: "Siddha, Suresh B" With smpnice, sched groups with highest priority tasks can mask the imbalance between the other sched groups with in the same domain. This patch fixes some of the listed down scenarios by not considering the sched groups which are lightly loaded. a) on a simple 4-way MP system, if we have one high priority and 4 normal priority tasks, with smpnice we would like to see the high priority task scheduled on one cpu, two other cpus getting one normal task each and the fourth cpu getting the remaining two normal tasks. but with current smpnice extra normal priority task keeps jumping from one cpu to another cpu having the normal priority task. This is because of the busiest_has_loaded_cpus, nr_loaded_cpus logic.. We are not including the cpu with high priority task in max_load calculations but including that in total and avg_load calcuations.. leading to max_load < avg_load and load balance between cpus running normal priority tasks(2 Vs 1) will always show imbalanace as one normal priority and the extra normal priority task will keep moving from one cpu to another cpu having normal priority task.. b) 4-way system with HT (8 logical processors). Package-P0 T0 has a highest priority task, T1 is idle. Package-P1 Both T0 and T1 have 1 normal priority task each.. P2 and P3 are idle. With this patch, one of the normal priority tasks on P1 will be moved to P2 or P3.. c) With the current weighted smp nice calculations, it doesn't always make sense to look at the highest weighted runqueue in the busy group.. Consider a load balance scenario on a DP with HT system, with Package-0 containing one high priority and one low priority, Package-1 containing one low priority(with other thread being idle).. Package-1 thinks that it need to take the low priority thread from Package-0. And find_busiest_queue() returns the cpu thread with highest priority task.. And ultimately(with help of active load balance) we move high priority task to Package-1. And same continues with Package-0 now, moving high priority task from package-1 to package-0.. Even without the presence of active load balance, load balance will fail to balance the above scenario.. Fix find_busiest_queue to use "imbalance" when it is lightly loaded. [kernel@kolivas.org: sched: store weighted load on up] [kernel@kolivas.org: sched: add discrete weighted cpu load function] [suresh.b.siddha@intel.com: sched: remove dead code] Signed-off-by: Peter Williams Cc: "Siddha, Suresh B" Cc: "Chen, Kenneth W" Acked-by: Ingo Molnar Cc: Nick Piggin Signed-off-by: Con Kolivas Cc: John Hawkes Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/sched.h | 8 +- kernel/sched.c | 284 ++++++++++++++++++++++++++++++++++++++------------ 2 files changed, 225 insertions(+), 67 deletions(-) (limited to 'include') diff --git a/include/linux/sched.h b/include/linux/sched.h index 122a25c1b99..74a1e39e0d3 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -123,6 +123,7 @@ extern unsigned long nr_running(void); extern unsigned long nr_uninterruptible(void); extern unsigned long nr_active(void); extern unsigned long nr_iowait(void); +extern unsigned long weighted_cpuload(const int cpu); /* @@ -558,9 +559,9 @@ enum idle_type /* * sched-domains (multiprocessor balancing) declarations: */ -#ifdef CONFIG_SMP #define SCHED_LOAD_SCALE 128UL /* increase resolution of load */ +#ifdef CONFIG_SMP #define SD_LOAD_BALANCE 1 /* Do load balancing on this domain. */ #define SD_BALANCE_NEWIDLE 2 /* Balance when about to become idle */ #define SD_BALANCE_EXEC 4 /* Balance on exec */ @@ -713,9 +714,12 @@ struct task_struct { int lock_depth; /* BKL lock depth */ -#if defined(CONFIG_SMP) && defined(__ARCH_WANT_UNLOCKED_CTXSW) +#ifdef CONFIG_SMP +#ifdef __ARCH_WANT_UNLOCKED_CTXSW int oncpu; #endif +#endif + int load_weight; /* for niceness load balancing purposes */ int prio, static_prio; struct list_head run_list; prio_array_t *array; diff --git a/kernel/sched.c b/kernel/sched.c index 678335a8b39..1847a4456a2 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -168,15 +168,21 @@ */ #define SCALE_PRIO(x, prio) \ - max(x * (MAX_PRIO - prio) / (MAX_USER_PRIO/2), MIN_TIMESLICE) + max(x * (MAX_PRIO - prio) / (MAX_USER_PRIO / 2), MIN_TIMESLICE) -static unsigned int task_timeslice(task_t *p) +static unsigned int static_prio_timeslice(int static_prio) { - if (p->static_prio < NICE_TO_PRIO(0)) - return SCALE_PRIO(DEF_TIMESLICE*4, p->static_prio); + if (static_prio < NICE_TO_PRIO(0)) + return SCALE_PRIO(DEF_TIMESLICE * 4, static_prio); else - return SCALE_PRIO(DEF_TIMESLICE, p->static_prio); + return SCALE_PRIO(DEF_TIMESLICE, static_prio); } + +static inline unsigned int task_timeslice(task_t *p) +{ + return static_prio_timeslice(p->static_prio); +} + #define task_hot(p, now, sd) ((long long) ((now) - (p)->last_ran) \ < (long long) (sd)->cache_hot_time) @@ -207,6 +213,7 @@ struct runqueue { * remote CPUs use both these fields when doing load calculation. */ unsigned long nr_running; + unsigned long raw_weighted_load; #ifdef CONFIG_SMP unsigned long cpu_load[3]; #endif @@ -661,6 +668,68 @@ static int effective_prio(task_t *p) return prio; } +/* + * To aid in avoiding the subversion of "niceness" due to uneven distribution + * of tasks with abnormal "nice" values across CPUs the contribution that + * each task makes to its run queue's load is weighted according to its + * scheduling class and "nice" value. For SCHED_NORMAL tasks this is just a + * scaled version of the new time slice allocation that they receive on time + * slice expiry etc. + */ + +/* + * Assume: static_prio_timeslice(NICE_TO_PRIO(0)) == DEF_TIMESLICE + * If static_prio_timeslice() is ever changed to break this assumption then + * this code will need modification + */ +#define TIME_SLICE_NICE_ZERO DEF_TIMESLICE +#define LOAD_WEIGHT(lp) \ + (((lp) * SCHED_LOAD_SCALE) / TIME_SLICE_NICE_ZERO) +#define PRIO_TO_LOAD_WEIGHT(prio) \ + LOAD_WEIGHT(static_prio_timeslice(prio)) +#define RTPRIO_TO_LOAD_WEIGHT(rp) \ + (PRIO_TO_LOAD_WEIGHT(MAX_RT_PRIO) + LOAD_WEIGHT(rp)) + +static void set_load_weight(task_t *p) +{ + if (rt_task(p)) { +#ifdef CONFIG_SMP + if (p == task_rq(p)->migration_thread) + /* + * The migration thread does the actual balancing. + * Giving its load any weight will skew balancing + * adversely. + */ + p->load_weight = 0; + else +#endif + p->load_weight = RTPRIO_TO_LOAD_WEIGHT(p->rt_priority); + } else + p->load_weight = PRIO_TO_LOAD_WEIGHT(p->static_prio); +} + +static inline void inc_raw_weighted_load(runqueue_t *rq, const task_t *p) +{ + rq->raw_weighted_load += p->load_weight; +} + +static inline void dec_raw_weighted_load(runqueue_t *rq, const task_t *p) +{ + rq->raw_weighted_load -= p->load_weight; +} + +static inline void inc_nr_running(task_t *p, runqueue_t *rq) +{ + rq->nr_running++; + inc_raw_weighted_load(rq, p); +} + +static inline void dec_nr_running(task_t *p, runqueue_t *rq) +{ + rq->nr_running--; + dec_raw_weighted_load(rq, p); +} + /* * __activate_task - move a task to the runqueue. */ @@ -671,7 +740,7 @@ static void __activate_task(task_t *p, runqueue_t *rq) if (batch_task(p)) target = rq->expired; enqueue_task(p, target); - rq->nr_running++; + inc_nr_running(p, rq); } /* @@ -680,7 +749,7 @@ static void __activate_task(task_t *p, runqueue_t *rq) static inline void __activate_idle_task(task_t *p, runqueue_t *rq) { enqueue_task_head(p, rq->active); - rq->nr_running++; + inc_nr_running(p, rq); } static int recalc_task_prio(task_t *p, unsigned long long now) @@ -804,7 +873,7 @@ static void activate_task(task_t *p, runqueue_t *rq, int local) */ static void deactivate_task(struct task_struct *p, runqueue_t *rq) { - rq->nr_running--; + dec_nr_running(p, rq); dequeue_task(p, p->array); p->array = NULL; } @@ -859,6 +928,12 @@ inline int task_curr(const task_t *p) return cpu_curr(task_cpu(p)) == p; } +/* Used instead of source_load when we know the type == 0 */ +unsigned long weighted_cpuload(const int cpu) +{ + return cpu_rq(cpu)->raw_weighted_load; +} + #ifdef CONFIG_SMP typedef struct { struct list_head list; @@ -948,7 +1023,8 @@ void kick_process(task_t *p) } /* - * Return a low guess at the load of a migration-source cpu. + * Return a low guess at the load of a migration-source cpu weighted + * according to the scheduling class and "nice" value. * * We want to under-estimate the load of migration sources, to * balance conservatively. @@ -956,24 +1032,36 @@ void kick_process(task_t *p) static inline unsigned long source_load(int cpu, int type) { runqueue_t *rq = cpu_rq(cpu); - unsigned long load_now = rq->nr_running * SCHED_LOAD_SCALE; + if (type == 0) - return load_now; + return rq->raw_weighted_load; - return min(rq->cpu_load[type-1], load_now); + return min(rq->cpu_load[type-1], rq->raw_weighted_load); } /* - * Return a high guess at the load of a migration-target cpu + * Return a high guess at the load of a migration-target cpu weighted + * according to the scheduling class and "nice" value. */ static inline unsigned long target_load(int cpu, int type) { runqueue_t *rq = cpu_rq(cpu); - unsigned long load_now = rq->nr_running * SCHED_LOAD_SCALE; + if (type == 0) - return load_now; + return rq->raw_weighted_load; - return max(rq->cpu_load[type-1], load_now); + return max(rq->cpu_load[type-1], rq->raw_weighted_load); +} + +/* + * Return the average load per task on the cpu's run queue + */ +static inline unsigned long cpu_avg_load_per_task(int cpu) +{ + runqueue_t *rq = cpu_rq(cpu); + unsigned long n = rq->nr_running; + + return n ? rq->raw_weighted_load / n : SCHED_LOAD_SCALE; } /* @@ -1046,7 +1134,7 @@ find_idlest_cpu(struct sched_group *group, struct task_struct *p, int this_cpu) cpus_and(tmp, group->cpumask, p->cpus_allowed); for_each_cpu_mask(i, tmp) { - load = source_load(i, 0); + load = weighted_cpuload(i); if (load < min_load || (load == min_load && i == this_cpu)) { min_load = load; @@ -1226,17 +1314,19 @@ static int try_to_wake_up(task_t *p, unsigned int state, int sync) if (this_sd->flags & SD_WAKE_AFFINE) { unsigned long tl = this_load; + unsigned long tl_per_task = cpu_avg_load_per_task(this_cpu); + /* * If sync wakeup then subtract the (maximum possible) * effect of the currently running task from the load * of the current CPU: */ if (sync) - tl -= SCHED_LOAD_SCALE; + tl -= current->load_weight; if ((tl <= load && - tl + target_load(cpu, idx) <= SCHED_LOAD_SCALE) || - 100*(tl + SCHED_LOAD_SCALE) <= imbalance*load) { + tl + target_load(cpu, idx) <= tl_per_task) || + 100*(tl + p->load_weight) <= imbalance*load) { /* * This domain has SD_WAKE_AFFINE and * p is cache cold in this domain, and @@ -1435,7 +1525,7 @@ void fastcall wake_up_new_task(task_t *p, unsigned long clone_flags) list_add_tail(&p->run_list, ¤t->run_list); p->array = current->array; p->array->nr_active++; - rq->nr_running++; + inc_nr_running(p, rq); } set_need_resched(); } else @@ -1802,9 +1892,9 @@ void pull_task(runqueue_t *src_rq, prio_array_t *src_array, task_t *p, runqueue_t *this_rq, prio_array_t *this_array, int this_cpu) { dequeue_task(p, src_array); - src_rq->nr_running--; + dec_nr_running(p, src_rq); set_task_cpu(p, this_cpu); - this_rq->nr_running++; + inc_nr_running(p, this_rq); enqueue_task(p, this_array); p->timestamp = (p->timestamp - src_rq->timestamp_last_tick) + this_rq->timestamp_last_tick; @@ -1852,24 +1942,27 @@ int can_migrate_task(task_t *p, runqueue_t *rq, int this_cpu, } /* - * move_tasks tries to move up to max_nr_move tasks from busiest to this_rq, - * as part of a balancing operation within "domain". Returns the number of - * tasks moved. + * move_tasks tries to move up to max_nr_move tasks and max_load_move weighted + * load from busiest to this_rq, as part of a balancing operation within + * "domain". Returns the number of tasks moved. * * Called with both runqueues locked. */ static int move_tasks(runqueue_t *this_rq, int this_cpu, runqueue_t *busiest, - unsigned long max_nr_move, struct sched_domain *sd, - enum idle_type idle, int *all_pinned) + unsigned long max_nr_move, unsigned long max_load_move, + struct sched_domain *sd, enum idle_type idle, + int *all_pinned) { prio_array_t *array, *dst_array; struct list_head *head, *curr; int idx, pulled = 0, pinned = 0; + long rem_load_move; task_t *tmp; - if (max_nr_move == 0) + if (max_nr_move == 0 || max_load_move == 0) goto out; + rem_load_move = max_load_move; pinned = 1; /* @@ -1910,7 +2003,8 @@ skip_queue: curr = curr->prev; - if (!can_migrate_task(tmp, busiest, this_cpu, sd, idle, &pinned)) { + if (tmp->load_weight > rem_load_move || + !can_migrate_task(tmp, busiest, this_cpu, sd, idle, &pinned)) { if (curr != head) goto skip_queue; idx++; @@ -1924,9 +2018,13 @@ skip_queue: pull_task(busiest, array, tmp, this_rq, dst_array, this_cpu); pulled++; + rem_load_move -= tmp->load_weight; - /* We only want to steal up to the prescribed number of tasks. */ - if (pulled < max_nr_move) { + /* + * We only want to steal up to the prescribed number of tasks + * and the prescribed amount of weighted load. + */ + if (pulled < max_nr_move && rem_load_move > 0) { if (curr != head) goto skip_queue; idx++; @@ -1947,7 +2045,7 @@ out: /* * find_busiest_group finds and returns the busiest CPU group within the - * domain. It calculates and returns the number of tasks which should be + * domain. It calculates and returns the amount of weighted load which should be * moved to restore balance via the imbalance parameter. */ static struct sched_group * @@ -1957,9 +2055,13 @@ find_busiest_group(struct sched_domain *sd, int this_cpu, struct sched_group *busiest = NULL, *this = NULL, *group = sd->groups; unsigned long max_load, avg_load, total_load, this_load, total_pwr; unsigned long max_pull; + unsigned long busiest_load_per_task, busiest_nr_running; + unsigned long this_load_per_task, this_nr_running; int load_idx; max_load = this_load = total_load = total_pwr = 0; + busiest_load_per_task = busiest_nr_running = 0; + this_load_per_task = this_nr_running = 0; if (idle == NOT_IDLE) load_idx = sd->busy_idx; else if (idle == NEWLY_IDLE) @@ -1971,13 +2073,16 @@ find_busiest_group(struct sched_domain *sd, int this_cpu, unsigned long load; int local_group; int i; + unsigned long sum_nr_running, sum_weighted_load; local_group = cpu_isset(this_cpu, group->cpumask); /* Tally up the load of all CPUs in the group */ - avg_load = 0; + sum_weighted_load = sum_nr_running = avg_load = 0; for_each_cpu_mask(i, group->cpumask) { + runqueue_t *rq = cpu_rq(i); + if (*sd_idle && !idle_cpu(i)) *sd_idle = 0; @@ -1988,6 +2093,8 @@ find_busiest_group(struct sched_domain *sd, int this_cpu, load = source_load(i, load_idx); avg_load += load; + sum_nr_running += rq->nr_running; + sum_weighted_load += rq->raw_weighted_load; } total_load += avg_load; @@ -1999,14 +2106,19 @@ find_busiest_group(struct sched_domain *sd, int this_cpu, if (local_group) { this_load = avg_load; this = group; - } else if (avg_load > max_load) { + this_nr_running = sum_nr_running; + this_load_per_task = sum_weighted_load; + } else if (avg_load > max_load && + sum_nr_running > group->cpu_power / SCHED_LOAD_SCALE) { max_load = avg_load; busiest = group; + busiest_nr_running = sum_nr_running; + busiest_load_per_task = sum_weighted_load; } group = group->next; } while (group != sd->groups); - if (!busiest || this_load >= max_load || max_load <= SCHED_LOAD_SCALE) + if (!busiest || this_load >= max_load || busiest_nr_running == 0) goto out_balanced; avg_load = (SCHED_LOAD_SCALE * total_load) / total_pwr; @@ -2015,6 +2127,7 @@ find_busiest_group(struct sched_domain *sd, int this_cpu, 100*max_load <= sd->imbalance_pct*this_load) goto out_balanced; + busiest_load_per_task /= busiest_nr_running; /* * We're trying to get all the cpus to the average_load, so we don't * want to push ourselves above the average load, nor do we wish to @@ -2026,21 +2139,50 @@ find_busiest_group(struct sched_domain *sd, int this_cpu, * by pulling tasks to us. Be careful of negative numbers as they'll * appear as very large values with unsigned longs. */ + if (max_load <= busiest_load_per_task) + goto out_balanced; + + /* + * In the presence of smp nice balancing, certain scenarios can have + * max load less than avg load(as we skip the groups at or below + * its cpu_power, while calculating max_load..) + */ + if (max_load < avg_load) { + *imbalance = 0; + goto small_imbalance; + } /* Don't want to pull so many tasks that a group would go idle */ - max_pull = min(max_load - avg_load, max_load - SCHED_LOAD_SCALE); + max_pull = min(max_load - avg_load, max_load - busiest_load_per_task); /* How much load to actually move to equalise the imbalance */ *imbalance = min(max_pull * busiest->cpu_power, (avg_load - this_load) * this->cpu_power) / SCHED_LOAD_SCALE; - if (*imbalance < SCHED_LOAD_SCALE) { - unsigned long pwr_now = 0, pwr_move = 0; + /* + * if *imbalance is less than the average load per runnable task + * there is no gaurantee that any tasks will be moved so we'll have + * a think about bumping its value to force at least one task to be + * moved + */ + if (*imbalance < busiest_load_per_task) { + unsigned long pwr_now, pwr_move; unsigned long tmp; + unsigned int imbn; + +small_imbalance: + pwr_move = pwr_now = 0; + imbn = 2; + if (this_nr_running) { + this_load_per_task /= this_nr_running; + if (busiest_load_per_task > this_load_per_task) + imbn = 1; + } else + this_load_per_task = SCHED_LOAD_SCALE; - if (max_load - this_load >= SCHED_LOAD_SCALE*2) { - *imbalance = 1; + if (max_load - this_load >= busiest_load_per_task * imbn) { + *imbalance = busiest_load_per_task; return busiest; } @@ -2050,35 +2192,34 @@ find_busiest_group(struct sched_domain *sd, int this_cpu, * moving them. */ - pwr_now += busiest->cpu_power*min(SCHED_LOAD_SCALE, max_load); - pwr_now += this->cpu_power*min(SCHED_LOAD_SCALE, this_load); + pwr_now += busiest->cpu_power * + min(busiest_load_per_task, max_load); + pwr_now += this->cpu_power * + min(this_load_per_task, this_load); pwr_now /= SCHED_LOAD_SCALE; /* Amount of load we'd subtract */ - tmp = SCHED_LOAD_SCALE*SCHED_LOAD_SCALE/busiest->cpu_power; + tmp = busiest_load_per_task*SCHED_LOAD_SCALE/busiest->cpu_power; if (max_load > tmp) - pwr_move += busiest->cpu_power*min(SCHED_LOAD_SCALE, - max_load - tmp); + pwr_move += busiest->cpu_power * + min(busiest_load_per_task, max_load - tmp); /* Amount of load we'd add */ if (max_load*busiest->cpu_power < - SCHED_LOAD_SCALE*SCHED_LOAD_SCALE) + busiest_load_per_task*SCHED_LOAD_SCALE) tmp = max_load*busiest->cpu_power/this->cpu_power; else - tmp = SCHED_LOAD_SCALE*SCHED_LOAD_SCALE/this->cpu_power; - pwr_move += this->cpu_power*min(SCHED_LOAD_SCALE, this_load + tmp); + tmp = busiest_load_per_task*SCHED_LOAD_SCALE/this->cpu_power; + pwr_move += this->cpu_power*min(this_load_per_task, this_load + tmp); pwr_move /= SCHED_LOAD_SCALE; /* Move if we gain throughput */ if (pwr_move <= pwr_now) goto out_balanced; - *imbalance = 1; - return busiest; + *imbalance = busiest_load_per_task; } - /* Get rid of the scaling factor, rounding down as we divide */ - *imbalance = *imbalance / SCHED_LOAD_SCALE; return busiest; out_balanced: @@ -2091,18 +2232,21 @@ out_balanced: * find_busiest_queue - find the busiest runqueue among the cpus in group. */ static runqueue_t *find_busiest_queue(struct sched_group *group, - enum idle_type idle) + enum idle_type idle, unsigned long imbalance) { - unsigned long load, max_load = 0; - runqueue_t *busiest = NULL; + unsigned long max_load = 0; + runqueue_t *busiest = NULL, *rqi; int i; for_each_cpu_mask(i, group->cpumask) { - load = source_load(i, 0); + rqi = cpu_rq(i); + + if (rqi->nr_running == 1 && rqi->raw_weighted_load > imbalance) + continue; - if (load > max_load) { - max_load = load; - busiest = cpu_rq(i); + if (rqi->raw_weighted_load > max_load) { + max_load = rqi->raw_weighted_load; + busiest = rqi; } } @@ -2115,6 +2259,7 @@ static runqueue_t *find_busiest_queue(struct sched_group *group, */ #define MAX_PINNED_INTERVAL 512 +#define minus_1_or_zero(n) ((n) > 0 ? (n) - 1 : 0) /* * Check this_cpu to ensure it is balanced within domain. Attempt to move * tasks if there is an imbalance. @@ -2142,7 +2287,7 @@ static int load_balance(int this_cpu, runqueue_t *this_rq, goto out_balanced; } - busiest = find_busiest_queue(group, idle); + busiest = find_busiest_queue(group, idle, imbalance); if (!busiest) { schedstat_inc(sd, lb_nobusyq[idle]); goto out_balanced; @@ -2162,6 +2307,7 @@ static int load_balance(int this_cpu, runqueue_t *this_rq, */ double_rq_lock(this_rq, busiest); nr_moved = move_tasks(this_rq, this_cpu, busiest, + minus_1_or_zero(busiest->nr_running), imbalance, sd, idle, &all_pinned); double_rq_unlock(this_rq, busiest); @@ -2265,7 +2411,7 @@ static int load_balance_newidle(int this_cpu, runqueue_t *this_rq, goto out_balanced; } - busiest = find_busiest_queue(group, NEWLY_IDLE); + busiest = find_busiest_queue(group, NEWLY_IDLE, imbalance); if (!busiest) { schedstat_inc(sd, lb_nobusyq[NEWLY_IDLE]); goto out_balanced; @@ -2280,6 +2426,7 @@ static int load_balance_newidle(int this_cpu, runqueue_t *this_rq, /* Attempt to move tasks */ double_lock_balance(this_rq, busiest); nr_moved = move_tasks(this_rq, this_cpu, busiest, + minus_1_or_zero(busiest->nr_running), imbalance, sd, NEWLY_IDLE, NULL); spin_unlock(&busiest->lock); } @@ -2361,7 +2508,8 @@ static void active_load_balance(runqueue_t *busiest_rq, int busiest_cpu) schedstat_inc(sd, alb_cnt); - if (move_tasks(target_rq, target_cpu, busiest_rq, 1, sd, SCHED_IDLE, NULL)) + if (move_tasks(target_rq, target_cpu, busiest_rq, 1, + RTPRIO_TO_LOAD_WEIGHT(100), sd, SCHED_IDLE, NULL)) schedstat_inc(sd, alb_pushed); else schedstat_inc(sd, alb_failed); @@ -2389,7 +2537,7 @@ static void rebalance_tick(int this_cpu, runqueue_t *this_rq, struct sched_domain *sd; int i; - this_load = this_rq->nr_running * SCHED_LOAD_SCALE; + this_load = this_rq->raw_weighted_load; /* Update our load */ for (i = 0; i < 3; i++) { unsigned long new_load = this_load; @@ -3441,17 +3589,21 @@ void set_user_nice(task_t *p, long nice) goto out_unlock; } array = p->array; - if (array) + if (array) { dequeue_task(p, array); + dec_raw_weighted_load(rq, p); + } old_prio = p->prio; new_prio = NICE_TO_PRIO(nice); delta = new_prio - old_prio; p->static_prio = NICE_TO_PRIO(nice); + set_load_weight(p); p->prio += delta; if (array) { enqueue_task(p, array); + inc_raw_weighted_load(rq, p); /* * If the task increased its priority or is running and * lowered its priority, then reschedule its CPU: @@ -3587,6 +3739,7 @@ static void __setscheduler(struct task_struct *p, int policy, int prio) if (policy == SCHED_BATCH) p->sleep_avg = 0; } + set_load_weight(p); } /** @@ -6106,6 +6259,7 @@ void __init sched_init(void) } } + set_load_weight(&init_task); /* * The boot idle thread does lazy MMU switching as well: */ -- cgit v1.2.3-70-g09d2 From 51888ca25a03125e742ef84d4ddfd74e139707a0 Mon Sep 17 00:00:00 2001 From: Srivatsa Vaddagiri Date: Tue, 27 Jun 2006 02:54:38 -0700 Subject: [PATCH] sched_domain: handle kmalloc failure Try to handle mem allocation failures in build_sched_domains by bailing out and cleaning up thus-far allocated memory. The patch has a direct consequence that we disable load balancing completely (even at sibling level) upon *any* memory allocation failure. [Lee.Schermerhorn@hp.com: bugfix] Signed-off-by: Srivatsa Vaddagir Cc: Nick Piggin Cc: Ingo Molnar Cc: "Siddha, Suresh B" Signed-off-by: Lee Schermerhorn Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/sched.h | 2 +- kernel/sched.c | 139 ++++++++++++++++++++++++++++---------------------- 2 files changed, 79 insertions(+), 62 deletions(-) (limited to 'include') diff --git a/include/linux/sched.h b/include/linux/sched.h index 74a1e39e0d3..ab8ffc54423 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -639,7 +639,7 @@ struct sched_domain { #endif }; -extern void partition_sched_domains(cpumask_t *partition1, +extern int partition_sched_domains(cpumask_t *partition1, cpumask_t *partition2); /* diff --git a/kernel/sched.c b/kernel/sched.c index 0ec84f57695..77a2ec55ef7 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -5820,11 +5820,56 @@ next_sg: } #endif +/* Free memory allocated for various sched_group structures */ +static void free_sched_groups(const cpumask_t *cpu_map) +{ +#ifdef CONFIG_NUMA + int i; + int cpu; + + for_each_cpu_mask(cpu, *cpu_map) { + struct sched_group *sched_group_allnodes + = sched_group_allnodes_bycpu[cpu]; + struct sched_group **sched_group_nodes + = sched_group_nodes_bycpu[cpu]; + + if (sched_group_allnodes) { + kfree(sched_group_allnodes); + sched_group_allnodes_bycpu[cpu] = NULL; + } + + if (!sched_group_nodes) + continue; + + for (i = 0; i < MAX_NUMNODES; i++) { + cpumask_t nodemask = node_to_cpumask(i); + struct sched_group *oldsg, *sg = sched_group_nodes[i]; + + cpus_and(nodemask, nodemask, *cpu_map); + if (cpus_empty(nodemask)) + continue; + + if (sg == NULL) + continue; + sg = sg->next; +next_sg: + oldsg = sg; + sg = sg->next; + kfree(oldsg); + if (oldsg != sched_group_nodes[i]) + goto next_sg; + } + kfree(sched_group_nodes); + sched_group_nodes_bycpu[cpu] = NULL; + } +#endif +} + /* * Build sched domains for a given set of cpus and attach the sched domains * to the individual cpus */ -void build_sched_domains(const cpumask_t *cpu_map) +static int build_sched_domains(const cpumask_t *cpu_map) { int i; #ifdef CONFIG_NUMA @@ -5834,11 +5879,11 @@ void build_sched_domains(const cpumask_t *cpu_map) /* * Allocate the per-node list of sched groups */ - sched_group_nodes = kmalloc(sizeof(struct sched_group*)*MAX_NUMNODES, + sched_group_nodes = kzalloc(sizeof(struct sched_group*)*MAX_NUMNODES, GFP_ATOMIC); if (!sched_group_nodes) { printk(KERN_WARNING "Can not alloc sched group node list\n"); - return; + return -ENOMEM; } sched_group_nodes_bycpu[first_cpu(*cpu_map)] = sched_group_nodes; #endif @@ -5864,7 +5909,7 @@ void build_sched_domains(const cpumask_t *cpu_map) if (!sched_group_allnodes) { printk(KERN_WARNING "Can not alloc allnodes sched group\n"); - break; + goto error; } sched_group_allnodes_bycpu[i] = sched_group_allnodes; @@ -5978,23 +6023,20 @@ void build_sched_domains(const cpumask_t *cpu_map) cpus_and(domainspan, domainspan, *cpu_map); sg = kmalloc(sizeof(struct sched_group), GFP_KERNEL); + if (!sg) { + printk(KERN_WARNING "Can not alloc domain group for " + "node %d\n", i); + goto error; + } sched_group_nodes[i] = sg; for_each_cpu_mask(j, nodemask) { struct sched_domain *sd; sd = &per_cpu(node_domains, j); sd->groups = sg; - if (sd->groups == NULL) { - /* Turn off balancing if we have no groups */ - sd->flags = 0; - } - } - if (!sg) { - printk(KERN_WARNING - "Can not alloc domain group for node %d\n", i); - continue; } sg->cpu_power = 0; sg->cpumask = nodemask; + sg->next = sg; cpus_or(covered, covered, nodemask); prev = sg; @@ -6017,15 +6059,15 @@ void build_sched_domains(const cpumask_t *cpu_map) if (!sg) { printk(KERN_WARNING "Can not alloc domain group for node %d\n", j); - break; + goto error; } sg->cpu_power = 0; sg->cpumask = tmp; + sg->next = prev->next; cpus_or(covered, covered, tmp); prev->next = sg; prev = sg; } - prev->next = sched_group_nodes[i]; } #endif @@ -6088,13 +6130,22 @@ void build_sched_domains(const cpumask_t *cpu_map) * Tune cache-hot values: */ calibrate_migration_costs(cpu_map); + + return 0; + +#ifdef CONFIG_NUMA +error: + free_sched_groups(cpu_map); + return -ENOMEM; +#endif } /* * Set up scheduler domains and groups. Callers must hold the hotplug lock. */ -static void arch_init_sched_domains(const cpumask_t *cpu_map) +static int arch_init_sched_domains(const cpumask_t *cpu_map) { cpumask_t cpu_default_map; + int err; /* * Setup mask for cpus without special case scheduling requirements. @@ -6103,51 +6154,14 @@ static void arch_init_sched_domains(const cpumask_t *cpu_map) */ cpus_andnot(cpu_default_map, *cpu_map, cpu_isolated_map); - build_sched_domains(&cpu_default_map); + err = build_sched_domains(&cpu_default_map); + + return err; } static void arch_destroy_sched_domains(const cpumask_t *cpu_map) { -#ifdef CONFIG_NUMA - int i; - int cpu; - - for_each_cpu_mask(cpu, *cpu_map) { - struct sched_group *sched_group_allnodes - = sched_group_allnodes_bycpu[cpu]; - struct sched_group **sched_group_nodes - = sched_group_nodes_bycpu[cpu]; - - if (sched_group_allnodes) { - kfree(sched_group_allnodes); - sched_group_allnodes_bycpu[cpu] = NULL; - } - - if (!sched_group_nodes) - continue; - - for (i = 0; i < MAX_NUMNODES; i++) { - cpumask_t nodemask = node_to_cpumask(i); - struct sched_group *oldsg, *sg = sched_group_nodes[i]; - - cpus_and(nodemask, nodemask, *cpu_map); - if (cpus_empty(nodemask)) - continue; - - if (sg == NULL) - continue; - sg = sg->next; -next_sg: - oldsg = sg; - sg = sg->next; - kfree(oldsg); - if (oldsg != sched_group_nodes[i]) - goto next_sg; - } - kfree(sched_group_nodes); - sched_group_nodes_bycpu[cpu] = NULL; - } -#endif + free_sched_groups(cpu_map); } /* @@ -6172,9 +6186,10 @@ static void detach_destroy_domains(const cpumask_t *cpu_map) * correct sched domains * Call with hotplug lock held */ -void partition_sched_domains(cpumask_t *partition1, cpumask_t *partition2) +int partition_sched_domains(cpumask_t *partition1, cpumask_t *partition2) { cpumask_t change_map; + int err = 0; cpus_and(*partition1, *partition1, cpu_online_map); cpus_and(*partition2, *partition2, cpu_online_map); @@ -6183,9 +6198,11 @@ void partition_sched_domains(cpumask_t *partition1, cpumask_t *partition2) /* Detach sched domains from all of the affected cpus */ detach_destroy_domains(&change_map); if (!cpus_empty(*partition1)) - build_sched_domains(partition1); - if (!cpus_empty(*partition2)) - build_sched_domains(partition2); + err = build_sched_domains(partition1); + if (!err && !cpus_empty(*partition2)) + err = build_sched_domains(partition2); + + return err; } #ifdef CONFIG_HOTPLUG_CPU -- cgit v1.2.3-70-g09d2 From 5c45bf279d378d436ce45825c0f136696c7b6109 Mon Sep 17 00:00:00 2001 From: "Siddha, Suresh B" Date: Tue, 27 Jun 2006 02:54:42 -0700 Subject: [PATCH] sched: mc/smt power savings sched policy sysfs entries 'sched_mc_power_savings' and 'sched_smt_power_savings' in /sys/devices/system/cpu/ control the MC/SMT power savings policy for the scheduler. Based on the values (1-enable, 0-disable) for these controls, sched groups cpu power will be determined for different domains. When power savings policy is enabled and under light load conditions, scheduler will minimize the physical packages/cpu cores carrying the load and thus conserving power(with a perf impact based on the workload characteristics... see OLS 2005 CMP kernel scheduler paper for more details..) Signed-off-by: Suresh Siddha Cc: Ingo Molnar Cc: Nick Piggin Cc: Con Kolivas Cc: "Chen, Kenneth W" Cc: "David S. Miller" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/i386/kernel/smpboot.c | 8 +- arch/x86_64/kernel/smpboot.c | 8 +- drivers/base/cpu.c | 10 +- include/asm-i386/topology.h | 5 + include/asm-ia64/topology.h | 1 + include/asm-powerpc/topology.h | 5 + include/asm-sparc64/topology.h | 3 + include/asm-x86_64/topology.h | 2 + include/linux/sched.h | 10 ++ include/linux/topology.h | 3 +- kernel/sched.c | 240 ++++++++++++++++++++++++++++++++++++----- 11 files changed, 262 insertions(+), 33 deletions(-) (limited to 'include') diff --git a/arch/i386/kernel/smpboot.c b/arch/i386/kernel/smpboot.c index ab5275beddf..89e7315e539 100644 --- a/arch/i386/kernel/smpboot.c +++ b/arch/i386/kernel/smpboot.c @@ -448,10 +448,12 @@ cpumask_t cpu_coregroup_map(int cpu) struct cpuinfo_x86 *c = cpu_data + cpu; /* * For perf, we return last level cache shared map. - * TBD: when power saving sched policy is added, we will return - * cpu_core_map when power saving policy is enabled + * And for power savings, we return cpu_core_map */ - return c->llc_shared_map; + if (sched_mc_power_savings || sched_smt_power_savings) + return cpu_core_map[cpu]; + else + return c->llc_shared_map; } /* representing cpus for which sibling maps can be computed */ diff --git a/arch/x86_64/kernel/smpboot.c b/arch/x86_64/kernel/smpboot.c index 4e9755179ec..540c0ccbccc 100644 --- a/arch/x86_64/kernel/smpboot.c +++ b/arch/x86_64/kernel/smpboot.c @@ -455,10 +455,12 @@ cpumask_t cpu_coregroup_map(int cpu) struct cpuinfo_x86 *c = cpu_data + cpu; /* * For perf, we return last level cache shared map. - * TBD: when power saving sched policy is added, we will return - * cpu_core_map when power saving policy is enabled + * And for power savings, we return cpu_core_map */ - return c->llc_shared_map; + if (sched_mc_power_savings || sched_smt_power_savings) + return cpu_core_map[cpu]; + else + return c->llc_shared_map; } /* representing cpus for which sibling maps can be computed */ diff --git a/drivers/base/cpu.c b/drivers/base/cpu.c index 3972d8ac978..4bef76a2f3f 100644 --- a/drivers/base/cpu.c +++ b/drivers/base/cpu.c @@ -143,5 +143,13 @@ EXPORT_SYMBOL_GPL(get_cpu_sysdev); int __init cpu_dev_init(void) { - return sysdev_class_register(&cpu_sysdev_class); + int err; + + err = sysdev_class_register(&cpu_sysdev_class); +#if defined(CONFIG_SCHED_MC) || defined(CONFIG_SCHED_SMT) + if (!err) + err = sched_create_sysfs_power_savings_entries(&cpu_sysdev_class); +#endif + + return err; } diff --git a/include/asm-i386/topology.h b/include/asm-i386/topology.h index aa4185ee81f..6adbd9b1ae8 100644 --- a/include/asm-i386/topology.h +++ b/include/asm-i386/topology.h @@ -112,4 +112,9 @@ extern unsigned long node_remap_size[]; extern cpumask_t cpu_coregroup_map(int cpu); +#ifdef CONFIG_SMP +#define mc_capable() (boot_cpu_data.x86_max_cores > 1) +#define smt_capable() (smp_num_siblings > 1) +#endif + #endif /* _ASM_I386_TOPOLOGY_H */ diff --git a/include/asm-ia64/topology.h b/include/asm-ia64/topology.h index 616b5ed2aa7..937c2125752 100644 --- a/include/asm-ia64/topology.h +++ b/include/asm-ia64/topology.h @@ -112,6 +112,7 @@ void build_cpu_to_node_map(void); #define topology_core_id(cpu) (cpu_data(cpu)->core_id) #define topology_core_siblings(cpu) (cpu_core_map[cpu]) #define topology_thread_siblings(cpu) (cpu_sibling_map[cpu]) +#define smt_capable() (smp_num_siblings > 1) #endif #include diff --git a/include/asm-powerpc/topology.h b/include/asm-powerpc/topology.h index 92f3e5507d2..bbc3844b086 100644 --- a/include/asm-powerpc/topology.h +++ b/include/asm-powerpc/topology.h @@ -93,5 +93,10 @@ static inline void sysfs_remove_device_from_node(struct sys_device *dev, #endif /* CONFIG_NUMA */ +#ifdef CONFIG_SMP +#include +#define smt_capable() (cpu_has_feature(CPU_FTR_SMT)) +#endif + #endif /* __KERNEL__ */ #endif /* _ASM_POWERPC_TOPOLOGY_H */ diff --git a/include/asm-sparc64/topology.h b/include/asm-sparc64/topology.h index 0e234e201bd..98a6c613589 100644 --- a/include/asm-sparc64/topology.h +++ b/include/asm-sparc64/topology.h @@ -1,6 +1,9 @@ #ifndef _ASM_SPARC64_TOPOLOGY_H #define _ASM_SPARC64_TOPOLOGY_H +#include +#define smt_capable() (tlb_type == hypervisor) + #include #endif /* _ASM_SPARC64_TOPOLOGY_H */ diff --git a/include/asm-x86_64/topology.h b/include/asm-x86_64/topology.h index c4e46e7fa7b..6e7a2e976b0 100644 --- a/include/asm-x86_64/topology.h +++ b/include/asm-x86_64/topology.h @@ -59,6 +59,8 @@ extern int __node_distance(int, int); #define topology_core_id(cpu) (cpu_data[cpu].cpu_core_id) #define topology_core_siblings(cpu) (cpu_core_map[cpu]) #define topology_thread_siblings(cpu) (cpu_sibling_map[cpu]) +#define mc_capable() (boot_cpu_data.x86_max_cores > 1) +#define smt_capable() (smp_num_siblings > 1) #endif #include diff --git a/include/linux/sched.h b/include/linux/sched.h index ab8ffc54423..0bc81a151e5 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -570,6 +570,11 @@ enum idle_type #define SD_WAKE_AFFINE 32 /* Wake task to waking CPU */ #define SD_WAKE_BALANCE 64 /* Perform balancing at task wakeup */ #define SD_SHARE_CPUPOWER 128 /* Domain members share cpu power */ +#define SD_POWERSAVINGS_BALANCE 256 /* Balance for power savings */ + +#define BALANCE_FOR_POWER ((sched_mc_power_savings || sched_smt_power_savings) \ + ? SD_POWERSAVINGS_BALANCE : 0) + struct sched_group { struct sched_group *next; /* Must be a circular list */ @@ -1412,6 +1417,11 @@ static inline void arch_pick_mmap_layout(struct mm_struct *mm) extern long sched_setaffinity(pid_t pid, cpumask_t new_mask); extern long sched_getaffinity(pid_t pid, cpumask_t *mask); +#include +extern int sched_mc_power_savings, sched_smt_power_savings; +extern struct sysdev_attribute attr_sched_mc_power_savings, attr_sched_smt_power_savings; +extern int sched_create_sysfs_power_savings_entries(struct sysdev_class *cls); + extern void normalize_rt_tasks(void); #ifdef CONFIG_PM diff --git a/include/linux/topology.h b/include/linux/topology.h index a305ae2e44b..ec1eca85290 100644 --- a/include/linux/topology.h +++ b/include/linux/topology.h @@ -134,7 +134,8 @@ .flags = SD_LOAD_BALANCE \ | SD_BALANCE_NEWIDLE \ | SD_BALANCE_EXEC \ - | SD_WAKE_AFFINE, \ + | SD_WAKE_AFFINE \ + | BALANCE_FOR_POWER, \ .last_balance = jiffies, \ .balance_interval = 1, \ .nr_balance_failed = 0, \ diff --git a/kernel/sched.c b/kernel/sched.c index 122b75584a1..54fa282657c 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -1162,6 +1162,11 @@ static int sched_balance_self(int cpu, int flag) struct sched_domain *tmp, *sd = NULL; for_each_domain(cpu, tmp) { + /* + * If power savings logic is enabled for a domain, stop there. + */ + if (tmp->flags & SD_POWERSAVINGS_BALANCE) + break; if (tmp->flags & flag) sd = tmp; } @@ -2082,6 +2087,12 @@ find_busiest_group(struct sched_domain *sd, int this_cpu, unsigned long busiest_load_per_task, busiest_nr_running; unsigned long this_load_per_task, this_nr_running; int load_idx; +#if defined(CONFIG_SCHED_MC) || defined(CONFIG_SCHED_SMT) + int power_savings_balance = 1; + unsigned long leader_nr_running = 0, min_load_per_task = 0; + unsigned long min_nr_running = ULONG_MAX; + struct sched_group *group_min = NULL, *group_leader = NULL; +#endif max_load = this_load = total_load = total_pwr = 0; busiest_load_per_task = busiest_nr_running = 0; @@ -2094,7 +2105,7 @@ find_busiest_group(struct sched_domain *sd, int this_cpu, load_idx = sd->idle_idx; do { - unsigned long load; + unsigned long load, group_capacity; int local_group; int i; unsigned long sum_nr_running, sum_weighted_load; @@ -2127,18 +2138,76 @@ find_busiest_group(struct sched_domain *sd, int this_cpu, /* Adjust by relative CPU power of the group */ avg_load = (avg_load * SCHED_LOAD_SCALE) / group->cpu_power; + group_capacity = group->cpu_power / SCHED_LOAD_SCALE; + if (local_group) { this_load = avg_load; this = group; this_nr_running = sum_nr_running; this_load_per_task = sum_weighted_load; } else if (avg_load > max_load && - sum_nr_running > group->cpu_power / SCHED_LOAD_SCALE) { + sum_nr_running > group_capacity) { max_load = avg_load; busiest = group; busiest_nr_running = sum_nr_running; busiest_load_per_task = sum_weighted_load; } + +#if defined(CONFIG_SCHED_MC) || defined(CONFIG_SCHED_SMT) + /* + * Busy processors will not participate in power savings + * balance. + */ + if (idle == NOT_IDLE || !(sd->flags & SD_POWERSAVINGS_BALANCE)) + goto group_next; + + /* + * If the local group is idle or completely loaded + * no need to do power savings balance at this domain + */ + if (local_group && (this_nr_running >= group_capacity || + !this_nr_running)) + power_savings_balance = 0; + + /* + * If a group is already running at full capacity or idle, + * don't include that group in power savings calculations + */ + if (!power_savings_balance || sum_nr_running >= group_capacity + || !sum_nr_running) + goto group_next; + + /* + * Calculate the group which has the least non-idle load. + * This is the group from where we need to pick up the load + * for saving power + */ + if ((sum_nr_running < min_nr_running) || + (sum_nr_running == min_nr_running && + first_cpu(group->cpumask) < + first_cpu(group_min->cpumask))) { + group_min = group; + min_nr_running = sum_nr_running; + min_load_per_task = sum_weighted_load / + sum_nr_running; + } + + /* + * Calculate the group which is almost near its + * capacity but still has some space to pick up some load + * from other group and save more power + */ + if (sum_nr_running <= group_capacity - 1) + if (sum_nr_running > leader_nr_running || + (sum_nr_running == leader_nr_running && + first_cpu(group->cpumask) > + first_cpu(group_leader->cpumask))) { + group_leader = group; + leader_nr_running = sum_nr_running; + } + +group_next: +#endif group = group->next; } while (group != sd->groups); @@ -2247,7 +2316,16 @@ small_imbalance: return busiest; out_balanced: +#if defined(CONFIG_SCHED_MC) || defined(CONFIG_SCHED_SMT) + if (idle == NOT_IDLE || !(sd->flags & SD_POWERSAVINGS_BALANCE)) + goto ret; + if (this == group_leader && group_leader != group_min) { + *imbalance = min_load_per_task; + return group_min; + } +ret: +#endif *imbalance = 0; return NULL; } @@ -2300,7 +2378,8 @@ static int load_balance(int this_cpu, runqueue_t *this_rq, int active_balance = 0; int sd_idle = 0; - if (idle != NOT_IDLE && sd->flags & SD_SHARE_CPUPOWER) + if (idle != NOT_IDLE && sd->flags & SD_SHARE_CPUPOWER && + !sched_smt_power_savings) sd_idle = 1; schedstat_inc(sd, lb_cnt[idle]); @@ -2389,7 +2468,8 @@ static int load_balance(int this_cpu, runqueue_t *this_rq, sd->balance_interval *= 2; } - if (!nr_moved && !sd_idle && sd->flags & SD_SHARE_CPUPOWER) + if (!nr_moved && !sd_idle && sd->flags & SD_SHARE_CPUPOWER && + !sched_smt_power_savings) return -1; return nr_moved; @@ -2404,7 +2484,7 @@ out_one_pinned: (sd->balance_interval < sd->max_interval)) sd->balance_interval *= 2; - if (!sd_idle && sd->flags & SD_SHARE_CPUPOWER) + if (!sd_idle && sd->flags & SD_SHARE_CPUPOWER && !sched_smt_power_savings) return -1; return 0; } @@ -2425,7 +2505,7 @@ static int load_balance_newidle(int this_cpu, runqueue_t *this_rq, int nr_moved = 0; int sd_idle = 0; - if (sd->flags & SD_SHARE_CPUPOWER) + if (sd->flags & SD_SHARE_CPUPOWER && !sched_smt_power_savings) sd_idle = 1; schedstat_inc(sd, lb_cnt[NEWLY_IDLE]); @@ -2466,7 +2546,7 @@ static int load_balance_newidle(int this_cpu, runqueue_t *this_rq, out_balanced: schedstat_inc(sd, lb_balanced[NEWLY_IDLE]); - if (!sd_idle && sd->flags & SD_SHARE_CPUPOWER) + if (!sd_idle && sd->flags & SD_SHARE_CPUPOWER && !sched_smt_power_savings) return -1; sd->nr_balance_failed = 0; return 0; @@ -5732,6 +5812,7 @@ static cpumask_t sched_domain_node_span(int node) } #endif +int sched_smt_power_savings = 0, sched_mc_power_savings = 0; /* * At the moment, CONFIG_SCHED_SMT is never defined, but leave it in so we * can switch it on easily if needed. @@ -6113,37 +6194,72 @@ static int build_sched_domains(const cpumask_t *cpu_map) #endif /* Calculate CPU power for physical packages and nodes */ +#ifdef CONFIG_SCHED_SMT for_each_cpu_mask(i, *cpu_map) { - int power; struct sched_domain *sd; -#ifdef CONFIG_SCHED_SMT sd = &per_cpu(cpu_domains, i); - power = SCHED_LOAD_SCALE; - sd->groups->cpu_power = power; + sd->groups->cpu_power = SCHED_LOAD_SCALE; + } #endif #ifdef CONFIG_SCHED_MC + for_each_cpu_mask(i, *cpu_map) { + int power; + struct sched_domain *sd; sd = &per_cpu(core_domains, i); - power = SCHED_LOAD_SCALE + (cpus_weight(sd->groups->cpumask)-1) + if (sched_smt_power_savings) + power = SCHED_LOAD_SCALE * cpus_weight(sd->groups->cpumask); + else + power = SCHED_LOAD_SCALE + (cpus_weight(sd->groups->cpumask)-1) * SCHED_LOAD_SCALE / 10; sd->groups->cpu_power = power; + } +#endif + for_each_cpu_mask(i, *cpu_map) { + struct sched_domain *sd; +#ifdef CONFIG_SCHED_MC sd = &per_cpu(phys_domains, i); + if (i != first_cpu(sd->groups->cpumask)) + continue; - /* - * This has to be < 2 * SCHED_LOAD_SCALE - * Lets keep it SCHED_LOAD_SCALE, so that - * while calculating NUMA group's cpu_power - * we can simply do - * numa_group->cpu_power += phys_group->cpu_power; - * - * See "only add power once for each physical pkg" - * comment below - */ - sd->groups->cpu_power = SCHED_LOAD_SCALE; + sd->groups->cpu_power = 0; + if (sched_mc_power_savings || sched_smt_power_savings) { + int j; + + for_each_cpu_mask(j, sd->groups->cpumask) { + struct sched_domain *sd1; + sd1 = &per_cpu(core_domains, j); + /* + * for each core we will add once + * to the group in physical domain + */ + if (j != first_cpu(sd1->groups->cpumask)) + continue; + + if (sched_smt_power_savings) + sd->groups->cpu_power += sd1->groups->cpu_power; + else + sd->groups->cpu_power += SCHED_LOAD_SCALE; + } + } else + /* + * This has to be < 2 * SCHED_LOAD_SCALE + * Lets keep it SCHED_LOAD_SCALE, so that + * while calculating NUMA group's cpu_power + * we can simply do + * numa_group->cpu_power += phys_group->cpu_power; + * + * See "only add power once for each physical pkg" + * comment below + */ + sd->groups->cpu_power = SCHED_LOAD_SCALE; #else + int power; sd = &per_cpu(phys_domains, i); - power = SCHED_LOAD_SCALE + SCHED_LOAD_SCALE * - (cpus_weight(sd->groups->cpumask)-1) / 10; + if (sched_smt_power_savings) + power = SCHED_LOAD_SCALE * cpus_weight(sd->groups->cpumask); + else + power = SCHED_LOAD_SCALE; sd->groups->cpu_power = power; #endif } @@ -6244,6 +6360,80 @@ int partition_sched_domains(cpumask_t *partition1, cpumask_t *partition2) return err; } +#if defined(CONFIG_SCHED_MC) || defined(CONFIG_SCHED_SMT) +int arch_reinit_sched_domains(void) +{ + int err; + + lock_cpu_hotplug(); + detach_destroy_domains(&cpu_online_map); + err = arch_init_sched_domains(&cpu_online_map); + unlock_cpu_hotplug(); + + return err; +} + +static ssize_t sched_power_savings_store(const char *buf, size_t count, int smt) +{ + int ret; + + if (buf[0] != '0' && buf[0] != '1') + return -EINVAL; + + if (smt) + sched_smt_power_savings = (buf[0] == '1'); + else + sched_mc_power_savings = (buf[0] == '1'); + + ret = arch_reinit_sched_domains(); + + return ret ? ret : count; +} + +int sched_create_sysfs_power_savings_entries(struct sysdev_class *cls) +{ + int err = 0; +#ifdef CONFIG_SCHED_SMT + if (smt_capable()) + err = sysfs_create_file(&cls->kset.kobj, + &attr_sched_smt_power_savings.attr); +#endif +#ifdef CONFIG_SCHED_MC + if (!err && mc_capable()) + err = sysfs_create_file(&cls->kset.kobj, + &attr_sched_mc_power_savings.attr); +#endif + return err; +} +#endif + +#ifdef CONFIG_SCHED_MC +static ssize_t sched_mc_power_savings_show(struct sys_device *dev, char *page) +{ + return sprintf(page, "%u\n", sched_mc_power_savings); +} +static ssize_t sched_mc_power_savings_store(struct sys_device *dev, const char *buf, size_t count) +{ + return sched_power_savings_store(buf, count, 0); +} +SYSDEV_ATTR(sched_mc_power_savings, 0644, sched_mc_power_savings_show, + sched_mc_power_savings_store); +#endif + +#ifdef CONFIG_SCHED_SMT +static ssize_t sched_smt_power_savings_show(struct sys_device *dev, char *page) +{ + return sprintf(page, "%u\n", sched_smt_power_savings); +} +static ssize_t sched_smt_power_savings_store(struct sys_device *dev, const char *buf, size_t count) +{ + return sched_power_savings_store(buf, count, 1); +} +SYSDEV_ATTR(sched_smt_power_savings, 0644, sched_smt_power_savings_show, + sched_smt_power_savings_store); +#endif + + #ifdef CONFIG_HOTPLUG_CPU /* * Force a reinitialization of the sched domains hierarchy. The domains -- cgit v1.2.3-70-g09d2 From e2970f2fb6950183a34e8545faa093eb49d186e1 Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Tue, 27 Jun 2006 02:54:47 -0700 Subject: [PATCH] pi-futex: futex code cleanups We are pleased to announce "lightweight userspace priority inheritance" (PI) support for futexes. The following patchset and glibc patch implements it, ontop of the robust-futexes patchset which is included in 2.6.16-mm1. We are calling it lightweight for 3 reasons: - in the user-space fastpath a PI-enabled futex involves no kernel work (or any other PI complexity) at all. No registration, no extra kernel calls - just pure fast atomic ops in userspace. - in the slowpath (in the lock-contention case), the system call and scheduling pattern is in fact better than that of normal futexes, due to the 'integrated' nature of FUTEX_LOCK_PI. [more about that further down] - the in-kernel PI implementation is streamlined around the mutex abstraction, with strict rules that keep the implementation relatively simple: only a single owner may own a lock (i.e. no read-write lock support), only the owner may unlock a lock, no recursive locking, etc. Priority Inheritance - why, oh why??? ------------------------------------- Many of you heard the horror stories about the evil PI code circling Linux for years, which makes no real sense at all and is only used by buggy applications and which has horrible overhead. Some of you have dreaded this very moment, when someone actually submits working PI code ;-) So why would we like to see PI support for futexes? We'd like to see it done purely for technological reasons. We dont think it's a buggy concept, we think it's useful functionality to offer to applications, which functionality cannot be achieved in other ways. We also think it's the right thing to do, and we think we've got the right arguments and the right numbers to prove that. We also believe that we can address all the counter-arguments as well. For these reasons (and the reasons outlined below) we are submitting this patch-set for upstream kernel inclusion. What are the benefits of PI? The short reply: ---------------- User-space PI helps achieving/improving determinism for user-space applications. In the best-case, it can help achieve determinism and well-bound latencies. Even in the worst-case, PI will improve the statistical distribution of locking related application delays. The longer reply: ----------------- Firstly, sharing locks between multiple tasks is a common programming technique that often cannot be replaced with lockless algorithms. As we can see it in the kernel [which is a quite complex program in itself], lockless structures are rather the exception than the norm - the current ratio of lockless vs. locky code for shared data structures is somewhere between 1:10 and 1:100. Lockless is hard, and the complexity of lockless algorithms often endangers to ability to do robust reviews of said code. I.e. critical RT apps often choose lock structures to protect critical data structures, instead of lockless algorithms. Furthermore, there are cases (like shared hardware, or other resource limits) where lockless access is mathematically impossible. Media players (such as Jack) are an example of reasonable application design with multiple tasks (with multiple priority levels) sharing short-held locks: for example, a highprio audio playback thread is combined with medium-prio construct-audio-data threads and low-prio display-colory-stuff threads. Add video and decoding to the mix and we've got even more priority levels. So once we accept that synchronization objects (locks) are an unavoidable fact of life, and once we accept that multi-task userspace apps have a very fair expectation of being able to use locks, we've got to think about how to offer the option of a deterministic locking implementation to user-space. Most of the technical counter-arguments against doing priority inheritance only apply to kernel-space locks. But user-space locks are different, there we cannot disable interrupts or make the task non-preemptible in a critical section, so the 'use spinlocks' argument does not apply (user-space spinlocks have the same priority inversion problems as other user-space locking constructs). Fact is, pretty much the only technique that currently enables good determinism for userspace locks (such as futex-based pthread mutexes) is priority inheritance: Currently (without PI), if a high-prio and a low-prio task shares a lock [this is a quite common scenario for most non-trivial RT applications], even if all critical sections are coded carefully to be deterministic (i.e. all critical sections are short in duration and only execute a limited number of instructions), the kernel cannot guarantee any deterministic execution of the high-prio task: any medium-priority task could preempt the low-prio task while it holds the shared lock and executes the critical section, and could delay it indefinitely. Implementation: --------------- As mentioned before, the userspace fastpath of PI-enabled pthread mutexes involves no kernel work at all - they behave quite similarly to normal futex-based locks: a 0 value means unlocked, and a value==TID means locked. (This is the same method as used by list-based robust futexes.) Userspace uses atomic ops to lock/unlock these mutexes without entering the kernel. To handle the slowpath, we have added two new futex ops: FUTEX_LOCK_PI FUTEX_UNLOCK_PI If the lock-acquire fastpath fails, [i.e. an atomic transition from 0 to TID fails], then FUTEX_LOCK_PI is called. The kernel does all the remaining work: if there is no futex-queue attached to the futex address yet then the code looks up the task that owns the futex [it has put its own TID into the futex value], and attaches a 'PI state' structure to the futex-queue. The pi_state includes an rt-mutex, which is a PI-aware, kernel-based synchronization object. The 'other' task is made the owner of the rt-mutex, and the FUTEX_WAITERS bit is atomically set in the futex value. Then this task tries to lock the rt-mutex, on which it blocks. Once it returns, it has the mutex acquired, and it sets the futex value to its own TID and returns. Userspace has no other work to perform - it now owns the lock, and futex value contains FUTEX_WAITERS|TID. If the unlock side fastpath succeeds, [i.e. userspace manages to do a TID -> 0 atomic transition of the futex value], then no kernel work is triggered. If the unlock fastpath fails (because the FUTEX_WAITERS bit is set), then FUTEX_UNLOCK_PI is called, and the kernel unlocks the futex on the behalf of userspace - and it also unlocks the attached pi_state->rt_mutex and thus wakes up any potential waiters. Note that under this approach, contrary to other PI-futex approaches, there is no prior 'registration' of a PI-futex. [which is not quite possible anyway, due to existing ABI properties of pthread mutexes.] Also, under this scheme, 'robustness' and 'PI' are two orthogonal properties of futexes, and all four combinations are possible: futex, robust-futex, PI-futex, robust+PI-futex. glibc support: -------------- Ulrich Drepper and Jakub Jelinek have written glibc support for PI-futexes (and robust futexes), enabling robust and PI (PTHREAD_PRIO_INHERIT) POSIX mutexes. (PTHREAD_PRIO_PROTECT support will be added later on too, no additional kernel changes are needed for that). [NOTE: The glibc patch is obviously inofficial and unsupported without matching upstream kernel functionality.] the patch-queue and the glibc patch can also be downloaded from: http://redhat.com/~mingo/PI-futex-patches/ Many thanks go to the people who helped us create this kernel feature: Steven Rostedt, Esben Nielsen, Benedikt Spranger, Daniel Walker, John Cooper, Arjan van de Ven, Oleg Nesterov and others. Credits for related prior projects goes to Dirk Grambow, Inaky Perez-Gonzalez, Bill Huey and many others. Clean up the futex code, before adding more features to it: - use u32 as the futex field type - that's the ABI - use __user and pointers to u32 instead of unsigned long - code style / comment style cleanups - rename hash-bucket name from 'bh' to 'hb'. I checked the pre and post futex.o object files to make sure this patch has no code effects. Signed-off-by: Ingo Molnar Signed-off-by: Thomas Gleixner Signed-off-by: Arjan van de Ven Cc: Ulrich Drepper Cc: Jakub Jelinek Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/futex.h | 5 +- include/linux/syscalls.h | 4 +- kernel/futex.c | 245 +++++++++++++++++++++++++---------------------- kernel/futex_compat.c | 3 +- 4 files changed, 135 insertions(+), 122 deletions(-) (limited to 'include') diff --git a/include/linux/futex.h b/include/linux/futex.h index 966a5b3da43..f05a3f46932 100644 --- a/include/linux/futex.h +++ b/include/linux/futex.h @@ -90,9 +90,8 @@ struct robust_list_head { */ #define ROBUST_LIST_LIMIT 2048 -long do_futex(unsigned long uaddr, int op, int val, - unsigned long timeout, unsigned long uaddr2, int val2, - int val3); +long do_futex(u32 __user *uaddr, int op, u32 val, unsigned long timeout, + u32 __user *uaddr2, u32 val2, u32 val3); extern int handle_futex_death(u32 __user *uaddr, struct task_struct *curr); diff --git a/include/linux/syscalls.h b/include/linux/syscalls.h index 33785b79d54..008f04c5673 100644 --- a/include/linux/syscalls.h +++ b/include/linux/syscalls.h @@ -174,9 +174,9 @@ asmlinkage long sys_waitid(int which, pid_t pid, int options, struct rusage __user *ru); asmlinkage long sys_waitpid(pid_t pid, int __user *stat_addr, int options); asmlinkage long sys_set_tid_address(int __user *tidptr); -asmlinkage long sys_futex(u32 __user *uaddr, int op, int val, +asmlinkage long sys_futex(u32 __user *uaddr, int op, u32 val, struct timespec __user *utime, u32 __user *uaddr2, - int val3); + u32 val3); asmlinkage long sys_init_module(void __user *umod, unsigned long len, const char __user *uargs); diff --git a/kernel/futex.c b/kernel/futex.c index e1a380c77a5..50356fb5d72 100644 --- a/kernel/futex.c +++ b/kernel/futex.c @@ -63,7 +63,7 @@ union futex_key { int offset; } shared; struct { - unsigned long uaddr; + unsigned long address; struct mm_struct *mm; int offset; } private; @@ -87,13 +87,13 @@ struct futex_q { struct list_head list; wait_queue_head_t waiters; - /* Which hash list lock to use. */ + /* Which hash list lock to use: */ spinlock_t *lock_ptr; - /* Key which the futex is hashed on. */ + /* Key which the futex is hashed on: */ union futex_key key; - /* For fd, sigio sent using these. */ + /* For fd, sigio sent using these: */ int fd; struct file *filp; }; @@ -144,8 +144,9 @@ static inline int match_futex(union futex_key *key1, union futex_key *key2) * * Should be called with ¤t->mm->mmap_sem but NOT any spinlocks. */ -static int get_futex_key(unsigned long uaddr, union futex_key *key) +static int get_futex_key(u32 __user *uaddr, union futex_key *key) { + unsigned long address = (unsigned long)uaddr; struct mm_struct *mm = current->mm; struct vm_area_struct *vma; struct page *page; @@ -154,16 +155,16 @@ static int get_futex_key(unsigned long uaddr, union futex_key *key) /* * The futex address must be "naturally" aligned. */ - key->both.offset = uaddr % PAGE_SIZE; + key->both.offset = address % PAGE_SIZE; if (unlikely((key->both.offset % sizeof(u32)) != 0)) return -EINVAL; - uaddr -= key->both.offset; + address -= key->both.offset; /* * The futex is hashed differently depending on whether * it's in a shared or private mapping. So check vma first. */ - vma = find_extend_vma(mm, uaddr); + vma = find_extend_vma(mm, address); if (unlikely(!vma)) return -EFAULT; @@ -184,7 +185,7 @@ static int get_futex_key(unsigned long uaddr, union futex_key *key) */ if (likely(!(vma->vm_flags & VM_MAYSHARE))) { key->private.mm = mm; - key->private.uaddr = uaddr; + key->private.address = address; return 0; } @@ -194,7 +195,7 @@ static int get_futex_key(unsigned long uaddr, union futex_key *key) key->shared.inode = vma->vm_file->f_dentry->d_inode; key->both.offset++; /* Bit 0 of offset indicates inode-based key. */ if (likely(!(vma->vm_flags & VM_NONLINEAR))) { - key->shared.pgoff = (((uaddr - vma->vm_start) >> PAGE_SHIFT) + key->shared.pgoff = (((address - vma->vm_start) >> PAGE_SHIFT) + vma->vm_pgoff); return 0; } @@ -205,7 +206,7 @@ static int get_futex_key(unsigned long uaddr, union futex_key *key) * from swap. But that's a lot of code to duplicate here * for a rare case, so we simply fetch the page. */ - err = get_user_pages(current, mm, uaddr, 1, 0, 0, &page, NULL); + err = get_user_pages(current, mm, address, 1, 0, 0, &page, NULL); if (err >= 0) { key->shared.pgoff = page->index << (PAGE_CACHE_SHIFT - PAGE_SHIFT); @@ -246,12 +247,12 @@ static void drop_key_refs(union futex_key *key) } } -static inline int get_futex_value_locked(int *dest, int __user *from) +static inline int get_futex_value_locked(u32 *dest, u32 __user *from) { int ret; inc_preempt_count(); - ret = __copy_from_user_inatomic(dest, from, sizeof(int)); + ret = __copy_from_user_inatomic(dest, from, sizeof(u32)); dec_preempt_count(); return ret ? -EFAULT : 0; @@ -288,12 +289,12 @@ static void wake_futex(struct futex_q *q) * Wake up all waiters hashed on the physical page that is mapped * to this virtual address: */ -static int futex_wake(unsigned long uaddr, int nr_wake) +static int futex_wake(u32 __user *uaddr, int nr_wake) { - union futex_key key; - struct futex_hash_bucket *bh; - struct list_head *head; + struct futex_hash_bucket *hb; struct futex_q *this, *next; + struct list_head *head; + union futex_key key; int ret; down_read(¤t->mm->mmap_sem); @@ -302,9 +303,9 @@ static int futex_wake(unsigned long uaddr, int nr_wake) if (unlikely(ret != 0)) goto out; - bh = hash_futex(&key); - spin_lock(&bh->lock); - head = &bh->chain; + hb = hash_futex(&key); + spin_lock(&hb->lock); + head = &hb->chain; list_for_each_entry_safe(this, next, head, list) { if (match_futex (&this->key, &key)) { @@ -314,7 +315,7 @@ static int futex_wake(unsigned long uaddr, int nr_wake) } } - spin_unlock(&bh->lock); + spin_unlock(&hb->lock); out: up_read(¤t->mm->mmap_sem); return ret; @@ -324,10 +325,12 @@ out: * Wake up all waiters hashed on the physical page that is mapped * to this virtual address: */ -static int futex_wake_op(unsigned long uaddr1, unsigned long uaddr2, int nr_wake, int nr_wake2, int op) +static int +futex_wake_op(u32 __user *uaddr1, u32 __user *uaddr2, + int nr_wake, int nr_wake2, int op) { union futex_key key1, key2; - struct futex_hash_bucket *bh1, *bh2; + struct futex_hash_bucket *hb1, *hb2; struct list_head *head; struct futex_q *this, *next; int ret, op_ret, attempt = 0; @@ -342,27 +345,29 @@ retryfull: if (unlikely(ret != 0)) goto out; - bh1 = hash_futex(&key1); - bh2 = hash_futex(&key2); + hb1 = hash_futex(&key1); + hb2 = hash_futex(&key2); retry: - if (bh1 < bh2) - spin_lock(&bh1->lock); - spin_lock(&bh2->lock); - if (bh1 > bh2) - spin_lock(&bh1->lock); + if (hb1 < hb2) + spin_lock(&hb1->lock); + spin_lock(&hb2->lock); + if (hb1 > hb2) + spin_lock(&hb1->lock); - op_ret = futex_atomic_op_inuser(op, (int __user *)uaddr2); + op_ret = futex_atomic_op_inuser(op, uaddr2); if (unlikely(op_ret < 0)) { - int dummy; + u32 dummy; - spin_unlock(&bh1->lock); - if (bh1 != bh2) - spin_unlock(&bh2->lock); + spin_unlock(&hb1->lock); + if (hb1 != hb2) + spin_unlock(&hb2->lock); #ifndef CONFIG_MMU - /* we don't get EFAULT from MMU faults if we don't have an MMU, - * but we might get them from range checking */ + /* + * we don't get EFAULT from MMU faults if we don't have an MMU, + * but we might get them from range checking + */ ret = op_ret; goto out; #endif @@ -372,23 +377,26 @@ retry: goto out; } - /* futex_atomic_op_inuser needs to both read and write + /* + * futex_atomic_op_inuser needs to both read and write * *(int __user *)uaddr2, but we can't modify it * non-atomically. Therefore, if get_user below is not * enough, we need to handle the fault ourselves, while - * still holding the mmap_sem. */ + * still holding the mmap_sem. + */ if (attempt++) { struct vm_area_struct * vma; struct mm_struct *mm = current->mm; + unsigned long address = (unsigned long)uaddr2; ret = -EFAULT; if (attempt >= 2 || - !(vma = find_vma(mm, uaddr2)) || - vma->vm_start > uaddr2 || + !(vma = find_vma(mm, address)) || + vma->vm_start > address || !(vma->vm_flags & VM_WRITE)) goto out; - switch (handle_mm_fault(mm, vma, uaddr2, 1)) { + switch (handle_mm_fault(mm, vma, address, 1)) { case VM_FAULT_MINOR: current->min_flt++; break; @@ -401,18 +409,20 @@ retry: goto retry; } - /* If we would have faulted, release mmap_sem, - * fault it in and start all over again. */ + /* + * If we would have faulted, release mmap_sem, + * fault it in and start all over again. + */ up_read(¤t->mm->mmap_sem); - ret = get_user(dummy, (int __user *)uaddr2); + ret = get_user(dummy, uaddr2); if (ret) return ret; goto retryfull; } - head = &bh1->chain; + head = &hb1->chain; list_for_each_entry_safe(this, next, head, list) { if (match_futex (&this->key, &key1)) { @@ -423,7 +433,7 @@ retry: } if (op_ret > 0) { - head = &bh2->chain; + head = &hb2->chain; op_ret = 0; list_for_each_entry_safe(this, next, head, list) { @@ -436,9 +446,9 @@ retry: ret += op_ret; } - spin_unlock(&bh1->lock); - if (bh1 != bh2) - spin_unlock(&bh2->lock); + spin_unlock(&hb1->lock); + if (hb1 != hb2) + spin_unlock(&hb2->lock); out: up_read(¤t->mm->mmap_sem); return ret; @@ -448,11 +458,11 @@ out: * Requeue all waiters hashed on one physical page to another * physical page. */ -static int futex_requeue(unsigned long uaddr1, unsigned long uaddr2, - int nr_wake, int nr_requeue, int *valp) +static int futex_requeue(u32 __user *uaddr1, u32 __user *uaddr2, + int nr_wake, int nr_requeue, u32 *cmpval) { union futex_key key1, key2; - struct futex_hash_bucket *bh1, *bh2; + struct futex_hash_bucket *hb1, *hb2; struct list_head *head1; struct futex_q *this, *next; int ret, drop_count = 0; @@ -467,68 +477,69 @@ static int futex_requeue(unsigned long uaddr1, unsigned long uaddr2, if (unlikely(ret != 0)) goto out; - bh1 = hash_futex(&key1); - bh2 = hash_futex(&key2); + hb1 = hash_futex(&key1); + hb2 = hash_futex(&key2); - if (bh1 < bh2) - spin_lock(&bh1->lock); - spin_lock(&bh2->lock); - if (bh1 > bh2) - spin_lock(&bh1->lock); + if (hb1 < hb2) + spin_lock(&hb1->lock); + spin_lock(&hb2->lock); + if (hb1 > hb2) + spin_lock(&hb1->lock); - if (likely(valp != NULL)) { - int curval; + if (likely(cmpval != NULL)) { + u32 curval; - ret = get_futex_value_locked(&curval, (int __user *)uaddr1); + ret = get_futex_value_locked(&curval, uaddr1); if (unlikely(ret)) { - spin_unlock(&bh1->lock); - if (bh1 != bh2) - spin_unlock(&bh2->lock); + spin_unlock(&hb1->lock); + if (hb1 != hb2) + spin_unlock(&hb2->lock); - /* If we would have faulted, release mmap_sem, fault + /* + * If we would have faulted, release mmap_sem, fault * it in and start all over again. */ up_read(¤t->mm->mmap_sem); - ret = get_user(curval, (int __user *)uaddr1); + ret = get_user(curval, uaddr1); if (!ret) goto retry; return ret; } - if (curval != *valp) { + if (curval != *cmpval) { ret = -EAGAIN; goto out_unlock; } } - head1 = &bh1->chain; + head1 = &hb1->chain; list_for_each_entry_safe(this, next, head1, list) { if (!match_futex (&this->key, &key1)) continue; if (++ret <= nr_wake) { wake_futex(this); } else { - list_move_tail(&this->list, &bh2->chain); - this->lock_ptr = &bh2->lock; + list_move_tail(&this->list, &hb2->chain); + this->lock_ptr = &hb2->lock; this->key = key2; get_key_refs(&key2); drop_count++; if (ret - nr_wake >= nr_requeue) break; - /* Make sure to stop if key1 == key2 */ - if (head1 == &bh2->chain && head1 != &next->list) + /* Make sure to stop if key1 == key2: */ + if (head1 == &hb2->chain && head1 != &next->list) head1 = &this->list; } } out_unlock: - spin_unlock(&bh1->lock); - if (bh1 != bh2) - spin_unlock(&bh2->lock); + spin_unlock(&hb1->lock); + if (hb1 != hb2) + spin_unlock(&hb2->lock); /* drop_key_refs() must be called outside the spinlocks. */ while (--drop_count >= 0) @@ -543,7 +554,7 @@ out: static inline struct futex_hash_bucket * queue_lock(struct futex_q *q, int fd, struct file *filp) { - struct futex_hash_bucket *bh; + struct futex_hash_bucket *hb; q->fd = fd; q->filp = filp; @@ -551,23 +562,23 @@ queue_lock(struct futex_q *q, int fd, struct file *filp) init_waitqueue_head(&q->waiters); get_key_refs(&q->key); - bh = hash_futex(&q->key); - q->lock_ptr = &bh->lock; + hb = hash_futex(&q->key); + q->lock_ptr = &hb->lock; - spin_lock(&bh->lock); - return bh; + spin_lock(&hb->lock); + return hb; } -static inline void __queue_me(struct futex_q *q, struct futex_hash_bucket *bh) +static inline void __queue_me(struct futex_q *q, struct futex_hash_bucket *hb) { - list_add_tail(&q->list, &bh->chain); - spin_unlock(&bh->lock); + list_add_tail(&q->list, &hb->chain); + spin_unlock(&hb->lock); } static inline void -queue_unlock(struct futex_q *q, struct futex_hash_bucket *bh) +queue_unlock(struct futex_q *q, struct futex_hash_bucket *hb) { - spin_unlock(&bh->lock); + spin_unlock(&hb->lock); drop_key_refs(&q->key); } @@ -579,16 +590,17 @@ queue_unlock(struct futex_q *q, struct futex_hash_bucket *bh) /* The key must be already stored in q->key. */ static void queue_me(struct futex_q *q, int fd, struct file *filp) { - struct futex_hash_bucket *bh; - bh = queue_lock(q, fd, filp); - __queue_me(q, bh); + struct futex_hash_bucket *hb; + + hb = queue_lock(q, fd, filp); + __queue_me(q, hb); } /* Return 1 if we were still queued (ie. 0 means we were woken) */ static int unqueue_me(struct futex_q *q) { - int ret = 0; spinlock_t *lock_ptr; + int ret = 0; /* In the common case we don't take the spinlock, which is nice. */ retry: @@ -622,12 +634,13 @@ static int unqueue_me(struct futex_q *q) return ret; } -static int futex_wait(unsigned long uaddr, int val, unsigned long time) +static int futex_wait(u32 __user *uaddr, u32 val, unsigned long time) { DECLARE_WAITQUEUE(wait, current); - int ret, curval; + struct futex_hash_bucket *hb; struct futex_q q; - struct futex_hash_bucket *bh; + u32 uval; + int ret; retry: down_read(¤t->mm->mmap_sem); @@ -636,7 +649,7 @@ static int futex_wait(unsigned long uaddr, int val, unsigned long time) if (unlikely(ret != 0)) goto out_release_sem; - bh = queue_lock(&q, -1, NULL); + hb = queue_lock(&q, -1, NULL); /* * Access the page AFTER the futex is queued. @@ -658,31 +671,31 @@ static int futex_wait(unsigned long uaddr, int val, unsigned long time) * We hold the mmap semaphore, so the mapping cannot have changed * since we looked it up in get_futex_key. */ - - ret = get_futex_value_locked(&curval, (int __user *)uaddr); + ret = get_futex_value_locked(&uval, uaddr); if (unlikely(ret)) { - queue_unlock(&q, bh); + queue_unlock(&q, hb); - /* If we would have faulted, release mmap_sem, fault it in and + /* + * If we would have faulted, release mmap_sem, fault it in and * start all over again. */ up_read(¤t->mm->mmap_sem); - ret = get_user(curval, (int __user *)uaddr); + ret = get_user(uval, uaddr); if (!ret) goto retry; return ret; } - if (curval != val) { + if (uval != val) { ret = -EWOULDBLOCK; - queue_unlock(&q, bh); + queue_unlock(&q, hb); goto out_release_sem; } /* Only actually queue if *uaddr contained val. */ - __queue_me(&q, bh); + __queue_me(&q, hb); /* * Now the futex is queued and we have checked the data, we @@ -720,8 +733,10 @@ static int futex_wait(unsigned long uaddr, int val, unsigned long time) return 0; if (time == 0) return -ETIMEDOUT; - /* We expect signal_pending(current), but another thread may - * have handled it for us already. */ + /* + * We expect signal_pending(current), but another thread may + * have handled it for us already. + */ return -EINTR; out_release_sem: @@ -735,6 +750,7 @@ static int futex_close(struct inode *inode, struct file *filp) unqueue_me(q); kfree(q); + return 0; } @@ -766,7 +782,7 @@ static struct file_operations futex_fops = { * Signal allows caller to avoid the race which would occur if they * set the sigio stuff up afterwards. */ -static int futex_fd(unsigned long uaddr, int signal) +static int futex_fd(u32 __user *uaddr, int signal) { struct futex_q *q; struct file *filp; @@ -937,7 +953,7 @@ retry: goto retry; if (uval & FUTEX_WAITERS) - futex_wake((unsigned long)uaddr, 1); + futex_wake(uaddr, 1); } return 0; } @@ -999,8 +1015,8 @@ void exit_robust_list(struct task_struct *curr) } } -long do_futex(unsigned long uaddr, int op, int val, unsigned long timeout, - unsigned long uaddr2, int val2, int val3) +long do_futex(u32 __user *uaddr, int op, u32 val, unsigned long timeout, + u32 __user *uaddr2, u32 val2, u32 val3) { int ret; @@ -1031,13 +1047,13 @@ long do_futex(unsigned long uaddr, int op, int val, unsigned long timeout, } -asmlinkage long sys_futex(u32 __user *uaddr, int op, int val, +asmlinkage long sys_futex(u32 __user *uaddr, int op, u32 val, struct timespec __user *utime, u32 __user *uaddr2, - int val3) + u32 val3) { struct timespec t; unsigned long timeout = MAX_SCHEDULE_TIMEOUT; - int val2 = 0; + u32 val2 = 0; if (utime && (op == FUTEX_WAIT)) { if (copy_from_user(&t, utime, sizeof(t)) != 0) @@ -1050,10 +1066,9 @@ asmlinkage long sys_futex(u32 __user *uaddr, int op, int val, * requeue parameter in 'utime' if op == FUTEX_REQUEUE. */ if (op >= FUTEX_REQUEUE) - val2 = (int) (unsigned long) utime; + val2 = (u32) (unsigned long) utime; - return do_futex((unsigned long)uaddr, op, val, timeout, - (unsigned long)uaddr2, val2, val3); + return do_futex(uaddr, op, val, timeout, uaddr2, val2, val3); } static int futexfs_get_sb(struct file_system_type *fs_type, diff --git a/kernel/futex_compat.c b/kernel/futex_compat.c index 1ab6a0ea3d1..7e57c31670a 100644 --- a/kernel/futex_compat.c +++ b/kernel/futex_compat.c @@ -139,6 +139,5 @@ asmlinkage long compat_sys_futex(u32 __user *uaddr, int op, u32 val, if (op >= FUTEX_REQUEUE) val2 = (int) (unsigned long) utime; - return do_futex((unsigned long)uaddr, op, val, timeout, - (unsigned long)uaddr2, val2, val3); + return do_futex(uaddr, op, val, timeout, uaddr2, val2, val3); } -- cgit v1.2.3-70-g09d2 From f9b8404cf8f8456dfa83459510762b700dc00385 Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Tue, 27 Jun 2006 02:54:49 -0700 Subject: [PATCH] pi-futex: introduce debug_check_no_locks_freed() Add debug_check_no_locks_freed(), as a central inline to add bad-lock-free-debugging functionality to. Signed-off-by: Ingo Molnar Signed-off-by: Thomas Gleixner Signed-off-by: Arjan van de Ven Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/i386/mm/pageattr.c | 4 ++-- include/linux/mm.h | 10 ++++++++-- mm/page_alloc.c | 4 ++-- mm/slab.c | 2 +- 4 files changed, 13 insertions(+), 7 deletions(-) (limited to 'include') diff --git a/arch/i386/mm/pageattr.c b/arch/i386/mm/pageattr.c index 0887b34bc59..353a836ed63 100644 --- a/arch/i386/mm/pageattr.c +++ b/arch/i386/mm/pageattr.c @@ -229,8 +229,8 @@ void kernel_map_pages(struct page *page, int numpages, int enable) if (PageHighMem(page)) return; if (!enable) - mutex_debug_check_no_locks_freed(page_address(page), - numpages * PAGE_SIZE); + debug_check_no_locks_freed(page_address(page), + numpages * PAGE_SIZE); /* the return value is ignored - the calls cannot fail, * large pages are disabled at boot time. diff --git a/include/linux/mm.h b/include/linux/mm.h index ff1fa87df8d..0ac255720f3 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -1030,13 +1030,19 @@ static inline void vm_stat_account(struct mm_struct *mm, } #endif /* CONFIG_PROC_FS */ +static inline void +debug_check_no_locks_freed(const void *from, unsigned long len) +{ + mutex_debug_check_no_locks_freed(from, len); +} + #ifndef CONFIG_DEBUG_PAGEALLOC static inline void kernel_map_pages(struct page *page, int numpages, int enable) { if (!PageHighMem(page) && !enable) - mutex_debug_check_no_locks_freed(page_address(page), - numpages * PAGE_SIZE); + debug_check_no_locks_freed(page_address(page), + numpages * PAGE_SIZE); } #endif diff --git a/mm/page_alloc.c b/mm/page_alloc.c index dafd12ec7a0..084a2de7e52 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -446,8 +446,8 @@ static void __free_pages_ok(struct page *page, unsigned int order) arch_free_page(page, order); if (!PageHighMem(page)) - mutex_debug_check_no_locks_freed(page_address(page), - PAGE_SIZE< Date: Tue, 27 Jun 2006 02:54:50 -0700 Subject: [PATCH] pi-futex: introduce WARN_ON_SMP Introduce a new WARN_ON variant: WARN_ON_SMP(cond). Signed-off-by: Ingo Molnar Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/asm-generic/bug.h | 6 ++++++ 1 file changed, 6 insertions(+) (limited to 'include') diff --git a/include/asm-generic/bug.h b/include/asm-generic/bug.h index 845cb67ad8e..8ceab7bcd8b 100644 --- a/include/asm-generic/bug.h +++ b/include/asm-generic/bug.h @@ -51,4 +51,10 @@ __ret; \ }) +#ifdef CONFIG_SMP +# define WARN_ON_SMP(x) WARN_ON(x) +#else +# define WARN_ON_SMP(x) do { } while (0) +#endif + #endif -- cgit v1.2.3-70-g09d2 From 77ba89c5cf28d5d98a3cae17f67a3e42b102cc25 Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Tue, 27 Jun 2006 02:54:51 -0700 Subject: [PATCH] pi-futex: add plist implementation Add the priority-sorted list (plist) implementation. Signed-off-by: Ingo Molnar Signed-off-by: Thomas Gleixner Signed-off-by: Arjan van de Ven Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/plist.h | 247 ++++++++++++++++++++++++++++++++++++++++++++++++++ lib/Kconfig | 6 ++ lib/Makefile | 1 + lib/plist.c | 118 ++++++++++++++++++++++++ 4 files changed, 372 insertions(+) create mode 100644 include/linux/plist.h create mode 100644 lib/plist.c (limited to 'include') diff --git a/include/linux/plist.h b/include/linux/plist.h new file mode 100644 index 00000000000..3404faef542 --- /dev/null +++ b/include/linux/plist.h @@ -0,0 +1,247 @@ +/* + * Descending-priority-sorted double-linked list + * + * (C) 2002-2003 Intel Corp + * Inaky Perez-Gonzalez . + * + * 2001-2005 (c) MontaVista Software, Inc. + * Daniel Walker + * + * (C) 2005 Thomas Gleixner + * + * Simplifications of the original code by + * Oleg Nesterov + * + * Licensed under the FSF's GNU Public License v2 or later. + * + * Based on simple lists (include/linux/list.h). + * + * This is a priority-sorted list of nodes; each node has a + * priority from INT_MIN (highest) to INT_MAX (lowest). + * + * Addition is O(K), removal is O(1), change of priority of a node is + * O(K) and K is the number of RT priority levels used in the system. + * (1 <= K <= 99) + * + * This list is really a list of lists: + * + * - The tier 1 list is the prio_list, different priority nodes. + * + * - The tier 2 list is the node_list, serialized nodes. + * + * Simple ASCII art explanation: + * + * |HEAD | + * | | + * |prio_list.prev|<------------------------------------| + * |prio_list.next|<->|pl|<->|pl|<--------------->|pl|<-| + * |10 | |10| |21| |21| |21| |40| (prio) + * | | | | | | | | | | | | + * | | | | | | | | | | | | + * |node_list.next|<->|nl|<->|nl|<->|nl|<->|nl|<->|nl|<-| + * |node_list.prev|<------------------------------------| + * + * The nodes on the prio_list list are sorted by priority to simplify + * the insertion of new nodes. There are no nodes with duplicate + * priorites on the list. + * + * The nodes on the node_list is ordered by priority and can contain + * entries which have the same priority. Those entries are ordered + * FIFO + * + * Addition means: look for the prio_list node in the prio_list + * for the priority of the node and insert it before the node_list + * entry of the next prio_list node. If it is the first node of + * that priority, add it to the prio_list in the right position and + * insert it into the serialized node_list list + * + * Removal means remove it from the node_list and remove it from + * the prio_list if the node_list list_head is non empty. In case + * of removal from the prio_list it must be checked whether other + * entries of the same priority are on the list or not. If there + * is another entry of the same priority then this entry has to + * replace the removed entry on the prio_list. If the entry which + * is removed is the only entry of this priority then a simple + * remove from both list is sufficient. + * + * INT_MIN is the highest priority, 0 is the medium highest, INT_MAX + * is lowest priority. + * + * No locking is done, up to the caller. + * + */ +#ifndef _LINUX_PLIST_H_ +#define _LINUX_PLIST_H_ + +#include +#include + +struct plist_head { + struct list_head prio_list; + struct list_head node_list; +#ifdef CONFIG_DEBUG_PI_LIST + spinlock_t *lock; +#endif +}; + +struct plist_node { + int prio; + struct plist_head plist; +}; + +#ifdef CONFIG_DEBUG_PI_LIST +# define PLIST_HEAD_LOCK_INIT(_lock) .lock = _lock +#else +# define PLIST_HEAD_LOCK_INIT(_lock) +#endif + +/** + * #PLIST_HEAD_INIT - static struct plist_head initializer + * + * @head: struct plist_head variable name + */ +#define PLIST_HEAD_INIT(head, _lock) \ +{ \ + .prio_list = LIST_HEAD_INIT((head).prio_list), \ + .node_list = LIST_HEAD_INIT((head).node_list), \ + PLIST_HEAD_LOCK_INIT(&(_lock)) \ +} + +/** + * #PLIST_NODE_INIT - static struct plist_node initializer + * + * @node: struct plist_node variable name + * @__prio: initial node priority + */ +#define PLIST_NODE_INIT(node, __prio) \ +{ \ + .prio = (__prio), \ + .plist = PLIST_HEAD_INIT((node).plist, NULL), \ +} + +/** + * plist_head_init - dynamic struct plist_head initializer + * + * @head: &struct plist_head pointer + */ +static inline void +plist_head_init(struct plist_head *head, spinlock_t *lock) +{ + INIT_LIST_HEAD(&head->prio_list); + INIT_LIST_HEAD(&head->node_list); +#ifdef CONFIG_DEBUG_PI_LIST + head->lock = lock; +#endif +} + +/** + * plist_node_init - Dynamic struct plist_node initializer + * + * @node: &struct plist_node pointer + * @prio: initial node priority + */ +static inline void plist_node_init(struct plist_node *node, int prio) +{ + node->prio = prio; + plist_head_init(&node->plist, NULL); +} + +extern void plist_add(struct plist_node *node, struct plist_head *head); +extern void plist_del(struct plist_node *node, struct plist_head *head); + +/** + * plist_for_each - iterate over the plist + * + * @pos1: the type * to use as a loop counter. + * @head: the head for your list. + */ +#define plist_for_each(pos, head) \ + list_for_each_entry(pos, &(head)->node_list, plist.node_list) + +/** + * plist_for_each_entry_safe - iterate over a plist of given type safe + * against removal of list entry + * + * @pos1: the type * to use as a loop counter. + * @n1: another type * to use as temporary storage + * @head: the head for your list. + */ +#define plist_for_each_safe(pos, n, head) \ + list_for_each_entry_safe(pos, n, &(head)->node_list, plist.node_list) + +/** + * plist_for_each_entry - iterate over list of given type + * + * @pos: the type * to use as a loop counter. + * @head: the head for your list. + * @member: the name of the list_struct within the struct. + */ +#define plist_for_each_entry(pos, head, mem) \ + list_for_each_entry(pos, &(head)->node_list, mem.plist.node_list) + +/** + * plist_for_each_entry_safe - iterate over list of given type safe against + * removal of list entry + * + * @pos: the type * to use as a loop counter. + * @n: another type * to use as temporary storage + * @head: the head for your list. + * @m: the name of the list_struct within the struct. + */ +#define plist_for_each_entry_safe(pos, n, head, m) \ + list_for_each_entry_safe(pos, n, &(head)->node_list, m.plist.node_list) + +/** + * plist_head_empty - return !0 if a plist_head is empty + * + * @head: &struct plist_head pointer + */ +static inline int plist_head_empty(const struct plist_head *head) +{ + return list_empty(&head->node_list); +} + +/** + * plist_node_empty - return !0 if plist_node is not on a list + * + * @node: &struct plist_node pointer + */ +static inline int plist_node_empty(const struct plist_node *node) +{ + return plist_head_empty(&node->plist); +} + +/* All functions below assume the plist_head is not empty. */ + +/** + * plist_first_entry - get the struct for the first entry + * + * @ptr: the &struct plist_head pointer. + * @type: the type of the struct this is embedded in. + * @member: the name of the list_struct within the struct. + */ +#ifdef CONFIG_DEBUG_PI_LIST +# define plist_first_entry(head, type, member) \ +({ \ + WARN_ON(plist_head_empty(head)); \ + container_of(plist_first(head), type, member); \ +}) +#else +# define plist_first_entry(head, type, member) \ + container_of(plist_first(head), type, member) +#endif + +/** + * plist_first - return the first node (and thus, highest priority) + * + * @head: the &struct plist_head pointer + * + * Assumes the plist is _not_ empty. + */ +static inline struct plist_node* plist_first(const struct plist_head *head) +{ + return list_entry(head->node_list.next, + struct plist_node, plist.node_list); +} + +#endif diff --git a/lib/Kconfig b/lib/Kconfig index 3de93357f5a..f6299342b88 100644 --- a/lib/Kconfig +++ b/lib/Kconfig @@ -86,4 +86,10 @@ config TEXTSEARCH_BM config TEXTSEARCH_FSM tristate +# +# plist support is select#ed if needed +# +config PLIST + boolean + endmenu diff --git a/lib/Makefile b/lib/Makefile index 79358ad1f11..10c13c9d782 100644 --- a/lib/Makefile +++ b/lib/Makefile @@ -25,6 +25,7 @@ lib-$(CONFIG_SEMAPHORE_SLEEPERS) += semaphore-sleepers.o lib-$(CONFIG_GENERIC_FIND_NEXT_BIT) += find_next_bit.o lib-$(CONFIG_GENERIC_HWEIGHT) += hweight.o obj-$(CONFIG_LOCK_KERNEL) += kernel_lock.o +obj-$(CONFIG_PLIST) += plist.o obj-$(CONFIG_DEBUG_PREEMPT) += smp_processor_id.o ifneq ($(CONFIG_HAVE_DEC_LOCK),y) diff --git a/lib/plist.c b/lib/plist.c new file mode 100644 index 00000000000..3074a02272f --- /dev/null +++ b/lib/plist.c @@ -0,0 +1,118 @@ +/* + * lib/plist.c + * + * Descending-priority-sorted double-linked list + * + * (C) 2002-2003 Intel Corp + * Inaky Perez-Gonzalez . + * + * 2001-2005 (c) MontaVista Software, Inc. + * Daniel Walker + * + * (C) 2005 Thomas Gleixner + * + * Simplifications of the original code by + * Oleg Nesterov + * + * Licensed under the FSF's GNU Public License v2 or later. + * + * Based on simple lists (include/linux/list.h). + * + * This file contains the add / del functions which are considered to + * be too large to inline. See include/linux/plist.h for further + * information. + */ + +#include +#include + +#ifdef CONFIG_DEBUG_PI_LIST + +static void plist_check_prev_next(struct list_head *t, struct list_head *p, + struct list_head *n) +{ + if (n->prev != p || p->next != n) { + printk("top: %p, n: %p, p: %p\n", t, t->next, t->prev); + printk("prev: %p, n: %p, p: %p\n", p, p->next, p->prev); + printk("next: %p, n: %p, p: %p\n", n, n->next, n->prev); + WARN_ON(1); + } +} + +static void plist_check_list(struct list_head *top) +{ + struct list_head *prev = top, *next = top->next; + + plist_check_prev_next(top, prev, next); + while (next != top) { + prev = next; + next = prev->next; + plist_check_prev_next(top, prev, next); + } +} + +static void plist_check_head(struct plist_head *head) +{ + WARN_ON(!head->lock); + if (head->lock) + WARN_ON_SMP(!spin_is_locked(head->lock)); + plist_check_list(&head->prio_list); + plist_check_list(&head->node_list); +} + +#else +# define plist_check_head(h) do { } while (0) +#endif + +/** + * plist_add - add @node to @head + * + * @node: &struct plist_node pointer + * @head: &struct plist_head pointer + */ +void plist_add(struct plist_node *node, struct plist_head *head) +{ + struct plist_node *iter; + + plist_check_head(head); + WARN_ON(!plist_node_empty(node)); + + list_for_each_entry(iter, &head->prio_list, plist.prio_list) { + if (node->prio < iter->prio) + goto lt_prio; + else if (node->prio == iter->prio) { + iter = list_entry(iter->plist.prio_list.next, + struct plist_node, plist.prio_list); + goto eq_prio; + } + } + +lt_prio: + list_add_tail(&node->plist.prio_list, &iter->plist.prio_list); +eq_prio: + list_add_tail(&node->plist.node_list, &iter->plist.node_list); + + plist_check_head(head); +} + +/** + * plist_del - Remove a @node from plist. + * + * @node: &struct plist_node pointer - entry to be removed + * @head: &struct plist_head pointer - list head + */ +void plist_del(struct plist_node *node, struct plist_head *head) +{ + plist_check_head(head); + + if (!list_empty(&node->plist.prio_list)) { + struct plist_node *next = plist_first(&node->plist); + + list_move_tail(&next->plist.prio_list, &node->plist.prio_list); + list_del_init(&node->plist.prio_list); + } + + list_del_init(&node->plist.node_list); + + plist_check_head(head); +} -- cgit v1.2.3-70-g09d2 From b29739f902ee76a05493fb7d2303490fc75364f4 Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Tue, 27 Jun 2006 02:54:51 -0700 Subject: [PATCH] pi-futex: scheduler support for pi Add framework to boost/unboost the priority of RT tasks. This consists of: - caching the 'normal' priority in ->normal_prio - providing a functions to set/get the priority of the task - make sched_setscheduler() aware of boosting The effective_prio() cleanups also fix a priority-calculation bug pointed out by Andrey Gelman, in set_user_nice(). has_rt_policy() fix: Peter Williams Signed-off-by: Ingo Molnar Signed-off-by: Thomas Gleixner Signed-off-by: Arjan van de Ven Cc: Andrey Gelman Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/init_task.h | 2 + include/linux/sched.h | 21 +++++- kernel/sched.c | 189 +++++++++++++++++++++++++++++++++++++++------- 3 files changed, 181 insertions(+), 31 deletions(-) (limited to 'include') diff --git a/include/linux/init_task.h b/include/linux/init_task.h index e127ef7e8da..678c1a90380 100644 --- a/include/linux/init_task.h +++ b/include/linux/init_task.h @@ -87,6 +87,7 @@ extern struct group_info init_groups; .lock_depth = -1, \ .prio = MAX_PRIO-20, \ .static_prio = MAX_PRIO-20, \ + .normal_prio = MAX_PRIO-20, \ .policy = SCHED_NORMAL, \ .cpus_allowed = CPU_MASK_ALL, \ .mm = NULL, \ @@ -122,6 +123,7 @@ extern struct group_info init_groups; .journal_info = NULL, \ .cpu_timers = INIT_CPU_TIMERS(tsk.cpu_timers), \ .fs_excl = ATOMIC_INIT(0), \ + .pi_lock = SPIN_LOCK_UNLOCKED, \ } diff --git a/include/linux/sched.h b/include/linux/sched.h index 0bc81a151e5..6f167645e7e 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -495,8 +495,11 @@ struct signal_struct { #define MAX_PRIO (MAX_RT_PRIO + 40) -#define rt_task(p) (unlikely((p)->prio < MAX_RT_PRIO)) +#define rt_prio(prio) unlikely((prio) < MAX_RT_PRIO) +#define rt_task(p) rt_prio((p)->prio) #define batch_task(p) (unlikely((p)->policy == SCHED_BATCH)) +#define has_rt_policy(p) \ + unlikely((p)->policy != SCHED_NORMAL && (p)->policy != SCHED_BATCH) /* * Some day this will be a full-fledged user tracking system.. @@ -725,7 +728,7 @@ struct task_struct { #endif #endif int load_weight; /* for niceness load balancing purposes */ - int prio, static_prio; + int prio, static_prio, normal_prio; struct list_head run_list; prio_array_t *array; @@ -852,6 +855,9 @@ struct task_struct { /* Protection of (de-)allocation: mm, files, fs, tty, keyrings */ spinlock_t alloc_lock; + /* Protection of the PI data structures: */ + spinlock_t pi_lock; + #ifdef CONFIG_DEBUG_MUTEXES /* mutex deadlock detection */ struct mutex_waiter *blocked_on; @@ -1018,6 +1024,17 @@ static inline void idle_task_exit(void) {} #endif extern void sched_idle_next(void); + +#ifdef CONFIG_RT_MUTEXES +extern int rt_mutex_getprio(task_t *p); +extern void rt_mutex_setprio(task_t *p, int prio); +#else +static inline int rt_mutex_getprio(task_t *p) +{ + return p->normal_prio; +} +#endif + extern void set_user_nice(task_t *p, long nice); extern int task_prio(const task_t *p); extern int task_nice(const task_t *p); diff --git a/kernel/sched.c b/kernel/sched.c index 15abf083324..08431f07a99 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -354,6 +354,25 @@ static inline void finish_lock_switch(runqueue_t *rq, task_t *prev) } #endif /* __ARCH_WANT_UNLOCKED_CTXSW */ +/* + * __task_rq_lock - lock the runqueue a given task resides on. + * Must be called interrupts disabled. + */ +static inline runqueue_t *__task_rq_lock(task_t *p) + __acquires(rq->lock) +{ + struct runqueue *rq; + +repeat_lock_task: + rq = task_rq(p); + spin_lock(&rq->lock); + if (unlikely(rq != task_rq(p))) { + spin_unlock(&rq->lock); + goto repeat_lock_task; + } + return rq; +} + /* * task_rq_lock - lock the runqueue a given task resides on and disable * interrupts. Note the ordering: we can safely lookup the task_rq without @@ -375,6 +394,12 @@ repeat_lock_task: return rq; } +static inline void __task_rq_unlock(runqueue_t *rq) + __releases(rq->lock) +{ + spin_unlock(&rq->lock); +} + static inline void task_rq_unlock(runqueue_t *rq, unsigned long *flags) __releases(rq->lock) { @@ -638,7 +663,7 @@ static inline void enqueue_task_head(struct task_struct *p, prio_array_t *array) } /* - * effective_prio - return the priority that is based on the static + * __normal_prio - return the priority that is based on the static * priority but is modified by bonuses/penalties. * * We scale the actual sleep average [0 .... MAX_SLEEP_AVG] @@ -651,13 +676,11 @@ static inline void enqueue_task_head(struct task_struct *p, prio_array_t *array) * * Both properties are important to certain workloads. */ -static int effective_prio(task_t *p) + +static inline int __normal_prio(task_t *p) { int bonus, prio; - if (rt_task(p)) - return p->prio; - bonus = CURRENT_BONUS(p) - MAX_BONUS / 2; prio = p->static_prio - bonus; @@ -692,7 +715,7 @@ static int effective_prio(task_t *p) static void set_load_weight(task_t *p) { - if (rt_task(p)) { + if (has_rt_policy(p)) { #ifdef CONFIG_SMP if (p == task_rq(p)->migration_thread) /* @@ -730,6 +753,44 @@ static inline void dec_nr_running(task_t *p, runqueue_t *rq) dec_raw_weighted_load(rq, p); } +/* + * Calculate the expected normal priority: i.e. priority + * without taking RT-inheritance into account. Might be + * boosted by interactivity modifiers. Changes upon fork, + * setprio syscalls, and whenever the interactivity + * estimator recalculates. + */ +static inline int normal_prio(task_t *p) +{ + int prio; + + if (has_rt_policy(p)) + prio = MAX_RT_PRIO-1 - p->rt_priority; + else + prio = __normal_prio(p); + return prio; +} + +/* + * Calculate the current priority, i.e. the priority + * taken into account by the scheduler. This value might + * be boosted by RT tasks, or might be boosted by + * interactivity modifiers. Will be RT if the task got + * RT-boosted. If not then it returns p->normal_prio. + */ +static int effective_prio(task_t *p) +{ + p->normal_prio = normal_prio(p); + /* + * If we are RT tasks or we were boosted to RT priority, + * keep the priority unchanged. Otherwise, update priority + * to the normal priority: + */ + if (!rt_prio(p->prio)) + return p->normal_prio; + return p->prio; +} + /* * __activate_task - move a task to the runqueue. */ @@ -752,6 +813,10 @@ static inline void __activate_idle_task(task_t *p, runqueue_t *rq) inc_nr_running(p, rq); } +/* + * Recalculate p->normal_prio and p->prio after having slept, + * updating the sleep-average too: + */ static int recalc_task_prio(task_t *p, unsigned long long now) { /* Caller must always ensure 'now >= p->timestamp' */ @@ -1448,6 +1513,12 @@ void fastcall sched_fork(task_t *p, int clone_flags) * event cannot wake it up and insert it on the runqueue either. */ p->state = TASK_RUNNING; + + /* + * Make sure we do not leak PI boosting priority to the child: + */ + p->prio = current->normal_prio; + INIT_LIST_HEAD(&p->run_list); p->array = NULL; #ifdef CONFIG_SCHEDSTATS @@ -1527,6 +1598,7 @@ void fastcall wake_up_new_task(task_t *p, unsigned long clone_flags) __activate_task(p, rq); else { p->prio = current->prio; + p->normal_prio = current->normal_prio; list_add_tail(&p->run_list, ¤t->run_list); p->array = current->array; p->array->nr_active++; @@ -3668,12 +3740,65 @@ long fastcall __sched sleep_on_timeout(wait_queue_head_t *q, long timeout) EXPORT_SYMBOL(sleep_on_timeout); +#ifdef CONFIG_RT_MUTEXES + +/* + * rt_mutex_setprio - set the current priority of a task + * @p: task + * @prio: prio value (kernel-internal form) + * + * This function changes the 'effective' priority of a task. It does + * not touch ->normal_prio like __setscheduler(). + * + * Used by the rt_mutex code to implement priority inheritance logic. + */ +void rt_mutex_setprio(task_t *p, int prio) +{ + unsigned long flags; + prio_array_t *array; + runqueue_t *rq; + int oldprio; + + BUG_ON(prio < 0 || prio > MAX_PRIO); + + rq = task_rq_lock(p, &flags); + + oldprio = p->prio; + array = p->array; + if (array) + dequeue_task(p, array); + p->prio = prio; + + if (array) { + /* + * If changing to an RT priority then queue it + * in the active array! + */ + if (rt_task(p)) + array = rq->active; + enqueue_task(p, array); + /* + * Reschedule if we are currently running on this runqueue and + * our priority decreased, or if we are not currently running on + * this runqueue and our priority is higher than the current's + */ + if (task_running(rq, p)) { + if (p->prio > oldprio) + resched_task(rq->curr); + } else if (TASK_PREEMPTS_CURR(p, rq)) + resched_task(rq->curr); + } + task_rq_unlock(rq, &flags); +} + +#endif + void set_user_nice(task_t *p, long nice) { unsigned long flags; prio_array_t *array; runqueue_t *rq; - int old_prio, new_prio, delta; + int old_prio, delta; if (TASK_NICE(p) == nice || nice < -20 || nice > 19) return; @@ -3688,7 +3813,7 @@ void set_user_nice(task_t *p, long nice) * it wont have any effect on scheduling until the task is * not SCHED_NORMAL/SCHED_BATCH: */ - if (rt_task(p)) { + if (has_rt_policy(p)) { p->static_prio = NICE_TO_PRIO(nice); goto out_unlock; } @@ -3698,12 +3823,11 @@ void set_user_nice(task_t *p, long nice) dec_raw_weighted_load(rq, p); } - old_prio = p->prio; - new_prio = NICE_TO_PRIO(nice); - delta = new_prio - old_prio; p->static_prio = NICE_TO_PRIO(nice); set_load_weight(p); - p->prio += delta; + old_prio = p->prio; + p->prio = effective_prio(p); + delta = p->prio - old_prio; if (array) { enqueue_task(p, array); @@ -3718,7 +3842,6 @@ void set_user_nice(task_t *p, long nice) out_unlock: task_rq_unlock(rq, &flags); } - EXPORT_SYMBOL(set_user_nice); /* @@ -3833,16 +3956,14 @@ static void __setscheduler(struct task_struct *p, int policy, int prio) BUG_ON(p->array); p->policy = policy; p->rt_priority = prio; - if (policy != SCHED_NORMAL && policy != SCHED_BATCH) { - p->prio = MAX_RT_PRIO-1 - p->rt_priority; - } else { - p->prio = p->static_prio; - /* - * SCHED_BATCH tasks are treated as perpetual CPU hogs: - */ - if (policy == SCHED_BATCH) - p->sleep_avg = 0; - } + p->normal_prio = normal_prio(p); + /* we are holding p->pi_lock already */ + p->prio = rt_mutex_getprio(p); + /* + * SCHED_BATCH tasks are treated as perpetual CPU hogs: + */ + if (policy == SCHED_BATCH) + p->sleep_avg = 0; set_load_weight(p); } @@ -3911,15 +4032,21 @@ recheck: retval = security_task_setscheduler(p, policy, param); if (retval) return retval; + /* + * make sure no PI-waiters arrive (or leave) while we are + * changing the priority of the task: + */ + spin_lock_irqsave(&p->pi_lock, flags); /* * To be able to change p->policy safely, the apropriate * runqueue lock must be held. */ - rq = task_rq_lock(p, &flags); + rq = __task_rq_lock(p); /* recheck policy now with rq lock held */ if (unlikely(oldpolicy != -1 && oldpolicy != p->policy)) { policy = oldpolicy = -1; - task_rq_unlock(rq, &flags); + __task_rq_unlock(rq); + spin_unlock_irqrestore(&p->pi_lock, flags); goto recheck; } array = p->array; @@ -3940,7 +4067,9 @@ recheck: } else if (TASK_PREEMPTS_CURR(p, rq)) resched_task(rq->curr); } - task_rq_unlock(rq, &flags); + __task_rq_unlock(rq); + spin_unlock_irqrestore(&p->pi_lock, flags); + return 0; } EXPORT_SYMBOL_GPL(sched_setscheduler); @@ -4575,7 +4704,7 @@ void __devinit init_idle(task_t *idle, int cpu) idle->timestamp = sched_clock(); idle->sleep_avg = 0; idle->array = NULL; - idle->prio = MAX_PRIO; + idle->prio = idle->normal_prio = MAX_PRIO; idle->state = TASK_RUNNING; idle->cpus_allowed = cpumask_of_cpu(cpu); set_task_cpu(idle, cpu); @@ -6582,7 +6711,8 @@ void normalize_rt_tasks(void) if (!rt_task(p)) continue; - rq = task_rq_lock(p, &flags); + spin_lock_irqsave(&p->pi_lock, flags); + rq = __task_rq_lock(p); array = p->array; if (array) @@ -6593,7 +6723,8 @@ void normalize_rt_tasks(void) resched_task(rq->curr); } - task_rq_unlock(rq, &flags); + __task_rq_unlock(rq); + spin_unlock_irqrestore(&p->pi_lock, flags); } read_unlock_irq(&tasklist_lock); } -- cgit v1.2.3-70-g09d2 From 23f78d4a03c53cbd75d87a795378ea540aa08c86 Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Tue, 27 Jun 2006 02:54:53 -0700 Subject: [PATCH] pi-futex: rt mutex core Core functions for the rt-mutex subsystem. Signed-off-by: Ingo Molnar Signed-off-by: Thomas Gleixner Signed-off-by: Arjan van de Ven Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/init_task.h | 1 + include/linux/rtmutex.h | 104 ++++++ include/linux/sched.h | 12 + include/linux/sysctl.h | 1 + init/Kconfig | 5 + kernel/Makefile | 1 + kernel/fork.c | 16 + kernel/rtmutex.c | 904 ++++++++++++++++++++++++++++++++++++++++++++++ kernel/rtmutex.h | 29 ++ kernel/rtmutex_common.h | 93 +++++ kernel/sysctl.c | 15 + 11 files changed, 1181 insertions(+) create mode 100644 include/linux/rtmutex.h create mode 100644 kernel/rtmutex.c create mode 100644 kernel/rtmutex.h create mode 100644 kernel/rtmutex_common.h (limited to 'include') diff --git a/include/linux/init_task.h b/include/linux/init_task.h index 678c1a90380..3a256957fb5 100644 --- a/include/linux/init_task.h +++ b/include/linux/init_task.h @@ -124,6 +124,7 @@ extern struct group_info init_groups; .cpu_timers = INIT_CPU_TIMERS(tsk.cpu_timers), \ .fs_excl = ATOMIC_INIT(0), \ .pi_lock = SPIN_LOCK_UNLOCKED, \ + INIT_RT_MUTEXES(tsk) \ } diff --git a/include/linux/rtmutex.h b/include/linux/rtmutex.h new file mode 100644 index 00000000000..12309c916c6 --- /dev/null +++ b/include/linux/rtmutex.h @@ -0,0 +1,104 @@ +/* + * RT Mutexes: blocking mutual exclusion locks with PI support + * + * started by Ingo Molnar and Thomas Gleixner: + * + * Copyright (C) 2004-2006 Red Hat, Inc., Ingo Molnar + * Copyright (C) 2006, Timesys Corp., Thomas Gleixner + * + * This file contains the public data structure and API definitions. + */ + +#ifndef __LINUX_RT_MUTEX_H +#define __LINUX_RT_MUTEX_H + +#include +#include +#include + +/* + * The rt_mutex structure + * + * @wait_lock: spinlock to protect the structure + * @wait_list: pilist head to enqueue waiters in priority order + * @owner: the mutex owner + */ +struct rt_mutex { + spinlock_t wait_lock; + struct plist_head wait_list; + struct task_struct *owner; +#ifdef CONFIG_DEBUG_RT_MUTEXES + int save_state; + struct list_head held_list_entry; + unsigned long acquire_ip; + const char *name, *file; + int line; + void *magic; +#endif +}; + +struct rt_mutex_waiter; +struct hrtimer_sleeper; + +#ifdef CONFIG_DEBUG_RT_MUTEXES +# define __DEBUG_RT_MUTEX_INITIALIZER(mutexname) \ + , .name = #mutexname, .file = __FILE__, .line = __LINE__ +# define rt_mutex_init(mutex) __rt_mutex_init(mutex, __FUNCTION__) + extern void rt_mutex_debug_task_free(struct task_struct *tsk); +#else +# define __DEBUG_RT_MUTEX_INITIALIZER(mutexname) +# define rt_mutex_init(mutex) __rt_mutex_init(mutex, NULL) +# define rt_mutex_debug_task_free(t) do { } while (0) +#endif + +#define __RT_MUTEX_INITIALIZER(mutexname) \ + { .wait_lock = SPIN_LOCK_UNLOCKED \ + , .wait_list = PLIST_HEAD_INIT(mutexname.wait_list, mutexname.wait_lock) \ + , .owner = NULL \ + __DEBUG_RT_MUTEX_INITIALIZER(mutexname)} + +#define DEFINE_RT_MUTEX(mutexname) \ + struct rt_mutex mutexname = __RT_MUTEX_INITIALIZER(mutexname) + +/*** + * rt_mutex_is_locked - is the mutex locked + * @lock: the mutex to be queried + * + * Returns 1 if the mutex is locked, 0 if unlocked. + */ +static inline int rt_mutex_is_locked(struct rt_mutex *lock) +{ + return lock->owner != NULL; +} + +extern void __rt_mutex_init(struct rt_mutex *lock, const char *name); +extern void rt_mutex_destroy(struct rt_mutex *lock); + +extern void rt_mutex_lock(struct rt_mutex *lock); +extern int rt_mutex_lock_interruptible(struct rt_mutex *lock, + int detect_deadlock); +extern int rt_mutex_timed_lock(struct rt_mutex *lock, + struct hrtimer_sleeper *timeout, + int detect_deadlock); + +extern int rt_mutex_trylock(struct rt_mutex *lock); + +extern void rt_mutex_unlock(struct rt_mutex *lock); + +#ifdef CONFIG_DEBUG_RT_MUTEXES +# define INIT_RT_MUTEX_DEBUG(tsk) \ + .held_list_head = LIST_HEAD_INIT(tsk.held_list_head), \ + .held_list_lock = SPIN_LOCK_UNLOCKED +#else +# define INIT_RT_MUTEX_DEBUG(tsk) +#endif + +#ifdef CONFIG_RT_MUTEXES +# define INIT_RT_MUTEXES(tsk) \ + .pi_waiters = PLIST_HEAD_INIT(tsk.pi_waiters, tsk.pi_lock), \ + INIT_RT_MUTEX_DEBUG(tsk) +#else +# define INIT_RT_MUTEXES(tsk) +#endif + +#endif diff --git a/include/linux/sched.h b/include/linux/sched.h index 6f167645e7e..6ea23c9af41 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -73,6 +73,7 @@ struct sched_param { #include #include #include +#include #include #include @@ -858,6 +859,17 @@ struct task_struct { /* Protection of the PI data structures: */ spinlock_t pi_lock; +#ifdef CONFIG_RT_MUTEXES + /* PI waiters blocked on a rt_mutex held by this task */ + struct plist_head pi_waiters; + /* Deadlock detection and priority inheritance handling */ + struct rt_mutex_waiter *pi_blocked_on; +# ifdef CONFIG_DEBUG_RT_MUTEXES + spinlock_t held_list_lock; + struct list_head held_list_head; +# endif +#endif + #ifdef CONFIG_DEBUG_MUTEXES /* mutex deadlock detection */ struct mutex_waiter *blocked_on; diff --git a/include/linux/sysctl.h b/include/linux/sysctl.h index bee12a7a057..46e4d8f2771 100644 --- a/include/linux/sysctl.h +++ b/include/linux/sysctl.h @@ -149,6 +149,7 @@ enum KERN_ACPI_VIDEO_FLAGS=71, /* int: flags for setting up video after ACPI sleep */ KERN_IA64_UNALIGNED=72, /* int: ia64 unaligned userland trap enable */ KERN_COMPAT_LOG=73, /* int: print compat layer messages */ + KERN_MAX_LOCK_DEPTH=74, }; diff --git a/init/Kconfig b/init/Kconfig index df55b366560..f70f2fd273c 100644 --- a/init/Kconfig +++ b/init/Kconfig @@ -339,9 +339,14 @@ config BASE_FULL kernel data structures. This saves memory on small machines, but may reduce performance. +config RT_MUTEXES + boolean + select PLIST + config FUTEX bool "Enable futex support" if EMBEDDED default y + select RT_MUTEXES help Disabling this option will cause the kernel to be built without support for "fast userspace mutexes". The resulting kernel may not diff --git a/kernel/Makefile b/kernel/Makefile index 752bd7d383a..21df9a338ff 100644 --- a/kernel/Makefile +++ b/kernel/Makefile @@ -16,6 +16,7 @@ obj-$(CONFIG_FUTEX) += futex.o ifeq ($(CONFIG_COMPAT),y) obj-$(CONFIG_FUTEX) += futex_compat.o endif +obj-$(CONFIG_RT_MUTEXES) += rtmutex.o obj-$(CONFIG_GENERIC_ISA_DMA) += dma.o obj-$(CONFIG_SMP) += cpu.o spinlock.o obj-$(CONFIG_DEBUG_SPINLOCK) += spinlock.o diff --git a/kernel/fork.c b/kernel/fork.c index 9b4e54ef022..b664a081fff 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -104,6 +104,7 @@ static kmem_cache_t *mm_cachep; void free_task(struct task_struct *tsk) { free_thread_info(tsk->thread_info); + rt_mutex_debug_task_free(tsk); free_task_struct(tsk); } EXPORT_SYMBOL(free_task); @@ -913,6 +914,19 @@ asmlinkage long sys_set_tid_address(int __user *tidptr) return current->pid; } +static inline void rt_mutex_init_task(struct task_struct *p) +{ +#ifdef CONFIG_RT_MUTEXES + spin_lock_init(&p->pi_lock); + plist_head_init(&p->pi_waiters, &p->pi_lock); + p->pi_blocked_on = NULL; +# ifdef CONFIG_DEBUG_RT_MUTEXES + spin_lock_init(&p->held_list_lock); + INIT_LIST_HEAD(&p->held_list_head); +# endif +#endif +} + /* * This creates a new process as a copy of the old one, * but does not actually start it yet. @@ -1034,6 +1048,8 @@ static task_t *copy_process(unsigned long clone_flags, mpol_fix_fork_child_flag(p); #endif + rt_mutex_init_task(p); + #ifdef CONFIG_DEBUG_MUTEXES p->blocked_on = NULL; /* not blocked yet */ #endif diff --git a/kernel/rtmutex.c b/kernel/rtmutex.c new file mode 100644 index 00000000000..937a474fae9 --- /dev/null +++ b/kernel/rtmutex.c @@ -0,0 +1,904 @@ +/* + * RT-Mutexes: simple blocking mutual exclusion locks with PI support + * + * started by Ingo Molnar and Thomas Gleixner. + * + * Copyright (C) 2004-2006 Red Hat, Inc., Ingo Molnar + * Copyright (C) 2005-2006 Timesys Corp., Thomas Gleixner + * Copyright (C) 2005 Kihon Technologies Inc., Steven Rostedt + * Copyright (C) 2006 Esben Nielsen + */ +#include +#include +#include +#include + +#include "rtmutex_common.h" + +#ifdef CONFIG_DEBUG_RT_MUTEXES +# include "rtmutex-debug.h" +#else +# include "rtmutex.h" +#endif + +/* + * lock->owner state tracking: + * + * lock->owner holds the task_struct pointer of the owner. Bit 0 and 1 + * are used to keep track of the "owner is pending" and "lock has + * waiters" state. + * + * owner bit1 bit0 + * NULL 0 0 lock is free (fast acquire possible) + * NULL 0 1 invalid state + * NULL 1 0 Transitional State* + * NULL 1 1 invalid state + * taskpointer 0 0 lock is held (fast release possible) + * taskpointer 0 1 task is pending owner + * taskpointer 1 0 lock is held and has waiters + * taskpointer 1 1 task is pending owner and lock has more waiters + * + * Pending ownership is assigned to the top (highest priority) + * waiter of the lock, when the lock is released. The thread is woken + * up and can now take the lock. Until the lock is taken (bit 0 + * cleared) a competing higher priority thread can steal the lock + * which puts the woken up thread back on the waiters list. + * + * The fast atomic compare exchange based acquire and release is only + * possible when bit 0 and 1 of lock->owner are 0. + * + * (*) There's a small time where the owner can be NULL and the + * "lock has waiters" bit is set. This can happen when grabbing the lock. + * To prevent a cmpxchg of the owner releasing the lock, we need to set this + * bit before looking at the lock, hence the reason this is a transitional + * state. + */ + +static void +rt_mutex_set_owner(struct rt_mutex *lock, struct task_struct *owner, + unsigned long mask) +{ + unsigned long val = (unsigned long)owner | mask; + + if (rt_mutex_has_waiters(lock)) + val |= RT_MUTEX_HAS_WAITERS; + + lock->owner = (struct task_struct *)val; +} + +static inline void clear_rt_mutex_waiters(struct rt_mutex *lock) +{ + lock->owner = (struct task_struct *) + ((unsigned long)lock->owner & ~RT_MUTEX_HAS_WAITERS); +} + +static void fixup_rt_mutex_waiters(struct rt_mutex *lock) +{ + if (!rt_mutex_has_waiters(lock)) + clear_rt_mutex_waiters(lock); +} + +/* + * We can speed up the acquire/release, if the architecture + * supports cmpxchg and if there's no debugging state to be set up + */ +#if defined(__HAVE_ARCH_CMPXCHG) && !defined(CONFIG_DEBUG_RT_MUTEXES) +# define rt_mutex_cmpxchg(l,c,n) (cmpxchg(&l->owner, c, n) == c) +static inline void mark_rt_mutex_waiters(struct rt_mutex *lock) +{ + unsigned long owner, *p = (unsigned long *) &lock->owner; + + do { + owner = *p; + } while (cmpxchg(p, owner, owner | RT_MUTEX_HAS_WAITERS) != owner); +} +#else +# define rt_mutex_cmpxchg(l,c,n) (0) +static inline void mark_rt_mutex_waiters(struct rt_mutex *lock) +{ + lock->owner = (struct task_struct *) + ((unsigned long)lock->owner | RT_MUTEX_HAS_WAITERS); +} +#endif + +/* + * Calculate task priority from the waiter list priority + * + * Return task->normal_prio when the waiter list is empty or when + * the waiter is not allowed to do priority boosting + */ +int rt_mutex_getprio(struct task_struct *task) +{ + if (likely(!task_has_pi_waiters(task))) + return task->normal_prio; + + return min(task_top_pi_waiter(task)->pi_list_entry.prio, + task->normal_prio); +} + +/* + * Adjust the priority of a task, after its pi_waiters got modified. + * + * This can be both boosting and unboosting. task->pi_lock must be held. + */ +static void __rt_mutex_adjust_prio(struct task_struct *task) +{ + int prio = rt_mutex_getprio(task); + + if (task->prio != prio) + rt_mutex_setprio(task, prio); +} + +/* + * Adjust task priority (undo boosting). Called from the exit path of + * rt_mutex_slowunlock() and rt_mutex_slowlock(). + * + * (Note: We do this outside of the protection of lock->wait_lock to + * allow the lock to be taken while or before we readjust the priority + * of task. We do not use the spin_xx_mutex() variants here as we are + * outside of the debug path.) + */ +static void rt_mutex_adjust_prio(struct task_struct *task) +{ + unsigned long flags; + + spin_lock_irqsave(&task->pi_lock, flags); + __rt_mutex_adjust_prio(task); + spin_unlock_irqrestore(&task->pi_lock, flags); +} + +/* + * Max number of times we'll walk the boosting chain: + */ +int max_lock_depth = 1024; + +/* + * Adjust the priority chain. Also used for deadlock detection. + * Decreases task's usage by one - may thus free the task. + * Returns 0 or -EDEADLK. + */ +static int rt_mutex_adjust_prio_chain(task_t *task, + int deadlock_detect, + struct rt_mutex *orig_lock, + struct rt_mutex_waiter *orig_waiter + __IP_DECL__) +{ + struct rt_mutex *lock; + struct rt_mutex_waiter *waiter, *top_waiter = orig_waiter; + int detect_deadlock, ret = 0, depth = 0; + unsigned long flags; + + detect_deadlock = debug_rt_mutex_detect_deadlock(orig_waiter, + deadlock_detect); + + /* + * The (de)boosting is a step by step approach with a lot of + * pitfalls. We want this to be preemptible and we want hold a + * maximum of two locks per step. So we have to check + * carefully whether things change under us. + */ + again: + if (++depth > max_lock_depth) { + static int prev_max; + + /* + * Print this only once. If the admin changes the limit, + * print a new message when reaching the limit again. + */ + if (prev_max != max_lock_depth) { + prev_max = max_lock_depth; + printk(KERN_WARNING "Maximum lock depth %d reached " + "task: %s (%d)\n", max_lock_depth, + current->comm, current->pid); + } + put_task_struct(task); + + return deadlock_detect ? -EDEADLK : 0; + } + retry: + /* + * Task can not go away as we did a get_task() before ! + */ + spin_lock_irqsave(&task->pi_lock, flags); + + waiter = task->pi_blocked_on; + /* + * Check whether the end of the boosting chain has been + * reached or the state of the chain has changed while we + * dropped the locks. + */ + if (!waiter || !waiter->task) + goto out_unlock_pi; + + if (top_waiter && (!task_has_pi_waiters(task) || + top_waiter != task_top_pi_waiter(task))) + goto out_unlock_pi; + + /* + * When deadlock detection is off then we check, if further + * priority adjustment is necessary. + */ + if (!detect_deadlock && waiter->list_entry.prio == task->prio) + goto out_unlock_pi; + + lock = waiter->lock; + if (!spin_trylock(&lock->wait_lock)) { + spin_unlock_irqrestore(&task->pi_lock, flags); + cpu_relax(); + goto retry; + } + + /* Deadlock detection */ + if (lock == orig_lock || rt_mutex_owner(lock) == current) { + debug_rt_mutex_deadlock(deadlock_detect, orig_waiter, lock); + spin_unlock(&lock->wait_lock); + ret = deadlock_detect ? -EDEADLK : 0; + goto out_unlock_pi; + } + + top_waiter = rt_mutex_top_waiter(lock); + + /* Requeue the waiter */ + plist_del(&waiter->list_entry, &lock->wait_list); + waiter->list_entry.prio = task->prio; + plist_add(&waiter->list_entry, &lock->wait_list); + + /* Release the task */ + spin_unlock_irqrestore(&task->pi_lock, flags); + put_task_struct(task); + + /* Grab the next task */ + task = rt_mutex_owner(lock); + spin_lock_irqsave(&task->pi_lock, flags); + + if (waiter == rt_mutex_top_waiter(lock)) { + /* Boost the owner */ + plist_del(&top_waiter->pi_list_entry, &task->pi_waiters); + waiter->pi_list_entry.prio = waiter->list_entry.prio; + plist_add(&waiter->pi_list_entry, &task->pi_waiters); + __rt_mutex_adjust_prio(task); + + } else if (top_waiter == waiter) { + /* Deboost the owner */ + plist_del(&waiter->pi_list_entry, &task->pi_waiters); + waiter = rt_mutex_top_waiter(lock); + waiter->pi_list_entry.prio = waiter->list_entry.prio; + plist_add(&waiter->pi_list_entry, &task->pi_waiters); + __rt_mutex_adjust_prio(task); + } + + get_task_struct(task); + spin_unlock_irqrestore(&task->pi_lock, flags); + + top_waiter = rt_mutex_top_waiter(lock); + spin_unlock(&lock->wait_lock); + + if (!detect_deadlock && waiter != top_waiter) + goto out_put_task; + + goto again; + + out_unlock_pi: + spin_unlock_irqrestore(&task->pi_lock, flags); + out_put_task: + put_task_struct(task); + return ret; +} + +/* + * Optimization: check if we can steal the lock from the + * assigned pending owner [which might not have taken the + * lock yet]: + */ +static inline int try_to_steal_lock(struct rt_mutex *lock) +{ + struct task_struct *pendowner = rt_mutex_owner(lock); + struct rt_mutex_waiter *next; + unsigned long flags; + + if (!rt_mutex_owner_pending(lock)) + return 0; + + if (pendowner == current) + return 1; + + spin_lock_irqsave(&pendowner->pi_lock, flags); + if (current->prio >= pendowner->prio) { + spin_unlock_irqrestore(&pendowner->pi_lock, flags); + return 0; + } + + /* + * Check if a waiter is enqueued on the pending owners + * pi_waiters list. Remove it and readjust pending owners + * priority. + */ + if (likely(!rt_mutex_has_waiters(lock))) { + spin_unlock_irqrestore(&pendowner->pi_lock, flags); + return 1; + } + + /* No chain handling, pending owner is not blocked on anything: */ + next = rt_mutex_top_waiter(lock); + plist_del(&next->pi_list_entry, &pendowner->pi_waiters); + __rt_mutex_adjust_prio(pendowner); + spin_unlock_irqrestore(&pendowner->pi_lock, flags); + + /* + * We are going to steal the lock and a waiter was + * enqueued on the pending owners pi_waiters queue. So + * we have to enqueue this waiter into + * current->pi_waiters list. This covers the case, + * where current is boosted because it holds another + * lock and gets unboosted because the booster is + * interrupted, so we would delay a waiter with higher + * priority as current->normal_prio. + * + * Note: in the rare case of a SCHED_OTHER task changing + * its priority and thus stealing the lock, next->task + * might be current: + */ + if (likely(next->task != current)) { + spin_lock_irqsave(¤t->pi_lock, flags); + plist_add(&next->pi_list_entry, ¤t->pi_waiters); + __rt_mutex_adjust_prio(current); + spin_unlock_irqrestore(¤t->pi_lock, flags); + } + return 1; +} + +/* + * Try to take an rt-mutex + * + * This fails + * - when the lock has a real owner + * - when a different pending owner exists and has higher priority than current + * + * Must be called with lock->wait_lock held. + */ +static int try_to_take_rt_mutex(struct rt_mutex *lock __IP_DECL__) +{ + /* + * We have to be careful here if the atomic speedups are + * enabled, such that, when + * - no other waiter is on the lock + * - the lock has been released since we did the cmpxchg + * the lock can be released or taken while we are doing the + * checks and marking the lock with RT_MUTEX_HAS_WAITERS. + * + * The atomic acquire/release aware variant of + * mark_rt_mutex_waiters uses a cmpxchg loop. After setting + * the WAITERS bit, the atomic release / acquire can not + * happen anymore and lock->wait_lock protects us from the + * non-atomic case. + * + * Note, that this might set lock->owner = + * RT_MUTEX_HAS_WAITERS in the case the lock is not contended + * any more. This is fixed up when we take the ownership. + * This is the transitional state explained at the top of this file. + */ + mark_rt_mutex_waiters(lock); + + if (rt_mutex_owner(lock) && !try_to_steal_lock(lock)) + return 0; + + /* We got the lock. */ + debug_rt_mutex_lock(lock __IP__); + + rt_mutex_set_owner(lock, current, 0); + + rt_mutex_deadlock_account_lock(lock, current); + + return 1; +} + +/* + * Task blocks on lock. + * + * Prepare waiter and propagate pi chain + * + * This must be called with lock->wait_lock held. + */ +static int task_blocks_on_rt_mutex(struct rt_mutex *lock, + struct rt_mutex_waiter *waiter, + int detect_deadlock + __IP_DECL__) +{ + struct rt_mutex_waiter *top_waiter = waiter; + task_t *owner = rt_mutex_owner(lock); + int boost = 0, res; + unsigned long flags; + + spin_lock_irqsave(¤t->pi_lock, flags); + __rt_mutex_adjust_prio(current); + waiter->task = current; + waiter->lock = lock; + plist_node_init(&waiter->list_entry, current->prio); + plist_node_init(&waiter->pi_list_entry, current->prio); + + /* Get the top priority waiter on the lock */ + if (rt_mutex_has_waiters(lock)) + top_waiter = rt_mutex_top_waiter(lock); + plist_add(&waiter->list_entry, &lock->wait_list); + + current->pi_blocked_on = waiter; + + spin_unlock_irqrestore(¤t->pi_lock, flags); + + if (waiter == rt_mutex_top_waiter(lock)) { + spin_lock_irqsave(&owner->pi_lock, flags); + plist_del(&top_waiter->pi_list_entry, &owner->pi_waiters); + plist_add(&waiter->pi_list_entry, &owner->pi_waiters); + + __rt_mutex_adjust_prio(owner); + if (owner->pi_blocked_on) { + boost = 1; + get_task_struct(owner); + } + spin_unlock_irqrestore(&owner->pi_lock, flags); + } + else if (debug_rt_mutex_detect_deadlock(waiter, detect_deadlock)) { + spin_lock_irqsave(&owner->pi_lock, flags); + if (owner->pi_blocked_on) { + boost = 1; + get_task_struct(owner); + } + spin_unlock_irqrestore(&owner->pi_lock, flags); + } + if (!boost) + return 0; + + spin_unlock(&lock->wait_lock); + + res = rt_mutex_adjust_prio_chain(owner, detect_deadlock, lock, + waiter __IP__); + + spin_lock(&lock->wait_lock); + + return res; +} + +/* + * Wake up the next waiter on the lock. + * + * Remove the top waiter from the current tasks waiter list and from + * the lock waiter list. Set it as pending owner. Then wake it up. + * + * Called with lock->wait_lock held. + */ +static void wakeup_next_waiter(struct rt_mutex *lock) +{ + struct rt_mutex_waiter *waiter; + struct task_struct *pendowner; + unsigned long flags; + + spin_lock_irqsave(¤t->pi_lock, flags); + + waiter = rt_mutex_top_waiter(lock); + plist_del(&waiter->list_entry, &lock->wait_list); + + /* + * Remove it from current->pi_waiters. We do not adjust a + * possible priority boost right now. We execute wakeup in the + * boosted mode and go back to normal after releasing + * lock->wait_lock. + */ + plist_del(&waiter->pi_list_entry, ¤t->pi_waiters); + pendowner = waiter->task; + waiter->task = NULL; + + rt_mutex_set_owner(lock, pendowner, RT_MUTEX_OWNER_PENDING); + + spin_unlock_irqrestore(¤t->pi_lock, flags); + + /* + * Clear the pi_blocked_on variable and enqueue a possible + * waiter into the pi_waiters list of the pending owner. This + * prevents that in case the pending owner gets unboosted a + * waiter with higher priority than pending-owner->normal_prio + * is blocked on the unboosted (pending) owner. + */ + spin_lock_irqsave(&pendowner->pi_lock, flags); + + WARN_ON(!pendowner->pi_blocked_on); + WARN_ON(pendowner->pi_blocked_on != waiter); + WARN_ON(pendowner->pi_blocked_on->lock != lock); + + pendowner->pi_blocked_on = NULL; + + if (rt_mutex_has_waiters(lock)) { + struct rt_mutex_waiter *next; + + next = rt_mutex_top_waiter(lock); + plist_add(&next->pi_list_entry, &pendowner->pi_waiters); + } + spin_unlock_irqrestore(&pendowner->pi_lock, flags); + + wake_up_process(pendowner); +} + +/* + * Remove a waiter from a lock + * + * Must be called with lock->wait_lock held + */ +static void remove_waiter(struct rt_mutex *lock, + struct rt_mutex_waiter *waiter __IP_DECL__) +{ + int first = (waiter == rt_mutex_top_waiter(lock)); + int boost = 0; + task_t *owner = rt_mutex_owner(lock); + unsigned long flags; + + spin_lock_irqsave(¤t->pi_lock, flags); + plist_del(&waiter->list_entry, &lock->wait_list); + waiter->task = NULL; + current->pi_blocked_on = NULL; + spin_unlock_irqrestore(¤t->pi_lock, flags); + + if (first && owner != current) { + + spin_lock_irqsave(&owner->pi_lock, flags); + + plist_del(&waiter->pi_list_entry, &owner->pi_waiters); + + if (rt_mutex_has_waiters(lock)) { + struct rt_mutex_waiter *next; + + next = rt_mutex_top_waiter(lock); + plist_add(&next->pi_list_entry, &owner->pi_waiters); + } + __rt_mutex_adjust_prio(owner); + + if (owner->pi_blocked_on) { + boost = 1; + get_task_struct(owner); + } + spin_unlock_irqrestore(&owner->pi_lock, flags); + } + + WARN_ON(!plist_node_empty(&waiter->pi_list_entry)); + + if (!boost) + return; + + spin_unlock(&lock->wait_lock); + + rt_mutex_adjust_prio_chain(owner, 0, lock, NULL __IP__); + + spin_lock(&lock->wait_lock); +} + +/* + * Slow path lock function: + */ +static int __sched +rt_mutex_slowlock(struct rt_mutex *lock, int state, + struct hrtimer_sleeper *timeout, + int detect_deadlock __IP_DECL__) +{ + struct rt_mutex_waiter waiter; + int ret = 0; + + debug_rt_mutex_init_waiter(&waiter); + waiter.task = NULL; + + spin_lock(&lock->wait_lock); + + /* Try to acquire the lock again: */ + if (try_to_take_rt_mutex(lock __IP__)) { + spin_unlock(&lock->wait_lock); + return 0; + } + + set_current_state(state); + + /* Setup the timer, when timeout != NULL */ + if (unlikely(timeout)) + hrtimer_start(&timeout->timer, timeout->timer.expires, + HRTIMER_ABS); + + for (;;) { + /* Try to acquire the lock: */ + if (try_to_take_rt_mutex(lock __IP__)) + break; + + /* + * TASK_INTERRUPTIBLE checks for signals and + * timeout. Ignored otherwise. + */ + if (unlikely(state == TASK_INTERRUPTIBLE)) { + /* Signal pending? */ + if (signal_pending(current)) + ret = -EINTR; + if (timeout && !timeout->task) + ret = -ETIMEDOUT; + if (ret) + break; + } + + /* + * waiter.task is NULL the first time we come here and + * when we have been woken up by the previous owner + * but the lock got stolen by a higher prio task. + */ + if (!waiter.task) { + ret = task_blocks_on_rt_mutex(lock, &waiter, + detect_deadlock __IP__); + /* + * If we got woken up by the owner then start loop + * all over without going into schedule to try + * to get the lock now: + */ + if (unlikely(!waiter.task)) + continue; + + if (unlikely(ret)) + break; + } + spin_unlock(&lock->wait_lock); + + debug_rt_mutex_print_deadlock(&waiter); + + schedule(); + + spin_lock(&lock->wait_lock); + set_current_state(state); + } + + set_current_state(TASK_RUNNING); + + if (unlikely(waiter.task)) + remove_waiter(lock, &waiter __IP__); + + /* + * try_to_take_rt_mutex() sets the waiter bit + * unconditionally. We might have to fix that up. + */ + fixup_rt_mutex_waiters(lock); + + spin_unlock(&lock->wait_lock); + + /* Remove pending timer: */ + if (unlikely(timeout)) + hrtimer_cancel(&timeout->timer); + + /* + * Readjust priority, when we did not get the lock. We might + * have been the pending owner and boosted. Since we did not + * take the lock, the PI boost has to go. + */ + if (unlikely(ret)) + rt_mutex_adjust_prio(current); + + debug_rt_mutex_free_waiter(&waiter); + + return ret; +} + +/* + * Slow path try-lock function: + */ +static inline int +rt_mutex_slowtrylock(struct rt_mutex *lock __IP_DECL__) +{ + int ret = 0; + + spin_lock(&lock->wait_lock); + + if (likely(rt_mutex_owner(lock) != current)) { + + ret = try_to_take_rt_mutex(lock __IP__); + /* + * try_to_take_rt_mutex() sets the lock waiters + * bit unconditionally. Clean this up. + */ + fixup_rt_mutex_waiters(lock); + } + + spin_unlock(&lock->wait_lock); + + return ret; +} + +/* + * Slow path to release a rt-mutex: + */ +static void __sched +rt_mutex_slowunlock(struct rt_mutex *lock) +{ + spin_lock(&lock->wait_lock); + + debug_rt_mutex_unlock(lock); + + rt_mutex_deadlock_account_unlock(current); + + if (!rt_mutex_has_waiters(lock)) { + lock->owner = NULL; + spin_unlock(&lock->wait_lock); + return; + } + + wakeup_next_waiter(lock); + + spin_unlock(&lock->wait_lock); + + /* Undo pi boosting if necessary: */ + rt_mutex_adjust_prio(current); +} + +/* + * debug aware fast / slowpath lock,trylock,unlock + * + * The atomic acquire/release ops are compiled away, when either the + * architecture does not support cmpxchg or when debugging is enabled. + */ +static inline int +rt_mutex_fastlock(struct rt_mutex *lock, int state, + int detect_deadlock, + int (*slowfn)(struct rt_mutex *lock, int state, + struct hrtimer_sleeper *timeout, + int detect_deadlock __IP_DECL__)) +{ + if (!detect_deadlock && likely(rt_mutex_cmpxchg(lock, NULL, current))) { + rt_mutex_deadlock_account_lock(lock, current); + return 0; + } else + return slowfn(lock, state, NULL, detect_deadlock __RET_IP__); +} + +static inline int +rt_mutex_timed_fastlock(struct rt_mutex *lock, int state, + struct hrtimer_sleeper *timeout, int detect_deadlock, + int (*slowfn)(struct rt_mutex *lock, int state, + struct hrtimer_sleeper *timeout, + int detect_deadlock __IP_DECL__)) +{ + if (!detect_deadlock && likely(rt_mutex_cmpxchg(lock, NULL, current))) { + rt_mutex_deadlock_account_lock(lock, current); + return 0; + } else + return slowfn(lock, state, timeout, detect_deadlock __RET_IP__); +} + +static inline int +rt_mutex_fasttrylock(struct rt_mutex *lock, + int (*slowfn)(struct rt_mutex *lock __IP_DECL__)) +{ + if (likely(rt_mutex_cmpxchg(lock, NULL, current))) { + rt_mutex_deadlock_account_lock(lock, current); + return 1; + } + return slowfn(lock __RET_IP__); +} + +static inline void +rt_mutex_fastunlock(struct rt_mutex *lock, + void (*slowfn)(struct rt_mutex *lock)) +{ + if (likely(rt_mutex_cmpxchg(lock, current, NULL))) + rt_mutex_deadlock_account_unlock(current); + else + slowfn(lock); +} + +/** + * rt_mutex_lock - lock a rt_mutex + * + * @lock: the rt_mutex to be locked + */ +void __sched rt_mutex_lock(struct rt_mutex *lock) +{ + might_sleep(); + + rt_mutex_fastlock(lock, TASK_UNINTERRUPTIBLE, 0, rt_mutex_slowlock); +} +EXPORT_SYMBOL_GPL(rt_mutex_lock); + +/** + * rt_mutex_lock_interruptible - lock a rt_mutex interruptible + * + * @lock: the rt_mutex to be locked + * @detect_deadlock: deadlock detection on/off + * + * Returns: + * 0 on success + * -EINTR when interrupted by a signal + * -EDEADLK when the lock would deadlock (when deadlock detection is on) + */ +int __sched rt_mutex_lock_interruptible(struct rt_mutex *lock, + int detect_deadlock) +{ + might_sleep(); + + return rt_mutex_fastlock(lock, TASK_INTERRUPTIBLE, + detect_deadlock, rt_mutex_slowlock); +} +EXPORT_SYMBOL_GPL(rt_mutex_lock_interruptible); + +/** + * rt_mutex_lock_interruptible_ktime - lock a rt_mutex interruptible + * the timeout structure is provided + * by the caller + * + * @lock: the rt_mutex to be locked + * @timeout: timeout structure or NULL (no timeout) + * @detect_deadlock: deadlock detection on/off + * + * Returns: + * 0 on success + * -EINTR when interrupted by a signal + * -ETIMEOUT when the timeout expired + * -EDEADLK when the lock would deadlock (when deadlock detection is on) + */ +int +rt_mutex_timed_lock(struct rt_mutex *lock, struct hrtimer_sleeper *timeout, + int detect_deadlock) +{ + might_sleep(); + + return rt_mutex_timed_fastlock(lock, TASK_INTERRUPTIBLE, timeout, + detect_deadlock, rt_mutex_slowlock); +} +EXPORT_SYMBOL_GPL(rt_mutex_timed_lock); + +/** + * rt_mutex_trylock - try to lock a rt_mutex + * + * @lock: the rt_mutex to be locked + * + * Returns 1 on success and 0 on contention + */ +int __sched rt_mutex_trylock(struct rt_mutex *lock) +{ + return rt_mutex_fasttrylock(lock, rt_mutex_slowtrylock); +} +EXPORT_SYMBOL_GPL(rt_mutex_trylock); + +/** + * rt_mutex_unlock - unlock a rt_mutex + * + * @lock: the rt_mutex to be unlocked + */ +void __sched rt_mutex_unlock(struct rt_mutex *lock) +{ + rt_mutex_fastunlock(lock, rt_mutex_slowunlock); +} +EXPORT_SYMBOL_GPL(rt_mutex_unlock); + +/*** + * rt_mutex_destroy - mark a mutex unusable + * @lock: the mutex to be destroyed + * + * This function marks the mutex uninitialized, and any subsequent + * use of the mutex is forbidden. The mutex must not be locked when + * this function is called. + */ +void rt_mutex_destroy(struct rt_mutex *lock) +{ + WARN_ON(rt_mutex_is_locked(lock)); +#ifdef CONFIG_DEBUG_RT_MUTEXES + lock->magic = NULL; +#endif +} + +EXPORT_SYMBOL_GPL(rt_mutex_destroy); + +/** + * __rt_mutex_init - initialize the rt lock + * + * @lock: the rt lock to be initialized + * + * Initialize the rt lock to unlocked state. + * + * Initializing of a locked rt lock is not allowed + */ +void __rt_mutex_init(struct rt_mutex *lock, const char *name) +{ + lock->owner = NULL; + spin_lock_init(&lock->wait_lock); + plist_head_init(&lock->wait_list, &lock->wait_lock); + + debug_rt_mutex_init(lock, name); +} +EXPORT_SYMBOL_GPL(__rt_mutex_init); diff --git a/kernel/rtmutex.h b/kernel/rtmutex.h new file mode 100644 index 00000000000..1e0fca13ff7 --- /dev/null +++ b/kernel/rtmutex.h @@ -0,0 +1,29 @@ +/* + * RT-Mutexes: blocking mutual exclusion locks with PI support + * + * started by Ingo Molnar and Thomas Gleixner: + * + * Copyright (C) 2004-2006 Red Hat, Inc., Ingo Molnar + * Copyright (C) 2006, Timesys Corp., Thomas Gleixner + * + * This file contains macros used solely by rtmutex.c. + * Non-debug version. + */ + +#define __IP_DECL__ +#define __IP__ +#define __RET_IP__ +#define rt_mutex_deadlock_check(l) (0) +#define rt_mutex_deadlock_account_lock(m, t) do { } while (0) +#define rt_mutex_deadlock_account_unlock(l) do { } while (0) +#define debug_rt_mutex_init_waiter(w) do { } while (0) +#define debug_rt_mutex_free_waiter(w) do { } while (0) +#define debug_rt_mutex_lock(l) do { } while (0) +#define debug_rt_mutex_proxy_lock(l,p) do { } while (0) +#define debug_rt_mutex_proxy_unlock(l) do { } while (0) +#define debug_rt_mutex_unlock(l) do { } while (0) +#define debug_rt_mutex_init(m, n) do { } while (0) +#define debug_rt_mutex_deadlock(d, a ,l) do { } while (0) +#define debug_rt_mutex_print_deadlock(w) do { } while (0) +#define debug_rt_mutex_detect_deadlock(w,d) (d) +#define debug_rt_mutex_reset_waiter(w) do { } while (0) diff --git a/kernel/rtmutex_common.h b/kernel/rtmutex_common.h new file mode 100644 index 00000000000..50eed60eb08 --- /dev/null +++ b/kernel/rtmutex_common.h @@ -0,0 +1,93 @@ +/* + * RT Mutexes: blocking mutual exclusion locks with PI support + * + * started by Ingo Molnar and Thomas Gleixner: + * + * Copyright (C) 2004-2006 Red Hat, Inc., Ingo Molnar + * Copyright (C) 2006, Timesys Corp., Thomas Gleixner + * + * This file contains the private data structure and API definitions. + */ + +#ifndef __KERNEL_RTMUTEX_COMMON_H +#define __KERNEL_RTMUTEX_COMMON_H + +#include + +/* + * This is the control structure for tasks blocked on a rt_mutex, + * which is allocated on the kernel stack on of the blocked task. + * + * @list_entry: pi node to enqueue into the mutex waiters list + * @pi_list_entry: pi node to enqueue into the mutex owner waiters list + * @task: task reference to the blocked task + */ +struct rt_mutex_waiter { + struct plist_node list_entry; + struct plist_node pi_list_entry; + struct task_struct *task; + struct rt_mutex *lock; +#ifdef CONFIG_DEBUG_RT_MUTEXES + unsigned long ip; + pid_t deadlock_task_pid; + struct rt_mutex *deadlock_lock; +#endif +}; + +/* + * Various helpers to access the waiters-plist: + */ +static inline int rt_mutex_has_waiters(struct rt_mutex *lock) +{ + return !plist_head_empty(&lock->wait_list); +} + +static inline struct rt_mutex_waiter * +rt_mutex_top_waiter(struct rt_mutex *lock) +{ + struct rt_mutex_waiter *w; + + w = plist_first_entry(&lock->wait_list, struct rt_mutex_waiter, + list_entry); + BUG_ON(w->lock != lock); + + return w; +} + +static inline int task_has_pi_waiters(struct task_struct *p) +{ + return !plist_head_empty(&p->pi_waiters); +} + +static inline struct rt_mutex_waiter * +task_top_pi_waiter(struct task_struct *p) +{ + return plist_first_entry(&p->pi_waiters, struct rt_mutex_waiter, + pi_list_entry); +} + +/* + * lock->owner state tracking: + */ +#define RT_MUTEX_OWNER_PENDING 1UL +#define RT_MUTEX_HAS_WAITERS 2UL +#define RT_MUTEX_OWNER_MASKALL 3UL + +static inline struct task_struct *rt_mutex_owner(struct rt_mutex *lock) +{ + return (struct task_struct *) + ((unsigned long)lock->owner & ~RT_MUTEX_OWNER_MASKALL); +} + +static inline struct task_struct *rt_mutex_real_owner(struct rt_mutex *lock) +{ + return (struct task_struct *) + ((unsigned long)lock->owner & ~RT_MUTEX_HAS_WAITERS); +} + +static inline unsigned long rt_mutex_owner_pending(struct rt_mutex *lock) +{ + return (unsigned long)lock->owner & RT_MUTEX_OWNER_PENDING; +} + +#endif diff --git a/kernel/sysctl.c b/kernel/sysctl.c index f54afed8426..93a2c539864 100644 --- a/kernel/sysctl.c +++ b/kernel/sysctl.c @@ -133,6 +133,10 @@ extern int acct_parm[]; extern int no_unaligned_warning; #endif +#ifdef CONFIG_RT_MUTEXES +extern int max_lock_depth; +#endif + static int parse_table(int __user *, int, void __user *, size_t __user *, void __user *, size_t, ctl_table *, void **); static int proc_doutsstring(ctl_table *table, int write, struct file *filp, @@ -688,6 +692,17 @@ static ctl_table kern_table[] = { .proc_handler = &proc_dointvec, }, #endif +#ifdef CONFIG_RT_MUTEXES + { + .ctl_name = KERN_MAX_LOCK_DEPTH, + .procname = "max_lock_depth", + .data = &max_lock_depth, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = &proc_dointvec, + }, +#endif + { .ctl_name = 0 } }; -- cgit v1.2.3-70-g09d2 From e7eebaf6a81b956c989f184ee4b27277c88f8afe Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Tue, 27 Jun 2006 02:54:55 -0700 Subject: [PATCH] pi-futex: rt mutex debug Runtime debugging functionality for rt-mutexes. Signed-off-by: Ingo Molnar Signed-off-by: Thomas Gleixner Signed-off-by: Arjan van de Ven Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/mm.h | 1 + include/linux/rtmutex.h | 15 +- kernel/Makefile | 1 + kernel/exit.c | 1 + kernel/rtmutex-debug.c | 513 ++++++++++++++++++++++++++++++++++++++++++++++++ kernel/rtmutex-debug.h | 37 ++++ lib/Kconfig.debug | 13 ++ mm/slab.c | 1 + 8 files changed, 581 insertions(+), 1 deletion(-) create mode 100644 kernel/rtmutex-debug.c create mode 100644 kernel/rtmutex-debug.h (limited to 'include') diff --git a/include/linux/mm.h b/include/linux/mm.h index 0ac255720f3..c41a1299b8c 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -1034,6 +1034,7 @@ static inline void debug_check_no_locks_freed(const void *from, unsigned long len) { mutex_debug_check_no_locks_freed(from, len); + rt_mutex_debug_check_no_locks_freed(from, len); } #ifndef CONFIG_DEBUG_PAGEALLOC diff --git a/include/linux/rtmutex.h b/include/linux/rtmutex.h index 12309c916c6..fa4a3b82ba7 100644 --- a/include/linux/rtmutex.h +++ b/include/linux/rtmutex.h @@ -40,6 +40,19 @@ struct rt_mutex { struct rt_mutex_waiter; struct hrtimer_sleeper; +#ifdef CONFIG_DEBUG_RT_MUTEXES + extern int rt_mutex_debug_check_no_locks_freed(const void *from, + unsigned long len); + extern void rt_mutex_debug_check_no_locks_held(struct task_struct *task); +#else + static inline int rt_mutex_debug_check_no_locks_freed(const void *from, + unsigned long len) + { + return 0; + } +# define rt_mutex_debug_check_no_locks_held(task) do { } while (0) +#endif + #ifdef CONFIG_DEBUG_RT_MUTEXES # define __DEBUG_RT_MUTEX_INITIALIZER(mutexname) \ , .name = #mutexname, .file = __FILE__, .line = __LINE__ @@ -48,7 +61,7 @@ struct hrtimer_sleeper; #else # define __DEBUG_RT_MUTEX_INITIALIZER(mutexname) # define rt_mutex_init(mutex) __rt_mutex_init(mutex, NULL) -# define rt_mutex_debug_task_free(t) do { } while (0) +# define rt_mutex_debug_task_free(t) do { } while (0) #endif #define __RT_MUTEX_INITIALIZER(mutexname) \ diff --git a/kernel/Makefile b/kernel/Makefile index 21df9a338ff..f9c92d34cde 100644 --- a/kernel/Makefile +++ b/kernel/Makefile @@ -17,6 +17,7 @@ ifeq ($(CONFIG_COMPAT),y) obj-$(CONFIG_FUTEX) += futex_compat.o endif obj-$(CONFIG_RT_MUTEXES) += rtmutex.o +obj-$(CONFIG_DEBUG_RT_MUTEXES) += rtmutex-debug.o obj-$(CONFIG_GENERIC_ISA_DMA) += dma.o obj-$(CONFIG_SMP) += cpu.o spinlock.o obj-$(CONFIG_DEBUG_SPINLOCK) += spinlock.o diff --git a/kernel/exit.c b/kernel/exit.c index 304ef637be6..3e8a0282e9a 100644 --- a/kernel/exit.c +++ b/kernel/exit.c @@ -929,6 +929,7 @@ fastcall NORET_TYPE void do_exit(long code) * If DEBUG_MUTEXES is on, make sure we are holding no locks: */ mutex_debug_check_no_locks_held(tsk); + rt_mutex_debug_check_no_locks_held(tsk); if (tsk->io_context) exit_io_context(); diff --git a/kernel/rtmutex-debug.c b/kernel/rtmutex-debug.c new file mode 100644 index 00000000000..4aa8a2c9f45 --- /dev/null +++ b/kernel/rtmutex-debug.c @@ -0,0 +1,513 @@ +/* + * RT-Mutexes: blocking mutual exclusion locks with PI support + * + * started by Ingo Molnar and Thomas Gleixner: + * + * Copyright (C) 2004-2006 Red Hat, Inc., Ingo Molnar + * Copyright (C) 2006 Timesys Corp., Thomas Gleixner + * + * This code is based on the rt.c implementation in the preempt-rt tree. + * Portions of said code are + * + * Copyright (C) 2004 LynuxWorks, Inc., Igor Manyilov, Bill Huey + * Copyright (C) 2006 Esben Nielsen + * Copyright (C) 2006 Kihon Technologies Inc., + * Steven Rostedt + * + * See rt.c in preempt-rt for proper credits and further information + */ +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include "rtmutex_common.h" + +#ifdef CONFIG_DEBUG_RT_MUTEXES +# include "rtmutex-debug.h" +#else +# include "rtmutex.h" +#endif + +# define TRACE_WARN_ON(x) WARN_ON(x) +# define TRACE_BUG_ON(x) BUG_ON(x) + +# define TRACE_OFF() \ +do { \ + if (rt_trace_on) { \ + rt_trace_on = 0; \ + console_verbose(); \ + if (spin_is_locked(¤t->pi_lock)) \ + spin_unlock(¤t->pi_lock); \ + if (spin_is_locked(¤t->held_list_lock)) \ + spin_unlock(¤t->held_list_lock); \ + } \ +} while (0) + +# define TRACE_OFF_NOLOCK() \ +do { \ + if (rt_trace_on) { \ + rt_trace_on = 0; \ + console_verbose(); \ + } \ +} while (0) + +# define TRACE_BUG_LOCKED() \ +do { \ + TRACE_OFF(); \ + BUG(); \ +} while (0) + +# define TRACE_WARN_ON_LOCKED(c) \ +do { \ + if (unlikely(c)) { \ + TRACE_OFF(); \ + WARN_ON(1); \ + } \ +} while (0) + +# define TRACE_BUG_ON_LOCKED(c) \ +do { \ + if (unlikely(c)) \ + TRACE_BUG_LOCKED(); \ +} while (0) + +#ifdef CONFIG_SMP +# define SMP_TRACE_BUG_ON_LOCKED(c) TRACE_BUG_ON_LOCKED(c) +#else +# define SMP_TRACE_BUG_ON_LOCKED(c) do { } while (0) +#endif + +/* + * deadlock detection flag. We turn it off when we detect + * the first problem because we dont want to recurse back + * into the tracing code when doing error printk or + * executing a BUG(): + */ +int rt_trace_on = 1; + +void deadlock_trace_off(void) +{ + rt_trace_on = 0; +} + +static void printk_task(task_t *p) +{ + if (p) + printk("%16s:%5d [%p, %3d]", p->comm, p->pid, p, p->prio); + else + printk(""); +} + +static void printk_task_short(task_t *p) +{ + if (p) + printk("%s/%d [%p, %3d]", p->comm, p->pid, p, p->prio); + else + printk(""); +} + +static void printk_lock(struct rt_mutex *lock, int print_owner) +{ + if (lock->name) + printk(" [%p] {%s}\n", + lock, lock->name); + else + printk(" [%p] {%s:%d}\n", + lock, lock->file, lock->line); + + if (print_owner && rt_mutex_owner(lock)) { + printk(".. ->owner: %p\n", lock->owner); + printk(".. held by: "); + printk_task(rt_mutex_owner(lock)); + printk("\n"); + } + if (rt_mutex_owner(lock)) { + printk("... acquired at: "); + print_symbol("%s\n", lock->acquire_ip); + } +} + +static void printk_waiter(struct rt_mutex_waiter *w) +{ + printk("-------------------------\n"); + printk("| waiter struct %p:\n", w); + printk("| w->list_entry: [DP:%p/%p|SP:%p/%p|PRI:%d]\n", + w->list_entry.plist.prio_list.prev, w->list_entry.plist.prio_list.next, + w->list_entry.plist.node_list.prev, w->list_entry.plist.node_list.next, + w->list_entry.prio); + printk("| w->pi_list_entry: [DP:%p/%p|SP:%p/%p|PRI:%d]\n", + w->pi_list_entry.plist.prio_list.prev, w->pi_list_entry.plist.prio_list.next, + w->pi_list_entry.plist.node_list.prev, w->pi_list_entry.plist.node_list.next, + w->pi_list_entry.prio); + printk("\n| lock:\n"); + printk_lock(w->lock, 1); + printk("| w->ti->task:\n"); + printk_task(w->task); + printk("| blocked at: "); + print_symbol("%s\n", w->ip); + printk("-------------------------\n"); +} + +static void show_task_locks(task_t *p) +{ + switch (p->state) { + case TASK_RUNNING: printk("R"); break; + case TASK_INTERRUPTIBLE: printk("S"); break; + case TASK_UNINTERRUPTIBLE: printk("D"); break; + case TASK_STOPPED: printk("T"); break; + case EXIT_ZOMBIE: printk("Z"); break; + case EXIT_DEAD: printk("X"); break; + default: printk("?"); break; + } + printk_task(p); + if (p->pi_blocked_on) { + struct rt_mutex *lock = p->pi_blocked_on->lock; + + printk(" blocked on:"); + printk_lock(lock, 1); + } else + printk(" (not blocked)\n"); +} + +void rt_mutex_show_held_locks(task_t *task, int verbose) +{ + struct list_head *curr, *cursor = NULL; + struct rt_mutex *lock; + task_t *t; + unsigned long flags; + int count = 0; + + if (!rt_trace_on) + return; + + if (verbose) { + printk("------------------------------\n"); + printk("| showing all locks held by: | ("); + printk_task_short(task); + printk("):\n"); + printk("------------------------------\n"); + } + +next: + spin_lock_irqsave(&task->held_list_lock, flags); + list_for_each(curr, &task->held_list_head) { + if (cursor && curr != cursor) + continue; + lock = list_entry(curr, struct rt_mutex, held_list_entry); + t = rt_mutex_owner(lock); + WARN_ON(t != task); + count++; + cursor = curr->next; + spin_unlock_irqrestore(&task->held_list_lock, flags); + + printk("\n#%03d: ", count); + printk_lock(lock, 0); + goto next; + } + spin_unlock_irqrestore(&task->held_list_lock, flags); + + printk("\n"); +} + +void rt_mutex_show_all_locks(void) +{ + task_t *g, *p; + int count = 10; + int unlock = 1; + + printk("\n"); + printk("----------------------\n"); + printk("| showing all tasks: |\n"); + printk("----------------------\n"); + + /* + * Here we try to get the tasklist_lock as hard as possible, + * if not successful after 2 seconds we ignore it (but keep + * trying). This is to enable a debug printout even if a + * tasklist_lock-holding task deadlocks or crashes. + */ +retry: + if (!read_trylock(&tasklist_lock)) { + if (count == 10) + printk("hm, tasklist_lock locked, retrying... "); + if (count) { + count--; + printk(" #%d", 10-count); + mdelay(200); + goto retry; + } + printk(" ignoring it.\n"); + unlock = 0; + } + if (count != 10) + printk(" locked it.\n"); + + do_each_thread(g, p) { + show_task_locks(p); + if (!unlock) + if (read_trylock(&tasklist_lock)) + unlock = 1; + } while_each_thread(g, p); + + printk("\n"); + + printk("-----------------------------------------\n"); + printk("| showing all locks held in the system: |\n"); + printk("-----------------------------------------\n"); + + do_each_thread(g, p) { + rt_mutex_show_held_locks(p, 0); + if (!unlock) + if (read_trylock(&tasklist_lock)) + unlock = 1; + } while_each_thread(g, p); + + + printk("=============================================\n\n"); + + if (unlock) + read_unlock(&tasklist_lock); +} + +void rt_mutex_debug_check_no_locks_held(task_t *task) +{ + struct rt_mutex_waiter *w; + struct list_head *curr; + struct rt_mutex *lock; + + if (!rt_trace_on) + return; + if (!rt_prio(task->normal_prio) && rt_prio(task->prio)) { + printk("BUG: PI priority boost leaked!\n"); + printk_task(task); + printk("\n"); + } + if (list_empty(&task->held_list_head)) + return; + + spin_lock(&task->pi_lock); + plist_for_each_entry(w, &task->pi_waiters, pi_list_entry) { + TRACE_OFF(); + + printk("hm, PI interest held at exit time? Task:\n"); + printk_task(task); + printk_waiter(w); + return; + } + spin_unlock(&task->pi_lock); + + list_for_each(curr, &task->held_list_head) { + lock = list_entry(curr, struct rt_mutex, held_list_entry); + + printk("BUG: %s/%d, lock held at task exit time!\n", + task->comm, task->pid); + printk_lock(lock, 1); + if (rt_mutex_owner(lock) != task) + printk("exiting task is not even the owner??\n"); + } +} + +int rt_mutex_debug_check_no_locks_freed(const void *from, unsigned long len) +{ + const void *to = from + len; + struct list_head *curr; + struct rt_mutex *lock; + unsigned long flags; + void *lock_addr; + + if (!rt_trace_on) + return 0; + + spin_lock_irqsave(¤t->held_list_lock, flags); + list_for_each(curr, ¤t->held_list_head) { + lock = list_entry(curr, struct rt_mutex, held_list_entry); + lock_addr = lock; + if (lock_addr < from || lock_addr >= to) + continue; + TRACE_OFF(); + + printk("BUG: %s/%d, active lock [%p(%p-%p)] freed!\n", + current->comm, current->pid, lock, from, to); + dump_stack(); + printk_lock(lock, 1); + if (rt_mutex_owner(lock) != current) + printk("freeing task is not even the owner??\n"); + return 1; + } + spin_unlock_irqrestore(¤t->held_list_lock, flags); + + return 0; +} + +void rt_mutex_debug_task_free(struct task_struct *task) +{ + WARN_ON(!plist_head_empty(&task->pi_waiters)); + WARN_ON(task->pi_blocked_on); +} + +/* + * We fill out the fields in the waiter to store the information about + * the deadlock. We print when we return. act_waiter can be NULL in + * case of a remove waiter operation. + */ +void debug_rt_mutex_deadlock(int detect, struct rt_mutex_waiter *act_waiter, + struct rt_mutex *lock) +{ + struct task_struct *task; + + if (!rt_trace_on || detect || !act_waiter) + return; + + task = rt_mutex_owner(act_waiter->lock); + if (task && task != current) { + act_waiter->deadlock_task_pid = task->pid; + act_waiter->deadlock_lock = lock; + } +} + +void debug_rt_mutex_print_deadlock(struct rt_mutex_waiter *waiter) +{ + struct task_struct *task; + + if (!waiter->deadlock_lock || !rt_trace_on) + return; + + task = find_task_by_pid(waiter->deadlock_task_pid); + if (!task) + return; + + TRACE_OFF_NOLOCK(); + + printk("\n============================================\n"); + printk( "[ BUG: circular locking deadlock detected! ]\n"); + printk( "--------------------------------------------\n"); + printk("%s/%d is deadlocking current task %s/%d\n\n", + task->comm, task->pid, current->comm, current->pid); + + printk("\n1) %s/%d is trying to acquire this lock:\n", + current->comm, current->pid); + printk_lock(waiter->lock, 1); + + printk("... trying at: "); + print_symbol("%s\n", waiter->ip); + + printk("\n2) %s/%d is blocked on this lock:\n", task->comm, task->pid); + printk_lock(waiter->deadlock_lock, 1); + + rt_mutex_show_held_locks(current, 1); + rt_mutex_show_held_locks(task, 1); + + printk("\n%s/%d's [blocked] stackdump:\n\n", task->comm, task->pid); + show_stack(task, NULL); + printk("\n%s/%d's [current] stackdump:\n\n", + current->comm, current->pid); + dump_stack(); + rt_mutex_show_all_locks(); + printk("[ turning off deadlock detection." + "Please report this trace. ]\n\n"); + local_irq_disable(); +} + +void debug_rt_mutex_lock(struct rt_mutex *lock __IP_DECL__) +{ + unsigned long flags; + + if (rt_trace_on) { + TRACE_WARN_ON_LOCKED(!list_empty(&lock->held_list_entry)); + + spin_lock_irqsave(¤t->held_list_lock, flags); + list_add_tail(&lock->held_list_entry, ¤t->held_list_head); + spin_unlock_irqrestore(¤t->held_list_lock, flags); + + lock->acquire_ip = ip; + } +} + +void debug_rt_mutex_unlock(struct rt_mutex *lock) +{ + unsigned long flags; + + if (rt_trace_on) { + TRACE_WARN_ON_LOCKED(rt_mutex_owner(lock) != current); + TRACE_WARN_ON_LOCKED(list_empty(&lock->held_list_entry)); + + spin_lock_irqsave(¤t->held_list_lock, flags); + list_del_init(&lock->held_list_entry); + spin_unlock_irqrestore(¤t->held_list_lock, flags); + } +} + +void debug_rt_mutex_proxy_lock(struct rt_mutex *lock, + struct task_struct *powner __IP_DECL__) +{ + unsigned long flags; + + if (rt_trace_on) { + TRACE_WARN_ON_LOCKED(!list_empty(&lock->held_list_entry)); + + spin_lock_irqsave(&powner->held_list_lock, flags); + list_add_tail(&lock->held_list_entry, &powner->held_list_head); + spin_unlock_irqrestore(&powner->held_list_lock, flags); + + lock->acquire_ip = ip; + } +} + +void debug_rt_mutex_proxy_unlock(struct rt_mutex *lock) +{ + unsigned long flags; + + if (rt_trace_on) { + struct task_struct *owner = rt_mutex_owner(lock); + + TRACE_WARN_ON_LOCKED(!owner); + TRACE_WARN_ON_LOCKED(list_empty(&lock->held_list_entry)); + + spin_lock_irqsave(&owner->held_list_lock, flags); + list_del_init(&lock->held_list_entry); + spin_unlock_irqrestore(&owner->held_list_lock, flags); + } +} + +void debug_rt_mutex_init_waiter(struct rt_mutex_waiter *waiter) +{ + memset(waiter, 0x11, sizeof(*waiter)); + plist_node_init(&waiter->list_entry, MAX_PRIO); + plist_node_init(&waiter->pi_list_entry, MAX_PRIO); +} + +void debug_rt_mutex_free_waiter(struct rt_mutex_waiter *waiter) +{ + TRACE_WARN_ON(!plist_node_empty(&waiter->list_entry)); + TRACE_WARN_ON(!plist_node_empty(&waiter->pi_list_entry)); + TRACE_WARN_ON(waiter->task); + memset(waiter, 0x22, sizeof(*waiter)); +} + +void debug_rt_mutex_init(struct rt_mutex *lock, const char *name) +{ + void *addr = lock; + + if (rt_trace_on) { + rt_mutex_debug_check_no_locks_freed(addr, + sizeof(struct rt_mutex)); + INIT_LIST_HEAD(&lock->held_list_entry); + lock->name = name; + } +} + +void rt_mutex_deadlock_account_lock(struct rt_mutex *lock, task_t *task) +{ +} + +void rt_mutex_deadlock_account_unlock(struct task_struct *task) +{ +} + diff --git a/kernel/rtmutex-debug.h b/kernel/rtmutex-debug.h new file mode 100644 index 00000000000..7612fbc62d7 --- /dev/null +++ b/kernel/rtmutex-debug.h @@ -0,0 +1,37 @@ +/* + * RT-Mutexes: blocking mutual exclusion locks with PI support + * + * started by Ingo Molnar and Thomas Gleixner: + * + * Copyright (C) 2004-2006 Red Hat, Inc., Ingo Molnar + * Copyright (C) 2006, Timesys Corp., Thomas Gleixner + * + * This file contains macros used solely by rtmutex.c. Debug version. + */ + +#define __IP_DECL__ , unsigned long ip +#define __IP__ , ip +#define __RET_IP__ , (unsigned long)__builtin_return_address(0) + +extern void +rt_mutex_deadlock_account_lock(struct rt_mutex *lock, struct task_struct *task); +extern void rt_mutex_deadlock_account_unlock(struct task_struct *task); +extern void debug_rt_mutex_init_waiter(struct rt_mutex_waiter *waiter); +extern void debug_rt_mutex_free_waiter(struct rt_mutex_waiter *waiter); +extern void debug_rt_mutex_init(struct rt_mutex *lock, const char *name); +extern void debug_rt_mutex_lock(struct rt_mutex *lock __IP_DECL__); +extern void debug_rt_mutex_unlock(struct rt_mutex *lock); +extern void debug_rt_mutex_proxy_lock(struct rt_mutex *lock, + struct task_struct *powner __IP_DECL__); +extern void debug_rt_mutex_proxy_unlock(struct rt_mutex *lock); +extern void debug_rt_mutex_deadlock(int detect, struct rt_mutex_waiter *waiter, + struct rt_mutex *lock); +extern void debug_rt_mutex_print_deadlock(struct rt_mutex_waiter *waiter); +# define debug_rt_mutex_reset_waiter(w) \ + do { (w)->deadlock_lock = NULL; } while (0) + +static inline int debug_rt_mutex_detect_deadlock(struct rt_mutex_waiter *waiter, + int detect) +{ + return (waiter != NULL); +} diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug index 8bab0102ac7..06d3ea1133c 100644 --- a/lib/Kconfig.debug +++ b/lib/Kconfig.debug @@ -107,6 +107,19 @@ config DEBUG_MUTEXES This allows mutex semantics violations and mutex related deadlocks (lockups) to be detected and reported automatically. +config DEBUG_RT_MUTEXES + bool "RT Mutex debugging, deadlock detection" + default y + depends on DEBUG_KERNEL && RT_MUTEXES + help + This allows rt mutex semantics violations and rt mutex related + deadlocks (lockups) to be detected and reported automatically. + +config DEBUG_PI_LIST + bool + default y + depends on DEBUG_RT_MUTEXES + config DEBUG_SPINLOCK bool "Spinlock debugging" depends on DEBUG_KERNEL diff --git a/mm/slab.c b/mm/slab.c index f378d027c68..233e39d14ca 100644 --- a/mm/slab.c +++ b/mm/slab.c @@ -107,6 +107,7 @@ #include #include #include +#include #include #include -- cgit v1.2.3-70-g09d2 From 61a87122869b6340a63b6f9f84097d3688604b90 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Tue, 27 Jun 2006 02:54:56 -0700 Subject: [PATCH] pi-futex: rt mutex tester RT-mutex tester: scriptable tester for rt mutexes, which allows userspace scripting of mutex unit-tests (and dynamic tests as well), using the actual rt-mutex implementation of the kernel. [akpm@osdl.org: fixlet] Signed-off-by: Thomas Gleixner Signed-off-by: Ingo Molnar Signed-off-by: Arjan van de Ven Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/sched.h | 1 + kernel/Makefile | 1 + kernel/rtmutex-tester.c | 436 +++++++++++++++++++++++++++ kernel/rtmutex.c | 3 +- kernel/rtmutex_common.h | 22 ++ lib/Kconfig.debug | 7 + scripts/rt-tester/check-all.sh | 21 ++ scripts/rt-tester/rt-tester.py | 222 ++++++++++++++ scripts/rt-tester/t2-l1-2rt-sameprio.tst | 101 +++++++ scripts/rt-tester/t2-l1-pi.tst | 84 ++++++ scripts/rt-tester/t2-l1-signal.tst | 79 +++++ scripts/rt-tester/t2-l2-2rt-deadlock.tst | 91 ++++++ scripts/rt-tester/t3-l1-pi-1rt.tst | 95 ++++++ scripts/rt-tester/t3-l1-pi-2rt.tst | 96 ++++++ scripts/rt-tester/t3-l1-pi-3rt.tst | 95 ++++++ scripts/rt-tester/t3-l1-pi-signal.tst | 101 +++++++ scripts/rt-tester/t3-l1-pi-steal.tst | 99 ++++++ scripts/rt-tester/t3-l2-pi.tst | 95 ++++++ scripts/rt-tester/t4-l2-pi-deboost.tst | 127 ++++++++ scripts/rt-tester/t5-l4-pi-boost-deboost.tst | 148 +++++++++ 20 files changed, 1923 insertions(+), 1 deletion(-) create mode 100644 kernel/rtmutex-tester.c create mode 100644 scripts/rt-tester/check-all.sh create mode 100644 scripts/rt-tester/rt-tester.py create mode 100644 scripts/rt-tester/t2-l1-2rt-sameprio.tst create mode 100644 scripts/rt-tester/t2-l1-pi.tst create mode 100644 scripts/rt-tester/t2-l1-signal.tst create mode 100644 scripts/rt-tester/t2-l2-2rt-deadlock.tst create mode 100644 scripts/rt-tester/t3-l1-pi-1rt.tst create mode 100644 scripts/rt-tester/t3-l1-pi-2rt.tst create mode 100644 scripts/rt-tester/t3-l1-pi-3rt.tst create mode 100644 scripts/rt-tester/t3-l1-pi-signal.tst create mode 100644 scripts/rt-tester/t3-l1-pi-steal.tst create mode 100644 scripts/rt-tester/t3-l2-pi.tst create mode 100644 scripts/rt-tester/t4-l2-pi-deboost.tst create mode 100644 scripts/rt-tester/t5-l4-pi-boost-deboost.tst (limited to 'include') diff --git a/include/linux/sched.h b/include/linux/sched.h index 6ea23c9af41..edadd13cf53 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -982,6 +982,7 @@ static inline void put_task_struct(struct task_struct *t) #define PF_SPREAD_PAGE 0x01000000 /* Spread page cache over cpuset */ #define PF_SPREAD_SLAB 0x02000000 /* Spread some slab caches over cpuset */ #define PF_MEMPOLICY 0x10000000 /* Non-default NUMA mempolicy */ +#define PF_MUTEX_TESTER 0x20000000 /* Thread belongs to the rt mutex tester */ /* * Only the _current_ task can read/write to tsk->flags, but other diff --git a/kernel/Makefile b/kernel/Makefile index f9c92d34cde..82fb182f6f6 100644 --- a/kernel/Makefile +++ b/kernel/Makefile @@ -18,6 +18,7 @@ obj-$(CONFIG_FUTEX) += futex_compat.o endif obj-$(CONFIG_RT_MUTEXES) += rtmutex.o obj-$(CONFIG_DEBUG_RT_MUTEXES) += rtmutex-debug.o +obj-$(CONFIG_RT_MUTEX_TESTER) += rtmutex-tester.o obj-$(CONFIG_GENERIC_ISA_DMA) += dma.o obj-$(CONFIG_SMP) += cpu.o spinlock.o obj-$(CONFIG_DEBUG_SPINLOCK) += spinlock.o diff --git a/kernel/rtmutex-tester.c b/kernel/rtmutex-tester.c new file mode 100644 index 00000000000..fe211ba3a5b --- /dev/null +++ b/kernel/rtmutex-tester.c @@ -0,0 +1,436 @@ +/* + * RT-Mutex-tester: scriptable tester for rt mutexes + * + * started by Thomas Gleixner: + * + * Copyright (C) 2006, Timesys Corp., Thomas Gleixner + * + */ +#include +#include +#include +#include +#include +#include +#include +#include + +#include "rtmutex.h" + +#define MAX_RT_TEST_THREADS 8 +#define MAX_RT_TEST_MUTEXES 8 + +static spinlock_t rttest_lock; +static atomic_t rttest_event; + +struct test_thread_data { + int opcode; + int opdata; + int mutexes[MAX_RT_TEST_MUTEXES]; + int bkl; + int event; + struct sys_device sysdev; +}; + +static struct test_thread_data thread_data[MAX_RT_TEST_THREADS]; +static task_t *threads[MAX_RT_TEST_THREADS]; +static struct rt_mutex mutexes[MAX_RT_TEST_MUTEXES]; + +enum test_opcodes { + RTTEST_NOP = 0, + RTTEST_SCHEDOT, /* 1 Sched other, data = nice */ + RTTEST_SCHEDRT, /* 2 Sched fifo, data = prio */ + RTTEST_LOCK, /* 3 Lock uninterruptible, data = lockindex */ + RTTEST_LOCKNOWAIT, /* 4 Lock uninterruptible no wait in wakeup, data = lockindex */ + RTTEST_LOCKINT, /* 5 Lock interruptible, data = lockindex */ + RTTEST_LOCKINTNOWAIT, /* 6 Lock interruptible no wait in wakeup, data = lockindex */ + RTTEST_LOCKCONT, /* 7 Continue locking after the wakeup delay */ + RTTEST_UNLOCK, /* 8 Unlock, data = lockindex */ + RTTEST_LOCKBKL, /* 9 Lock BKL */ + RTTEST_UNLOCKBKL, /* 10 Unlock BKL */ + RTTEST_SIGNAL, /* 11 Signal other test thread, data = thread id */ + RTTEST_RESETEVENT = 98, /* 98 Reset event counter */ + RTTEST_RESET = 99, /* 99 Reset all pending operations */ +}; + +static int handle_op(struct test_thread_data *td, int lockwakeup) +{ + struct sched_param schedpar; + int i, id, ret = -EINVAL; + + switch(td->opcode) { + + case RTTEST_NOP: + return 0; + + case RTTEST_SCHEDOT: + schedpar.sched_priority = 0; + ret = sched_setscheduler(current, SCHED_NORMAL, &schedpar); + if (!ret) + set_user_nice(current, 0); + return ret; + + case RTTEST_SCHEDRT: + schedpar.sched_priority = td->opdata; + return sched_setscheduler(current, SCHED_FIFO, &schedpar); + + case RTTEST_LOCKCONT: + td->mutexes[td->opdata] = 1; + td->event = atomic_add_return(1, &rttest_event); + return 0; + + case RTTEST_RESET: + for (i = 0; i < MAX_RT_TEST_MUTEXES; i++) { + if (td->mutexes[i] == 4) { + rt_mutex_unlock(&mutexes[i]); + td->mutexes[i] = 0; + } + } + + if (!lockwakeup && td->bkl == 4) { + unlock_kernel(); + td->bkl = 0; + } + return 0; + + case RTTEST_RESETEVENT: + atomic_set(&rttest_event, 0); + return 0; + + default: + if (lockwakeup) + return ret; + } + + switch(td->opcode) { + + case RTTEST_LOCK: + case RTTEST_LOCKNOWAIT: + id = td->opdata; + if (id < 0 || id >= MAX_RT_TEST_MUTEXES) + return ret; + + td->mutexes[id] = 1; + td->event = atomic_add_return(1, &rttest_event); + rt_mutex_lock(&mutexes[id]); + td->event = atomic_add_return(1, &rttest_event); + td->mutexes[id] = 4; + return 0; + + case RTTEST_LOCKINT: + case RTTEST_LOCKINTNOWAIT: + id = td->opdata; + if (id < 0 || id >= MAX_RT_TEST_MUTEXES) + return ret; + + td->mutexes[id] = 1; + td->event = atomic_add_return(1, &rttest_event); + ret = rt_mutex_lock_interruptible(&mutexes[id], 0); + td->event = atomic_add_return(1, &rttest_event); + td->mutexes[id] = ret ? 0 : 4; + return ret ? -EINTR : 0; + + case RTTEST_UNLOCK: + id = td->opdata; + if (id < 0 || id >= MAX_RT_TEST_MUTEXES || td->mutexes[id] != 4) + return ret; + + td->event = atomic_add_return(1, &rttest_event); + rt_mutex_unlock(&mutexes[id]); + td->event = atomic_add_return(1, &rttest_event); + td->mutexes[id] = 0; + return 0; + + case RTTEST_LOCKBKL: + if (td->bkl) + return 0; + td->bkl = 1; + lock_kernel(); + td->bkl = 4; + return 0; + + case RTTEST_UNLOCKBKL: + if (td->bkl != 4) + break; + unlock_kernel(); + td->bkl = 0; + return 0; + + default: + break; + } + return ret; +} + +/* + * Schedule replacement for rtsem_down(). Only called for threads with + * PF_MUTEX_TESTER set. + * + * This allows us to have finegrained control over the event flow. + * + */ +void schedule_rt_mutex_test(struct rt_mutex *mutex) +{ + int tid, op, dat; + struct test_thread_data *td; + + /* We have to lookup the task */ + for (tid = 0; tid < MAX_RT_TEST_THREADS; tid++) { + if (threads[tid] == current) + break; + } + + BUG_ON(tid == MAX_RT_TEST_THREADS); + + td = &thread_data[tid]; + + op = td->opcode; + dat = td->opdata; + + switch (op) { + case RTTEST_LOCK: + case RTTEST_LOCKINT: + case RTTEST_LOCKNOWAIT: + case RTTEST_LOCKINTNOWAIT: + if (mutex != &mutexes[dat]) + break; + + if (td->mutexes[dat] != 1) + break; + + td->mutexes[dat] = 2; + td->event = atomic_add_return(1, &rttest_event); + break; + + case RTTEST_LOCKBKL: + default: + break; + } + + schedule(); + + + switch (op) { + case RTTEST_LOCK: + case RTTEST_LOCKINT: + if (mutex != &mutexes[dat]) + return; + + if (td->mutexes[dat] != 2) + return; + + td->mutexes[dat] = 3; + td->event = atomic_add_return(1, &rttest_event); + break; + + case RTTEST_LOCKNOWAIT: + case RTTEST_LOCKINTNOWAIT: + if (mutex != &mutexes[dat]) + return; + + if (td->mutexes[dat] != 2) + return; + + td->mutexes[dat] = 1; + td->event = atomic_add_return(1, &rttest_event); + return; + + case RTTEST_LOCKBKL: + return; + default: + return; + } + + td->opcode = 0; + + for (;;) { + set_current_state(TASK_INTERRUPTIBLE); + + if (td->opcode > 0) { + int ret; + + set_current_state(TASK_RUNNING); + ret = handle_op(td, 1); + set_current_state(TASK_INTERRUPTIBLE); + if (td->opcode == RTTEST_LOCKCONT) + break; + td->opcode = ret; + } + + /* Wait for the next command to be executed */ + schedule(); + } + + /* Restore previous command and data */ + td->opcode = op; + td->opdata = dat; +} + +static int test_func(void *data) +{ + struct test_thread_data *td = data; + int ret; + + current->flags |= PF_MUTEX_TESTER; + allow_signal(SIGHUP); + + for(;;) { + + set_current_state(TASK_INTERRUPTIBLE); + + if (td->opcode > 0) { + set_current_state(TASK_RUNNING); + ret = handle_op(td, 0); + set_current_state(TASK_INTERRUPTIBLE); + td->opcode = ret; + } + + /* Wait for the next command to be executed */ + schedule(); + + if (signal_pending(current)) + flush_signals(current); + + if(kthread_should_stop()) + break; + } + return 0; +} + +/** + * sysfs_test_command - interface for test commands + * @dev: thread reference + * @buf: command for actual step + * @count: length of buffer + * + * command syntax: + * + * opcode:data + */ +static ssize_t sysfs_test_command(struct sys_device *dev, const char *buf, + size_t count) +{ + struct test_thread_data *td; + char cmdbuf[32]; + int op, dat, tid; + + td = container_of(dev, struct test_thread_data, sysdev); + tid = td->sysdev.id; + + /* strings from sysfs write are not 0 terminated! */ + if (count >= sizeof(cmdbuf)) + return -EINVAL; + + /* strip of \n: */ + if (buf[count-1] == '\n') + count--; + if (count < 1) + return -EINVAL; + + memcpy(cmdbuf, buf, count); + cmdbuf[count] = 0; + + if (sscanf(cmdbuf, "%d:%d", &op, &dat) != 2) + return -EINVAL; + + switch (op) { + case RTTEST_SIGNAL: + send_sig(SIGHUP, threads[tid], 0); + break; + + default: + if (td->opcode > 0) + return -EBUSY; + td->opdata = dat; + td->opcode = op; + wake_up_process(threads[tid]); + } + + return count; +} + +/** + * sysfs_test_status - sysfs interface for rt tester + * @dev: thread to query + * @buf: char buffer to be filled with thread status info + */ +static ssize_t sysfs_test_status(struct sys_device *dev, char *buf) +{ + struct test_thread_data *td; + char *curr = buf; + task_t *tsk; + int i; + + td = container_of(dev, struct test_thread_data, sysdev); + tsk = threads[td->sysdev.id]; + + spin_lock(&rttest_lock); + + curr += sprintf(curr, + "O: %4d, E:%8d, S: 0x%08lx, P: %4d, N: %4d, B: %p, K: %d, M:", + td->opcode, td->event, tsk->state, + (MAX_RT_PRIO - 1) - tsk->prio, + (MAX_RT_PRIO - 1) - tsk->normal_prio, + tsk->pi_blocked_on, td->bkl); + + for (i = MAX_RT_TEST_MUTEXES - 1; i >=0 ; i--) + curr += sprintf(curr, "%d", td->mutexes[i]); + + spin_unlock(&rttest_lock); + + curr += sprintf(curr, ", T: %p, R: %p\n", tsk, + mutexes[td->sysdev.id].owner); + + return curr - buf; +} + +static SYSDEV_ATTR(status, 0600, sysfs_test_status, NULL); +static SYSDEV_ATTR(command, 0600, NULL, sysfs_test_command); + +static struct sysdev_class rttest_sysclass = { + set_kset_name("rttest"), +}; + +static int init_test_thread(int id) +{ + thread_data[id].sysdev.cls = &rttest_sysclass; + thread_data[id].sysdev.id = id; + + threads[id] = kthread_run(test_func, &thread_data[id], "rt-test-%d", id); + if (IS_ERR(threads[id])) + return PTR_ERR(threads[id]); + + return sysdev_register(&thread_data[id].sysdev); +} + +static int init_rttest(void) +{ + int ret, i; + + spin_lock_init(&rttest_lock); + + for (i = 0; i < MAX_RT_TEST_MUTEXES; i++) + rt_mutex_init(&mutexes[i]); + + ret = sysdev_class_register(&rttest_sysclass); + if (ret) + return ret; + + for (i = 0; i < MAX_RT_TEST_THREADS; i++) { + ret = init_test_thread(i); + if (ret) + break; + ret = sysdev_create_file(&thread_data[i].sysdev, &attr_status); + if (ret) + break; + ret = sysdev_create_file(&thread_data[i].sysdev, &attr_command); + if (ret) + break; + } + + printk("Initializing RT-Tester: %s\n", ret ? "Failed" : "OK" ); + + return ret; +} + +device_initcall(init_rttest); diff --git a/kernel/rtmutex.c b/kernel/rtmutex.c index 937a474fae9..39c8ca0cf52 100644 --- a/kernel/rtmutex.c +++ b/kernel/rtmutex.c @@ -640,7 +640,8 @@ rt_mutex_slowlock(struct rt_mutex *lock, int state, debug_rt_mutex_print_deadlock(&waiter); - schedule(); + if (waiter.task) + schedule_rt_mutex(lock); spin_lock(&lock->wait_lock); set_current_state(state); diff --git a/kernel/rtmutex_common.h b/kernel/rtmutex_common.h index 50eed60eb08..e068024eeff 100644 --- a/kernel/rtmutex_common.h +++ b/kernel/rtmutex_common.h @@ -14,6 +14,28 @@ #include +/* + * The rtmutex in kernel tester is independent of rtmutex debugging. We + * call schedule_rt_mutex_test() instead of schedule() for the tasks which + * belong to the tester. That way we can delay the wakeup path of those + * threads to provoke lock stealing and testing of complex boosting scenarios. + */ +#ifdef CONFIG_RT_MUTEX_TESTER + +extern void schedule_rt_mutex_test(struct rt_mutex *lock); + +#define schedule_rt_mutex(_lock) \ + do { \ + if (!(current->flags & PF_MUTEX_TESTER)) \ + schedule(); \ + else \ + schedule_rt_mutex_test(_lock); \ + } while (0) + +#else +# define schedule_rt_mutex(_lock) schedule() +#endif + /* * This is the control structure for tasks blocked on a rt_mutex, * which is allocated on the kernel stack on of the blocked task. diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug index 06d3ea1133c..6afe0af0a19 100644 --- a/lib/Kconfig.debug +++ b/lib/Kconfig.debug @@ -120,6 +120,13 @@ config DEBUG_PI_LIST default y depends on DEBUG_RT_MUTEXES +config RT_MUTEX_TESTER + bool "Built-in scriptable tester for rt-mutexes" + depends on RT_MUTEXES + default n + help + This option enables a rt-mutex tester. + config DEBUG_SPINLOCK bool "Spinlock debugging" depends on DEBUG_KERNEL diff --git a/scripts/rt-tester/check-all.sh b/scripts/rt-tester/check-all.sh new file mode 100644 index 00000000000..ac45c3e65a3 --- /dev/null +++ b/scripts/rt-tester/check-all.sh @@ -0,0 +1,21 @@ + + +function testit () +{ + printf "%-30s: " $1 + ./rt-tester.py $1 | grep Pass +} + +testit t2-l1-2rt-sameprio.tst +testit t2-l1-pi.tst +testit t2-l1-signal.tst +#testit t2-l2-2rt-deadlock.tst +testit t3-l1-pi-1rt.tst +testit t3-l1-pi-2rt.tst +testit t3-l1-pi-3rt.tst +testit t3-l1-pi-signal.tst +testit t3-l1-pi-steal.tst +testit t3-l2-pi.tst +testit t4-l2-pi-deboost.tst +testit t5-l4-pi-boost-deboost.tst + diff --git a/scripts/rt-tester/rt-tester.py b/scripts/rt-tester/rt-tester.py new file mode 100644 index 00000000000..4c79660793c --- /dev/null +++ b/scripts/rt-tester/rt-tester.py @@ -0,0 +1,222 @@ +#!/usr/bin/env python +# +# rt-mutex tester +# +# (C) 2006 Thomas Gleixner +# +# This program is free software; you can redistribute it and/or modify +# it under the terms of the GNU General Public License version 2 as +# published by the Free Software Foundation. +# +import os +import sys +import getopt +import shutil +import string + +# Globals +quiet = 0 +test = 0 +comments = 0 + +sysfsprefix = "/sys/devices/system/rttest/rttest" +statusfile = "/status" +commandfile = "/command" + +# Command opcodes +cmd_opcodes = { + "schedother" : "1", + "schedfifo" : "2", + "lock" : "3", + "locknowait" : "4", + "lockint" : "5", + "lockintnowait" : "6", + "lockcont" : "7", + "unlock" : "8", + "lockbkl" : "9", + "unlockbkl" : "10", + "signal" : "11", + "resetevent" : "98", + "reset" : "99", + } + +test_opcodes = { + "prioeq" : ["P" , "eq" , None], + "priolt" : ["P" , "lt" , None], + "priogt" : ["P" , "gt" , None], + "nprioeq" : ["N" , "eq" , None], + "npriolt" : ["N" , "lt" , None], + "npriogt" : ["N" , "gt" , None], + "unlocked" : ["M" , "eq" , 0], + "trylock" : ["M" , "eq" , 1], + "blocked" : ["M" , "eq" , 2], + "blockedwake" : ["M" , "eq" , 3], + "locked" : ["M" , "eq" , 4], + "opcodeeq" : ["O" , "eq" , None], + "opcodelt" : ["O" , "lt" , None], + "opcodegt" : ["O" , "gt" , None], + "eventeq" : ["E" , "eq" , None], + "eventlt" : ["E" , "lt" , None], + "eventgt" : ["E" , "gt" , None], + } + +# Print usage information +def usage(): + print "rt-tester.py <-c -h -q -t> " + print " -c display comments after first command" + print " -h help" + print " -q quiet mode" + print " -t test mode (syntax check)" + print " testfile: read test specification from testfile" + print " otherwise from stdin" + return + +# Print progress when not in quiet mode +def progress(str): + if not quiet: + print str + +# Analyse a status value +def analyse(val, top, arg): + + intval = int(val) + + if top[0] == "M": + intval = intval / (10 ** int(arg)) + intval = intval % 10 + argval = top[2] + elif top[0] == "O": + argval = int(cmd_opcodes.get(arg, arg)) + else: + argval = int(arg) + + # progress("%d %s %d" %(intval, top[1], argval)) + + if top[1] == "eq" and intval == argval: + return 1 + if top[1] == "lt" and intval < argval: + return 1 + if top[1] == "gt" and intval > argval: + return 1 + return 0 + +# Parse the commandline +try: + (options, arguments) = getopt.getopt(sys.argv[1:],'chqt') +except getopt.GetoptError, ex: + usage() + sys.exit(1) + +# Parse commandline options +for option, value in options: + if option == "-c": + comments = 1 + elif option == "-q": + quiet = 1 + elif option == "-t": + test = 1 + elif option == '-h': + usage() + sys.exit(0) + +# Select the input source +if arguments: + try: + fd = open(arguments[0]) + except Exception,ex: + sys.stderr.write("File not found %s\n" %(arguments[0])) + sys.exit(1) +else: + fd = sys.stdin + +linenr = 0 + +# Read the test patterns +while 1: + + linenr = linenr + 1 + line = fd.readline() + if not len(line): + break + + line = line.strip() + parts = line.split(":") + + if not parts or len(parts) < 1: + continue + + if len(parts[0]) == 0: + continue + + if parts[0].startswith("#"): + if comments > 1: + progress(line) + continue + + if comments == 1: + comments = 2 + + progress(line) + + cmd = parts[0].strip().lower() + opc = parts[1].strip().lower() + tid = parts[2].strip() + dat = parts[3].strip() + + try: + # Test or wait for a status value + if cmd == "t" or cmd == "w": + testop = test_opcodes[opc] + + fname = "%s%s%s" %(sysfsprefix, tid, statusfile) + if test: + print fname + continue + + while 1: + query = 1 + fsta = open(fname, 'r') + status = fsta.readline().strip() + fsta.close() + stat = status.split(",") + for s in stat: + s = s.strip() + if s.startswith(testop[0]): + # Seperate status value + val = s[2:].strip() + query = analyse(val, testop, dat) + break + if query or cmd == "t": + break + + progress(" " + status) + + if not query: + sys.stderr.write("Test failed in line %d\n" %(linenr)) + sys.exit(1) + + # Issue a command to the tester + elif cmd == "c": + cmdnr = cmd_opcodes[opc] + # Build command string and sys filename + cmdstr = "%s:%s" %(cmdnr, dat) + fname = "%s%s%s" %(sysfsprefix, tid, commandfile) + if test: + print fname + continue + fcmd = open(fname, 'w') + fcmd.write(cmdstr) + fcmd.close() + + except Exception,ex: + sys.stderr.write(str(ex)) + sys.stderr.write("\nSyntax error in line %d\n" %(linenr)) + if not test: + fd.close() + sys.exit(1) + +# Normal exit pass +print "Pass" +sys.exit(0) + + diff --git a/scripts/rt-tester/t2-l1-2rt-sameprio.tst b/scripts/rt-tester/t2-l1-2rt-sameprio.tst new file mode 100644 index 00000000000..a2b6f2aae75 --- /dev/null +++ b/scripts/rt-tester/t2-l1-2rt-sameprio.tst @@ -0,0 +1,101 @@ +# +# RT-Mutex test +# +# Op: C(ommand)/T(est)/W(ait) +# | opcode +# | | threadid: 0-7 +# | | | opcode argument +# | | | | +# C: lock: 0: 0 +# +# Commands +# +# opcode opcode argument +# schedother nice value +# schedfifo priority +# lock lock nr (0-7) +# locknowait lock nr (0-7) +# lockint lock nr (0-7) +# lockintnowait lock nr (0-7) +# lockcont lock nr (0-7) +# unlock lock nr (0-7) +# lockbkl lock nr (0-7) +# unlockbkl lock nr (0-7) +# signal 0 +# reset 0 +# resetevent 0 +# +# Tests / Wait +# +# opcode opcode argument +# +# prioeq priority +# priolt priority +# priogt priority +# nprioeq normal priority +# npriolt normal priority +# npriogt normal priority +# locked lock nr (0-7) +# blocked lock nr (0-7) +# blockedwake lock nr (0-7) +# unlocked lock nr (0-7) +# lockedbkl dont care +# blockedbkl dont care +# unlockedbkl dont care +# opcodeeq command opcode or number +# opcodelt number +# opcodegt number +# eventeq number +# eventgt number +# eventlt number + +# +# 2 threads 1 lock +# +C: resetevent: 0: 0 +W: opcodeeq: 0: 0 + +# Set schedulers +C: schedfifo: 0: 80 +W: opcodeeq: 0: 0 +C: schedfifo: 1: 80 +W: opcodeeq: 1: 0 + +# T0 lock L0 +C: locknowait: 0: 0 +C: locknowait: 1: 0 +W: locked: 0: 0 +W: blocked: 1: 0 +T: prioeq: 0: 80 + +# T0 unlock L0 +C: unlock: 0: 0 +W: locked: 1: 0 + +# Verify T0 +W: unlocked: 0: 0 +T: prioeq: 0: 80 + +# Unlock +C: unlock: 1: 0 +W: unlocked: 1: 0 + +# T1,T0 lock L0 +C: locknowait: 1: 0 +C: locknowait: 0: 0 +W: locked: 1: 0 +W: blocked: 0: 0 +T: prioeq: 1: 80 + +# T1 unlock L0 +C: unlock: 1: 0 +W: locked: 0: 0 + +# Verify T1 +W: unlocked: 1: 0 +T: prioeq: 1: 80 + +# Unlock and exit +C: unlock: 0: 0 +W: unlocked: 0: 0 + diff --git a/scripts/rt-tester/t2-l1-pi.tst b/scripts/rt-tester/t2-l1-pi.tst new file mode 100644 index 00000000000..aa4c8940acd --- /dev/null +++ b/scripts/rt-tester/t2-l1-pi.tst @@ -0,0 +1,84 @@ +# +# RT-Mutex test +# +# Op: C(ommand)/T(est)/W(ait) +# | opcode +# | | threadid: 0-7 +# | | | opcode argument +# | | | | +# C: lock: 0: 0 +# +# Commands +# +# opcode opcode argument +# schedother nice value +# schedfifo priority +# lock lock nr (0-7) +# locknowait lock nr (0-7) +# lockint lock nr (0-7) +# lockintnowait lock nr (0-7) +# lockcont lock nr (0-7) +# unlock lock nr (0-7) +# lockbkl lock nr (0-7) +# unlockbkl lock nr (0-7) +# signal 0 +# reset 0 +# resetevent 0 +# +# Tests / Wait +# +# opcode opcode argument +# +# prioeq priority +# priolt priority +# priogt priority +# nprioeq normal priority +# npriolt normal priority +# npriogt normal priority +# locked lock nr (0-7) +# blocked lock nr (0-7) +# blockedwake lock nr (0-7) +# unlocked lock nr (0-7) +# lockedbkl dont care +# blockedbkl dont care +# unlockedbkl dont care +# opcodeeq command opcode or number +# opcodelt number +# opcodegt number +# eventeq number +# eventgt number +# eventlt number + +# +# 2 threads 1 lock with priority inversion +# +C: resetevent: 0: 0 +W: opcodeeq: 0: 0 + +# Set schedulers +C: schedother: 0: 0 +W: opcodeeq: 0: 0 +C: schedfifo: 1: 80 +W: opcodeeq: 1: 0 + +# T0 lock L0 +C: locknowait: 0: 0 +W: locked: 0: 0 + +# T1 lock L0 +C: locknowait: 1: 0 +W: blocked: 1: 0 +T: prioeq: 0: 80 + +# T0 unlock L0 +C: unlock: 0: 0 +W: locked: 1: 0 + +# Verify T1 +W: unlocked: 0: 0 +T: priolt: 0: 1 + +# Unlock and exit +C: unlock: 1: 0 +W: unlocked: 1: 0 + diff --git a/scripts/rt-tester/t2-l1-signal.tst b/scripts/rt-tester/t2-l1-signal.tst new file mode 100644 index 00000000000..c47ba06af20 --- /dev/null +++ b/scripts/rt-tester/t2-l1-signal.tst @@ -0,0 +1,79 @@ +# +# RT-Mutex test +# +# Op: C(ommand)/T(est)/W(ait) +# | opcode +# | | threadid: 0-7 +# | | | opcode argument +# | | | | +# C: lock: 0: 0 +# +# Commands +# +# opcode opcode argument +# schedother nice value +# schedfifo priority +# lock lock nr (0-7) +# locknowait lock nr (0-7) +# lockint lock nr (0-7) +# lockintnowait lock nr (0-7) +# lockcont lock nr (0-7) +# unlock lock nr (0-7) +# lockbkl lock nr (0-7) +# unlockbkl lock nr (0-7) +# signal 0 +# reset 0 +# resetevent 0 +# +# Tests / Wait +# +# opcode opcode argument +# +# prioeq priority +# priolt priority +# priogt priority +# nprioeq normal priority +# npriolt normal priority +# npriogt normal priority +# locked lock nr (0-7) +# blocked lock nr (0-7) +# blockedwake lock nr (0-7) +# unlocked lock nr (0-7) +# lockedbkl dont care +# blockedbkl dont care +# unlockedbkl dont care +# opcodeeq command opcode or number +# opcodelt number +# opcodegt number +# eventeq number +# eventgt number +# eventlt number + +# +# 2 threads 1 lock with priority inversion +# +C: resetevent: 0: 0 +W: opcodeeq: 0: 0 + +# Set schedulers +C: schedother: 0: 0 +W: opcodeeq: 0: 0 +C: schedother: 1: 0 +W: opcodeeq: 1: 0 + +# T0 lock L0 +C: locknowait: 0: 0 +W: locked: 0: 0 + +# T1 lock L0 +C: lockintnowait: 1: 0 +W: blocked: 1: 0 + +# Interrupt T1 +C: signal: 1: 0 +W: unlocked: 1: 0 +T: opcodeeq: 1: -4 + +# Unlock and exit +C: unlock: 0: 0 +W: unlocked: 0: 0 diff --git a/scripts/rt-tester/t2-l2-2rt-deadlock.tst b/scripts/rt-tester/t2-l2-2rt-deadlock.tst new file mode 100644 index 00000000000..0cee476b206 --- /dev/null +++ b/scripts/rt-tester/t2-l2-2rt-deadlock.tst @@ -0,0 +1,91 @@ +# +# RT-Mutex test +# +# Op: C(ommand)/T(est)/W(ait) +# | opcode +# | | threadid: 0-7 +# | | | opcode argument +# | | | | +# C: lock: 0: 0 +# +# Commands +# +# opcode opcode argument +# schedother nice value +# schedfifo priority +# lock lock nr (0-7) +# locknowait lock nr (0-7) +# lockint lock nr (0-7) +# lockintnowait lock nr (0-7) +# lockcont lock nr (0-7) +# unlock lock nr (0-7) +# lockbkl lock nr (0-7) +# unlockbkl lock nr (0-7) +# signal 0 +# reset 0 +# resetevent 0 +# +# Tests / Wait +# +# opcode opcode argument +# +# prioeq priority +# priolt priority +# priogt priority +# nprioeq normal priority +# npriolt normal priority +# npriogt normal priority +# locked lock nr (0-7) +# blocked lock nr (0-7) +# blockedwake lock nr (0-7) +# unlocked lock nr (0-7) +# lockedbkl dont care +# blockedbkl dont care +# unlockedbkl dont care +# opcodeeq command opcode or number +# opcodelt number +# opcodegt number +# eventeq number +# eventgt number +# eventlt number + +# +# 2 threads 2 lock +# +C: resetevent: 0: 0 +W: opcodeeq: 0: 0 + +# Set schedulers +C: schedfifo: 0: 80 +W: opcodeeq: 0: 0 +C: schedfifo: 1: 80 +W: opcodeeq: 1: 0 + +# T0 lock L0 +C: locknowait: 0: 0 +W: locked: 0: 0 + +# T1 lock L1 +C: locknowait: 1: 1 +W: locked: 1: 1 + +# T0 lock L1 +C: lockintnowait: 0: 1 +W: blocked: 0: 1 + +# T1 lock L0 +C: lockintnowait: 1: 0 +W: blocked: 1: 0 + +# Make deadlock go away +C: signal: 1: 0 +W: unlocked: 1: 0 +C: signal: 0: 0 +W: unlocked: 0: 1 + +# Unlock and exit +C: unlock: 0: 0 +W: unlocked: 0: 0 +C: unlock: 1: 1 +W: unlocked: 1: 1 + diff --git a/scripts/rt-tester/t3-l1-pi-1rt.tst b/scripts/rt-tester/t3-l1-pi-1rt.tst new file mode 100644 index 00000000000..a5eaf7ed5d5 --- /dev/null +++ b/scripts/rt-tester/t3-l1-pi-1rt.tst @@ -0,0 +1,95 @@ +# +# rt-mutex test +# +# Op: C(ommand)/T(est)/W(ait) +# | opcode +# | | threadid: 0-7 +# | | | opcode argument +# | | | | +# C: lock: 0: 0 +# +# Commands +# +# opcode opcode argument +# schedother nice value +# schedfifo priority +# lock lock nr (0-7) +# locknowait lock nr (0-7) +# lockint lock nr (0-7) +# lockintnowait lock nr (0-7) +# lockcont lock nr (0-7) +# unlock lock nr (0-7) +# lockbkl lock nr (0-7) +# unlockbkl lock nr (0-7) +# signal thread to signal (0-7) +# reset 0 +# resetevent 0 +# +# Tests / Wait +# +# opcode opcode argument +# +# prioeq priority +# priolt priority +# priogt priority +# nprioeq normal priority +# npriolt normal priority +# npriogt normal priority +# locked lock nr (0-7) +# blocked lock nr (0-7) +# blockedwake lock nr (0-7) +# unlocked lock nr (0-7) +# lockedbkl dont care +# blockedbkl dont care +# unlockedbkl dont care +# opcodeeq command opcode or number +# opcodelt number +# opcodegt number +# eventeq number +# eventgt number +# eventlt number + +# +# 3 threads 1 lock PI +# +C: resetevent: 0: 0 +W: opcodeeq: 0: 0 + +# Set schedulers +C: schedother: 0: 0 +W: opcodeeq: 0: 0 +C: schedother: 1: 0 +W: opcodeeq: 1: 0 +C: schedfifo: 2: 82 +W: opcodeeq: 2: 0 + +# T0 lock L0 +C: locknowait: 0: 0 +W: locked: 0: 0 + +# T1 lock L0 +C: locknowait: 1: 0 +W: blocked: 1: 0 +T: priolt: 0: 1 + +# T2 lock L0 +C: locknowait: 2: 0 +W: blocked: 2: 0 +T: prioeq: 0: 82 + +# T0 unlock L0 +C: unlock: 0: 0 + +# Wait until T2 got the lock +W: locked: 2: 0 +W: unlocked: 0: 0 +T: priolt: 0: 1 + +# T2 unlock L0 +C: unlock: 2: 0 + +W: unlocked: 2: 0 +W: locked: 1: 0 + +C: unlock: 1: 0 +W: unlocked: 1: 0 diff --git a/scripts/rt-tester/t3-l1-pi-2rt.tst b/scripts/rt-tester/t3-l1-pi-2rt.tst new file mode 100644 index 00000000000..c622391a8af --- /dev/null +++ b/scripts/rt-tester/t3-l1-pi-2rt.tst @@ -0,0 +1,96 @@ +# +# rt-mutex test +# +# Op: C(ommand)/T(est)/W(ait) +# | opcode +# | | threadid: 0-7 +# | | | opcode argument +# | | | | +# C: lock: 0: 0 +# +# Commands +# +# opcode opcode argument +# schedother nice value +# schedfifo priority +# lock lock nr (0-7) +# locknowait lock nr (0-7) +# lockint lock nr (0-7) +# lockintnowait lock nr (0-7) +# lockcont lock nr (0-7) +# unlock lock nr (0-7) +# lockbkl lock nr (0-7) +# unlockbkl lock nr (0-7) +# signal thread to signal (0-7) +# reset 0 +# resetevent 0 +# +# Tests / Wait +# +# opcode opcode argument +# +# prioeq priority +# priolt priority +# priogt priority +# nprioeq normal priority +# npriolt normal priority +# npriogt normal priority +# locked lock nr (0-7) +# blocked lock nr (0-7) +# blockedwake lock nr (0-7) +# unlocked lock nr (0-7) +# lockedbkl dont care +# blockedbkl dont care +# unlockedbkl dont care +# opcodeeq command opcode or number +# opcodelt number +# opcodegt number +# eventeq number +# eventgt number +# eventlt number + +# +# 3 threads 1 lock PI +# +C: resetevent: 0: 0 +W: opcodeeq: 0: 0 + +# Set schedulers +C: schedother: 0: 0 +W: opcodeeq: 0: 0 +C: schedfifo: 1: 81 +W: opcodeeq: 1: 0 +C: schedfifo: 2: 82 +W: opcodeeq: 2: 0 + +# T0 lock L0 +C: locknowait: 0: 0 +W: locked: 0: 0 + +# T1 lock L0 +C: locknowait: 1: 0 +W: blocked: 1: 0 +T: prioeq: 0: 81 + +# T2 lock L0 +C: locknowait: 2: 0 +W: blocked: 2: 0 +T: prioeq: 0: 82 +T: prioeq: 1: 81 + +# T0 unlock L0 +C: unlock: 0: 0 + +# Wait until T2 got the lock +W: locked: 2: 0 +W: unlocked: 0: 0 +T: priolt: 0: 1 + +# T2 unlock L0 +C: unlock: 2: 0 + +W: unlocked: 2: 0 +W: locked: 1: 0 + +C: unlock: 1: 0 +W: unlocked: 1: 0 diff --git a/scripts/rt-tester/t3-l1-pi-3rt.tst b/scripts/rt-tester/t3-l1-pi-3rt.tst new file mode 100644 index 00000000000..b5057fb13e0 --- /dev/null +++ b/scripts/rt-tester/t3-l1-pi-3rt.tst @@ -0,0 +1,95 @@ +# +# rt-mutex test +# +# Op: C(ommand)/T(est)/W(ait) +# | opcode +# | | threadid: 0-7 +# | | | opcode argument +# | | | | +# C: lock: 0: 0 +# +# Commands +# +# opcode opcode argument +# schedother nice value +# schedfifo priority +# lock lock nr (0-7) +# locknowait lock nr (0-7) +# lockint lock nr (0-7) +# lockintnowait lock nr (0-7) +# lockcont lock nr (0-7) +# unlock lock nr (0-7) +# lockbkl lock nr (0-7) +# unlockbkl lock nr (0-7) +# signal thread to signal (0-7) +# reset 0 +# resetevent 0 +# +# Tests / Wait +# +# opcode opcode argument +# +# prioeq priority +# priolt priority +# priogt priority +# nprioeq normal priority +# npriolt normal priority +# npriogt normal priority +# locked lock nr (0-7) +# blocked lock nr (0-7) +# blockedwake lock nr (0-7) +# unlocked lock nr (0-7) +# lockedbkl dont care +# blockedbkl dont care +# unlockedbkl dont care +# opcodeeq command opcode or number +# opcodelt number +# opcodegt number +# eventeq number +# eventgt number +# eventlt number + +# +# 3 threads 1 lock PI +# +C: resetevent: 0: 0 +W: opcodeeq: 0: 0 + +# Set schedulers +C: schedfifo: 0: 80 +W: opcodeeq: 0: 0 +C: schedfifo: 1: 81 +W: opcodeeq: 1: 0 +C: schedfifo: 2: 82 +W: opcodeeq: 2: 0 + +# T0 lock L0 +C: locknowait: 0: 0 +W: locked: 0: 0 + +# T1 lock L0 +C: locknowait: 1: 0 +W: blocked: 1: 0 +T: prioeq: 0: 81 + +# T2 lock L0 +C: locknowait: 2: 0 +W: blocked: 2: 0 +T: prioeq: 0: 82 + +# T0 unlock L0 +C: unlock: 0: 0 + +# Wait until T2 got the lock +W: locked: 2: 0 +W: unlocked: 0: 0 +T: prioeq: 0: 80 + +# T2 unlock L0 +C: unlock: 2: 0 + +W: locked: 1: 0 +W: unlocked: 2: 0 + +C: unlock: 1: 0 +W: unlocked: 1: 0 diff --git a/scripts/rt-tester/t3-l1-pi-signal.tst b/scripts/rt-tester/t3-l1-pi-signal.tst new file mode 100644 index 00000000000..3e427daa33f --- /dev/null +++ b/scripts/rt-tester/t3-l1-pi-signal.tst @@ -0,0 +1,101 @@ +# +# rt-mutex test +# +# Op: C(ommand)/T(est)/W(ait) +# | opcode +# | | threadid: 0-7 +# | | | opcode argument +# | | | | +# C: lock: 0: 0 +# +# Commands +# +# opcode opcode argument +# schedother nice value +# schedfifo priority +# lock lock nr (0-7) +# locknowait lock nr (0-7) +# lockint lock nr (0-7) +# lockintnowait lock nr (0-7) +# lockcont lock nr (0-7) +# unlock lock nr (0-7) +# lockbkl lock nr (0-7) +# unlockbkl lock nr (0-7) +# signal thread to signal (0-7) +# reset 0 +# resetevent 0 +# +# Tests / Wait +# +# opcode opcode argument +# +# prioeq priority +# priolt priority +# priogt priority +# nprioeq normal priority +# npriolt normal priority +# npriogt normal priority +# locked lock nr (0-7) +# blocked lock nr (0-7) +# blockedwake lock nr (0-7) +# unlocked lock nr (0-7) +# lockedbkl dont care +# blockedbkl dont care +# unlockedbkl dont care +# opcodeeq command opcode or number +# opcodelt number +# opcodegt number +# eventeq number +# eventgt number +# eventlt number + +# Reset event counter +C: resetevent: 0: 0 +W: opcodeeq: 0: 0 + +# Set priorities +C: schedother: 0: 0 +W: opcodeeq: 0: 0 +C: schedfifo: 1: 80 +W: opcodeeq: 1: 0 +C: schedfifo: 2: 81 +W: opcodeeq: 2: 0 + +# T0 lock L0 +C: lock: 0: 0 +W: locked: 0: 0 + +# T1 lock L0, no wait in the wakeup path +C: locknowait: 1: 0 +W: blocked: 1: 0 +T: prioeq: 0: 80 +T: prioeq: 1: 80 + +# T2 lock L0 interruptible, no wait in the wakeup path +C: lockintnowait: 2: 0 +W: blocked: 2: 0 +T: prioeq: 0: 81 +T: prioeq: 1: 80 + +# Interrupt T2 +C: signal: 2: 2 +W: unlocked: 2: 0 +T: prioeq: 1: 80 +T: prioeq: 0: 80 + +T: locked: 0: 0 +T: blocked: 1: 0 + +# T0 unlock L0 +C: unlock: 0: 0 + +# Wait until T1 has locked L0 and exit +W: locked: 1: 0 +W: unlocked: 0: 0 +T: priolt: 0: 1 + +C: unlock: 1: 0 +W: unlocked: 1: 0 + + + diff --git a/scripts/rt-tester/t3-l1-pi-steal.tst b/scripts/rt-tester/t3-l1-pi-steal.tst new file mode 100644 index 00000000000..72c24a9e4be --- /dev/null +++ b/scripts/rt-tester/t3-l1-pi-steal.tst @@ -0,0 +1,99 @@ +# +# rt-mutex test +# +# Op: C(ommand)/T(est)/W(ait) +# | opcode +# | | threadid: 0-7 +# | | | opcode argument +# | | | | +# C: lock: 0: 0 +# +# Commands +# +# opcode opcode argument +# schedother nice value +# schedfifo priority +# lock lock nr (0-7) +# locknowait lock nr (0-7) +# lockint lock nr (0-7) +# lockintnowait lock nr (0-7) +# lockcont lock nr (0-7) +# unlock lock nr (0-7) +# lockbkl lock nr (0-7) +# unlockbkl lock nr (0-7) +# signal thread to signal (0-7) +# reset 0 +# resetevent 0 +# +# Tests / Wait +# +# opcode opcode argument +# +# prioeq priority +# priolt priority +# priogt priority +# nprioeq normal priority +# npriolt normal priority +# npriogt normal priority +# locked lock nr (0-7) +# blocked lock nr (0-7) +# blockedwake lock nr (0-7) +# unlocked lock nr (0-7) +# lockedbkl dont care +# blockedbkl dont care +# unlockedbkl dont care +# opcodeeq command opcode or number +# opcodelt number +# opcodegt number +# eventeq number +# eventgt number +# eventlt number + +# +# 3 threads 1 lock PI steal pending ownership +# +C: resetevent: 0: 0 +W: opcodeeq: 0: 0 + +# Set schedulers +C: schedother: 0: 0 +W: opcodeeq: 0: 0 +C: schedfifo: 1: 80 +W: opcodeeq: 1: 0 +C: schedfifo: 2: 81 +W: opcodeeq: 2: 0 + +# T0 lock L0 +C: lock: 0: 0 +W: locked: 0: 0 + +# T1 lock L0 +C: lock: 1: 0 +W: blocked: 1: 0 +T: prioeq: 0: 80 + +# T0 unlock L0 +C: unlock: 0: 0 + +# Wait until T1 is in the wakeup loop +W: blockedwake: 1: 0 +T: priolt: 0: 1 + +# T2 lock L0 +C: lock: 2: 0 +# T1 leave wakeup loop +C: lockcont: 1: 0 + +# T2 must have the lock and T1 must be blocked +W: locked: 2: 0 +W: blocked: 1: 0 + +# T2 unlock L0 +C: unlock: 2: 0 + +# Wait until T1 is in the wakeup loop and let it run +W: blockedwake: 1: 0 +C: lockcont: 1: 0 +W: locked: 1: 0 +C: unlock: 1: 0 +W: unlocked: 1: 0 diff --git a/scripts/rt-tester/t3-l2-pi.tst b/scripts/rt-tester/t3-l2-pi.tst new file mode 100644 index 00000000000..2ba0dced48f --- /dev/null +++ b/scripts/rt-tester/t3-l2-pi.tst @@ -0,0 +1,95 @@ +# +# rt-mutex test +# +# Op: C(ommand)/T(est)/W(ait) +# | opcode +# | | threadid: 0-7 +# | | | opcode argument +# | | | | +# C: lock: 0: 0 +# +# Commands +# +# opcode opcode argument +# schedother nice value +# schedfifo priority +# lock lock nr (0-7) +# locknowait lock nr (0-7) +# lockint lock nr (0-7) +# lockintnowait lock nr (0-7) +# lockcont lock nr (0-7) +# unlock lock nr (0-7) +# lockbkl lock nr (0-7) +# unlockbkl lock nr (0-7) +# signal thread to signal (0-7) +# reset 0 +# resetevent 0 +# +# Tests / Wait +# +# opcode opcode argument +# +# prioeq priority +# priolt priority +# priogt priority +# nprioeq normal priority +# npriolt normal priority +# npriogt normal priority +# locked lock nr (0-7) +# blocked lock nr (0-7) +# blockedwake lock nr (0-7) +# unlocked lock nr (0-7) +# lockedbkl dont care +# blockedbkl dont care +# unlockedbkl dont care +# opcodeeq command opcode or number +# opcodelt number +# opcodegt number +# eventeq number +# eventgt number +# eventlt number + +# +# 3 threads 2 lock PI +# +C: resetevent: 0: 0 +W: opcodeeq: 0: 0 + +# Set schedulers +C: schedother: 0: 0 +W: opcodeeq: 0: 0 +C: schedother: 1: 0 +W: opcodeeq: 1: 0 +C: schedfifo: 2: 82 +W: opcodeeq: 2: 0 + +# T0 lock L0 +C: locknowait: 0: 0 +W: locked: 0: 0 + +# T1 lock L0 +C: locknowait: 1: 0 +W: blocked: 1: 0 +T: priolt: 0: 1 + +# T2 lock L0 +C: locknowait: 2: 0 +W: blocked: 2: 0 +T: prioeq: 0: 82 + +# T0 unlock L0 +C: unlock: 0: 0 + +# Wait until T2 got the lock +W: locked: 2: 0 +W: unlocked: 0: 0 +T: priolt: 0: 1 + +# T2 unlock L0 +C: unlock: 2: 0 + +W: unlocked: 2: 0 +W: locked: 1: 0 + +C: unlock: 1: 0 +W: unlocked: 1: 0 diff --git a/scripts/rt-tester/t4-l2-pi-deboost.tst b/scripts/rt-tester/t4-l2-pi-deboost.tst new file mode 100644 index 00000000000..01f1a80fa02 --- /dev/null +++ b/scripts/rt-tester/t4-l2-pi-deboost.tst @@ -0,0 +1,127 @@ +# +# rt-mutex test +# +# Op: C(ommand)/T(est)/W(ait) +# | opcode +# | | threadid: 0-7 +# | | | opcode argument +# | | | | +# C: lock: 0: 0 +# +# Commands +# +# opcode opcode argument +# schedother nice value +# schedfifo priority +# lock lock nr (0-7) +# locknowait lock nr (0-7) +# lockint lock nr (0-7) +# lockintnowait lock nr (0-7) +# lockcont lock nr (0-7) +# unlock lock nr (0-7) +# lockbkl lock nr (0-7) +# unlockbkl lock nr (0-7) +# signal thread to signal (0-7) +# reset 0 +# resetevent 0 +# +# Tests / Wait +# +# opcode opcode argument +# +# prioeq priority +# priolt priority +# priogt priority +# nprioeq normal priority +# npriolt normal priority +# npriogt normal priority +# locked lock nr (0-7) +# blocked lock nr (0-7) +# blockedwake lock nr (0-7) +# unlocked lock nr (0-7) +# lockedbkl dont care +# blockedbkl dont care +# unlockedbkl dont care +# opcodeeq command opcode or number +# opcodelt number +# opcodegt number +# eventeq number +# eventgt number +# eventlt number + +# +# 4 threads 2 lock PI +# +C: resetevent: 0: 0 +W: opcodeeq: 0: 0 + +# Set schedulers +C: schedother: 0: 0 +W: opcodeeq: 0: 0 +C: schedother: 1: 0 +W: opcodeeq: 1: 0 +C: schedfifo: 2: 82 +W: opcodeeq: 2: 0 +C: schedfifo: 3: 83 +W: opcodeeq: 3: 0 + +# T0 lock L0 +C: locknowait: 0: 0 +W: locked: 0: 0 + +# T1 lock L1 +C: locknowait: 1: 1 +W: locked: 1: 1 + +# T3 lock L0 +C: lockintnowait: 3: 0 +W: blocked: 3: 0 +T: prioeq: 0: 83 + +# T0 lock L1 +C: lock: 0: 1 +W: blocked: 0: 1 +T: prioeq: 1: 83 + +# T1 unlock L1 +C: unlock: 1: 1 + +# Wait until T0 is in the wakeup code +W: blockedwake: 0: 1 + +# Verify that T1 is unboosted +W: unlocked: 1: 1 +T: priolt: 1: 1 + +# T2 lock L1 (T0 is boosted and pending owner !) +C: locknowait: 2: 1 +W: blocked: 2: 1 +T: prioeq: 0: 83 + +# Interrupt T3 and wait until T3 returned +C: signal: 3: 0 +W: unlocked: 3: 0 + +# Verify prio of T0 (still pending owner, +# but T2 is enqueued due to the previous boost by T3 +T: prioeq: 0: 82 + +# Let T0 continue +C: lockcont: 0: 1 +W: locked: 0: 1 + +# Unlock L1 and let T2 get L1 +C: unlock: 0: 1 +W: locked: 2: 1 + +# Verify that T0 is unboosted +W: unlocked: 0: 1 +T: priolt: 0: 1 + +# Unlock everything and exit +C: unlock: 2: 1 +W: unlocked: 2: 1 + +C: unlock: 0: 0 +W: unlocked: 0: 0 + diff --git a/scripts/rt-tester/t5-l4-pi-boost-deboost.tst b/scripts/rt-tester/t5-l4-pi-boost-deboost.tst new file mode 100644 index 00000000000..efa0788c118 --- /dev/null +++ b/scripts/rt-tester/t5-l4-pi-boost-deboost.tst @@ -0,0 +1,148 @@ +# +# rt-mutex test +# +# Op: C(ommand)/T(est)/W(ait) +# | opcode +# | | threadid: 0-7 +# | | | opcode argument +# | | | | +# C: lock: 0: 0 +# +# Commands +# +# opcode opcode argument +# schedother nice value +# schedfifo priority +# lock lock nr (0-7) +# locknowait lock nr (0-7) +# lockint lock nr (0-7) +# lockintnowait lock nr (0-7) +# lockcont lock nr (0-7) +# unlock lock nr (0-7) +# lockbkl lock nr (0-7) +# unlockbkl lock nr (0-7) +# signal thread to signal (0-7) +# reset 0 +# resetevent 0 +# +# Tests / Wait +# +# opcode opcode argument +# +# prioeq priority +# priolt priority +# priogt priority +# nprioeq normal priority +# npriolt normal priority +# npriogt normal priority +# locked lock nr (0-7) +# blocked lock nr (0-7) +# blockedwake lock nr (0-7) +# unlocked lock nr (0-7) +# lockedbkl dont care +# blockedbkl dont care +# unlockedbkl dont care +# opcodeeq command opcode or number +# opcodelt number +# opcodegt number +# eventeq number +# eventgt number +# eventlt number + +# +# 5 threads 4 lock PI +# +C: resetevent: 0: 0 +W: opcodeeq: 0: 0 + +# Set schedulers +C: schedother: 0: 0 +W: opcodeeq: 0: 0 +C: schedfifo: 1: 81 +W: opcodeeq: 1: 0 +C: schedfifo: 2: 82 +W: opcodeeq: 2: 0 +C: schedfifo: 3: 83 +W: opcodeeq: 3: 0 +C: schedfifo: 4: 84 +W: opcodeeq: 4: 0 + +# T0 lock L0 +C: locknowait: 0: 0 +W: locked: 0: 0 + +# T1 lock L1 +C: locknowait: 1: 1 +W: locked: 1: 1 + +# T1 lock L0 +C: lockintnowait: 1: 0 +W: blocked: 1: 0 +T: prioeq: 0: 81 + +# T2 lock L2 +C: locknowait: 2: 2 +W: locked: 2: 2 + +# T2 lock L1 +C: lockintnowait: 2: 1 +W: blocked: 2: 1 +T: prioeq: 0: 82 +T: prioeq: 1: 82 + +# T3 lock L3 +C: locknowait: 3: 3 +W: locked: 3: 3 + +# T3 lock L2 +C: lockintnowait: 3: 2 +W: blocked: 3: 2 +T: prioeq: 0: 83 +T: prioeq: 1: 83 +T: prioeq: 2: 83 + +# T4 lock L3 +C: lockintnowait: 4: 3 +W: blocked: 4: 3 +T: prioeq: 0: 84 +T: prioeq: 1: 84 +T: prioeq: 2: 84 +T: prioeq: 3: 84 + +# Signal T4 +C: signal: 4: 0 +W: unlocked: 4: 3 +T: prioeq: 0: 83 +T: prioeq: 1: 83 +T: prioeq: 2: 83 +T: prioeq: 3: 83 + +# Signal T3 +C: signal: 3: 0 +W: unlocked: 3: 2 +T: prioeq: 0: 82 +T: prioeq: 1: 82 +T: prioeq: 2: 82 + +# Signal T2 +C: signal: 2: 0 +W: unlocked: 2: 1 +T: prioeq: 0: 81 +T: prioeq: 1: 81 + +# Signal T1 +C: signal: 1: 0 +W: unlocked: 1: 0 +T: priolt: 0: 1 + +# Unlock and exit +C: unlock: 3: 3 +C: unlock: 2: 2 +C: unlock: 1: 1 +C: unlock: 0: 0 + +W: unlocked: 3: 3 +W: unlocked: 2: 2 +W: unlocked: 1: 1 +W: unlocked: 0: 0 + -- cgit v1.2.3-70-g09d2 From c87e2837be82df479a6bae9f155c43516d2feebc Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Tue, 27 Jun 2006 02:54:58 -0700 Subject: [PATCH] pi-futex: futex_lock_pi/futex_unlock_pi support This adds the actual pi-futex implementation, based on rt-mutexes. [dino@in.ibm.com: fix an oops-causing race] Signed-off-by: Ingo Molnar Signed-off-by: Thomas Gleixner Signed-off-by: Arjan van de Ven Signed-off-by: Dinakar Guniguntala Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/futex.h | 7 + include/linux/sched.h | 3 + kernel/exit.c | 8 + kernel/fork.c | 3 + kernel/futex.c | 829 +++++++++++++++++++++++++++++++++++++++++++++--- kernel/futex_compat.c | 11 +- kernel/rtmutex_common.h | 8 + 7 files changed, 828 insertions(+), 41 deletions(-) (limited to 'include') diff --git a/include/linux/futex.h b/include/linux/futex.h index f05a3f46932..34c3a215f2c 100644 --- a/include/linux/futex.h +++ b/include/linux/futex.h @@ -12,6 +12,9 @@ #define FUTEX_REQUEUE 3 #define FUTEX_CMP_REQUEUE 4 #define FUTEX_WAKE_OP 5 +#define FUTEX_LOCK_PI 6 +#define FUTEX_UNLOCK_PI 7 +#define FUTEX_TRYLOCK_PI 8 /* * Support for robust futexes: the kernel cleans up held futexes at @@ -97,10 +100,14 @@ extern int handle_futex_death(u32 __user *uaddr, struct task_struct *curr); #ifdef CONFIG_FUTEX extern void exit_robust_list(struct task_struct *curr); +extern void exit_pi_state_list(struct task_struct *curr); #else static inline void exit_robust_list(struct task_struct *curr) { } +static inline void exit_pi_state_list(struct task_struct *curr) +{ +} #endif #define FUTEX_OP_SET 0 /* *(int *)UADDR2 = OPARG; */ diff --git a/include/linux/sched.h b/include/linux/sched.h index edadd13cf53..b4e6be7de5a 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -84,6 +84,7 @@ struct sched_param { #include struct exec_domain; +struct futex_pi_state; /* * List of flags we want to share for kernel threads, @@ -915,6 +916,8 @@ struct task_struct { #ifdef CONFIG_COMPAT struct compat_robust_list_head __user *compat_robust_list; #endif + struct list_head pi_state_list; + struct futex_pi_state *pi_state_cache; atomic_t fs_excl; /* holding fs exclusive resources */ struct rcu_head rcu; diff --git a/kernel/exit.c b/kernel/exit.c index 3e8a0282e9a..ab06b9f88f6 100644 --- a/kernel/exit.c +++ b/kernel/exit.c @@ -925,6 +925,14 @@ fastcall NORET_TYPE void do_exit(long code) mpol_free(tsk->mempolicy); tsk->mempolicy = NULL; #endif + /* + * This must happen late, after the PID is not + * hashed anymore: + */ + if (unlikely(!list_empty(&tsk->pi_state_list))) + exit_pi_state_list(tsk); + if (unlikely(current->pi_state_cache)) + kfree(current->pi_state_cache); /* * If DEBUG_MUTEXES is on, make sure we are holding no locks: */ diff --git a/kernel/fork.c b/kernel/fork.c index b664a081fff..628198a4f28 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -1092,6 +1092,9 @@ static task_t *copy_process(unsigned long clone_flags, #ifdef CONFIG_COMPAT p->compat_robust_list = NULL; #endif + INIT_LIST_HEAD(&p->pi_state_list); + p->pi_state_cache = NULL; + /* * sigaltstack should be cleared when sharing the same VM */ diff --git a/kernel/futex.c b/kernel/futex.c index 50356fb5d72..b305b7f8dad 100644 --- a/kernel/futex.c +++ b/kernel/futex.c @@ -12,6 +12,10 @@ * (C) Copyright 2006 Red Hat Inc, All Rights Reserved * Thanks to Thomas Gleixner for suggestions, analysis and fixes. * + * PI-futex support started by Ingo Molnar and Thomas Gleixner + * Copyright (C) 2006 Red Hat, Inc., Ingo Molnar + * Copyright (C) 2006 Timesys Corp., Thomas Gleixner + * * Thanks to Ben LaHaise for yelling "hashed waitqueues" loudly * enough at me, Linus for the original (flawed) idea, Matthew * Kirkwood for proof-of-concept implementation. @@ -46,6 +50,8 @@ #include #include +#include "rtmutex_common.h" + #define FUTEX_HASHBITS (CONFIG_BASE_SMALL ? 4 : 8) /* @@ -74,6 +80,27 @@ union futex_key { } both; }; +/* + * Priority Inheritance state: + */ +struct futex_pi_state { + /* + * list of 'owned' pi_state instances - these have to be + * cleaned up in do_exit() if the task exits prematurely: + */ + struct list_head list; + + /* + * The PI object: + */ + struct rt_mutex pi_mutex; + + struct task_struct *owner; + atomic_t refcount; + + union futex_key key; +}; + /* * We use this hashed waitqueue instead of a normal wait_queue_t, so * we can wake only the relevant ones (hashed queues may be shared). @@ -96,6 +123,10 @@ struct futex_q { /* For fd, sigio sent using these: */ int fd; struct file *filp; + + /* Optional priority inheritance state: */ + struct futex_pi_state *pi_state; + struct task_struct *task; }; /* @@ -258,6 +289,232 @@ static inline int get_futex_value_locked(u32 *dest, u32 __user *from) return ret ? -EFAULT : 0; } +/* + * Fault handling. Called with current->mm->mmap_sem held. + */ +static int futex_handle_fault(unsigned long address, int attempt) +{ + struct vm_area_struct * vma; + struct mm_struct *mm = current->mm; + + if (attempt >= 2 || !(vma = find_vma(mm, address)) || + vma->vm_start > address || !(vma->vm_flags & VM_WRITE)) + return -EFAULT; + + switch (handle_mm_fault(mm, vma, address, 1)) { + case VM_FAULT_MINOR: + current->min_flt++; + break; + case VM_FAULT_MAJOR: + current->maj_flt++; + break; + default: + return -EFAULT; + } + return 0; +} + +/* + * PI code: + */ +static int refill_pi_state_cache(void) +{ + struct futex_pi_state *pi_state; + + if (likely(current->pi_state_cache)) + return 0; + + pi_state = kmalloc(sizeof(*pi_state), GFP_KERNEL); + + if (!pi_state) + return -ENOMEM; + + memset(pi_state, 0, sizeof(*pi_state)); + INIT_LIST_HEAD(&pi_state->list); + /* pi_mutex gets initialized later */ + pi_state->owner = NULL; + atomic_set(&pi_state->refcount, 1); + + current->pi_state_cache = pi_state; + + return 0; +} + +static struct futex_pi_state * alloc_pi_state(void) +{ + struct futex_pi_state *pi_state = current->pi_state_cache; + + WARN_ON(!pi_state); + current->pi_state_cache = NULL; + + return pi_state; +} + +static void free_pi_state(struct futex_pi_state *pi_state) +{ + if (!atomic_dec_and_test(&pi_state->refcount)) + return; + + /* + * If pi_state->owner is NULL, the owner is most probably dying + * and has cleaned up the pi_state already + */ + if (pi_state->owner) { + spin_lock_irq(&pi_state->owner->pi_lock); + list_del_init(&pi_state->list); + spin_unlock_irq(&pi_state->owner->pi_lock); + + rt_mutex_proxy_unlock(&pi_state->pi_mutex, pi_state->owner); + } + + if (current->pi_state_cache) + kfree(pi_state); + else { + /* + * pi_state->list is already empty. + * clear pi_state->owner. + * refcount is at 0 - put it back to 1. + */ + pi_state->owner = NULL; + atomic_set(&pi_state->refcount, 1); + current->pi_state_cache = pi_state; + } +} + +/* + * Look up the task based on what TID userspace gave us. + * We dont trust it. + */ +static struct task_struct * futex_find_get_task(pid_t pid) +{ + struct task_struct *p; + + read_lock(&tasklist_lock); + p = find_task_by_pid(pid); + if (!p) + goto out_unlock; + if ((current->euid != p->euid) && (current->euid != p->uid)) { + p = NULL; + goto out_unlock; + } + if (p->state == EXIT_ZOMBIE || p->exit_state == EXIT_ZOMBIE) { + p = NULL; + goto out_unlock; + } + get_task_struct(p); +out_unlock: + read_unlock(&tasklist_lock); + + return p; +} + +/* + * This task is holding PI mutexes at exit time => bad. + * Kernel cleans up PI-state, but userspace is likely hosed. + * (Robust-futex cleanup is separate and might save the day for userspace.) + */ +void exit_pi_state_list(struct task_struct *curr) +{ + struct futex_hash_bucket *hb; + struct list_head *next, *head = &curr->pi_state_list; + struct futex_pi_state *pi_state; + union futex_key key; + + /* + * We are a ZOMBIE and nobody can enqueue itself on + * pi_state_list anymore, but we have to be careful + * versus waiters unqueueing themselfs + */ + spin_lock_irq(&curr->pi_lock); + while (!list_empty(head)) { + + next = head->next; + pi_state = list_entry(next, struct futex_pi_state, list); + key = pi_state->key; + spin_unlock_irq(&curr->pi_lock); + + hb = hash_futex(&key); + spin_lock(&hb->lock); + + spin_lock_irq(&curr->pi_lock); + if (head->next != next) { + spin_unlock(&hb->lock); + continue; + } + + list_del_init(&pi_state->list); + + WARN_ON(pi_state->owner != curr); + + pi_state->owner = NULL; + spin_unlock_irq(&curr->pi_lock); + + rt_mutex_unlock(&pi_state->pi_mutex); + + spin_unlock(&hb->lock); + + spin_lock_irq(&curr->pi_lock); + } + spin_unlock_irq(&curr->pi_lock); +} + +static int +lookup_pi_state(u32 uval, struct futex_hash_bucket *hb, struct futex_q *me) +{ + struct futex_pi_state *pi_state = NULL; + struct futex_q *this, *next; + struct list_head *head; + struct task_struct *p; + pid_t pid; + + head = &hb->chain; + + list_for_each_entry_safe(this, next, head, list) { + if (match_futex (&this->key, &me->key)) { + /* + * Another waiter already exists - bump up + * the refcount and return its pi_state: + */ + pi_state = this->pi_state; + atomic_inc(&pi_state->refcount); + me->pi_state = pi_state; + + return 0; + } + } + + /* + * We are the first waiter - try to look up the real owner and + * attach the new pi_state to it: + */ + pid = uval & FUTEX_TID_MASK; + p = futex_find_get_task(pid); + if (!p) + return -ESRCH; + + pi_state = alloc_pi_state(); + + /* + * Initialize the pi_mutex in locked state and make 'p' + * the owner of it: + */ + rt_mutex_init_proxy_locked(&pi_state->pi_mutex, p); + + /* Store the key for possible exit cleanups: */ + pi_state->key = me->key; + + spin_lock_irq(&p->pi_lock); + list_add(&pi_state->list, &p->pi_state_list); + pi_state->owner = p; + spin_unlock_irq(&p->pi_lock); + + put_task_struct(p); + + me->pi_state = pi_state; + + return 0; +} + /* * The hash bucket lock must be held when this is called. * Afterwards, the futex_q must not be accessed. @@ -285,6 +542,70 @@ static void wake_futex(struct futex_q *q) q->lock_ptr = NULL; } +static int wake_futex_pi(u32 __user *uaddr, u32 uval, struct futex_q *this) +{ + struct task_struct *new_owner; + struct futex_pi_state *pi_state = this->pi_state; + u32 curval, newval; + + if (!pi_state) + return -EINVAL; + + new_owner = rt_mutex_next_owner(&pi_state->pi_mutex); + + /* + * This happens when we have stolen the lock and the original + * pending owner did not enqueue itself back on the rt_mutex. + * Thats not a tragedy. We know that way, that a lock waiter + * is on the fly. We make the futex_q waiter the pending owner. + */ + if (!new_owner) + new_owner = this->task; + + /* + * We pass it to the next owner. (The WAITERS bit is always + * kept enabled while there is PI state around. We must also + * preserve the owner died bit.) + */ + newval = (uval & FUTEX_OWNER_DIED) | FUTEX_WAITERS | new_owner->pid; + + inc_preempt_count(); + curval = futex_atomic_cmpxchg_inatomic(uaddr, uval, newval); + dec_preempt_count(); + + if (curval == -EFAULT) + return -EFAULT; + if (curval != uval) + return -EINVAL; + + list_del_init(&pi_state->owner->pi_state_list); + list_add(&pi_state->list, &new_owner->pi_state_list); + pi_state->owner = new_owner; + rt_mutex_unlock(&pi_state->pi_mutex); + + return 0; +} + +static int unlock_futex_pi(u32 __user *uaddr, u32 uval) +{ + u32 oldval; + + /* + * There is no waiter, so we unlock the futex. The owner died + * bit has not to be preserved here. We are the owner: + */ + inc_preempt_count(); + oldval = futex_atomic_cmpxchg_inatomic(uaddr, uval, 0); + dec_preempt_count(); + + if (oldval == -EFAULT) + return oldval; + if (oldval != uval) + return -EAGAIN; + + return 0; +} + /* * Wake up all waiters hashed on the physical page that is mapped * to this virtual address: @@ -309,6 +630,8 @@ static int futex_wake(u32 __user *uaddr, int nr_wake) list_for_each_entry_safe(this, next, head, list) { if (match_futex (&this->key, &key)) { + if (this->pi_state) + return -EINVAL; wake_futex(this); if (++ret >= nr_wake) break; @@ -385,27 +708,9 @@ retry: * still holding the mmap_sem. */ if (attempt++) { - struct vm_area_struct * vma; - struct mm_struct *mm = current->mm; - unsigned long address = (unsigned long)uaddr2; - - ret = -EFAULT; - if (attempt >= 2 || - !(vma = find_vma(mm, address)) || - vma->vm_start > address || - !(vma->vm_flags & VM_WRITE)) + if (futex_handle_fault((unsigned long)uaddr2, + attempt)) goto out; - - switch (handle_mm_fault(mm, vma, address, 1)) { - case VM_FAULT_MINOR: - current->min_flt++; - break; - case VM_FAULT_MAJOR: - current->maj_flt++; - break; - default: - goto out; - } goto retry; } @@ -572,6 +877,7 @@ queue_lock(struct futex_q *q, int fd, struct file *filp) static inline void __queue_me(struct futex_q *q, struct futex_hash_bucket *hb) { list_add_tail(&q->list, &hb->chain); + q->task = current; spin_unlock(&hb->lock); } @@ -626,6 +932,9 @@ static int unqueue_me(struct futex_q *q) } WARN_ON(list_empty(&q->list)); list_del(&q->list); + + BUG_ON(q->pi_state); + spin_unlock(lock_ptr); ret = 1; } @@ -634,16 +943,36 @@ static int unqueue_me(struct futex_q *q) return ret; } +/* + * PI futexes can not be requeued and must remove themself from the + * hash bucket. The hash bucket lock is held on entry and dropped here. + */ +static void unqueue_me_pi(struct futex_q *q, struct futex_hash_bucket *hb) +{ + WARN_ON(list_empty(&q->list)); + list_del(&q->list); + + BUG_ON(!q->pi_state); + free_pi_state(q->pi_state); + q->pi_state = NULL; + + spin_unlock(&hb->lock); + + drop_key_refs(&q->key); +} + static int futex_wait(u32 __user *uaddr, u32 val, unsigned long time) { - DECLARE_WAITQUEUE(wait, current); + struct task_struct *curr = current; + DECLARE_WAITQUEUE(wait, curr); struct futex_hash_bucket *hb; struct futex_q q; u32 uval; int ret; + q.pi_state = NULL; retry: - down_read(¤t->mm->mmap_sem); + down_read(&curr->mm->mmap_sem); ret = get_futex_key(uaddr, &q.key); if (unlikely(ret != 0)) @@ -680,7 +1009,7 @@ static int futex_wait(u32 __user *uaddr, u32 val, unsigned long time) * If we would have faulted, release mmap_sem, fault it in and * start all over again. */ - up_read(¤t->mm->mmap_sem); + up_read(&curr->mm->mmap_sem); ret = get_user(uval, uaddr); @@ -688,11 +1017,9 @@ static int futex_wait(u32 __user *uaddr, u32 val, unsigned long time) goto retry; return ret; } - if (uval != val) { - ret = -EWOULDBLOCK; - queue_unlock(&q, hb); - goto out_release_sem; - } + ret = -EWOULDBLOCK; + if (uval != val) + goto out_unlock_release_sem; /* Only actually queue if *uaddr contained val. */ __queue_me(&q, hb); @@ -700,8 +1027,8 @@ static int futex_wait(u32 __user *uaddr, u32 val, unsigned long time) /* * Now the futex is queued and we have checked the data, we * don't want to hold mmap_sem while we sleep. - */ - up_read(¤t->mm->mmap_sem); + */ + up_read(&curr->mm->mmap_sem); /* * There might have been scheduling since the queue_me(), as we @@ -739,8 +1066,415 @@ static int futex_wait(u32 __user *uaddr, u32 val, unsigned long time) */ return -EINTR; + out_unlock_release_sem: + queue_unlock(&q, hb); + out_release_sem: + up_read(&curr->mm->mmap_sem); + return ret; +} + +/* + * Userspace tried a 0 -> TID atomic transition of the futex value + * and failed. The kernel side here does the whole locking operation: + * if there are waiters then it will block, it does PI, etc. (Due to + * races the kernel might see a 0 value of the futex too.) + */ +static int do_futex_lock_pi(u32 __user *uaddr, int detect, int trylock, + struct hrtimer_sleeper *to) +{ + struct task_struct *curr = current; + struct futex_hash_bucket *hb; + u32 uval, newval, curval; + struct futex_q q; + int ret, attempt = 0; + + if (refill_pi_state_cache()) + return -ENOMEM; + + q.pi_state = NULL; + retry: + down_read(&curr->mm->mmap_sem); + + ret = get_futex_key(uaddr, &q.key); + if (unlikely(ret != 0)) + goto out_release_sem; + + hb = queue_lock(&q, -1, NULL); + + retry_locked: + /* + * To avoid races, we attempt to take the lock here again + * (by doing a 0 -> TID atomic cmpxchg), while holding all + * the locks. It will most likely not succeed. + */ + newval = current->pid; + + inc_preempt_count(); + curval = futex_atomic_cmpxchg_inatomic(uaddr, 0, newval); + dec_preempt_count(); + + if (unlikely(curval == -EFAULT)) + goto uaddr_faulted; + + /* We own the lock already */ + if (unlikely((curval & FUTEX_TID_MASK) == current->pid)) { + if (!detect && 0) + force_sig(SIGKILL, current); + ret = -EDEADLK; + goto out_unlock_release_sem; + } + + /* + * Surprise - we got the lock. Just return + * to userspace: + */ + if (unlikely(!curval)) + goto out_unlock_release_sem; + + uval = curval; + newval = uval | FUTEX_WAITERS; + + inc_preempt_count(); + curval = futex_atomic_cmpxchg_inatomic(uaddr, uval, newval); + dec_preempt_count(); + + if (unlikely(curval == -EFAULT)) + goto uaddr_faulted; + if (unlikely(curval != uval)) + goto retry_locked; + + /* + * We dont have the lock. Look up the PI state (or create it if + * we are the first waiter): + */ + ret = lookup_pi_state(uval, hb, &q); + + if (unlikely(ret)) { + /* + * There were no waiters and the owner task lookup + * failed. When the OWNER_DIED bit is set, then we + * know that this is a robust futex and we actually + * take the lock. This is safe as we are protected by + * the hash bucket lock. We also set the waiters bit + * unconditionally here, to simplify glibc handling of + * multiple tasks racing to acquire the lock and + * cleanup the problems which were left by the dead + * owner. + */ + if (curval & FUTEX_OWNER_DIED) { + uval = newval; + newval = current->pid | + FUTEX_OWNER_DIED | FUTEX_WAITERS; + + inc_preempt_count(); + curval = futex_atomic_cmpxchg_inatomic(uaddr, + uval, newval); + dec_preempt_count(); + + if (unlikely(curval == -EFAULT)) + goto uaddr_faulted; + if (unlikely(curval != uval)) + goto retry_locked; + ret = 0; + } + goto out_unlock_release_sem; + } + + /* + * Only actually queue now that the atomic ops are done: + */ + __queue_me(&q, hb); + + /* + * Now the futex is queued and we have checked the data, we + * don't want to hold mmap_sem while we sleep. + */ + up_read(&curr->mm->mmap_sem); + + WARN_ON(!q.pi_state); + /* + * Block on the PI mutex: + */ + if (!trylock) + ret = rt_mutex_timed_lock(&q.pi_state->pi_mutex, to, 1); + else { + ret = rt_mutex_trylock(&q.pi_state->pi_mutex); + /* Fixup the trylock return value: */ + ret = ret ? 0 : -EWOULDBLOCK; + } + + down_read(&curr->mm->mmap_sem); + hb = queue_lock(&q, -1, NULL); + + /* + * Got the lock. We might not be the anticipated owner if we + * did a lock-steal - fix up the PI-state in that case. + */ + if (!ret && q.pi_state->owner != curr) { + u32 newtid = current->pid | FUTEX_WAITERS; + + /* Owner died? */ + if (q.pi_state->owner != NULL) { + spin_lock_irq(&q.pi_state->owner->pi_lock); + list_del_init(&q.pi_state->list); + spin_unlock_irq(&q.pi_state->owner->pi_lock); + } else + newtid |= FUTEX_OWNER_DIED; + + q.pi_state->owner = current; + + spin_lock_irq(¤t->pi_lock); + list_add(&q.pi_state->list, ¤t->pi_state_list); + spin_unlock_irq(¤t->pi_lock); + + /* Unqueue and drop the lock */ + unqueue_me_pi(&q, hb); + up_read(&curr->mm->mmap_sem); + /* + * We own it, so we have to replace the pending owner + * TID. This must be atomic as we have preserve the + * owner died bit here. + */ + ret = get_user(uval, uaddr); + while (!ret) { + newval = (uval & FUTEX_OWNER_DIED) | newtid; + curval = futex_atomic_cmpxchg_inatomic(uaddr, + uval, newval); + if (curval == -EFAULT) + ret = -EFAULT; + if (curval == uval) + break; + uval = curval; + } + } else { + /* + * Catch the rare case, where the lock was released + * when we were on the way back before we locked + * the hash bucket. + */ + if (ret && q.pi_state->owner == curr) { + if (rt_mutex_trylock(&q.pi_state->pi_mutex)) + ret = 0; + } + /* Unqueue and drop the lock */ + unqueue_me_pi(&q, hb); + up_read(&curr->mm->mmap_sem); + } + + if (!detect && ret == -EDEADLK && 0) + force_sig(SIGKILL, current); + + return ret; + + out_unlock_release_sem: + queue_unlock(&q, hb); + + out_release_sem: + up_read(&curr->mm->mmap_sem); + return ret; + + uaddr_faulted: + /* + * We have to r/w *(int __user *)uaddr, but we can't modify it + * non-atomically. Therefore, if get_user below is not + * enough, we need to handle the fault ourselves, while + * still holding the mmap_sem. + */ + if (attempt++) { + if (futex_handle_fault((unsigned long)uaddr, attempt)) + goto out_unlock_release_sem; + + goto retry_locked; + } + + queue_unlock(&q, hb); + up_read(&curr->mm->mmap_sem); + + ret = get_user(uval, uaddr); + if (!ret && (uval != -EFAULT)) + goto retry; + + return ret; +} + +/* + * Restart handler + */ +static long futex_lock_pi_restart(struct restart_block *restart) +{ + struct hrtimer_sleeper timeout, *to = NULL; + int ret; + + restart->fn = do_no_restart_syscall; + + if (restart->arg2 || restart->arg3) { + to = &timeout; + hrtimer_init(&to->timer, CLOCK_REALTIME, HRTIMER_ABS); + hrtimer_init_sleeper(to, current); + to->timer.expires.tv64 = ((u64)restart->arg1 << 32) | + (u64) restart->arg0; + } + + pr_debug("lock_pi restart: %p, %d (%d)\n", + (u32 __user *)restart->arg0, current->pid); + + ret = do_futex_lock_pi((u32 __user *)restart->arg0, restart->arg1, + 0, to); + + if (ret != -EINTR) + return ret; + + restart->fn = futex_lock_pi_restart; + + /* The other values are filled in */ + return -ERESTART_RESTARTBLOCK; +} + +/* + * Called from the syscall entry below. + */ +static int futex_lock_pi(u32 __user *uaddr, int detect, unsigned long sec, + long nsec, int trylock) +{ + struct hrtimer_sleeper timeout, *to = NULL; + struct restart_block *restart; + int ret; + + if (sec != MAX_SCHEDULE_TIMEOUT) { + to = &timeout; + hrtimer_init(&to->timer, CLOCK_REALTIME, HRTIMER_ABS); + hrtimer_init_sleeper(to, current); + to->timer.expires = ktime_set(sec, nsec); + } + + ret = do_futex_lock_pi(uaddr, detect, trylock, to); + + if (ret != -EINTR) + return ret; + + pr_debug("lock_pi interrupted: %p, %d (%d)\n", uaddr, current->pid); + + restart = ¤t_thread_info()->restart_block; + restart->fn = futex_lock_pi_restart; + restart->arg0 = (unsigned long) uaddr; + restart->arg1 = detect; + if (to) { + restart->arg2 = to->timer.expires.tv64 & 0xFFFFFFFF; + restart->arg3 = to->timer.expires.tv64 >> 32; + } else + restart->arg2 = restart->arg3 = 0; + + return -ERESTART_RESTARTBLOCK; +} + +/* + * Userspace attempted a TID -> 0 atomic transition, and failed. + * This is the in-kernel slowpath: we look up the PI state (if any), + * and do the rt-mutex unlock. + */ +static int futex_unlock_pi(u32 __user *uaddr) +{ + struct futex_hash_bucket *hb; + struct futex_q *this, *next; + u32 uval; + struct list_head *head; + union futex_key key; + int ret, attempt = 0; + +retry: + if (get_user(uval, uaddr)) + return -EFAULT; + /* + * We release only a lock we actually own: + */ + if ((uval & FUTEX_TID_MASK) != current->pid) + return -EPERM; + /* + * First take all the futex related locks: + */ + down_read(¤t->mm->mmap_sem); + + ret = get_futex_key(uaddr, &key); + if (unlikely(ret != 0)) + goto out; + + hb = hash_futex(&key); + spin_lock(&hb->lock); + +retry_locked: + /* + * To avoid races, try to do the TID -> 0 atomic transition + * again. If it succeeds then we can return without waking + * anyone else up: + */ + inc_preempt_count(); + uval = futex_atomic_cmpxchg_inatomic(uaddr, current->pid, 0); + dec_preempt_count(); + + if (unlikely(uval == -EFAULT)) + goto pi_faulted; + /* + * Rare case: we managed to release the lock atomically, + * no need to wake anyone else up: + */ + if (unlikely(uval == current->pid)) + goto out_unlock; + + /* + * Ok, other tasks may need to be woken up - check waiters + * and do the wakeup if necessary: + */ + head = &hb->chain; + + list_for_each_entry_safe(this, next, head, list) { + if (!match_futex (&this->key, &key)) + continue; + ret = wake_futex_pi(uaddr, uval, this); + /* + * The atomic access to the futex value + * generated a pagefault, so retry the + * user-access and the wakeup: + */ + if (ret == -EFAULT) + goto pi_faulted; + goto out_unlock; + } + /* + * No waiters - kernel unlocks the futex: + */ + ret = unlock_futex_pi(uaddr, uval); + if (ret == -EFAULT) + goto pi_faulted; + +out_unlock: + spin_unlock(&hb->lock); +out: + up_read(¤t->mm->mmap_sem); + + return ret; + +pi_faulted: + /* + * We have to r/w *(int __user *)uaddr, but we can't modify it + * non-atomically. Therefore, if get_user below is not + * enough, we need to handle the fault ourselves, while + * still holding the mmap_sem. + */ + if (attempt++) { + if (futex_handle_fault((unsigned long)uaddr, attempt)) + goto out_unlock; + + goto retry_locked; + } + + spin_unlock(&hb->lock); up_read(¤t->mm->mmap_sem); + + ret = get_user(uval, uaddr); + if (!ret && (uval != -EFAULT)) + goto retry; + return ret; } @@ -819,6 +1553,7 @@ static int futex_fd(u32 __user *uaddr, int signal) err = -ENOMEM; goto error; } + q->pi_state = NULL; down_read(¤t->mm->mmap_sem); err = get_futex_key(uaddr, &q->key); @@ -856,7 +1591,7 @@ error: * Implementation: user-space maintains a per-thread list of locks it * is holding. Upon do_exit(), the kernel carefully walks this list, * and marks all locks that are owned by this thread with the - * FUTEX_OWNER_DEAD bit, and wakes up a waiter (if any). The list is + * FUTEX_OWNER_DIED bit, and wakes up a waiter (if any). The list is * always manipulated with the lock held, so the list is private and * per-thread. Userspace also maintains a per-thread 'list_op_pending' * field, to allow the kernel to clean up if the thread dies after @@ -931,7 +1666,7 @@ err_unlock: */ int handle_futex_death(u32 __user *uaddr, struct task_struct *curr) { - u32 uval; + u32 uval, nval; retry: if (get_user(uval, uaddr)) @@ -948,8 +1683,12 @@ retry: * thread-death.) The rest of the cleanup is done in * userspace. */ - if (futex_atomic_cmpxchg_inatomic(uaddr, uval, - uval | FUTEX_OWNER_DIED) != uval) + nval = futex_atomic_cmpxchg_inatomic(uaddr, uval, + uval | FUTEX_OWNER_DIED); + if (nval == -EFAULT) + return -1; + + if (nval != uval) goto retry; if (uval & FUTEX_WAITERS) @@ -994,7 +1733,7 @@ void exit_robust_list(struct task_struct *curr) while (entry != &head->list) { /* * A pending lock might already be on the list, so - * dont process it twice: + * don't process it twice: */ if (entry != pending) if (handle_futex_death((void *)entry + futex_offset, @@ -1040,6 +1779,15 @@ long do_futex(u32 __user *uaddr, int op, u32 val, unsigned long timeout, case FUTEX_WAKE_OP: ret = futex_wake_op(uaddr, uaddr2, val, val2, val3); break; + case FUTEX_LOCK_PI: + ret = futex_lock_pi(uaddr, val, timeout, val2, 0); + break; + case FUTEX_UNLOCK_PI: + ret = futex_unlock_pi(uaddr); + break; + case FUTEX_TRYLOCK_PI: + ret = futex_lock_pi(uaddr, 0, timeout, val2, 1); + break; default: ret = -ENOSYS; } @@ -1055,17 +1803,22 @@ asmlinkage long sys_futex(u32 __user *uaddr, int op, u32 val, unsigned long timeout = MAX_SCHEDULE_TIMEOUT; u32 val2 = 0; - if (utime && (op == FUTEX_WAIT)) { + if (utime && (op == FUTEX_WAIT || op == FUTEX_LOCK_PI)) { if (copy_from_user(&t, utime, sizeof(t)) != 0) return -EFAULT; if (!timespec_valid(&t)) return -EINVAL; - timeout = timespec_to_jiffies(&t) + 1; + if (op == FUTEX_WAIT) + timeout = timespec_to_jiffies(&t) + 1; + else { + timeout = t.tv_sec; + val2 = t.tv_nsec; + } } /* * requeue parameter in 'utime' if op == FUTEX_REQUEUE. */ - if (op >= FUTEX_REQUEUE) + if (op == FUTEX_REQUEUE || op == FUTEX_CMP_REQUEUE) val2 = (u32) (unsigned long) utime; return do_futex(uaddr, op, val, timeout, uaddr2, val2, val3); diff --git a/kernel/futex_compat.c b/kernel/futex_compat.c index 7e57c31670a..d1d92b441fb 100644 --- a/kernel/futex_compat.c +++ b/kernel/futex_compat.c @@ -129,14 +129,19 @@ asmlinkage long compat_sys_futex(u32 __user *uaddr, int op, u32 val, unsigned long timeout = MAX_SCHEDULE_TIMEOUT; int val2 = 0; - if (utime && (op == FUTEX_WAIT)) { + if (utime && (op == FUTEX_WAIT || op == FUTEX_LOCK_PI)) { if (get_compat_timespec(&t, utime)) return -EFAULT; if (!timespec_valid(&t)) return -EINVAL; - timeout = timespec_to_jiffies(&t) + 1; + if (op == FUTEX_WAIT) + timeout = timespec_to_jiffies(&t) + 1; + else { + timeout = t.tv_sec; + val2 = t.tv_nsec; + } } - if (op >= FUTEX_REQUEUE) + if (op == FUTEX_REQUEUE || op == FUTEX_CMP_REQUEUE) val2 = (int) (unsigned long) utime; return do_futex(uaddr, op, val, timeout, uaddr2, val2, val3); diff --git a/kernel/rtmutex_common.h b/kernel/rtmutex_common.h index e068024eeff..9c75856e791 100644 --- a/kernel/rtmutex_common.h +++ b/kernel/rtmutex_common.h @@ -112,4 +112,12 @@ static inline unsigned long rt_mutex_owner_pending(struct rt_mutex *lock) return (unsigned long)lock->owner & RT_MUTEX_OWNER_PENDING; } +/* + * PI-futex support (proxy locking functions, etc.): + */ +extern struct task_struct *rt_mutex_next_owner(struct rt_mutex *lock); +extern void rt_mutex_init_proxy_locked(struct rt_mutex *lock, + struct task_struct *proxy_owner); +extern void rt_mutex_proxy_unlock(struct rt_mutex *lock, + struct task_struct *proxy_owner); #endif -- cgit v1.2.3-70-g09d2 From 95e02ca9bb5324360e7dea1ea1c563036d84a5e6 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Tue, 27 Jun 2006 02:55:02 -0700 Subject: [PATCH] rtmutex: Propagate priority settings into PI lock chains When the priority of a task, which is blocked on a lock, changes we must propagate this change into the PI lock chain. Therefor the chain walk code is changed to get rid of the references to current to avoid false positives in the deadlock detector, as setscheduler might be called by a task which holds the lock on which the task whose priority is changed is blocked. Also add some comments about the get/put_task_struct usage to avoid confusion. Signed-off-by: Thomas Gleixner Signed-off-by: Ingo Molnar Cc: Steven Rostedt Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/sched.h | 2 ++ kernel/rtmutex.c | 42 ++++++++++++++++++++++++++++++++++++------ kernel/sched.c | 2 ++ 3 files changed, 40 insertions(+), 6 deletions(-) (limited to 'include') diff --git a/include/linux/sched.h b/include/linux/sched.h index b4e6be7de5a..821f0481ebe 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -1044,11 +1044,13 @@ extern void sched_idle_next(void); #ifdef CONFIG_RT_MUTEXES extern int rt_mutex_getprio(task_t *p); extern void rt_mutex_setprio(task_t *p, int prio); +extern void rt_mutex_adjust_pi(task_t *p); #else static inline int rt_mutex_getprio(task_t *p) { return p->normal_prio; } +# define rt_mutex_adjust_pi(p) do { } while (0) #endif extern void set_user_nice(task_t *p, long nice); diff --git a/kernel/rtmutex.c b/kernel/rtmutex.c index 3fc0f0680ca..45d61016da5 100644 --- a/kernel/rtmutex.c +++ b/kernel/rtmutex.c @@ -160,7 +160,8 @@ int max_lock_depth = 1024; static int rt_mutex_adjust_prio_chain(task_t *task, int deadlock_detect, struct rt_mutex *orig_lock, - struct rt_mutex_waiter *orig_waiter + struct rt_mutex_waiter *orig_waiter, + struct task_struct *top_task __IP_DECL__) { struct rt_mutex *lock; @@ -189,7 +190,7 @@ static int rt_mutex_adjust_prio_chain(task_t *task, prev_max = max_lock_depth; printk(KERN_WARNING "Maximum lock depth %d reached " "task: %s (%d)\n", max_lock_depth, - current->comm, current->pid); + top_task->comm, top_task->pid); } put_task_struct(task); @@ -229,7 +230,7 @@ static int rt_mutex_adjust_prio_chain(task_t *task, } /* Deadlock detection */ - if (lock == orig_lock || rt_mutex_owner(lock) == current) { + if (lock == orig_lock || rt_mutex_owner(lock) == top_task) { debug_rt_mutex_deadlock(deadlock_detect, orig_waiter, lock); spin_unlock(&lock->wait_lock); ret = deadlock_detect ? -EDEADLK : 0; @@ -433,6 +434,7 @@ static int task_blocks_on_rt_mutex(struct rt_mutex *lock, __rt_mutex_adjust_prio(owner); if (owner->pi_blocked_on) { boost = 1; + /* gets dropped in rt_mutex_adjust_prio_chain()! */ get_task_struct(owner); } spin_unlock_irqrestore(&owner->pi_lock, flags); @@ -441,6 +443,7 @@ static int task_blocks_on_rt_mutex(struct rt_mutex *lock, spin_lock_irqsave(&owner->pi_lock, flags); if (owner->pi_blocked_on) { boost = 1; + /* gets dropped in rt_mutex_adjust_prio_chain()! */ get_task_struct(owner); } spin_unlock_irqrestore(&owner->pi_lock, flags); @@ -450,8 +453,8 @@ static int task_blocks_on_rt_mutex(struct rt_mutex *lock, spin_unlock(&lock->wait_lock); - res = rt_mutex_adjust_prio_chain(owner, detect_deadlock, lock, - waiter __IP__); + res = rt_mutex_adjust_prio_chain(owner, detect_deadlock, lock, waiter, + current __IP__); spin_lock(&lock->wait_lock); @@ -552,6 +555,7 @@ static void remove_waiter(struct rt_mutex *lock, if (owner->pi_blocked_on) { boost = 1; + /* gets dropped in rt_mutex_adjust_prio_chain()! */ get_task_struct(owner); } spin_unlock_irqrestore(&owner->pi_lock, flags); @@ -564,11 +568,36 @@ static void remove_waiter(struct rt_mutex *lock, spin_unlock(&lock->wait_lock); - rt_mutex_adjust_prio_chain(owner, 0, lock, NULL __IP__); + rt_mutex_adjust_prio_chain(owner, 0, lock, NULL, current __IP__); spin_lock(&lock->wait_lock); } +/* + * Recheck the pi chain, in case we got a priority setting + * + * Called from sched_setscheduler + */ +void rt_mutex_adjust_pi(struct task_struct *task) +{ + struct rt_mutex_waiter *waiter; + unsigned long flags; + + spin_lock_irqsave(&task->pi_lock, flags); + + waiter = task->pi_blocked_on; + if (!waiter || waiter->list_entry.prio == task->prio) { + spin_unlock_irqrestore(&task->pi_lock, flags); + return; + } + + /* gets dropped in rt_mutex_adjust_prio_chain()! */ + get_task_struct(task); + spin_unlock_irqrestore(&task->pi_lock, flags); + + rt_mutex_adjust_prio_chain(task, 0, NULL, NULL, task __RET_IP__); +} + /* * Slow path lock function: */ @@ -636,6 +665,7 @@ rt_mutex_slowlock(struct rt_mutex *lock, int state, if (unlikely(ret)) break; } + spin_unlock(&lock->wait_lock); debug_rt_mutex_print_deadlock(&waiter); diff --git a/kernel/sched.c b/kernel/sched.c index 7a30addfd23..2629c1711fd 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -4070,6 +4070,8 @@ recheck: __task_rq_unlock(rq); spin_unlock_irqrestore(&p->pi_lock, flags); + rt_mutex_adjust_pi(p); + return 0; } EXPORT_SYMBOL_GPL(sched_setscheduler); -- cgit v1.2.3-70-g09d2 From 456229a91d2bdf884f0c01b33f1ecee762abba7d Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Tue, 27 Jun 2006 02:55:07 -0700 Subject: [PATCH] drivers/char/ipmi/ipmi_msghandler.c: make proc_ipmi_root static Make struct proc_ipmi_root static. Besides this, tremove removes an unused #ifdef CONFIG_PROC_FS from include/linux/ipmi.h. Acked-by: Corey Minyard Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/char/ipmi/ipmi_msghandler.c | 3 +-- include/linux/ipmi.h | 4 ---- 2 files changed, 1 insertion(+), 6 deletions(-) (limited to 'include') diff --git a/drivers/char/ipmi/ipmi_msghandler.c b/drivers/char/ipmi/ipmi_msghandler.c index b03ddab1bef..83ed6ae466a 100644 --- a/drivers/char/ipmi/ipmi_msghandler.c +++ b/drivers/char/ipmi/ipmi_msghandler.c @@ -57,8 +57,7 @@ static int ipmi_init_msghandler(void); static int initialized = 0; #ifdef CONFIG_PROC_FS -struct proc_dir_entry *proc_ipmi_root = NULL; -EXPORT_SYMBOL(proc_ipmi_root); +static struct proc_dir_entry *proc_ipmi_root = NULL; #endif /* CONFIG_PROC_FS */ #define MAX_EVENTS_IN_QUEUE 25 diff --git a/include/linux/ipmi.h b/include/linux/ipmi.h index 5653b2f23b6..d09fbeabf1d 100644 --- a/include/linux/ipmi.h +++ b/include/linux/ipmi.h @@ -210,11 +210,7 @@ struct kernel_ipmi_msg #include #include #include - -#ifdef CONFIG_PROC_FS #include -extern struct proc_dir_entry *proc_ipmi_root; -#endif /* CONFIG_PROC_FS */ /* Opaque type for a IPMI message user. One of these is needed to send and receive messages. */ -- cgit v1.2.3-70-g09d2 From 999567d829b191b2cb7fd8dee31d5bccc81d7c55 Mon Sep 17 00:00:00 2001 From: Greg Ungerer Date: Tue, 27 Jun 2006 13:22:07 +1000 Subject: [PATCH] m68knommu: remove NO_FORMAT_VECi from ptrace.h header Remove NO_FORMAT_VEC conditional check. It is not used or defined anywhere. Signed-off-by: Greg Ungerer Signed-off-by: Linus Torvalds --- include/asm-m68knommu/ptrace.h | 2 -- 1 file changed, 2 deletions(-) (limited to 'include') diff --git a/include/asm-m68knommu/ptrace.h b/include/asm-m68knommu/ptrace.h index 1e19c457de7..47258e86e8c 100644 --- a/include/asm-m68knommu/ptrace.h +++ b/include/asm-m68knommu/ptrace.h @@ -46,11 +46,9 @@ struct pt_regs { #else unsigned short sr; unsigned long pc; -#ifndef NO_FORMAT_VEC unsigned format : 4; /* frame format specifier */ unsigned vector : 12; /* vector offset */ #endif -#endif }; /* -- cgit v1.2.3-70-g09d2 From f93d6d071fdbf27141c5331d6cd988664650bc1f Mon Sep 17 00:00:00 2001 From: Jon Loeliger Date: Thu, 22 Jun 2006 13:02:58 -0500 Subject: [POWERPC] Remove obsolete #include . Signed-off-by: Jon Loeliger Signed-off-by: Paul Mackerras --- arch/powerpc/platforms/86xx/mpc8641_hpcn.h | 1 - arch/powerpc/platforms/86xx/mpc86xx_hpcn.c | 1 - arch/powerpc/platforms/86xx/mpc86xx_smp.c | 1 - arch/powerpc/platforms/86xx/pci.c | 1 - include/asm-powerpc/mpc86xx.h | 1 - 5 files changed, 5 deletions(-) (limited to 'include') diff --git a/arch/powerpc/platforms/86xx/mpc8641_hpcn.h b/arch/powerpc/platforms/86xx/mpc8641_hpcn.h index 5042253758b..5d2bcf78cef 100644 --- a/arch/powerpc/platforms/86xx/mpc8641_hpcn.h +++ b/arch/powerpc/platforms/86xx/mpc8641_hpcn.h @@ -14,7 +14,6 @@ #ifndef __MPC8641_HPCN_H__ #define __MPC8641_HPCN_H__ -#include #include /* PCI interrupt controller */ diff --git a/arch/powerpc/platforms/86xx/mpc86xx_hpcn.c b/arch/powerpc/platforms/86xx/mpc86xx_hpcn.c index 483c21df181..a9a8539ec0a 100644 --- a/arch/powerpc/platforms/86xx/mpc86xx_hpcn.c +++ b/arch/powerpc/platforms/86xx/mpc86xx_hpcn.c @@ -12,7 +12,6 @@ * option) any later version. */ -#include #include #include #include diff --git a/arch/powerpc/platforms/86xx/mpc86xx_smp.c b/arch/powerpc/platforms/86xx/mpc86xx_smp.c index 944ec4b7141..167f0401a1c 100644 --- a/arch/powerpc/platforms/86xx/mpc86xx_smp.c +++ b/arch/powerpc/platforms/86xx/mpc86xx_smp.c @@ -10,7 +10,6 @@ * option) any later version. */ -#include #include #include #include diff --git a/arch/powerpc/platforms/86xx/pci.c b/arch/powerpc/platforms/86xx/pci.c index 5180df7c75b..6dfdcb87587 100644 --- a/arch/powerpc/platforms/86xx/pci.c +++ b/arch/powerpc/platforms/86xx/pci.c @@ -12,7 +12,6 @@ * option) any later version. */ -#include #include #include #include diff --git a/include/asm-powerpc/mpc86xx.h b/include/asm-powerpc/mpc86xx.h index d0a6718d188..1ed72f6b121 100644 --- a/include/asm-powerpc/mpc86xx.h +++ b/include/asm-powerpc/mpc86xx.h @@ -15,7 +15,6 @@ #ifndef __ASM_POWERPC_MPC86xx_H__ #define __ASM_POWERPC_MPC86xx_H__ -#include #include #ifdef CONFIG_PPC_86xx -- cgit v1.2.3-70-g09d2 From 5f50867b4f1938ab80d249206efbec37bba48c39 Mon Sep 17 00:00:00 2001 From: Haren Myneni Date: Thu, 22 Jun 2006 23:35:10 -0700 Subject: [POWERPC] kdump: Reserve the existing TCE mappings left by the first kernel During kdump boot, noticed some machines checkstop on dma protection fault for ongoing DMA left in the first kernel. Instead of initializing TCE entries in iommu_init() for the kdump boot, this patch fixes this issue by walking through the each TCE table and checks whether the entries are in use by the first kernel. If so, reserve those entries by setting the corresponding bit in tbl->it_map such that these entries will not be available for the kdump boot. However it could be possible that all TCE entries might be used up due to the driver bug that does continuous mapping. My observation is around 1700 TCE entries are used on some systems (Ex: P4) at some point of time during kdump boot and saving dump (either write into the disk or sending to remote machine). Hence, this patch will make sure that minimum of 2048 entries will be available such that kdump boot could be successful in some cases. Signed-off-by: Haren Myneni Acked-by: Olof Johansson Signed-off-by: Paul Mackerras --- arch/powerpc/kernel/iommu.c | 30 ++++++++++++++++++++++++++++++ arch/powerpc/platforms/pseries/iommu.c | 33 +++++++++++++++++++++++++++++++++ include/asm-powerpc/kdump.h | 2 ++ include/asm-powerpc/machdep.h | 2 ++ 4 files changed, 67 insertions(+) (limited to 'include') diff --git a/arch/powerpc/kernel/iommu.c b/arch/powerpc/kernel/iommu.c index 7cb77c20fc5..3d677ac9965 100644 --- a/arch/powerpc/kernel/iommu.c +++ b/arch/powerpc/kernel/iommu.c @@ -38,6 +38,7 @@ #include #include #include +#include #define DBG(...) @@ -440,8 +441,37 @@ struct iommu_table *iommu_init_table(struct iommu_table *tbl, int nid) tbl->it_largehint = tbl->it_halfpoint; spin_lock_init(&tbl->it_lock); +#ifdef CONFIG_CRASH_DUMP + if (ppc_md.tce_get) { + unsigned long index, tceval; + unsigned long tcecount = 0; + + /* + * Reserve the existing mappings left by the first kernel. + */ + for (index = 0; index < tbl->it_size; index++) { + tceval = ppc_md.tce_get(tbl, index + tbl->it_offset); + /* + * Freed TCE entry contains 0x7fffffffffffffff on JS20 + */ + if (tceval && (tceval != 0x7fffffffffffffffUL)) { + __set_bit(index, tbl->it_map); + tcecount++; + } + } + if ((tbl->it_size - tcecount) < KDUMP_MIN_TCE_ENTRIES) { + printk(KERN_WARNING "TCE table is full; "); + printk(KERN_WARNING "freeing %d entries for the kdump boot\n", + KDUMP_MIN_TCE_ENTRIES); + for (index = tbl->it_size - KDUMP_MIN_TCE_ENTRIES; + index < tbl->it_size; index++) + __clear_bit(index, tbl->it_map); + } + } +#else /* Clear the hardware table in case firmware left allocations in it */ ppc_md.tce_free(tbl, tbl->it_offset, tbl->it_size); +#endif if (!welcomed) { printk(KERN_INFO "IOMMU table initialized, virtual merging %s\n", diff --git a/arch/powerpc/platforms/pseries/iommu.c b/arch/powerpc/platforms/pseries/iommu.c index d03a8b078f9..8cfb5706790 100644 --- a/arch/powerpc/platforms/pseries/iommu.c +++ b/arch/powerpc/platforms/pseries/iommu.c @@ -92,6 +92,15 @@ static void tce_free_pSeries(struct iommu_table *tbl, long index, long npages) *(tcep++) = 0; } +static unsigned long tce_get_pseries(struct iommu_table *tbl, long index) +{ + u64 *tcep; + + index <<= TCE_PAGE_FACTOR; + tcep = ((u64 *)tbl->it_base) + index; + + return *tcep; +} static void tce_build_pSeriesLP(struct iommu_table *tbl, long tcenum, long npages, unsigned long uaddr, @@ -235,6 +244,25 @@ static void tce_freemulti_pSeriesLP(struct iommu_table *tbl, long tcenum, long n } } +static unsigned long tce_get_pSeriesLP(struct iommu_table *tbl, long tcenum) +{ + u64 rc; + unsigned long tce_ret; + + tcenum <<= TCE_PAGE_FACTOR; + rc = plpar_tce_get((u64)tbl->it_index, (u64)tcenum << 12, &tce_ret); + + if (rc && printk_ratelimit()) { + printk("tce_get_pSeriesLP: plpar_tce_get failed. rc=%ld\n", + rc); + printk("\tindex = 0x%lx\n", (u64)tbl->it_index); + printk("\ttcenum = 0x%lx\n", (u64)tcenum); + show_stack(current, (unsigned long *)__get_SP()); + } + + return tce_ret; +} + static void iommu_table_setparms(struct pci_controller *phb, struct device_node *dn, struct iommu_table *tbl) @@ -254,7 +282,10 @@ static void iommu_table_setparms(struct pci_controller *phb, } tbl->it_base = (unsigned long)__va(*basep); + +#ifndef CONFIG_CRASH_DUMP memset((void *)tbl->it_base, 0, *sizep); +#endif tbl->it_busno = phb->bus->number; @@ -560,11 +591,13 @@ void iommu_init_early_pSeries(void) ppc_md.tce_build = tce_build_pSeriesLP; ppc_md.tce_free = tce_free_pSeriesLP; } + ppc_md.tce_get = tce_get_pSeriesLP; ppc_md.iommu_bus_setup = iommu_bus_setup_pSeriesLP; ppc_md.iommu_dev_setup = iommu_dev_setup_pSeriesLP; } else { ppc_md.tce_build = tce_build_pSeries; ppc_md.tce_free = tce_free_pSeries; + ppc_md.tce_get = tce_get_pseries; ppc_md.iommu_bus_setup = iommu_bus_setup_pSeries; ppc_md.iommu_dev_setup = iommu_dev_setup_pSeries; } diff --git a/include/asm-powerpc/kdump.h b/include/asm-powerpc/kdump.h index 5a5c3b5ab1e..dc1574c945f 100644 --- a/include/asm-powerpc/kdump.h +++ b/include/asm-powerpc/kdump.h @@ -15,6 +15,8 @@ #define KDUMP_TRAMPOLINE_START 0x0100 #define KDUMP_TRAMPOLINE_END 0x3000 +#define KDUMP_MIN_TCE_ENTRIES 2048 + #else /* !CONFIG_CRASH_DUMP */ #define PHYSICAL_START 0x0 diff --git a/include/asm-powerpc/machdep.h b/include/asm-powerpc/machdep.h index 73db1f71329..eba133d149a 100644 --- a/include/asm-powerpc/machdep.h +++ b/include/asm-powerpc/machdep.h @@ -81,6 +81,8 @@ struct machdep_calls { void (*tce_free)(struct iommu_table *tbl, long index, long npages); + unsigned long (*tce_get)(struct iommu_table *tbl, + long index); void (*tce_flush)(struct iommu_table *tbl); void (*iommu_dev_setup)(struct pci_dev *dev); void (*iommu_bus_setup)(struct pci_bus *bus); -- cgit v1.2.3-70-g09d2 From 7d0daae4ae1a3e80d78b83cddf414a3b98a962f4 Mon Sep 17 00:00:00 2001 From: Michael Ellerman Date: Fri, 23 Jun 2006 18:16:38 +1000 Subject: [POWERPC] powerpc: Initialise ppc_md htab pointers earlier Initialise the ppc_md htab callbacks earlier, in the probe routines. This allows us to call htab_finish_init() from htab_initialize(), and makes it private to hash_utils_64.c. Move htab_finish_init() and make_bl() above htab_initialize() to avoid forward declarations. Signed-off-by: Michael Ellerman Signed-off-by: Paul Mackerras --- arch/powerpc/kernel/setup_64.c | 6 +-- arch/powerpc/mm/hash_native_64.c | 3 +- arch/powerpc/mm/hash_utils_64.c | 72 +++++++++++++++++---------------- arch/powerpc/platforms/cell/setup.c | 12 +++--- arch/powerpc/platforms/iseries/htab.c | 4 +- arch/powerpc/platforms/iseries/setup.c | 7 +--- arch/powerpc/platforms/maple/setup.c | 7 +--- arch/powerpc/platforms/powermac/setup.c | 9 +---- arch/powerpc/platforms/pseries/lpar.c | 4 +- arch/powerpc/platforms/pseries/setup.c | 10 ++--- include/asm-powerpc/mmu.h | 1 - 11 files changed, 58 insertions(+), 77 deletions(-) (limited to 'include') diff --git a/arch/powerpc/kernel/setup_64.c b/arch/powerpc/kernel/setup_64.c index 78f3a5fd43f..6f213a9e36b 100644 --- a/arch/powerpc/kernel/setup_64.c +++ b/arch/powerpc/kernel/setup_64.c @@ -358,11 +358,7 @@ void __init setup_system(void) * Fill the ppc64_caches & systemcfg structures with informations * retrieved from the device-tree. Need to be called before * finish_device_tree() since the later requires some of the - * informations filled up here to properly parse the interrupt - * tree. - * It also sets up the cache line sizes which allows to call - * routines like flush_icache_range (used by the hash init - * later on). + * informations filled up here to properly parse the interrupt tree. */ initialize_cache_info(); diff --git a/arch/powerpc/mm/hash_native_64.c b/arch/powerpc/mm/hash_native_64.c index a0f3cbd00d3..c90f124f3c7 100644 --- a/arch/powerpc/mm/hash_native_64.c +++ b/arch/powerpc/mm/hash_native_64.c @@ -520,7 +520,7 @@ static inline int tlb_batching_enabled(void) } #endif -void hpte_init_native(void) +void __init hpte_init_native(void) { ppc_md.hpte_invalidate = native_hpte_invalidate; ppc_md.hpte_updatepp = native_hpte_updatepp; @@ -530,5 +530,4 @@ void hpte_init_native(void) ppc_md.hpte_clear_all = native_hpte_clear; if (tlb_batching_enabled()) ppc_md.flush_hash_range = native_flush_hash_range; - htab_finish_init(); } diff --git a/arch/powerpc/mm/hash_utils_64.c b/arch/powerpc/mm/hash_utils_64.c index d03fd2b4445..9cefe6a7aeb 100644 --- a/arch/powerpc/mm/hash_utils_64.c +++ b/arch/powerpc/mm/hash_utils_64.c @@ -413,6 +413,41 @@ void create_section_mapping(unsigned long start, unsigned long end) } #endif /* CONFIG_MEMORY_HOTPLUG */ +static inline void make_bl(unsigned int *insn_addr, void *func) +{ + unsigned long funcp = *((unsigned long *)func); + int offset = funcp - (unsigned long)insn_addr; + + *insn_addr = (unsigned int)(0x48000001 | (offset & 0x03fffffc)); + flush_icache_range((unsigned long)insn_addr, 4+ + (unsigned long)insn_addr); +} + +static void __init htab_finish_init(void) +{ + extern unsigned int *htab_call_hpte_insert1; + extern unsigned int *htab_call_hpte_insert2; + extern unsigned int *htab_call_hpte_remove; + extern unsigned int *htab_call_hpte_updatepp; + +#ifdef CONFIG_PPC_64K_PAGES + extern unsigned int *ht64_call_hpte_insert1; + extern unsigned int *ht64_call_hpte_insert2; + extern unsigned int *ht64_call_hpte_remove; + extern unsigned int *ht64_call_hpte_updatepp; + + make_bl(ht64_call_hpte_insert1, ppc_md.hpte_insert); + make_bl(ht64_call_hpte_insert2, ppc_md.hpte_insert); + make_bl(ht64_call_hpte_remove, ppc_md.hpte_remove); + make_bl(ht64_call_hpte_updatepp, ppc_md.hpte_updatepp); +#endif /* CONFIG_PPC_64K_PAGES */ + + make_bl(htab_call_hpte_insert1, ppc_md.hpte_insert); + make_bl(htab_call_hpte_insert2, ppc_md.hpte_insert); + make_bl(htab_call_hpte_remove, ppc_md.hpte_remove); + make_bl(htab_call_hpte_updatepp, ppc_md.hpte_updatepp); +} + void __init htab_initialize(void) { unsigned long table; @@ -525,6 +560,8 @@ void __init htab_initialize(void) mmu_linear_psize)); } + htab_finish_init(); + DBG(" <- htab_initialize()\n"); } #undef KB @@ -787,16 +824,6 @@ void flush_hash_range(unsigned long number, int local) } } -static inline void make_bl(unsigned int *insn_addr, void *func) -{ - unsigned long funcp = *((unsigned long *)func); - int offset = funcp - (unsigned long)insn_addr; - - *insn_addr = (unsigned int)(0x48000001 | (offset & 0x03fffffc)); - flush_icache_range((unsigned long)insn_addr, 4+ - (unsigned long)insn_addr); -} - /* * low_hash_fault is called when we the low level hash code failed * to instert a PTE due to an hypervisor error @@ -815,28 +842,3 @@ void low_hash_fault(struct pt_regs *regs, unsigned long address) } bad_page_fault(regs, address, SIGBUS); } - -void __init htab_finish_init(void) -{ - extern unsigned int *htab_call_hpte_insert1; - extern unsigned int *htab_call_hpte_insert2; - extern unsigned int *htab_call_hpte_remove; - extern unsigned int *htab_call_hpte_updatepp; - -#ifdef CONFIG_PPC_64K_PAGES - extern unsigned int *ht64_call_hpte_insert1; - extern unsigned int *ht64_call_hpte_insert2; - extern unsigned int *ht64_call_hpte_remove; - extern unsigned int *ht64_call_hpte_updatepp; - - make_bl(ht64_call_hpte_insert1, ppc_md.hpte_insert); - make_bl(ht64_call_hpte_insert2, ppc_md.hpte_insert); - make_bl(ht64_call_hpte_remove, ppc_md.hpte_remove); - make_bl(ht64_call_hpte_updatepp, ppc_md.hpte_updatepp); -#endif /* CONFIG_PPC_64K_PAGES */ - - make_bl(htab_call_hpte_insert1, ppc_md.hpte_insert); - make_bl(htab_call_hpte_insert2, ppc_md.hpte_insert); - make_bl(htab_call_hpte_remove, ppc_md.hpte_remove); - make_bl(htab_call_hpte_updatepp, ppc_md.hpte_updatepp); -} diff --git a/arch/powerpc/platforms/cell/setup.c b/arch/powerpc/platforms/cell/setup.c index 3d1831d331e..2436eaab701 100644 --- a/arch/powerpc/platforms/cell/setup.c +++ b/arch/powerpc/platforms/cell/setup.c @@ -125,8 +125,6 @@ static void __init cell_init_early(void) { DBG(" -> cell_init_early()\n"); - hpte_init_native(); - cell_init_iommu(); ppc64_interrupt_controller = IC_CELL_PIC; @@ -139,11 +137,13 @@ static int __init cell_probe(void) { unsigned long root = of_get_flat_dt_root(); - if (of_flat_dt_is_compatible(root, "IBM,CBEA") || - of_flat_dt_is_compatible(root, "IBM,CPBW-1.0")) - return 1; + if (!of_flat_dt_is_compatible(root, "IBM,CBEA") && + !of_flat_dt_is_compatible(root, "IBM,CPBW-1.0")) + return 0; + + hpte_init_native(); - return 0; + return 1; } /* diff --git a/arch/powerpc/platforms/iseries/htab.c b/arch/powerpc/platforms/iseries/htab.c index 30bdcf3925d..ed44dfceaa4 100644 --- a/arch/powerpc/platforms/iseries/htab.c +++ b/arch/powerpc/platforms/iseries/htab.c @@ -242,13 +242,11 @@ static void iSeries_hpte_invalidate(unsigned long slot, unsigned long va, local_irq_restore(flags); } -void hpte_init_iSeries(void) +void __init hpte_init_iSeries(void) { ppc_md.hpte_invalidate = iSeries_hpte_invalidate; ppc_md.hpte_updatepp = iSeries_hpte_updatepp; ppc_md.hpte_updateboltedpp = iSeries_hpte_updateboltedpp; ppc_md.hpte_insert = iSeries_hpte_insert; ppc_md.hpte_remove = iSeries_hpte_remove; - - htab_finish_init(); } diff --git a/arch/powerpc/platforms/iseries/setup.c b/arch/powerpc/platforms/iseries/setup.c index 973f5aa7165..66c77e4f8ec 100644 --- a/arch/powerpc/platforms/iseries/setup.c +++ b/arch/powerpc/platforms/iseries/setup.c @@ -318,11 +318,6 @@ static void __init iSeries_init_early(void) iSeries_recal_tb = get_tb(); iSeries_recal_titan = HvCallXm_loadTod(); - /* - * Initialize the hash table management pointers - */ - hpte_init_iSeries(); - /* * Initialize the DMA/TCE management */ @@ -671,6 +666,8 @@ static int __init iseries_probe(void) */ virt_irq_max = 255; + hpte_init_iSeries(); + return 1; } diff --git a/arch/powerpc/platforms/maple/setup.c b/arch/powerpc/platforms/maple/setup.c index a0505ea48a8..4e32a5417fd 100644 --- a/arch/powerpc/platforms/maple/setup.c +++ b/arch/powerpc/platforms/maple/setup.c @@ -199,11 +199,6 @@ static void __init maple_init_early(void) { DBG(" -> maple_init_early\n"); - /* Initialize hash table, from now on, we can take hash faults - * and call ioremap - */ - hpte_init_native(); - /* Setup interrupt mapping options */ ppc64_interrupt_controller = IC_OPEN_PIC; @@ -272,6 +267,8 @@ static int __init maple_probe(void) */ alloc_dart_table(); + hpte_init_native(); + return 1; } diff --git a/arch/powerpc/platforms/powermac/setup.c b/arch/powerpc/platforms/powermac/setup.c index 9cc7db7a8bd..89c5775f83b 100644 --- a/arch/powerpc/platforms/powermac/setup.c +++ b/arch/powerpc/platforms/powermac/setup.c @@ -600,13 +600,6 @@ pmac_halt(void) */ static void __init pmac_init_early(void) { -#ifdef CONFIG_PPC64 - /* Initialize hash table, from now on, we can take hash faults - * and call ioremap - */ - hpte_init_native(); -#endif - /* Enable early btext debug if requested */ if (strstr(cmd_line, "btextdbg")) { udbg_adb_init_early(); @@ -683,6 +676,8 @@ static int __init pmac_probe(void) * part of the cacheable linar mapping */ alloc_dart_table(); + + hpte_init_native(); #endif #ifdef CONFIG_PPC32 diff --git a/arch/powerpc/platforms/pseries/lpar.c b/arch/powerpc/platforms/pseries/lpar.c index 634b7d06d3c..27480705996 100644 --- a/arch/powerpc/platforms/pseries/lpar.c +++ b/arch/powerpc/platforms/pseries/lpar.c @@ -513,7 +513,7 @@ void pSeries_lpar_flush_hash_range(unsigned long number, int local) spin_unlock_irqrestore(&pSeries_lpar_tlbie_lock, flags); } -void hpte_init_lpar(void) +void __init hpte_init_lpar(void) { ppc_md.hpte_invalidate = pSeries_lpar_hpte_invalidate; ppc_md.hpte_updatepp = pSeries_lpar_hpte_updatepp; @@ -522,6 +522,4 @@ void hpte_init_lpar(void) ppc_md.hpte_remove = pSeries_lpar_hpte_remove; ppc_md.flush_hash_range = pSeries_lpar_flush_hash_range; ppc_md.hpte_clear_all = pSeries_lpar_hptab_clear; - - htab_finish_init(); } diff --git a/arch/powerpc/platforms/pseries/setup.c b/arch/powerpc/platforms/pseries/setup.c index 1e28518c612..b3197ff156c 100644 --- a/arch/powerpc/platforms/pseries/setup.c +++ b/arch/powerpc/platforms/pseries/setup.c @@ -322,11 +322,6 @@ static void __init pSeries_init_early(void) DBG(" -> pSeries_init_early()\n"); fw_feature_init(); - - if (firmware_has_feature(FW_FEATURE_LPAR)) - hpte_init_lpar(); - else - hpte_init_native(); if (firmware_has_feature(FW_FEATURE_LPAR)) find_udbg_vterm(); @@ -384,6 +379,11 @@ static int __init pSeries_probe_hypertas(unsigned long node, if (of_get_flat_dt_prop(node, "ibm,hypertas-functions", NULL) != NULL) powerpc_firmware_features |= FW_FEATURE_LPAR; + if (firmware_has_feature(FW_FEATURE_LPAR)) + hpte_init_lpar(); + else + hpte_init_native(); + return 1; } diff --git a/include/asm-powerpc/mmu.h b/include/asm-powerpc/mmu.h index 3a5ebe229af..c3fc7a28e3c 100644 --- a/include/asm-powerpc/mmu.h +++ b/include/asm-powerpc/mmu.h @@ -238,7 +238,6 @@ extern int hash_huge_page(struct mm_struct *mm, unsigned long access, unsigned long ea, unsigned long vsid, int local, unsigned long trap); -extern void htab_finish_init(void); extern int htab_bolt_mapping(unsigned long vstart, unsigned long vend, unsigned long pstart, unsigned long mode, int psize); -- cgit v1.2.3-70-g09d2 From aa98c50dcb5d5b85d2a4c26d54fa1e3c31c11e4b Mon Sep 17 00:00:00 2001 From: Michael Ellerman Date: Fri, 23 Jun 2006 18:17:32 +1000 Subject: [POWERPC] Make kexec_setup() a regular initcall There's no reason kexec_setup() needs to be called explicitly from setup_system(), it can just be a regular initcall. Signed-off-by: Michael Ellerman Signed-off-by: Paul Mackerras --- arch/powerpc/kernel/machine_kexec_64.c | 4 +++- arch/powerpc/kernel/setup_64.c | 4 ---- include/asm-powerpc/kexec.h | 1 - 3 files changed, 3 insertions(+), 6 deletions(-) (limited to 'include') diff --git a/arch/powerpc/kernel/machine_kexec_64.c b/arch/powerpc/kernel/machine_kexec_64.c index a8fa04ef27c..b438d45a068 100644 --- a/arch/powerpc/kernel/machine_kexec_64.c +++ b/arch/powerpc/kernel/machine_kexec_64.c @@ -378,11 +378,13 @@ static void __init export_crashk_values(void) of_node_put(node); } -void __init kexec_setup(void) +static int __init kexec_setup(void) { export_htab_values(); export_crashk_values(); + return 0; } +__initcall(kexec_setup); static int __init early_parse_crashk(char *p) { diff --git a/arch/powerpc/kernel/setup_64.c b/arch/powerpc/kernel/setup_64.c index 6f213a9e36b..cc26530145d 100644 --- a/arch/powerpc/kernel/setup_64.c +++ b/arch/powerpc/kernel/setup_64.c @@ -350,10 +350,6 @@ void __init setup_system(void) */ unflatten_device_tree(); -#ifdef CONFIG_KEXEC - kexec_setup(); /* requires unflattened device tree. */ -#endif - /* * Fill the ppc64_caches & systemcfg structures with informations * retrieved from the device-tree. Need to be called before diff --git a/include/asm-powerpc/kexec.h b/include/asm-powerpc/kexec.h index efe8872ec58..234bd684c7f 100644 --- a/include/asm-powerpc/kexec.h +++ b/include/asm-powerpc/kexec.h @@ -112,7 +112,6 @@ static inline void crash_setup_regs(struct pt_regs *newregs, #ifdef __powerpc64__ extern void kexec_smp_wait(void); /* get and clear naca physid, wait for master to copy new code to 0 */ -extern void __init kexec_setup(void); extern int crashing_cpu; extern void crash_send_ipi(void (*crash_ipi_callback)(struct pt_regs *)); #endif /* __powerpc64 __ */ -- cgit v1.2.3-70-g09d2 From 458148c00b97864a27ecf528a1d45a8e5ebd9bbc Mon Sep 17 00:00:00 2001 From: Michael Ellerman Date: Fri, 23 Jun 2006 18:20:13 +1000 Subject: [POWERPC] Setup RTAS values earlier, to enable rtas_call() earlier Althought RTAS is instantiated when we enter the kernel, we can't actually call into it until we know its entry point address. Currently we grab that in rtas_initialize(), however that's quite late in the boot sequence. To enable rtas_call() earlier, we can grab the RTAS entry etc. values while we're scanning the flattened device tree. There's existing code to retrieve the values from /chosen, however we don't store them there anymore, so remove that code. Signed-off-by: Michael Ellerman Signed-off-by: Paul Mackerras --- arch/powerpc/kernel/prom.c | 23 +++++------------------ arch/powerpc/kernel/rtas.c | 22 ++++++++++++++++++++++ include/asm-powerpc/rtas.h | 3 +++ 3 files changed, 30 insertions(+), 18 deletions(-) (limited to 'include') diff --git a/arch/powerpc/kernel/prom.c b/arch/powerpc/kernel/prom.c index efed4bc2b45..ce02c056ac3 100644 --- a/arch/powerpc/kernel/prom.c +++ b/arch/powerpc/kernel/prom.c @@ -1125,24 +1125,6 @@ static int __init early_init_dt_scan_chosen(unsigned long node, tce_alloc_end = *lprop; #endif -#ifdef CONFIG_PPC_RTAS - /* To help early debugging via the front panel, we retrieve a minimal - * set of RTAS infos now if available - */ - { - u64 *basep, *entryp, *sizep; - - basep = of_get_flat_dt_prop(node, "linux,rtas-base", NULL); - entryp = of_get_flat_dt_prop(node, "linux,rtas-entry", NULL); - sizep = of_get_flat_dt_prop(node, "linux,rtas-size", NULL); - if (basep && entryp && sizep) { - rtas.base = *basep; - rtas.entry = *entryp; - rtas.size = *sizep; - } - } -#endif /* CONFIG_PPC_RTAS */ - #ifdef CONFIG_KEXEC lprop = (u64*)of_get_flat_dt_prop(node, "linux,crashkernel-base", NULL); if (lprop) @@ -1327,6 +1309,11 @@ void __init early_init_devtree(void *params) /* Setup flat device-tree pointer */ initial_boot_params = params; +#ifdef CONFIG_PPC_RTAS + /* Some machines might need RTAS info for debugging, grab it now. */ + of_scan_flat_dt(early_init_dt_scan_rtas, NULL); +#endif + /* Retrieve various informations from the /chosen node of the * device-tree, including the platform type, initrd location and * size, TCE reserve, and more ... diff --git a/arch/powerpc/kernel/rtas.c b/arch/powerpc/kernel/rtas.c index b6aed765966..061d8afd246 100644 --- a/arch/powerpc/kernel/rtas.c +++ b/arch/powerpc/kernel/rtas.c @@ -801,3 +801,25 @@ void __init rtas_initialize(void) rtas_last_error_token = rtas_token("rtas-last-error"); #endif } + +int __init early_init_dt_scan_rtas(unsigned long node, + const char *uname, int depth, void *data) +{ + u32 *basep, *entryp, *sizep; + + if (depth != 1 || strcmp(uname, "rtas") != 0) + return 0; + + basep = of_get_flat_dt_prop(node, "linux,rtas-base", NULL); + entryp = of_get_flat_dt_prop(node, "linux,rtas-entry", NULL); + sizep = of_get_flat_dt_prop(node, "rtas-size", NULL); + + if (basep && entryp && sizep) { + rtas.base = *basep; + rtas.entry = *entryp; + rtas.size = *sizep; + } + + /* break now */ + return 1; +} diff --git a/include/asm-powerpc/rtas.h b/include/asm-powerpc/rtas.h index 02e213e3d69..a33c6acffa6 100644 --- a/include/asm-powerpc/rtas.h +++ b/include/asm-powerpc/rtas.h @@ -181,6 +181,9 @@ extern int rtas_set_rtc_time(struct rtc_time *rtc_time); extern unsigned int rtas_busy_delay_time(int status); extern unsigned int rtas_busy_delay(int status); +extern int early_init_dt_scan_rtas(unsigned long node, + const char *uname, int depth, void *data); + extern void pSeries_log_error(char *buf, unsigned int err_type, int fatal); /* Error types logged. */ -- cgit v1.2.3-70-g09d2 From cc46bb98c0d52695f26c239de701050660f5c79f Mon Sep 17 00:00:00 2001 From: Michael Ellerman Date: Fri, 23 Jun 2006 18:20:16 +1000 Subject: [POWERPC] Add udbg support for RTAS console Add udbg hooks for the RTAS console, based on the RTAS put-term-char and get-term-char calls. Along with my previous patches, this should enable debugging as soon as early_init_dt_scan_rtas() is called. Signed-off-by: Michael Ellerman Signed-off-by: Paul Mackerras --- arch/powerpc/Kconfig | 4 +++ arch/powerpc/Kconfig.debug | 9 +++++- arch/powerpc/kernel/rtas.c | 72 +++++++++++++++++++++++++++++++++++++++++++++- arch/powerpc/kernel/udbg.c | 7 +++-- include/asm-powerpc/udbg.h | 3 +- 5 files changed, 90 insertions(+), 5 deletions(-) (limited to 'include') diff --git a/arch/powerpc/Kconfig b/arch/powerpc/Kconfig index c5e01c4b637..deea4c8e1f4 100644 --- a/arch/powerpc/Kconfig +++ b/arch/powerpc/Kconfig @@ -420,6 +420,10 @@ config PPC_IBM_CELL_BLADE select MMIO_NVRAM select PPC_UDBG_16550 +config UDBG_RTAS_CONSOLE + bool + default n + config XICS depends on PPC_PSERIES bool diff --git a/arch/powerpc/Kconfig.debug b/arch/powerpc/Kconfig.debug index c69006ae824..e29ef77d3b0 100644 --- a/arch/powerpc/Kconfig.debug +++ b/arch/powerpc/Kconfig.debug @@ -134,12 +134,19 @@ config PPC_EARLY_DEBUG_G5 help Select this to enable early debugging for Apple G5 machines. -config PPC_EARLY_DEBUG_RTAS +config PPC_EARLY_DEBUG_RTAS_PANEL bool "RTAS Panel" depends on PPC_RTAS help Select this to enable early debugging via the RTAS panel. +config PPC_EARLY_DEBUG_RTAS_CONSOLE + bool "RTAS Console" + depends on PPC_RTAS + select UDBG_RTAS_CONSOLE + help + Select this to enable early debugging via the RTAS console. + config PPC_EARLY_DEBUG_MAPLE bool "Maple real mode" depends on PPC_MAPLE diff --git a/arch/powerpc/kernel/rtas.c b/arch/powerpc/kernel/rtas.c index 061d8afd246..4a4cb559840 100644 --- a/arch/powerpc/kernel/rtas.c +++ b/arch/powerpc/kernel/rtas.c @@ -109,11 +109,71 @@ static void call_rtas_display_status_delay(char c) } } -void __init udbg_init_rtas(void) +void __init udbg_init_rtas_panel(void) { udbg_putc = call_rtas_display_status_delay; } +#ifdef CONFIG_UDBG_RTAS_CONSOLE + +/* If you think you're dying before early_init_dt_scan_rtas() does its + * work, you can hard code the token values for your firmware here and + * hardcode rtas.base/entry etc. + */ +static unsigned int rtas_putchar_token = RTAS_UNKNOWN_SERVICE; +static unsigned int rtas_getchar_token = RTAS_UNKNOWN_SERVICE; + +static void udbg_rtascon_putc(char c) +{ + int tries; + + if (!rtas.base) + return; + + /* Add CRs before LFs */ + if (c == '\n') + udbg_rtascon_putc('\r'); + + /* if there is more than one character to be displayed, wait a bit */ + for (tries = 0; tries < 16; tries++) { + if (rtas_call(rtas_putchar_token, 1, 1, NULL, c) == 0) + break; + udelay(1000); + } +} + +static int udbg_rtascon_getc_poll(void) +{ + int c; + + if (!rtas.base) + return -1; + + if (rtas_call(rtas_getchar_token, 0, 2, &c)) + return -1; + + return c; +} + +static int udbg_rtascon_getc(void) +{ + int c; + + while ((c = udbg_rtascon_getc_poll()) == -1) + ; + + return c; +} + + +void __init udbg_init_rtas_console(void) +{ + udbg_putc = udbg_rtascon_putc; + udbg_getc = udbg_rtascon_getc; + udbg_getc_poll = udbg_rtascon_getc_poll; +} +#endif /* CONFIG_UDBG_RTAS_CONSOLE */ + void rtas_progress(char *s, unsigned short hex) { struct device_node *root; @@ -820,6 +880,16 @@ int __init early_init_dt_scan_rtas(unsigned long node, rtas.size = *sizep; } +#ifdef CONFIG_UDBG_RTAS_CONSOLE + basep = of_get_flat_dt_prop(node, "put-term-char", NULL); + if (basep) + rtas_putchar_token = *basep; + + basep = of_get_flat_dt_prop(node, "get-term-char", NULL); + if (basep) + rtas_getchar_token = *basep; +#endif + /* break now */ return 1; } diff --git a/arch/powerpc/kernel/udbg.c b/arch/powerpc/kernel/udbg.c index 67d9fd9ae2b..759afd5e0d8 100644 --- a/arch/powerpc/kernel/udbg.c +++ b/arch/powerpc/kernel/udbg.c @@ -34,9 +34,12 @@ void __init udbg_early_init(void) #elif defined(CONFIG_PPC_EARLY_DEBUG_G5) /* For use on Apple G5 machines */ udbg_init_pmac_realmode(); -#elif defined(CONFIG_PPC_EARLY_DEBUG_RTAS) +#elif defined(CONFIG_PPC_EARLY_DEBUG_RTAS_PANEL) /* RTAS panel debug */ - udbg_init_rtas(); + udbg_init_rtas_panel(); +#elif defined(CONFIG_PPC_EARLY_DEBUG_RTAS_CONSOLE) + /* RTAS console debug */ + udbg_init_rtas_console(); #elif defined(CONFIG_PPC_EARLY_DEBUG_MAPLE) /* Maple real mode debug */ udbg_init_maple_realmode(); diff --git a/include/asm-powerpc/udbg.h b/include/asm-powerpc/udbg.h index 19a1517ac43..55e57844fa7 100644 --- a/include/asm-powerpc/udbg.h +++ b/include/asm-powerpc/udbg.h @@ -42,7 +42,8 @@ extern void __init udbg_init_debug_lpar(void); extern void __init udbg_init_pmac_realmode(void); extern void __init udbg_init_maple_realmode(void); extern void __init udbg_init_iseries(void); -extern void __init udbg_init_rtas(void); +extern void __init udbg_init_rtas_panel(void); +extern void __init udbg_init_rtas_console(void); #endif /* __KERNEL__ */ #endif /* _ASM_POWERPC_UDBG_H */ -- cgit v1.2.3-70-g09d2 From c0ce7d0886cf0c2579c604eac41a7e125bc0e96d Mon Sep 17 00:00:00 2001 From: David Wilder Date: Fri, 23 Jun 2006 15:29:34 -0700 Subject: [POWERPC] Add the use of the firmware soft-reset-nmi to kdump. With this patch, kdump uses the firmware soft-reset NMI for two purposes: 1) Initiate the kdump (take a crash dump) by issuing a soft-reset. 2) Break a CPU out of a deadlock condition that is detected during kdump processing. When a soft-reset is initiated each CPU will enter system_reset_exception() and set its corresponding bit in the global bit-array cpus_in_sr then call die(). When die() finds the CPU's bit set in cpu_in_sr crash_kexec() is called to initiate a crash dump. The first CPU to enter crash_kexec() is called the "crashing CPU". All other CPUs are "secondary CPUs". The secondary CPU's pass through to crash_kexec_secondary() and sleep. The crashing CPU waits for all CPUs to enter via soft-reset then boots the kdump kernel (see crash_soft_reset_check()) When the system crashes due to a panic or exception, crash_kexec() is called by panic() or die(). The crashing CPU sends an IPI to all other CPUs to notify them of the pending shutdown. If a CPU is in a deadlock or hung state with interrupts disabled, the IPI will not be delivered. The result being, that the kdump kernel is not booted. This problem is solved with the use of a firmware generated soft-reset. After the crashing_cpu has issued the IPI, it waits for 10 sec for all CPUs to enter crash_ipi_callback(). A CPU signifies its entry to crash_ipi_callback() by setting its corresponding bit in the cpus_in_crash bit array. After 10 sec, if one or more CPUs have not set their bit in cpus_in_crash we assume that the CPU(s) is deadlocked. The operator is then prompted to generate a soft-reset to break the deadlock. Each CPU enters the soft reset handler as described above. Two conditions must be handled at this point: 1) The system crashed because the operator generated a soft-reset. See 2) The system had crashed before the soft-reset was generated ( in the case of a Panic or oops). The first CPU to enter crash_kexec() uses the state of the kexec_lock to determine this state. If kexec_lock is already held then condition 2 is true and crash_kexec_secondary() is called, else; this CPU is flagged as the crashing CPU, the kexec_lock is acquired and crash_kexec() proceeds as described above. Each additional CPUs responding to the soft-reset will pass through crash_kexec() to kexec_secondary(). All secondary CPUs call crash_ipi_callback() readying them self's for the shutdown. When ready they clear their bit in cpus_in_sr. The crashing CPU waits in kexec_secondary() until all other CPUs have cleared their bits in cpus_in_sr. The kexec kernel boot is then started. Signed-off-by: Haren Myneni Signed-off-by: David Wilder Signed-off-by: Paul Mackerras --- arch/powerpc/kernel/crash.c | 147 +++++++++++++++++++++++++++++++++++--------- arch/powerpc/kernel/traps.c | 27 ++++---- include/asm-powerpc/kexec.h | 8 +++ kernel/kexec.c | 6 +- 4 files changed, 142 insertions(+), 46 deletions(-) (limited to 'include') diff --git a/arch/powerpc/kernel/crash.c b/arch/powerpc/kernel/crash.c index e253a45dcf1..b537cfa4e09 100644 --- a/arch/powerpc/kernel/crash.c +++ b/arch/powerpc/kernel/crash.c @@ -24,9 +24,11 @@ #include #include #include +#include #include #include +#include #include #include #include @@ -41,6 +43,7 @@ /* This keeps a track of which one is crashing cpu. */ int crashing_cpu = -1; +static cpumask_t cpus_in_crash = CPU_MASK_NONE; static u32 *append_elf_note(u32 *buf, char *name, unsigned type, void *data, size_t data_len) @@ -98,34 +101,66 @@ static void crash_save_this_cpu(struct pt_regs *regs, int cpu) } #ifdef CONFIG_SMP -static atomic_t waiting_for_crash_ipi; +static atomic_t enter_on_soft_reset = ATOMIC_INIT(0); void crash_ipi_callback(struct pt_regs *regs) { int cpu = smp_processor_id(); - if (cpu == crashing_cpu) - return; - if (!cpu_online(cpu)) return; - if (ppc_md.kexec_cpu_down) - ppc_md.kexec_cpu_down(1, 1); - local_irq_disable(); + if (!cpu_isset(cpu, cpus_in_crash)) + crash_save_this_cpu(regs, cpu); + cpu_set(cpu, cpus_in_crash); - crash_save_this_cpu(regs, cpu); - atomic_dec(&waiting_for_crash_ipi); + /* + * Entered via soft-reset - could be the kdump + * process is invoked using soft-reset or user activated + * it if some CPU did not respond to an IPI. + * For soft-reset, the secondary CPU can enter this func + * twice. 1 - using IPI, and 2. soft-reset. + * Tell the kexec CPU that entered via soft-reset and ready + * to go down. + */ + if (cpu_isset(cpu, cpus_in_sr)) { + cpu_clear(cpu, cpus_in_sr); + atomic_inc(&enter_on_soft_reset); + } + + /* + * Starting the kdump boot. + * This barrier is needed to make sure that all CPUs are stopped. + * If not, soft-reset will be invoked to bring other CPUs. + */ + while (!cpu_isset(crashing_cpu, cpus_in_crash)) + cpu_relax(); + + if (ppc_md.kexec_cpu_down) + ppc_md.kexec_cpu_down(1, 1); kexec_smp_wait(); /* NOTREACHED */ } -static void crash_kexec_prepare_cpus(void) +/* + * Wait until all CPUs are entered via soft-reset. + */ +static void crash_soft_reset_check(int cpu) +{ + unsigned int ncpus = num_online_cpus() - 1;/* Excluding the panic cpu */ + + cpu_clear(cpu, cpus_in_sr); + while (atomic_read(&enter_on_soft_reset) != ncpus) + cpu_relax(); +} + + +static void crash_kexec_prepare_cpus(int cpu) { unsigned int msecs; - atomic_set(&waiting_for_crash_ipi, num_online_cpus() - 1); + unsigned int ncpus = num_online_cpus() - 1;/* Excluding the panic cpu */ crash_send_ipi(crash_ipi_callback); smp_wmb(); @@ -133,14 +168,13 @@ static void crash_kexec_prepare_cpus(void) /* * FIXME: Until we will have the way to stop other CPUSs reliabally, * the crash CPU will send an IPI and wait for other CPUs to - * respond. If not, proceed the kexec boot even though we failed to - * capture other CPU states. + * respond. * Delay of at least 10 seconds. */ - printk(KERN_ALERT "Sending IPI to other cpus...\n"); + printk(KERN_EMERG "Sending IPI to other cpus...\n"); msecs = 10000; - while ((atomic_read(&waiting_for_crash_ipi) > 0) && (--msecs > 0)) { - barrier(); + while ((cpus_weight(cpus_in_crash) < ncpus) && (--msecs > 0)) { + cpu_relax(); mdelay(1); } @@ -149,18 +183,71 @@ static void crash_kexec_prepare_cpus(void) /* * FIXME: In case if we do not get all CPUs, one possibility: ask the * user to do soft reset such that we get all. - * IPI handler is already set by the panic cpu initially. Therefore, - * all cpus could invoke this handler from die() and the panic CPU - * will call machine_kexec() directly from this handler to do - * kexec boot. + * Soft-reset will be used until better mechanism is implemented. + */ + if (cpus_weight(cpus_in_crash) < ncpus) { + printk(KERN_EMERG "done waiting: %d cpu(s) not responding\n", + ncpus - cpus_weight(cpus_in_crash)); + printk(KERN_EMERG "Activate soft-reset to stop other cpu(s)\n"); + cpus_in_sr = CPU_MASK_NONE; + atomic_set(&enter_on_soft_reset, 0); + while (cpus_weight(cpus_in_crash) < ncpus) + cpu_relax(); + } + /* + * Make sure all CPUs are entered via soft-reset if the kdump is + * invoked using soft-reset. */ - if (atomic_read(&waiting_for_crash_ipi)) - printk(KERN_ALERT "done waiting: %d cpus not responding\n", - atomic_read(&waiting_for_crash_ipi)); + if (cpu_isset(cpu, cpus_in_sr)) + crash_soft_reset_check(cpu); /* Leave the IPI callback set */ } + +/* + * This function will be called by secondary cpus or by kexec cpu + * if soft-reset is activated to stop some CPUs. + */ +void crash_kexec_secondary(struct pt_regs *regs) +{ + int cpu = smp_processor_id(); + unsigned long flags; + int msecs = 5; + + local_irq_save(flags); + /* Wait 5ms if the kexec CPU is not entered yet. */ + while (crashing_cpu < 0) { + if (--msecs < 0) { + /* + * Either kdump image is not loaded or + * kdump process is not started - Probably xmon + * exited using 'x'(exit and recover) or + * kexec_should_crash() failed for all running tasks. + */ + cpu_clear(cpu, cpus_in_sr); + local_irq_restore(flags); + return; + } + mdelay(1); + cpu_relax(); + } + if (cpu == crashing_cpu) { + /* + * Panic CPU will enter this func only via soft-reset. + * Wait until all secondary CPUs entered and + * then start kexec boot. + */ + crash_soft_reset_check(cpu); + cpu_set(crashing_cpu, cpus_in_crash); + if (ppc_md.kexec_cpu_down) + ppc_md.kexec_cpu_down(1, 0); + machine_kexec(kexec_crash_image); + /* NOTREACHED */ + } + crash_ipi_callback(regs); +} + #else -static void crash_kexec_prepare_cpus(void) +static void crash_kexec_prepare_cpus(int cpu) { /* * move the secondarys to us so that we can copy @@ -171,6 +258,10 @@ static void crash_kexec_prepare_cpus(void) smp_release_cpus(); } +void crash_kexec_secondary(struct pt_regs *regs) +{ + cpus_in_sr = CPU_MASK_NONE; +} #endif void default_machine_crash_shutdown(struct pt_regs *regs) @@ -199,14 +290,14 @@ void default_machine_crash_shutdown(struct pt_regs *regs) desc->handler->disable(irq); } - if (ppc_md.kexec_cpu_down) - ppc_md.kexec_cpu_down(1, 0); - /* * Make a note of crashing cpu. Will be used in machine_kexec * such that another IPI will not be sent. */ crashing_cpu = smp_processor_id(); - crash_kexec_prepare_cpus(); crash_save_this_cpu(regs, crashing_cpu); + crash_kexec_prepare_cpus(crashing_cpu); + cpu_set(crashing_cpu, cpus_in_crash); + if (ppc_md.kexec_cpu_down) + ppc_md.kexec_cpu_down(1, 0); } diff --git a/arch/powerpc/kernel/traps.c b/arch/powerpc/kernel/traps.c index 52f5659534f..fa6bd97b6b9 100644 --- a/arch/powerpc/kernel/traps.c +++ b/arch/powerpc/kernel/traps.c @@ -52,9 +52,13 @@ #include #include #endif +#include #ifdef CONFIG_PPC64 /* XXX */ #define _IO_BASE pci_io_base +#ifdef CONFIG_KEXEC +cpumask_t cpus_in_sr = CPU_MASK_NONE; +#endif #endif #ifdef CONFIG_DEBUGGER @@ -97,7 +101,7 @@ static DEFINE_SPINLOCK(die_lock); int die(const char *str, struct pt_regs *regs, long err) { - static int die_counter, crash_dump_start = 0; + static int die_counter; if (debugger(regs)) return 1; @@ -137,21 +141,12 @@ int die(const char *str, struct pt_regs *regs, long err) print_modules(); show_regs(regs); bust_spinlocks(0); + spin_unlock_irq(&die_lock); - if (!crash_dump_start && kexec_should_crash(current)) { - crash_dump_start = 1; - spin_unlock_irq(&die_lock); + if (kexec_should_crash(current) || + kexec_sr_activated(smp_processor_id())) crash_kexec(regs); - /* NOTREACHED */ - } - spin_unlock_irq(&die_lock); - if (crash_dump_start) - /* - * Only for soft-reset: Other CPUs will be responded to an IPI - * sent by first kexec CPU. - */ - for(;;) - ; + crash_kexec_secondary(regs); if (in_interrupt()) panic("Fatal exception in interrupt"); @@ -215,6 +210,10 @@ void system_reset_exception(struct pt_regs *regs) return; } +#ifdef CONFIG_KEXEC + cpu_set(smp_processor_id(), cpus_in_sr); +#endif + die("System Reset", regs, SIGABRT); /* Must die if the interrupt is not recoverable */ diff --git a/include/asm-powerpc/kexec.h b/include/asm-powerpc/kexec.h index 234bd684c7f..8f7fd5cfec3 100644 --- a/include/asm-powerpc/kexec.h +++ b/include/asm-powerpc/kexec.h @@ -114,6 +114,11 @@ extern void kexec_smp_wait(void); /* get and clear naca physid, wait for master to copy new code to 0 */ extern int crashing_cpu; extern void crash_send_ipi(void (*crash_ipi_callback)(struct pt_regs *)); +extern cpumask_t cpus_in_sr; +static inline int kexec_sr_activated(int cpu) +{ + return cpu_isset(cpu,cpus_in_sr); +} #endif /* __powerpc64 __ */ struct kimage; @@ -123,10 +128,13 @@ extern int default_machine_kexec_prepare(struct kimage *image); extern void default_machine_crash_shutdown(struct pt_regs *regs); extern void machine_kexec_simple(struct kimage *image); +extern void crash_kexec_secondary(struct pt_regs *regs); extern int overlaps_crashkernel(unsigned long start, unsigned long size); extern void reserve_crashkernel(void); #else /* !CONFIG_KEXEC */ +static inline int kexec_sr_activated(int cpu) { return 0; } +static inline void crash_kexec_secondary(struct pt_regs *regs) { } static inline int overlaps_crashkernel(unsigned long start, unsigned long size) { diff --git a/kernel/kexec.c b/kernel/kexec.c index 58f0f382597..50087ecf337 100644 --- a/kernel/kexec.c +++ b/kernel/kexec.c @@ -1042,7 +1042,6 @@ asmlinkage long compat_sys_kexec_load(unsigned long entry, void crash_kexec(struct pt_regs *regs) { - struct kimage *image; int locked; @@ -1056,12 +1055,11 @@ void crash_kexec(struct pt_regs *regs) */ locked = xchg(&kexec_lock, 1); if (!locked) { - image = xchg(&kexec_crash_image, NULL); - if (image) { + if (kexec_crash_image) { struct pt_regs fixed_regs; crash_setup_regs(&fixed_regs, regs); machine_crash_shutdown(&fixed_regs); - machine_kexec(image); + machine_kexec(kexec_crash_image); } xchg(&kexec_lock, 0); } -- cgit v1.2.3-70-g09d2 From 612f02d6d6eadd9c5dde5a5b97b55a3a620d28e5 Mon Sep 17 00:00:00 2001 From: Stephen Rothwell Date: Wed, 28 Jun 2006 11:49:10 +1000 Subject: [POWERPC] Clean up it_lp_queue.h No more StudlyCaps. Remove from a couple of places it is no longer needed. Use C style comments. Signed-off-by: Stephen Rothwell Signed-off-by: Paul Mackerras --- arch/powerpc/kernel/paca.c | 1 - arch/powerpc/platforms/iseries/lpevents.c | 55 ++++++++++++++++--------------- arch/powerpc/platforms/iseries/proc.c | 1 - include/asm-powerpc/iseries/it_lp_queue.h | 40 +++++++++++----------- 4 files changed, 48 insertions(+), 49 deletions(-) (limited to 'include') diff --git a/arch/powerpc/kernel/paca.c b/arch/powerpc/kernel/paca.c index f505a8827e3..a0bb354c1c0 100644 --- a/arch/powerpc/kernel/paca.c +++ b/arch/powerpc/kernel/paca.c @@ -16,7 +16,6 @@ #include #include #include -#include #include #include diff --git a/arch/powerpc/platforms/iseries/lpevents.c b/arch/powerpc/platforms/iseries/lpevents.c index 8ca7b939635..2a9f81ea27d 100644 --- a/arch/powerpc/platforms/iseries/lpevents.c +++ b/arch/powerpc/platforms/iseries/lpevents.c @@ -51,20 +51,21 @@ static unsigned lpEventHandlerPaths[HvLpEvent_Type_NumTypes]; static struct HvLpEvent * get_next_hvlpevent(void) { struct HvLpEvent * event; - event = (struct HvLpEvent *)hvlpevent_queue.xSlicCurEventPtr; + event = (struct HvLpEvent *)hvlpevent_queue.hq_current_event; if (hvlpevent_is_valid(event)) { /* rmb() needed only for weakly consistent machines (regatta) */ rmb(); /* Set pointer to next potential event */ - hvlpevent_queue.xSlicCurEventPtr += ((event->xSizeMinus1 + - LpEventAlign) / LpEventAlign) * LpEventAlign; + hvlpevent_queue.hq_current_event += ((event->xSizeMinus1 + + IT_LP_EVENT_ALIGN) / IT_LP_EVENT_ALIGN) * + IT_LP_EVENT_ALIGN; /* Wrap to beginning if no room at end */ - if (hvlpevent_queue.xSlicCurEventPtr > - hvlpevent_queue.xSlicLastValidEventPtr) { - hvlpevent_queue.xSlicCurEventPtr = - hvlpevent_queue.xSlicEventStackPtr; + if (hvlpevent_queue.hq_current_event > + hvlpevent_queue.hq_last_event) { + hvlpevent_queue.hq_current_event = + hvlpevent_queue.hq_event_stack; } } else { event = NULL; @@ -82,10 +83,10 @@ int hvlpevent_is_pending(void) if (smp_processor_id() >= spread_lpevents) return 0; - next_event = (struct HvLpEvent *)hvlpevent_queue.xSlicCurEventPtr; + next_event = (struct HvLpEvent *)hvlpevent_queue.hq_current_event; return hvlpevent_is_valid(next_event) || - hvlpevent_queue.xPlicOverflowIntPending; + hvlpevent_queue.hq_overflow_pending; } static void hvlpevent_clear_valid(struct HvLpEvent * event) @@ -95,18 +96,18 @@ static void hvlpevent_clear_valid(struct HvLpEvent * event) * ie. on 64-byte boundaries. */ struct HvLpEvent *tmp; - unsigned extra = ((event->xSizeMinus1 + LpEventAlign) / - LpEventAlign) - 1; + unsigned extra = ((event->xSizeMinus1 + IT_LP_EVENT_ALIGN) / + IT_LP_EVENT_ALIGN) - 1; switch (extra) { case 3: - tmp = (struct HvLpEvent*)((char*)event + 3 * LpEventAlign); + tmp = (struct HvLpEvent*)((char*)event + 3 * IT_LP_EVENT_ALIGN); hvlpevent_invalidate(tmp); case 2: - tmp = (struct HvLpEvent*)((char*)event + 2 * LpEventAlign); + tmp = (struct HvLpEvent*)((char*)event + 2 * IT_LP_EVENT_ALIGN); hvlpevent_invalidate(tmp); case 1: - tmp = (struct HvLpEvent*)((char*)event + 1 * LpEventAlign); + tmp = (struct HvLpEvent*)((char*)event + 1 * IT_LP_EVENT_ALIGN); hvlpevent_invalidate(tmp); } @@ -120,7 +121,7 @@ void process_hvlpevents(struct pt_regs *regs) struct HvLpEvent * event; /* If we have recursed, just return */ - if (!spin_trylock(&hvlpevent_queue.lock)) + if (!spin_trylock(&hvlpevent_queue.hq_lock)) return; for (;;) { @@ -148,17 +149,17 @@ void process_hvlpevents(struct pt_regs *regs) printk(KERN_INFO "Unexpected Lp Event type=%d\n", event->xType ); hvlpevent_clear_valid(event); - } else if (hvlpevent_queue.xPlicOverflowIntPending) + } else if (hvlpevent_queue.hq_overflow_pending) /* * No more valid events. If overflow events are * pending process them */ - HvCallEvent_getOverflowLpEvents(hvlpevent_queue.xIndex); + HvCallEvent_getOverflowLpEvents(hvlpevent_queue.hq_index); else break; } - spin_unlock(&hvlpevent_queue.lock); + spin_unlock(&hvlpevent_queue.hq_lock); } static int set_spread_lpevents(char *str) @@ -184,20 +185,20 @@ void setup_hvlpevent_queue(void) { void *eventStack; - spin_lock_init(&hvlpevent_queue.lock); + spin_lock_init(&hvlpevent_queue.hq_lock); /* Allocate a page for the Event Stack. */ - eventStack = alloc_bootmem_pages(LpEventStackSize); - memset(eventStack, 0, LpEventStackSize); + eventStack = alloc_bootmem_pages(IT_LP_EVENT_STACK_SIZE); + memset(eventStack, 0, IT_LP_EVENT_STACK_SIZE); /* Invoke the hypervisor to initialize the event stack */ - HvCallEvent_setLpEventStack(0, eventStack, LpEventStackSize); + HvCallEvent_setLpEventStack(0, eventStack, IT_LP_EVENT_STACK_SIZE); - hvlpevent_queue.xSlicEventStackPtr = (char *)eventStack; - hvlpevent_queue.xSlicCurEventPtr = (char *)eventStack; - hvlpevent_queue.xSlicLastValidEventPtr = (char *)eventStack + - (LpEventStackSize - LpEventMaxSize); - hvlpevent_queue.xIndex = 0; + hvlpevent_queue.hq_event_stack = eventStack; + hvlpevent_queue.hq_current_event = eventStack; + hvlpevent_queue.hq_last_event = (char *)eventStack + + (IT_LP_EVENT_STACK_SIZE - IT_LP_EVENT_MAX_SIZE); + hvlpevent_queue.hq_index = 0; } /* Register a handler for an LpEvent type */ diff --git a/arch/powerpc/platforms/iseries/proc.c b/arch/powerpc/platforms/iseries/proc.c index e68b6b5fa89..c241413629a 100644 --- a/arch/powerpc/platforms/iseries/proc.c +++ b/arch/powerpc/platforms/iseries/proc.c @@ -24,7 +24,6 @@ #include #include #include -#include #include #include "processor_vpd.h" diff --git a/include/asm-powerpc/iseries/it_lp_queue.h b/include/asm-powerpc/iseries/it_lp_queue.h index b7c6fc12cce..284c5a7db3a 100644 --- a/include/asm-powerpc/iseries/it_lp_queue.h +++ b/include/asm-powerpc/iseries/it_lp_queue.h @@ -29,20 +29,20 @@ struct HvLpEvent; -#define ITMaxLpQueues 8 +#define IT_LP_MAX_QUEUES 8 -#define NotUsed 0 // Queue will not be used by PLIC -#define DedicatedIo 1 // Queue dedicated to IO processor specified -#define DedicatedLp 2 // Queue dedicated to LP specified -#define Shared 3 // Queue shared for both IO and LP +#define IT_LP_NOT_USED 0 /* Queue will not be used by PLIC */ +#define IT_LP_DEDICATED_IO 1 /* Queue dedicated to IO processor specified */ +#define IT_LP_DEDICATED_LP 2 /* Queue dedicated to LP specified */ +#define IT_LP_SHARED 3 /* Queue shared for both IO and LP */ -#define LpEventStackSize 4096 -#define LpEventMaxSize 256 -#define LpEventAlign 64 +#define IT_LP_EVENT_STACK_SIZE 4096 +#define IT_LP_EVENT_MAX_SIZE 256 +#define IT_LP_EVENT_ALIGN 64 struct hvlpevent_queue { /* - * The xSlicCurEventPtr is the pointer to the next event stack entry + * The hq_current_event is the pointer to the next event stack entry * that will become valid. The OS must peek at this entry to determine * if it is valid. PLIC will set the valid indicator as the very last * store into that entry. @@ -52,23 +52,23 @@ struct hvlpevent_queue { * location again. * * If the event stack fills and there are overflow events, then PLIC - * will set the xPlicOverflowIntPending flag in which case the OS will + * will set the hq_overflow_pending flag in which case the OS will * have to fetch the additional LP events once they have drained the * event stack. * * The first 16-bytes are known by both the OS and PLIC. The remainder * of the cache line is for use by the OS. */ - u8 xPlicOverflowIntPending;// 0x00 Overflow events are pending - u8 xPlicStatus; // 0x01 DedicatedIo or DedicatedLp or NotUsed - u16 xSlicLogicalProcIndex; // 0x02 Logical Proc Index for correlation - u8 xPlicRsvd[12]; // 0x04 - char *xSlicCurEventPtr; // 0x10 - char *xSlicLastValidEventPtr; // 0x18 - char *xSlicEventStackPtr; // 0x20 - u8 xIndex; // 0x28 unique sequential index. - u8 xSlicRsvd[3]; // 0x29-2b - spinlock_t lock; + u8 hq_overflow_pending; /* 0x00 Overflow events are pending */ + u8 hq_status; /* 0x01 DedicatedIo or DedicatedLp or NotUsed */ + u16 hq_proc_index; /* 0x02 Logical Proc Index for correlation */ + u8 hq_reserved1[12]; /* 0x04 */ + char *hq_current_event; /* 0x10 */ + char *hq_last_event; /* 0x18 */ + char *hq_event_stack; /* 0x20 */ + u8 hq_index; /* 0x28 unique sequential index. */ + u8 hq_reserved2[3]; /* 0x29-2b */ + spinlock_t hq_lock; }; extern struct hvlpevent_queue hvlpevent_queue; -- cgit v1.2.3-70-g09d2 From 9b47569a9d410199aea6386917fd0662809c60be Mon Sep 17 00:00:00 2001 From: Stephen Rothwell Date: Wed, 28 Jun 2006 11:51:41 +1000 Subject: [POWERPC] update asm-powerpc/time.h If we ever build a combined kernel including iSeries, then this will be needed. Signed-off-by: Stephen Rothwell Signed-off-by: Paul Mackerras --- include/asm-powerpc/time.h | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/include/asm-powerpc/time.h b/include/asm-powerpc/time.h index 4463148c659..dcde4410348 100644 --- a/include/asm-powerpc/time.h +++ b/include/asm-powerpc/time.h @@ -18,8 +18,9 @@ #include #include -#ifdef CONFIG_PPC64 +#ifdef CONFIG_PPC_ISERIES #include +#include #include #endif @@ -177,7 +178,8 @@ static inline void set_dec(int val) #ifdef CONFIG_PPC_ISERIES int cur_dec; - if (get_lppaca()->shared_proc) { + if (firmware_has_feature(FW_FEATURE_ISERIES) && + get_lppaca()->shared_proc) { get_lppaca()->virtual_decr = val; cur_dec = get_dec(); if (cur_dec > val) -- cgit v1.2.3-70-g09d2 From 3965f8c59778b0d37460ec268f92c0c382546b5c Mon Sep 17 00:00:00 2001 From: Paul Mackerras Date: Wed, 28 Jun 2006 13:50:39 +1000 Subject: [POWERPC] Simplify the code defining the 64-bit CPU features Signed-off-by: Paul Mackerras --- include/asm-powerpc/cputable.h | 48 ++++++++++++++++++------------------------ 1 file changed, 20 insertions(+), 28 deletions(-) (limited to 'include') diff --git a/include/asm-powerpc/cputable.h b/include/asm-powerpc/cputable.h index fab41c280aa..1ba3c998361 100644 --- a/include/asm-powerpc/cputable.h +++ b/include/asm-powerpc/cputable.h @@ -117,38 +117,30 @@ extern void do_cpu_ftr_fixups(unsigned long offset); #define CPU_FTR_PPC_LE ASM_CONST(0x0000000000200000) #define CPU_FTR_REAL_LE ASM_CONST(0x0000000000400000) +/* + * Add the 64-bit processor unique features in the top half of the word; + * on 32-bit, make the names available but defined to be 0. + */ #ifdef __powerpc64__ -/* Add the 64b processor unique features in the top half of the word */ -#define CPU_FTR_SLB ASM_CONST(0x0000000100000000) -#define CPU_FTR_16M_PAGE ASM_CONST(0x0000000200000000) -#define CPU_FTR_TLBIEL ASM_CONST(0x0000000400000000) -#define CPU_FTR_NOEXECUTE ASM_CONST(0x0000000800000000) -#define CPU_FTR_IABR ASM_CONST(0x0000002000000000) -#define CPU_FTR_MMCRA ASM_CONST(0x0000004000000000) -#define CPU_FTR_CTRL ASM_CONST(0x0000008000000000) -#define CPU_FTR_SMT ASM_CONST(0x0000010000000000) -#define CPU_FTR_COHERENT_ICACHE ASM_CONST(0x0000020000000000) -#define CPU_FTR_LOCKLESS_TLBIE ASM_CONST(0x0000040000000000) -#define CPU_FTR_CI_LARGE_PAGE ASM_CONST(0x0000100000000000) -#define CPU_FTR_PAUSE_ZERO ASM_CONST(0x0000200000000000) -#define CPU_FTR_PURR ASM_CONST(0x0000400000000000) +#define LONG_ASM_CONST(x) ASM_CONST(x) #else -/* ensure on 32b processors the flags are available for compiling but - * don't do anything */ -#define CPU_FTR_SLB ASM_CONST(0x0) -#define CPU_FTR_16M_PAGE ASM_CONST(0x0) -#define CPU_FTR_TLBIEL ASM_CONST(0x0) -#define CPU_FTR_NOEXECUTE ASM_CONST(0x0) -#define CPU_FTR_IABR ASM_CONST(0x0) -#define CPU_FTR_MMCRA ASM_CONST(0x0) -#define CPU_FTR_CTRL ASM_CONST(0x0) -#define CPU_FTR_SMT ASM_CONST(0x0) -#define CPU_FTR_COHERENT_ICACHE ASM_CONST(0x0) -#define CPU_FTR_LOCKLESS_TLBIE ASM_CONST(0x0) -#define CPU_FTR_CI_LARGE_PAGE ASM_CONST(0x0) -#define CPU_FTR_PURR ASM_CONST(0x0) +#define LONG_ASM_CONST(x) 0 #endif +#define CPU_FTR_SLB LONG_ASM_CONST(0x0000000100000000) +#define CPU_FTR_16M_PAGE LONG_ASM_CONST(0x0000000200000000) +#define CPU_FTR_TLBIEL LONG_ASM_CONST(0x0000000400000000) +#define CPU_FTR_NOEXECUTE LONG_ASM_CONST(0x0000000800000000) +#define CPU_FTR_IABR LONG_ASM_CONST(0x0000002000000000) +#define CPU_FTR_MMCRA LONG_ASM_CONST(0x0000004000000000) +#define CPU_FTR_CTRL LONG_ASM_CONST(0x0000008000000000) +#define CPU_FTR_SMT LONG_ASM_CONST(0x0000010000000000) +#define CPU_FTR_COHERENT_ICACHE LONG_ASM_CONST(0x0000020000000000) +#define CPU_FTR_LOCKLESS_TLBIE LONG_ASM_CONST(0x0000040000000000) +#define CPU_FTR_CI_LARGE_PAGE LONG_ASM_CONST(0x0000100000000000) +#define CPU_FTR_PAUSE_ZERO LONG_ASM_CONST(0x0000200000000000) +#define CPU_FTR_PURR LONG_ASM_CONST(0x0000400000000000) + #ifndef __ASSEMBLY__ #define CPU_FTR_PPCAS_ARCH_V2_BASE (CPU_FTR_SLB | \ -- cgit v1.2.3-70-g09d2 From 9ad494f62444ee37209e85173377c67612e66ef1 Mon Sep 17 00:00:00 2001 From: Kumar Gala Date: Wed, 28 Jun 2006 00:37:45 -0500 Subject: powerpc: minor cleanups for mpc86xx * Remove duplicated cputable entry for 8641 (matches w/7448) * Removed __init from function prototypes in mpc86xx.h * Moved pci fixups into board specific code * Moved mpc86xx_exclude_device to generic mpc86xx pci code * Fixed sparse warnings in mpc86xx_smp.c * Removed board specific header include from asm-powerpc/mpc86xx.h Signed-off-by: Kumar Gala --- arch/powerpc/kernel/cputable.c | 12 --- arch/powerpc/platforms/86xx/mpc86xx.h | 8 +- arch/powerpc/platforms/86xx/mpc86xx_hpcn.c | 128 +++++++++++++++++++++++++-- arch/powerpc/platforms/86xx/mpc86xx_smp.c | 9 +- arch/powerpc/platforms/86xx/pci.c | 136 ++--------------------------- include/asm-powerpc/mpc86xx.h | 4 - 6 files changed, 138 insertions(+), 159 deletions(-) (limited to 'include') diff --git a/arch/powerpc/kernel/cputable.c b/arch/powerpc/kernel/cputable.c index 1c114880dc0..abf7d42a8b0 100644 --- a/arch/powerpc/kernel/cputable.c +++ b/arch/powerpc/kernel/cputable.c @@ -722,18 +722,6 @@ struct cpu_spec cpu_specs[] = { .oprofile_type = PPC_OPROFILE_G4, .platform = "ppc7450", }, - { /* 8641 */ - .pvr_mask = 0xffffffff, - .pvr_value = 0x80040010, - .cpu_name = "8641", - .cpu_features = CPU_FTRS_7447A, - .cpu_user_features = COMMON_USER | PPC_FEATURE_HAS_ALTIVEC_COMP, - .icache_bsize = 32, - .dcache_bsize = 32, - .num_pmcs = 6, - .cpu_setup = __setup_cpu_745x - }, - { /* 82xx (8240, 8245, 8260 are all 603e cores) */ .pvr_mask = 0x7fff0000, .pvr_value = 0x00810000, diff --git a/arch/powerpc/platforms/86xx/mpc86xx.h b/arch/powerpc/platforms/86xx/mpc86xx.h index e3c9e4f417d..2834462590b 100644 --- a/arch/powerpc/platforms/86xx/mpc86xx.h +++ b/arch/powerpc/platforms/86xx/mpc86xx.h @@ -15,11 +15,13 @@ * mpc86xx_* files. Mostly for use by mpc86xx_setup(). */ -extern int __init add_bridge(struct device_node *dev); +extern int add_bridge(struct device_node *dev); -extern void __init setup_indirect_pcie(struct pci_controller *hose, +extern int mpc86xx_exclude_device(u_char bus, u_char devfn); + +extern void setup_indirect_pcie(struct pci_controller *hose, u32 cfg_addr, u32 cfg_data); -extern void __init setup_indirect_pcie_nomap(struct pci_controller *hose, +extern void setup_indirect_pcie_nomap(struct pci_controller *hose, void __iomem *cfg_addr, void __iomem *cfg_data); diff --git a/arch/powerpc/platforms/86xx/mpc86xx_hpcn.c b/arch/powerpc/platforms/86xx/mpc86xx_hpcn.c index 483c21df181..ac7f41802e9 100644 --- a/arch/powerpc/platforms/86xx/mpc86xx_hpcn.c +++ b/arch/powerpc/platforms/86xx/mpc86xx_hpcn.c @@ -36,6 +36,7 @@ #include #include "mpc86xx.h" +#include "mpc8641_hpcn.h" #ifndef CONFIG_PCI unsigned long isa_io_base = 0; @@ -186,17 +187,130 @@ mpc86xx_map_irq(struct pci_dev *dev, unsigned char idsel, unsigned char pin) return PCI_IRQ_TABLE_LOOKUP + I8259_OFFSET; } +static void __devinit quirk_ali1575(struct pci_dev *dev) +{ + unsigned short temp; + + /* + * ALI1575 interrupts route table setup: + * + * IRQ pin IRQ# + * PIRQA ---- 3 + * PIRQB ---- 4 + * PIRQC ---- 5 + * PIRQD ---- 6 + * PIRQE ---- 9 + * PIRQF ---- 10 + * PIRQG ---- 11 + * PIRQH ---- 12 + * + * interrupts for PCI slot0 -- PIRQA / PIRQB / PIRQC / PIRQD + * PCI slot1 -- PIRQB / PIRQC / PIRQD / PIRQA + */ + pci_write_config_dword(dev, 0x48, 0xb9317542); + + /* USB 1.1 OHCI controller 1, interrupt: PIRQE */ + pci_write_config_byte(dev, 0x86, 0x0c); + + /* USB 1.1 OHCI controller 2, interrupt: PIRQF */ + pci_write_config_byte(dev, 0x87, 0x0d); + + /* USB 1.1 OHCI controller 3, interrupt: PIRQH */ + pci_write_config_byte(dev, 0x88, 0x0f); + + /* USB 2.0 controller, interrupt: PIRQ7 */ + pci_write_config_byte(dev, 0x74, 0x06); + + /* Audio controller, interrupt: PIRQE */ + pci_write_config_byte(dev, 0x8a, 0x0c); + + /* Modem controller, interrupt: PIRQF */ + pci_write_config_byte(dev, 0x8b, 0x0d); + + /* HD audio controller, interrupt: PIRQG */ + pci_write_config_byte(dev, 0x8c, 0x0e); + + /* Serial ATA interrupt: PIRQD */ + pci_write_config_byte(dev, 0x8d, 0x0b); + + /* SMB interrupt: PIRQH */ + pci_write_config_byte(dev, 0x8e, 0x0f); + + /* PMU ACPI SCI interrupt: PIRQH */ + pci_write_config_byte(dev, 0x8f, 0x0f); + + /* Primary PATA IDE IRQ: 14 + * Secondary PATA IDE IRQ: 15 + */ + pci_write_config_byte(dev, 0x44, 0x3d); + pci_write_config_byte(dev, 0x75, 0x0f); + + /* Set IRQ14 and IRQ15 to legacy IRQs */ + pci_read_config_word(dev, 0x46, &temp); + temp |= 0xc000; + pci_write_config_word(dev, 0x46, temp); + + /* Set i8259 interrupt trigger + * IRQ 3: Level + * IRQ 4: Level + * IRQ 5: Level + * IRQ 6: Level + * IRQ 7: Level + * IRQ 9: Level + * IRQ 10: Level + * IRQ 11: Level + * IRQ 12: Level + * IRQ 14: Edge + * IRQ 15: Edge + */ + outb(0xfa, 0x4d0); + outb(0x1e, 0x4d1); +} -int -mpc86xx_exclude_device(u_char bus, u_char devfn) +static void __devinit quirk_uli5288(struct pci_dev *dev) { -#if !defined(CONFIG_PCI) - if (bus == 0 && PCI_SLOT(devfn) == 0) - return PCIBIOS_DEVICE_NOT_FOUND; -#endif + unsigned char c; + + pci_read_config_byte(dev,0x83,&c); + c |= 0x80; + pci_write_config_byte(dev, 0x83, c); + + pci_write_config_byte(dev, 0x09, 0x01); + pci_write_config_byte(dev, 0x0a, 0x06); + + pci_read_config_byte(dev,0x83,&c); + c &= 0x7f; + pci_write_config_byte(dev, 0x83, c); - return PCIBIOS_SUCCESSFUL; + pci_read_config_byte(dev,0x84,&c); + c |= 0x01; + pci_write_config_byte(dev, 0x84, c); } + +static void __devinit quirk_uli5229(struct pci_dev *dev) +{ + unsigned short temp; + pci_write_config_word(dev, 0x04, 0x0405); + pci_read_config_word(dev, 0x4a, &temp); + temp |= 0x1000; + pci_write_config_word(dev, 0x4a, temp); +} + +static void __devinit early_uli5249(struct pci_dev *dev) +{ + unsigned char temp; + pci_write_config_word(dev, 0x04, 0x0007); + pci_read_config_byte(dev, 0x7c, &temp); + pci_write_config_byte(dev, 0x7c, 0x80); + pci_write_config_byte(dev, 0x09, 0x01); + pci_write_config_byte(dev, 0x7c, temp); + dev->class |= 0x1; +} + +DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_AL, 0x1575, quirk_ali1575); +DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_AL, 0x5288, quirk_uli5288); +DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_AL, 0x5229, quirk_uli5229); +DECLARE_PCI_FIXUP_EARLY(PCI_VENDOR_ID_AL, 0x5249, early_uli5249); #endif /* CONFIG_PCI */ diff --git a/arch/powerpc/platforms/86xx/mpc86xx_smp.c b/arch/powerpc/platforms/86xx/mpc86xx_smp.c index 944ec4b7141..9cca3d1a67f 100644 --- a/arch/powerpc/platforms/86xx/mpc86xx_smp.c +++ b/arch/powerpc/platforms/86xx/mpc86xx_smp.c @@ -34,8 +34,8 @@ extern unsigned long __secondary_hold_acknowledge; static void __init smp_86xx_release_core(int nr) { - void *mcm_vaddr; - unsigned long vaddr, pcr; + __be32 __iomem *mcm_vaddr; + unsigned long pcr; if (nr < 0 || nr >= NR_CPUS) return; @@ -45,10 +45,9 @@ smp_86xx_release_core(int nr) */ mcm_vaddr = ioremap(get_immrbase() + MPC86xx_MCM_OFFSET, MPC86xx_MCM_SIZE); - vaddr = (unsigned long)mcm_vaddr + MCM_PORT_CONFIG_OFFSET; - pcr = in_be32((volatile unsigned *)vaddr); + pcr = in_be32(mcm_vaddr + (MCM_PORT_CONFIG_OFFSET >> 2)); pcr |= 1 << (nr + 24); - out_be32((volatile unsigned *)vaddr, pcr); + out_be32(mcm_vaddr + (MCM_PORT_CONFIG_OFFSET >> 2), pcr); } diff --git a/arch/powerpc/platforms/86xx/pci.c b/arch/powerpc/platforms/86xx/pci.c index 5180df7c75b..0d8b34089b9 100644 --- a/arch/powerpc/platforms/86xx/pci.c +++ b/arch/powerpc/platforms/86xx/pci.c @@ -122,15 +122,12 @@ static void __init setup_pcie_atmu(struct pci_controller *hose, struct resource static void __init mpc86xx_setup_pcie(struct pci_controller *hose, u32 pcie_offset, u32 pcie_size) { - volatile struct ccsr_pex *pcie; u16 cmd; unsigned int temps; DBG("PCIE host controller register offset 0x%08x, size 0x%08x.\n", pcie_offset, pcie_size); - pcie = ioremap(pcie_offset, pcie_size); - early_read_config_word(hose, 0, 0, PCI_COMMAND, &cmd); cmd |= PCI_COMMAND_SERR | PCI_COMMAND_MASTER | PCI_COMMAND_MEMORY | PCI_COMMAND_IO; @@ -144,6 +141,14 @@ mpc86xx_setup_pcie(struct pci_controller *hose, u32 pcie_offset, u32 pcie_size) early_write_config_dword(hose, 0, 0, PCI_PRIMARY_BUS, temps); } +int mpc86xx_exclude_device(u_char bus, u_char devfn) +{ + if (bus == 0 && PCI_SLOT(devfn) == 0) + return PCIBIOS_DEVICE_NOT_FOUND; + + return PCIBIOS_SUCCESSFUL; +} + int __init add_bridge(struct device_node *dev) { int len; @@ -198,128 +203,3 @@ int __init add_bridge(struct device_node *dev) return 0; } - -static void __devinit quirk_ali1575(struct pci_dev *dev) -{ - unsigned short temp; - - /* - * ALI1575 interrupts route table setup: - * - * IRQ pin IRQ# - * PIRQA ---- 3 - * PIRQB ---- 4 - * PIRQC ---- 5 - * PIRQD ---- 6 - * PIRQE ---- 9 - * PIRQF ---- 10 - * PIRQG ---- 11 - * PIRQH ---- 12 - * - * interrupts for PCI slot0 -- PIRQA / PIRQB / PIRQC / PIRQD - * PCI slot1 -- PIRQB / PIRQC / PIRQD / PIRQA - */ - pci_write_config_dword(dev, 0x48, 0xb9317542); - - /* USB 1.1 OHCI controller 1, interrupt: PIRQE */ - pci_write_config_byte(dev, 0x86, 0x0c); - - /* USB 1.1 OHCI controller 2, interrupt: PIRQF */ - pci_write_config_byte(dev, 0x87, 0x0d); - - /* USB 1.1 OHCI controller 3, interrupt: PIRQH */ - pci_write_config_byte(dev, 0x88, 0x0f); - - /* USB 2.0 controller, interrupt: PIRQ7 */ - pci_write_config_byte(dev, 0x74, 0x06); - - /* Audio controller, interrupt: PIRQE */ - pci_write_config_byte(dev, 0x8a, 0x0c); - - /* Modem controller, interrupt: PIRQF */ - pci_write_config_byte(dev, 0x8b, 0x0d); - - /* HD audio controller, interrupt: PIRQG */ - pci_write_config_byte(dev, 0x8c, 0x0e); - - /* Serial ATA interrupt: PIRQD */ - pci_write_config_byte(dev, 0x8d, 0x0b); - - /* SMB interrupt: PIRQH */ - pci_write_config_byte(dev, 0x8e, 0x0f); - - /* PMU ACPI SCI interrupt: PIRQH */ - pci_write_config_byte(dev, 0x8f, 0x0f); - - /* Primary PATA IDE IRQ: 14 - * Secondary PATA IDE IRQ: 15 - */ - pci_write_config_byte(dev, 0x44, 0x3d); - pci_write_config_byte(dev, 0x75, 0x0f); - - /* Set IRQ14 and IRQ15 to legacy IRQs */ - pci_read_config_word(dev, 0x46, &temp); - temp |= 0xc000; - pci_write_config_word(dev, 0x46, temp); - - /* Set i8259 interrupt trigger - * IRQ 3: Level - * IRQ 4: Level - * IRQ 5: Level - * IRQ 6: Level - * IRQ 7: Level - * IRQ 9: Level - * IRQ 10: Level - * IRQ 11: Level - * IRQ 12: Level - * IRQ 14: Edge - * IRQ 15: Edge - */ - outb(0xfa, 0x4d0); - outb(0x1e, 0x4d1); -} - -static void __devinit quirk_uli5288(struct pci_dev *dev) -{ - unsigned char c; - - pci_read_config_byte(dev,0x83,&c); - c |= 0x80; - pci_write_config_byte(dev, 0x83, c); - - pci_write_config_byte(dev, 0x09, 0x01); - pci_write_config_byte(dev, 0x0a, 0x06); - - pci_read_config_byte(dev,0x83,&c); - c &= 0x7f; - pci_write_config_byte(dev, 0x83, c); - - pci_read_config_byte(dev,0x84,&c); - c |= 0x01; - pci_write_config_byte(dev, 0x84, c); -} - -static void __devinit quirk_uli5229(struct pci_dev *dev) -{ - unsigned short temp; - pci_write_config_word(dev, 0x04, 0x0405); - pci_read_config_word(dev, 0x4a, &temp); - temp |= 0x1000; - pci_write_config_word(dev, 0x4a, temp); -} - -static void __devinit early_uli5249(struct pci_dev *dev) -{ - unsigned char temp; - pci_write_config_word(dev, 0x04, 0x0007); - pci_read_config_byte(dev, 0x7c, &temp); - pci_write_config_byte(dev, 0x7c, 0x80); - pci_write_config_byte(dev, 0x09, 0x01); - pci_write_config_byte(dev, 0x7c, temp); - dev->class |= 0x1; -} - -DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_AL, 0x1575, quirk_ali1575); -DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_AL, 0x5288, quirk_uli5288); -DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_AL, 0x5229, quirk_uli5229); -DECLARE_PCI_FIXUP_EARLY(PCI_VENDOR_ID_AL, 0x5249, early_uli5249); diff --git a/include/asm-powerpc/mpc86xx.h b/include/asm-powerpc/mpc86xx.h index d0a6718d188..00d72a7284e 100644 --- a/include/asm-powerpc/mpc86xx.h +++ b/include/asm-powerpc/mpc86xx.h @@ -20,10 +20,6 @@ #ifdef CONFIG_PPC_86xx -#ifdef CONFIG_MPC8641_HPCN -#include -#endif - #define _IO_BASE isa_io_base #define _ISA_MEM_BASE isa_mem_base #ifdef CONFIG_PCI -- cgit v1.2.3-70-g09d2 From 02fcb974369d21c3df61d4ac5a2e1a58dce86495 Mon Sep 17 00:00:00 2001 From: Russell King Date: Wed, 21 Jun 2006 14:44:52 +0100 Subject: [ARM] Remove the __arch_* layer from uaccess.h Back in the days when we had armo (26-bit) and armv (32-bit) combined, we had an additional layer to the uaccess macros to ensure correct typing. Since we no longer have 26-bit in this tree, we no longer need this layer, so eliminate it. Signed-off-by: Russell King --- arch/arm/kernel/armksyms.c | 10 ++++----- arch/arm/lib/clear_user.S | 4 ++-- arch/arm/lib/copy_from_user.S | 4 ++-- arch/arm/lib/copy_to_user.S | 4 ++-- arch/arm/lib/strncpy_from_user.S | 2 +- arch/arm/lib/strnlen_user.S | 4 ++-- arch/arm/lib/uaccess.S | 8 +++---- include/asm-arm/uaccess.h | 45 ++++++++++++---------------------------- 8 files changed, 31 insertions(+), 50 deletions(-) (limited to 'include') diff --git a/arch/arm/kernel/armksyms.c b/arch/arm/kernel/armksyms.c index c49b5d4d7fc..f8bb7abd3e9 100644 --- a/arch/arm/kernel/armksyms.c +++ b/arch/arm/kernel/armksyms.c @@ -109,11 +109,11 @@ EXPORT_SYMBOL(memchr); EXPORT_SYMBOL(__memzero); /* user mem (segment) */ -EXPORT_SYMBOL(__arch_copy_from_user); -EXPORT_SYMBOL(__arch_copy_to_user); -EXPORT_SYMBOL(__arch_clear_user); -EXPORT_SYMBOL(__arch_strnlen_user); -EXPORT_SYMBOL(__arch_strncpy_from_user); +EXPORT_SYMBOL(__copy_from_user); +EXPORT_SYMBOL(__copy_to_user); +EXPORT_SYMBOL(__clear_user); +EXPORT_SYMBOL(__strnlen_user); +EXPORT_SYMBOL(__strncpy_from_user); EXPORT_SYMBOL(__get_user_1); EXPORT_SYMBOL(__get_user_2); diff --git a/arch/arm/lib/clear_user.S b/arch/arm/lib/clear_user.S index ea435ae2e4a..ecb28dcdaf7 100644 --- a/arch/arm/lib/clear_user.S +++ b/arch/arm/lib/clear_user.S @@ -12,13 +12,13 @@ .text -/* Prototype: int __arch_clear_user(void *addr, size_t sz) +/* Prototype: int __clear_user(void *addr, size_t sz) * Purpose : clear some user memory * Params : addr - user memory address to clear * : sz - number of bytes to clear * Returns : number of bytes NOT cleared */ -ENTRY(__arch_clear_user) +ENTRY(__clear_user) stmfd sp!, {r1, lr} mov r2, #0 cmp r1, #4 diff --git a/arch/arm/lib/copy_from_user.S b/arch/arm/lib/copy_from_user.S index 7497393a0e8..6b7363ce749 100644 --- a/arch/arm/lib/copy_from_user.S +++ b/arch/arm/lib/copy_from_user.S @@ -16,7 +16,7 @@ /* * Prototype: * - * size_t __arch_copy_from_user(void *to, const void *from, size_t n) + * size_t __copy_from_user(void *to, const void *from, size_t n) * * Purpose: * @@ -83,7 +83,7 @@ .text -ENTRY(__arch_copy_from_user) +ENTRY(__copy_from_user) #include "copy_template.S" diff --git a/arch/arm/lib/copy_to_user.S b/arch/arm/lib/copy_to_user.S index 4a6d8ea1402..5224d94688d 100644 --- a/arch/arm/lib/copy_to_user.S +++ b/arch/arm/lib/copy_to_user.S @@ -16,7 +16,7 @@ /* * Prototype: * - * size_t __arch_copy_to_user(void *to, const void *from, size_t n) + * size_t __copy_to_user(void *to, const void *from, size_t n) * * Purpose: * @@ -86,7 +86,7 @@ .text -ENTRY(__arch_copy_to_user) +ENTRY(__copy_to_user) #include "copy_template.S" diff --git a/arch/arm/lib/strncpy_from_user.S b/arch/arm/lib/strncpy_from_user.S index 35649f04fca..36e3741a377 100644 --- a/arch/arm/lib/strncpy_from_user.S +++ b/arch/arm/lib/strncpy_from_user.S @@ -20,7 +20,7 @@ * returns the number of characters copied (strlen of copied string), * -EFAULT on exception, or "len" if we fill the whole buffer */ -ENTRY(__arch_strncpy_from_user) +ENTRY(__strncpy_from_user) mov ip, r1 1: subs r2, r2, #1 USER( ldrplbt r3, [r1], #1) diff --git a/arch/arm/lib/strnlen_user.S b/arch/arm/lib/strnlen_user.S index 3668a15991e..18d8fa4f925 100644 --- a/arch/arm/lib/strnlen_user.S +++ b/arch/arm/lib/strnlen_user.S @@ -14,13 +14,13 @@ .text .align 5 -/* Prototype: unsigned long __arch_strnlen_user(const char *str, long n) +/* Prototype: unsigned long __strnlen_user(const char *str, long n) * Purpose : get length of a string in user memory * Params : str - address of string in user memory * Returns : length of string *including terminator* * or zero on exception, or n + 1 if too long */ -ENTRY(__arch_strnlen_user) +ENTRY(__strnlen_user) mov r2, r0 1: USER( ldrbt r3, [r0], #1) diff --git a/arch/arm/lib/uaccess.S b/arch/arm/lib/uaccess.S index 1f1545d737b..b48bd6d5fd8 100644 --- a/arch/arm/lib/uaccess.S +++ b/arch/arm/lib/uaccess.S @@ -19,7 +19,7 @@ #define PAGE_SHIFT 12 -/* Prototype: int __arch_copy_to_user(void *to, const char *from, size_t n) +/* Prototype: int __copy_to_user(void *to, const char *from, size_t n) * Purpose : copy a block to user memory from kernel memory * Params : to - user memory * : from - kernel memory @@ -39,7 +39,7 @@ USER( strgtbt r3, [r0], #1) @ May fault sub r2, r2, ip b .Lc2u_dest_aligned -ENTRY(__arch_copy_to_user) +ENTRY(__copy_to_user) stmfd sp!, {r2, r4 - r7, lr} cmp r2, #4 blt .Lc2u_not_enough @@ -283,7 +283,7 @@ USER( strgtbt r3, [r0], #1) @ May fault 9001: ldmfd sp!, {r0, r4 - r7, pc} .previous -/* Prototype: unsigned long __arch_copy_from_user(void *to,const void *from,unsigned long n); +/* Prototype: unsigned long __copy_from_user(void *to,const void *from,unsigned long n); * Purpose : copy a block from user memory to kernel memory * Params : to - kernel memory * : from - user memory @@ -302,7 +302,7 @@ USER( ldrgtbt r3, [r1], #1) @ May fault sub r2, r2, ip b .Lcfu_dest_aligned -ENTRY(__arch_copy_from_user) +ENTRY(__copy_from_user) stmfd sp!, {r0, r2, r4 - r7, lr} cmp r2, #4 blt .Lcfu_not_enough diff --git a/include/asm-arm/uaccess.h b/include/asm-arm/uaccess.h index 064f0f5e8e2..f909dc75301 100644 --- a/include/asm-arm/uaccess.h +++ b/include/asm-arm/uaccess.h @@ -353,66 +353,47 @@ do { \ : "r" (x), "i" (-EFAULT) \ : "cc") -extern unsigned long __arch_copy_from_user(void *to, const void __user *from, unsigned long n); -extern unsigned long __arch_copy_to_user(void __user *to, const void *from, unsigned long n); -extern unsigned long __arch_clear_user(void __user *addr, unsigned long n); -extern unsigned long __arch_strncpy_from_user(char *to, const char __user *from, unsigned long count); -extern unsigned long __arch_strnlen_user(const char __user *s, long n); + +extern unsigned long __copy_from_user(void *to, const void __user *from, unsigned long n); +extern unsigned long __copy_to_user(void __user *to, const void *from, unsigned long n); +extern unsigned long __clear_user(void __user *addr, unsigned long n); +extern unsigned long __strncpy_from_user(char *to, const char __user *from, unsigned long count); +extern unsigned long __strnlen_user(const char __user *s, long n); static inline unsigned long copy_from_user(void *to, const void __user *from, unsigned long n) { if (access_ok(VERIFY_READ, from, n)) - n = __arch_copy_from_user(to, from, n); + n = __copy_from_user(to, from, n); else /* security hole - plug it */ memzero(to, n); return n; } -static inline unsigned long __copy_from_user(void *to, const void __user *from, unsigned long n) -{ - return __arch_copy_from_user(to, from, n); -} - static inline unsigned long copy_to_user(void __user *to, const void *from, unsigned long n) { if (access_ok(VERIFY_WRITE, to, n)) - n = __arch_copy_to_user(to, from, n); + n = __copy_to_user(to, from, n); return n; } -static inline unsigned long __copy_to_user(void __user *to, const void *from, unsigned long n) -{ - return __arch_copy_to_user(to, from, n); -} - #define __copy_to_user_inatomic __copy_to_user #define __copy_from_user_inatomic __copy_from_user -static inline unsigned long clear_user (void __user *to, unsigned long n) +static inline unsigned long clear_user(void __user *to, unsigned long n) { if (access_ok(VERIFY_WRITE, to, n)) - n = __arch_clear_user(to, n); + n = __clear_user(to, n); return n; } -static inline unsigned long __clear_user (void __user *to, unsigned long n) -{ - return __arch_clear_user(to, n); -} - -static inline long strncpy_from_user (char *dst, const char __user *src, long count) +static inline long strncpy_from_user(char *dst, const char __user *src, long count) { long res = -EFAULT; if (access_ok(VERIFY_READ, src, 1)) - res = __arch_strncpy_from_user(dst, src, count); + res = __strncpy_from_user(dst, src, count); return res; } -static inline long __strncpy_from_user (char *dst, const char __user *src, long count) -{ - return __arch_strncpy_from_user(dst, src, count); -} - #define strlen_user(s) strnlen_user(s, ~0UL >> 1) static inline long strnlen_user(const char __user *s, long n) @@ -420,7 +401,7 @@ static inline long strnlen_user(const char __user *s, long n) unsigned long res = 0; if (__addr_ok(s)) - res = __arch_strnlen_user(s, n); + res = __strnlen_user(s, n); return res; } -- cgit v1.2.3-70-g09d2 From 2b8b4913efa20f76718f524accf40e6d02a3bf0a Mon Sep 17 00:00:00 2001 From: Ben Dooks Date: Tue, 27 Jun 2006 22:53:04 +0100 Subject: [ARM] 3661/1: S3C2412: Fix compilation if CPU_S3C2410 only Patch from Ben Dooks If only the S3C2412 based machines are selected, then the regs-dsc.h does not export the S3C2412_DSC registers as it is wrapped in CONFIG_CPU_S3C2440. Signed-off-by: Ben Dooks Signed-off-by: Russell King --- include/asm-arm/arch-s3c2410/regs-dsc.h | 16 +++++++--------- 1 file changed, 7 insertions(+), 9 deletions(-) (limited to 'include') diff --git a/include/asm-arm/arch-s3c2410/regs-dsc.h b/include/asm-arm/arch-s3c2410/regs-dsc.h index 84aca61cbaa..a0a12487516 100644 --- a/include/asm-arm/arch-s3c2410/regs-dsc.h +++ b/include/asm-arm/arch-s3c2410/regs-dsc.h @@ -7,25 +7,23 @@ * it under the terms of the GNU General Public License version 2 as * published by the Free Software Foundation. * - * S3C2440 Signal Drive Strength Control - * - * Changelog: - * 11-Aug-2004 BJD Created file - * 25-Aug-2004 BJD Added the _SELECT_* defs for using with functions + * S3C2440/S3C2412 Signal Drive Strength Control */ #ifndef __ASM_ARCH_REGS_DSC_H #define __ASM_ARCH_REGS_DSC_H "2440-dsc" -#ifdef CONFIG_CPU_S3C2440 +#if defined(CONFIG_CPU_S3C2412) +#define S3C2412_DSC0 S3C2410_GPIOREG(0xdc) +#define S3C2412_DSC1 S3C2410_GPIOREG(0xe0) +#endif + +#if defined(CONFIG_CPU_S3C2440) #define S3C2440_DSC0 S3C2410_GPIOREG(0xc4) #define S3C2440_DSC1 S3C2410_GPIOREG(0xc8) -#define S3C2412_DSC0 S3C2410_GPIOREG(0xdc) -#define S3C2412_DSC1 S3C2410_GPIOREG(0xe0) - #define S3C2440_SELECT_DSC0 (0) #define S3C2440_SELECT_DSC1 (1<<31) -- cgit v1.2.3-70-g09d2 From 313f53e3823c00bd0aa3601f01897573f2fc326d Mon Sep 17 00:00:00 2001 From: Andrew Victor Date: Tue, 27 Jun 2006 22:54:39 +0100 Subject: [ARM] 3660/1: Remove legacy defines Patch from Andrew Victor Remove the remaining legacy __virt_to_bus__is_a_macro and __bus_to_virt__is_a_macro defines in some ARM platforms. Signed-off-by: Andrew Victor Signed-off-by: Russell King --- include/asm-arm/arch-at91rm9200/memory.h | 2 -- include/asm-arm/arch-h720x/memory.h | 2 -- include/asm-arm/arch-imx/memory.h | 6 ++---- 3 files changed, 2 insertions(+), 8 deletions(-) (limited to 'include') diff --git a/include/asm-arm/arch-at91rm9200/memory.h b/include/asm-arm/arch-at91rm9200/memory.h index 3c327c40437..f985069e6d0 100644 --- a/include/asm-arm/arch-at91rm9200/memory.h +++ b/include/asm-arm/arch-at91rm9200/memory.h @@ -33,9 +33,7 @@ * bus_to_virt: Used to convert an address for DMA operations * to an address that the kernel can use. */ -#define __virt_to_bus__is_a_macro #define __virt_to_bus(x) __virt_to_phys(x) -#define __bus_to_virt__is_a_macro #define __bus_to_virt(x) __phys_to_virt(x) #endif diff --git a/include/asm-arm/arch-h720x/memory.h b/include/asm-arm/arch-h720x/memory.h index 4a1bfd78a0f..53e923dba76 100644 --- a/include/asm-arm/arch-h720x/memory.h +++ b/include/asm-arm/arch-h720x/memory.h @@ -23,9 +23,7 @@ * There is something to do here later !, Mar 2000, Jungjun Kim */ -#define __virt_to_bus__is_a_macro #define __virt_to_bus(x) __virt_to_phys(x) -#define __bus_to_virt__is_a_macro #define __bus_to_virt(x) __phys_to_virt(x) #endif diff --git a/include/asm-arm/arch-imx/memory.h b/include/asm-arm/arch-imx/memory.h index d09ae32cd2f..5ad90127915 100644 --- a/include/asm-arm/arch-imx/memory.h +++ b/include/asm-arm/arch-imx/memory.h @@ -30,9 +30,7 @@ * bus_to_virt: Used to convert an address for DMA operations * to an address that the kernel can use. */ -#define __virt_to_bus__is_a_macro -#define __virt_to_bus(x) (x - PAGE_OFFSET + PHYS_OFFSET) -#define __bus_to_virt__is_a_macro -#define __bus_to_virt(x) (x - PHYS_OFFSET + PAGE_OFFSET) +#define __virt_to_bus(x) (x - PAGE_OFFSET + PHYS_OFFSET) +#define __bus_to_virt(x) (x - PHYS_OFFSET + PAGE_OFFSET) #endif -- cgit v1.2.3-70-g09d2 From 02c4293194432c7466efe9477126d6179d6fda07 Mon Sep 17 00:00:00 2001 From: Lennert Buytenhek Date: Tue, 27 Jun 2006 22:56:16 +0100 Subject: [ARM] 3662/1: ixp23xx: don't include asm/hardware.h in uncompress.h Patch from Lennert Buytenhek ixp23xx was including asm/hardware.h in its version of uncompress.h, to get at the physical address of the debug UART, but this include was causing various inline functions that are totally unrelated to the decompressor, defined in headers in include/asm-arm/arch-ixp23xx, to be included in the decompressor image. Include asm/arch/ixp23xx.h instead, and move the sole inline function in ixp23xx.h to another header. Signed-off-by: Lennert Buytenhek Signed-off-by: Russell King --- include/asm-arm/arch-ixp23xx/ixp23xx.h | 11 ----------- include/asm-arm/arch-ixp23xx/platform.h | 10 ++++++++++ include/asm-arm/arch-ixp23xx/uncompress.h | 2 +- 3 files changed, 11 insertions(+), 12 deletions(-) (limited to 'include') diff --git a/include/asm-arm/arch-ixp23xx/ixp23xx.h b/include/asm-arm/arch-ixp23xx/ixp23xx.h index d0a72201ee9..3927b1d61b1 100644 --- a/include/asm-arm/arch-ixp23xx/ixp23xx.h +++ b/include/asm-arm/arch-ixp23xx/ixp23xx.h @@ -295,15 +295,4 @@ #define IXP23XX_PCI_CPP_ADDR_BITS IXP23XX_PCI_CSR(0x0160) -#ifndef __ASSEMBLY__ -/* - * Is system memory on the XSI or CPP bus? - */ -static inline unsigned ixp23xx_cpp_boot(void) -{ - return (*IXP23XX_EXP_CFG0 & IXP23XX_EXP_CFG0_XSI_NOT_PRES); -} -#endif - - #endif diff --git a/include/asm-arm/arch-ixp23xx/platform.h b/include/asm-arm/arch-ixp23xx/platform.h index 19a73b39c86..56e16d66645 100644 --- a/include/asm-arm/arch-ixp23xx/platform.h +++ b/include/asm-arm/arch-ixp23xx/platform.h @@ -43,5 +43,15 @@ extern struct sys_timer ixp23xx_timer; #define IXP23XX_UART_XTAL 14745600 +#ifndef __ASSEMBLY__ +/* + * Is system memory on the XSI or CPP bus? + */ +static inline unsigned ixp23xx_cpp_boot(void) +{ + return (*IXP23XX_EXP_CFG0 & IXP23XX_EXP_CFG0_XSI_NOT_PRES); +} +#endif + #endif diff --git a/include/asm-arm/arch-ixp23xx/uncompress.h b/include/asm-arm/arch-ixp23xx/uncompress.h index 013575e6a9a..16c1110f230 100644 --- a/include/asm-arm/arch-ixp23xx/uncompress.h +++ b/include/asm-arm/arch-ixp23xx/uncompress.h @@ -11,7 +11,7 @@ #ifndef __ASM_ARCH_UNCOMPRESS_H #define __ASM_ARCH_UNCOMPRESS_H -#include +#include #include #define UART_BASE ((volatile u32 *)IXP23XX_UART1_PHYS) -- cgit v1.2.3-70-g09d2 From 3bec6ded282b331552587267d67a06ed7fd95ddd Mon Sep 17 00:00:00 2001 From: Lennert Buytenhek Date: Tue, 27 Jun 2006 22:56:18 +0100 Subject: [ARM] 3664/1: crunch: add signal frame save/restore Patch from Lennert Buytenhek This patch makes the kernel save Crunch state in userland signal frames, so that any userland signal handler can safely use the Crunch coprocessor without corrupting the Crunch state of the code it preempted. Signed-off-by: Lennert Buytenhek Signed-off-by: Russell King --- arch/arm/kernel/signal.c | 39 +++++++++++++++++++++++++++++++++++++++ include/asm-arm/ucontext.h | 14 ++++++++++++++ 2 files changed, 53 insertions(+) (limited to 'include') diff --git a/arch/arm/kernel/signal.c b/arch/arm/kernel/signal.c index 1ce05ec086c..83a8d3c95eb 100644 --- a/arch/arm/kernel/signal.c +++ b/arch/arm/kernel/signal.c @@ -132,6 +132,37 @@ sys_sigaction(int sig, const struct old_sigaction __user *act, return ret; } +#ifdef CONFIG_CRUNCH +static int preserve_crunch_context(struct crunch_sigframe *frame) +{ + char kbuf[sizeof(*frame) + 8]; + struct crunch_sigframe *kframe; + + /* the crunch context must be 64 bit aligned */ + kframe = (struct crunch_sigframe *)((unsigned long)(kbuf + 8) & ~7); + kframe->magic = CRUNCH_MAGIC; + kframe->size = CRUNCH_STORAGE_SIZE; + crunch_task_copy(current_thread_info(), &kframe->storage); + return __copy_to_user(frame, kframe, sizeof(*frame)); +} + +static int restore_crunch_context(struct crunch_sigframe *frame) +{ + char kbuf[sizeof(*frame) + 8]; + struct crunch_sigframe *kframe; + + /* the crunch context must be 64 bit aligned */ + kframe = (struct crunch_sigframe *)((unsigned long)(kbuf + 8) & ~7); + if (__copy_from_user(kframe, frame, sizeof(*frame))) + return -1; + if (kframe->magic != CRUNCH_MAGIC || + kframe->size != CRUNCH_STORAGE_SIZE) + return -1; + crunch_task_restore(current_thread_info(), &kframe->storage); + return 0; +} +#endif + #ifdef CONFIG_IWMMXT static int preserve_iwmmxt_context(struct iwmmxt_sigframe *frame) @@ -214,6 +245,10 @@ static int restore_sigframe(struct pt_regs *regs, struct sigframe __user *sf) err |= !valid_user_regs(regs); aux = (struct aux_sigframe __user *) sf->uc.uc_regspace; +#ifdef CONFIG_CRUNCH + if (err == 0) + err |= restore_crunch_context(&aux->crunch); +#endif #ifdef CONFIG_IWMMXT if (err == 0 && test_thread_flag(TIF_USING_IWMMXT)) err |= restore_iwmmxt_context(&aux->iwmmxt); @@ -333,6 +368,10 @@ setup_sigframe(struct sigframe __user *sf, struct pt_regs *regs, sigset_t *set) err |= __copy_to_user(&sf->uc.uc_sigmask, set, sizeof(*set)); aux = (struct aux_sigframe __user *) sf->uc.uc_regspace; +#ifdef CONFIG_CRUNCH + if (err == 0) + err |= preserve_crunch_context(&aux->crunch); +#endif #ifdef CONFIG_IWMMXT if (err == 0 && test_thread_flag(TIF_USING_IWMMXT)) err |= preserve_iwmmxt_context(&aux->iwmmxt); diff --git a/include/asm-arm/ucontext.h b/include/asm-arm/ucontext.h index 9e6f7ca9f5a..bf65e9f4525 100644 --- a/include/asm-arm/ucontext.h +++ b/include/asm-arm/ucontext.h @@ -35,6 +35,17 @@ struct ucontext { * bytes, to prevent unpredictable padding in the signal frame. */ +#ifdef CONFIG_CRUNCH +#define CRUNCH_MAGIC 0x5065cf03 +#define CRUNCH_STORAGE_SIZE (CRUNCH_SIZE + 8) + +struct crunch_sigframe { + unsigned long magic; + unsigned long size; + struct crunch_state storage; +} __attribute__((__aligned__(8))); +#endif + #ifdef CONFIG_IWMMXT /* iwmmxt_area is 0x98 bytes long, preceeded by 8 bytes of signature */ #define IWMMXT_MAGIC 0x12ef842a @@ -74,6 +85,9 @@ struct vfp_sigframe * one of these. */ struct aux_sigframe { +#ifdef CONFIG_CRUNCH + struct crunch_sigframe crunch; +#endif #ifdef CONFIG_IWMMXT struct iwmmxt_sigframe iwmmxt; #endif -- cgit v1.2.3-70-g09d2 From 5429b060df6d556f396b78364ad017686015bc34 Mon Sep 17 00:00:00 2001 From: Lennert Buytenhek Date: Tue, 27 Jun 2006 22:56:19 +0100 Subject: [ARM] 3665/1: crunch: add ptrace support Patch from Lennert Buytenhek This patch makes it possible to get/set a task's Crunch state via the ptrace(2) system call. Signed-off-by: Lennert Buytenhek Signed-off-by: Russell King --- arch/arm/kernel/ptrace.c | 36 ++++++++++++++++++++++++++++++++++++ include/asm-arm/ptrace.h | 5 +++++ 2 files changed, 41 insertions(+) (limited to 'include') diff --git a/arch/arm/kernel/ptrace.c b/arch/arm/kernel/ptrace.c index a1d1b2906e8..c40bdc77005 100644 --- a/arch/arm/kernel/ptrace.c +++ b/arch/arm/kernel/ptrace.c @@ -634,6 +634,32 @@ static int ptrace_setwmmxregs(struct task_struct *tsk, void __user *ufp) #endif +#ifdef CONFIG_CRUNCH +/* + * Get the child Crunch state. + */ +static int ptrace_getcrunchregs(struct task_struct *tsk, void __user *ufp) +{ + struct thread_info *thread = task_thread_info(tsk); + + crunch_task_disable(thread); /* force it to ram */ + return copy_to_user(ufp, &thread->crunchstate, CRUNCH_SIZE) + ? -EFAULT : 0; +} + +/* + * Set the child Crunch state. + */ +static int ptrace_setcrunchregs(struct task_struct *tsk, void __user *ufp) +{ + struct thread_info *thread = task_thread_info(tsk); + + crunch_task_release(thread); /* force a reload */ + return copy_from_user(&thread->crunchstate, ufp, CRUNCH_SIZE) + ? -EFAULT : 0; +} +#endif + long arch_ptrace(struct task_struct *child, long request, long addr, long data) { unsigned long tmp; @@ -765,6 +791,16 @@ long arch_ptrace(struct task_struct *child, long request, long addr, long data) child->ptrace_message = data; break; +#ifdef CONFIG_CRUNCH + case PTRACE_GETCRUNCHREGS: + ret = ptrace_getcrunchregs(child, (void __user *)data); + break; + + case PTRACE_SETCRUNCHREGS: + ret = ptrace_setcrunchregs(child, (void __user *)data); + break; +#endif + default: ret = ptrace_request(child, request, addr, data); break; diff --git a/include/asm-arm/ptrace.h b/include/asm-arm/ptrace.h index 2bebe3dc0a3..5a8ef787dbf 100644 --- a/include/asm-arm/ptrace.h +++ b/include/asm-arm/ptrace.h @@ -25,6 +25,11 @@ #define PTRACE_SET_SYSCALL 23 +/* PTRACE_SYSCALL is 24 */ + +#define PTRACE_GETCRUNCHREGS 25 +#define PTRACE_SETCRUNCHREGS 26 + /* * PSR bits */ -- cgit v1.2.3-70-g09d2 From c17fad11f3105ca4d5bbb2686725aad208f5ead4 Mon Sep 17 00:00:00 2001 From: Lennert Buytenhek Date: Tue, 27 Jun 2006 23:03:03 +0100 Subject: [ARM] 3370/2: ep93xx: add crunch support Patch from Lennert Buytenhek Add the necessary kernel bits for crunch task switching. Signed-off-by: Lennert Buytenhek Signed-off-by: Russell King --- arch/arm/kernel/Makefile | 3 + arch/arm/kernel/asm-offsets.c | 3 + arch/arm/kernel/crunch-bits.S | 305 ++++++++++++++++++++++++++++++++++++++++++ arch/arm/kernel/crunch.c | 83 ++++++++++++ arch/arm/kernel/entry-armv.S | 6 + arch/arm/mach-ep93xx/Kconfig | 5 + include/asm-arm/fpstate.h | 8 ++ include/asm-arm/thread_info.h | 6 + 8 files changed, 419 insertions(+) create mode 100644 arch/arm/kernel/crunch-bits.S create mode 100644 arch/arm/kernel/crunch.c (limited to 'include') diff --git a/arch/arm/kernel/Makefile b/arch/arm/kernel/Makefile index a601b8b55f3..7cffbaef064 100644 --- a/arch/arm/kernel/Makefile +++ b/arch/arm/kernel/Makefile @@ -22,6 +22,9 @@ obj-$(CONFIG_PCI) += bios32.o obj-$(CONFIG_SMP) += smp.o obj-$(CONFIG_OABI_COMPAT) += sys_oabi-compat.o +obj-$(CONFIG_CRUNCH) += crunch.o crunch-bits.o +AFLAGS_crunch-bits.o := -Wa,-mcpu=ep9312 + obj-$(CONFIG_IWMMXT) += iwmmxt.o AFLAGS_iwmmxt.o := -Wa,-mcpu=iwmmxt diff --git a/arch/arm/kernel/asm-offsets.c b/arch/arm/kernel/asm-offsets.c index 396efba9bac..447ede5143a 100644 --- a/arch/arm/kernel/asm-offsets.c +++ b/arch/arm/kernel/asm-offsets.c @@ -59,6 +59,9 @@ int main(void) DEFINE(TI_VFPSTATE, offsetof(struct thread_info, vfpstate)); #ifdef CONFIG_IWMMXT DEFINE(TI_IWMMXT_STATE, offsetof(struct thread_info, fpstate.iwmmxt)); +#endif +#ifdef CONFIG_CRUNCH + DEFINE(TI_CRUNCH_STATE, offsetof(struct thread_info, crunchstate)); #endif BLANK(); DEFINE(S_R0, offsetof(struct pt_regs, ARM_r0)); diff --git a/arch/arm/kernel/crunch-bits.S b/arch/arm/kernel/crunch-bits.S new file mode 100644 index 00000000000..a26886758c6 --- /dev/null +++ b/arch/arm/kernel/crunch-bits.S @@ -0,0 +1,305 @@ +/* + * arch/arm/kernel/crunch-bits.S + * Cirrus MaverickCrunch context switching and handling + * + * Copyright (C) 2006 Lennert Buytenhek + * + * Shamelessly stolen from the iWMMXt code by Nicolas Pitre, which is + * Copyright (c) 2003-2004, MontaVista Software, Inc. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ + +#include +#include +#include +#include +#include + +/* + * We can't use hex constants here due to a bug in gas. + */ +#define CRUNCH_MVDX0 0 +#define CRUNCH_MVDX1 8 +#define CRUNCH_MVDX2 16 +#define CRUNCH_MVDX3 24 +#define CRUNCH_MVDX4 32 +#define CRUNCH_MVDX5 40 +#define CRUNCH_MVDX6 48 +#define CRUNCH_MVDX7 56 +#define CRUNCH_MVDX8 64 +#define CRUNCH_MVDX9 72 +#define CRUNCH_MVDX10 80 +#define CRUNCH_MVDX11 88 +#define CRUNCH_MVDX12 96 +#define CRUNCH_MVDX13 104 +#define CRUNCH_MVDX14 112 +#define CRUNCH_MVDX15 120 +#define CRUNCH_MVAX0L 128 +#define CRUNCH_MVAX0M 132 +#define CRUNCH_MVAX0H 136 +#define CRUNCH_MVAX1L 140 +#define CRUNCH_MVAX1M 144 +#define CRUNCH_MVAX1H 148 +#define CRUNCH_MVAX2L 152 +#define CRUNCH_MVAX2M 156 +#define CRUNCH_MVAX2H 160 +#define CRUNCH_MVAX3L 164 +#define CRUNCH_MVAX3M 168 +#define CRUNCH_MVAX3H 172 +#define CRUNCH_DSPSC 176 + +#define CRUNCH_SIZE 184 + + .text + +/* + * Lazy switching of crunch coprocessor context + * + * r10 = struct thread_info pointer + * r9 = ret_from_exception + * lr = undefined instr exit + * + * called from prefetch exception handler with interrupts disabled + */ +ENTRY(crunch_task_enable) + ldr r8, =(EP93XX_APB_VIRT_BASE + 0x00130000) @ syscon addr + + ldr r1, [r8, #0x80] + tst r1, #0x00800000 @ access to crunch enabled? + movne pc, lr @ if so no business here + mov r3, #0xaa @ unlock syscon swlock + str r3, [r8, #0xc0] + orr r1, r1, #0x00800000 @ enable access to crunch + str r1, [r8, #0x80] + + ldr r3, =crunch_owner + add r0, r10, #TI_CRUNCH_STATE @ get task crunch save area + ldr r2, [sp, #60] @ current task pc value + ldr r1, [r3] @ get current crunch owner + str r0, [r3] @ this task now owns crunch + sub r2, r2, #4 @ adjust pc back + str r2, [sp, #60] + + ldr r2, [r8, #0x80] + mov r2, r2 @ flush out enable (@@@) + + teq r1, #0 @ test for last ownership + mov lr, r9 @ normal exit from exception + beq crunch_load @ no owner, skip save + +crunch_save: + cfstr64 mvdx0, [r1, #CRUNCH_MVDX0] @ save 64b registers + cfstr64 mvdx1, [r1, #CRUNCH_MVDX1] + cfstr64 mvdx2, [r1, #CRUNCH_MVDX2] + cfstr64 mvdx3, [r1, #CRUNCH_MVDX3] + cfstr64 mvdx4, [r1, #CRUNCH_MVDX4] + cfstr64 mvdx5, [r1, #CRUNCH_MVDX5] + cfstr64 mvdx6, [r1, #CRUNCH_MVDX6] + cfstr64 mvdx7, [r1, #CRUNCH_MVDX7] + cfstr64 mvdx8, [r1, #CRUNCH_MVDX8] + cfstr64 mvdx9, [r1, #CRUNCH_MVDX9] + cfstr64 mvdx10, [r1, #CRUNCH_MVDX10] + cfstr64 mvdx11, [r1, #CRUNCH_MVDX11] + cfstr64 mvdx12, [r1, #CRUNCH_MVDX12] + cfstr64 mvdx13, [r1, #CRUNCH_MVDX13] + cfstr64 mvdx14, [r1, #CRUNCH_MVDX14] + cfstr64 mvdx15, [r1, #CRUNCH_MVDX15] + +#ifdef __ARMEB__ +#error fix me for ARMEB +#endif + + cfmv32al mvfx0, mvax0 @ save 72b accumulators + cfstr32 mvfx0, [r1, #CRUNCH_MVAX0L] + cfmv32am mvfx0, mvax0 + cfstr32 mvfx0, [r1, #CRUNCH_MVAX0M] + cfmv32ah mvfx0, mvax0 + cfstr32 mvfx0, [r1, #CRUNCH_MVAX0H] + cfmv32al mvfx0, mvax1 + cfstr32 mvfx0, [r1, #CRUNCH_MVAX1L] + cfmv32am mvfx0, mvax1 + cfstr32 mvfx0, [r1, #CRUNCH_MVAX1M] + cfmv32ah mvfx0, mvax1 + cfstr32 mvfx0, [r1, #CRUNCH_MVAX1H] + cfmv32al mvfx0, mvax2 + cfstr32 mvfx0, [r1, #CRUNCH_MVAX2L] + cfmv32am mvfx0, mvax2 + cfstr32 mvfx0, [r1, #CRUNCH_MVAX2M] + cfmv32ah mvfx0, mvax2 + cfstr32 mvfx0, [r1, #CRUNCH_MVAX2H] + cfmv32al mvfx0, mvax3 + cfstr32 mvfx0, [r1, #CRUNCH_MVAX3L] + cfmv32am mvfx0, mvax3 + cfstr32 mvfx0, [r1, #CRUNCH_MVAX3M] + cfmv32ah mvfx0, mvax3 + cfstr32 mvfx0, [r1, #CRUNCH_MVAX3H] + + cfmv32sc mvdx0, dspsc @ save status word + cfstr64 mvdx0, [r1, #CRUNCH_DSPSC] + + teq r0, #0 @ anything to load? + cfldr64eq mvdx0, [r1, #CRUNCH_MVDX0] @ mvdx0 was clobbered + moveq pc, lr + +crunch_load: + cfldr64 mvdx0, [r0, #CRUNCH_DSPSC] @ load status word + cfmvsc32 dspsc, mvdx0 + + cfldr32 mvfx0, [r0, #CRUNCH_MVAX0L] @ load 72b accumulators + cfmval32 mvax0, mvfx0 + cfldr32 mvfx0, [r0, #CRUNCH_MVAX0M] + cfmvam32 mvax0, mvfx0 + cfldr32 mvfx0, [r0, #CRUNCH_MVAX0H] + cfmvah32 mvax0, mvfx0 + cfldr32 mvfx0, [r0, #CRUNCH_MVAX1L] + cfmval32 mvax1, mvfx0 + cfldr32 mvfx0, [r0, #CRUNCH_MVAX1M] + cfmvam32 mvax1, mvfx0 + cfldr32 mvfx0, [r0, #CRUNCH_MVAX1H] + cfmvah32 mvax1, mvfx0 + cfldr32 mvfx0, [r0, #CRUNCH_MVAX2L] + cfmval32 mvax2, mvfx0 + cfldr32 mvfx0, [r0, #CRUNCH_MVAX2M] + cfmvam32 mvax2, mvfx0 + cfldr32 mvfx0, [r0, #CRUNCH_MVAX2H] + cfmvah32 mvax2, mvfx0 + cfldr32 mvfx0, [r0, #CRUNCH_MVAX3L] + cfmval32 mvax3, mvfx0 + cfldr32 mvfx0, [r0, #CRUNCH_MVAX3M] + cfmvam32 mvax3, mvfx0 + cfldr32 mvfx0, [r0, #CRUNCH_MVAX3H] + cfmvah32 mvax3, mvfx0 + + cfldr64 mvdx0, [r0, #CRUNCH_MVDX0] @ load 64b registers + cfldr64 mvdx1, [r0, #CRUNCH_MVDX1] + cfldr64 mvdx2, [r0, #CRUNCH_MVDX2] + cfldr64 mvdx3, [r0, #CRUNCH_MVDX3] + cfldr64 mvdx4, [r0, #CRUNCH_MVDX4] + cfldr64 mvdx5, [r0, #CRUNCH_MVDX5] + cfldr64 mvdx6, [r0, #CRUNCH_MVDX6] + cfldr64 mvdx7, [r0, #CRUNCH_MVDX7] + cfldr64 mvdx8, [r0, #CRUNCH_MVDX8] + cfldr64 mvdx9, [r0, #CRUNCH_MVDX9] + cfldr64 mvdx10, [r0, #CRUNCH_MVDX10] + cfldr64 mvdx11, [r0, #CRUNCH_MVDX11] + cfldr64 mvdx12, [r0, #CRUNCH_MVDX12] + cfldr64 mvdx13, [r0, #CRUNCH_MVDX13] + cfldr64 mvdx14, [r0, #CRUNCH_MVDX14] + cfldr64 mvdx15, [r0, #CRUNCH_MVDX15] + + mov pc, lr + +/* + * Back up crunch regs to save area and disable access to them + * (mainly for gdb or sleep mode usage) + * + * r0 = struct thread_info pointer of target task or NULL for any + */ +ENTRY(crunch_task_disable) + stmfd sp!, {r4, r5, lr} + + mrs ip, cpsr + orr r2, ip, #PSR_I_BIT @ disable interrupts + msr cpsr_c, r2 + + ldr r4, =(EP93XX_APB_VIRT_BASE + 0x00130000) @ syscon addr + + ldr r3, =crunch_owner + add r2, r0, #TI_CRUNCH_STATE @ get task crunch save area + ldr r1, [r3] @ get current crunch owner + teq r1, #0 @ any current owner? + beq 1f @ no: quit + teq r0, #0 @ any owner? + teqne r1, r2 @ or specified one? + bne 1f @ no: quit + + ldr r5, [r4, #0x80] @ enable access to crunch + mov r2, #0xaa + str r2, [r4, #0xc0] + orr r5, r5, #0x00800000 + str r5, [r4, #0x80] + + mov r0, #0 @ nothing to load + str r0, [r3] @ no more current owner + ldr r2, [r4, #0x80] @ flush out enable (@@@) + mov r2, r2 + bl crunch_save + + mov r2, #0xaa @ disable access to crunch + str r2, [r4, #0xc0] + bic r5, r5, #0x00800000 + str r5, [r4, #0x80] + ldr r5, [r4, #0x80] @ flush out enable (@@@) + mov r5, r5 + +1: msr cpsr_c, ip @ restore interrupt mode + ldmfd sp!, {r4, r5, pc} + +/* + * Copy crunch state to given memory address + * + * r0 = struct thread_info pointer of target task + * r1 = memory address where to store crunch state + * + * this is called mainly in the creation of signal stack frames + */ +ENTRY(crunch_task_copy) + mrs ip, cpsr + orr r2, ip, #PSR_I_BIT @ disable interrupts + msr cpsr_c, r2 + + ldr r3, =crunch_owner + add r2, r0, #TI_CRUNCH_STATE @ get task crunch save area + ldr r3, [r3] @ get current crunch owner + teq r2, r3 @ does this task own it... + beq 1f + + @ current crunch values are in the task save area + msr cpsr_c, ip @ restore interrupt mode + mov r0, r1 + mov r1, r2 + mov r2, #CRUNCH_SIZE + b memcpy + +1: @ this task owns crunch regs -- grab a copy from there + mov r0, #0 @ nothing to load + mov r3, lr @ preserve return address + bl crunch_save + msr cpsr_c, ip @ restore interrupt mode + mov pc, r3 + +/* + * Restore crunch state from given memory address + * + * r0 = struct thread_info pointer of target task + * r1 = memory address where to get crunch state from + * + * this is used to restore crunch state when unwinding a signal stack frame + */ +ENTRY(crunch_task_restore) + mrs ip, cpsr + orr r2, ip, #PSR_I_BIT @ disable interrupts + msr cpsr_c, r2 + + ldr r3, =crunch_owner + add r2, r0, #TI_CRUNCH_STATE @ get task crunch save area + ldr r3, [r3] @ get current crunch owner + teq r2, r3 @ does this task own it... + beq 1f + + @ this task doesn't own crunch regs -- use its save area + msr cpsr_c, ip @ restore interrupt mode + mov r0, r2 + mov r2, #CRUNCH_SIZE + b memcpy + +1: @ this task owns crunch regs -- load them directly + mov r0, r1 + mov r1, #0 @ nothing to save + mov r3, lr @ preserve return address + bl crunch_load + msr cpsr_c, ip @ restore interrupt mode + mov pc, r3 diff --git a/arch/arm/kernel/crunch.c b/arch/arm/kernel/crunch.c new file mode 100644 index 00000000000..748175921f9 --- /dev/null +++ b/arch/arm/kernel/crunch.c @@ -0,0 +1,83 @@ +/* + * arch/arm/kernel/crunch.c + * Cirrus MaverickCrunch context switching and handling + * + * Copyright (C) 2006 Lennert Buytenhek + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +struct crunch_state *crunch_owner; + +void crunch_task_release(struct thread_info *thread) +{ + local_irq_disable(); + if (crunch_owner == &thread->crunchstate) + crunch_owner = NULL; + local_irq_enable(); +} + +static int crunch_enabled(u32 devcfg) +{ + return !!(devcfg & EP93XX_SYSCON_DEVICE_CONFIG_CRUNCH_ENABLE); +} + +static int crunch_do(struct notifier_block *self, unsigned long cmd, void *t) +{ + struct thread_info *thread = (struct thread_info *)t; + struct crunch_state *crunch_state; + u32 devcfg; + + crunch_state = &thread->crunchstate; + + switch (cmd) { + case THREAD_NOTIFY_FLUSH: + memset(crunch_state, 0, sizeof(*crunch_state)); + + /* + * FALLTHROUGH: Ensure we don't try to overwrite our newly + * initialised state information on the first fault. + */ + + case THREAD_NOTIFY_RELEASE: + crunch_task_release(thread); + break; + + case THREAD_NOTIFY_SWITCH: + devcfg = __raw_readl(EP93XX_SYSCON_DEVICE_CONFIG); + if (crunch_enabled(devcfg) || crunch_owner == crunch_state) { + devcfg ^= EP93XX_SYSCON_DEVICE_CONFIG_CRUNCH_ENABLE; + __raw_writel(0xaa, EP93XX_SYSCON_SWLOCK); + __raw_writel(devcfg, EP93XX_SYSCON_DEVICE_CONFIG); + } + break; + } + + return NOTIFY_DONE; +} + +static struct notifier_block crunch_notifier_block = { + .notifier_call = crunch_do, +}; + +static int __init crunch_init(void) +{ + thread_register_notifier(&crunch_notifier_block); + + return 0; +} + +late_initcall(crunch_init); diff --git a/arch/arm/kernel/entry-armv.S b/arch/arm/kernel/entry-armv.S index 86c92523a34..6423a38839b 100644 --- a/arch/arm/kernel/entry-armv.S +++ b/arch/arm/kernel/entry-armv.S @@ -492,9 +492,15 @@ call_fpe: b do_fpe @ CP#1 (FPE) b do_fpe @ CP#2 (FPE) mov pc, lr @ CP#3 +#ifdef CONFIG_CRUNCH + b crunch_task_enable @ CP#4 (MaverickCrunch) + b crunch_task_enable @ CP#5 (MaverickCrunch) + b crunch_task_enable @ CP#6 (MaverickCrunch) +#else mov pc, lr @ CP#4 mov pc, lr @ CP#5 mov pc, lr @ CP#6 +#endif mov pc, lr @ CP#7 mov pc, lr @ CP#8 mov pc, lr @ CP#9 diff --git a/arch/arm/mach-ep93xx/Kconfig b/arch/arm/mach-ep93xx/Kconfig index cec5a21ca4e..293c1f5c618 100644 --- a/arch/arm/mach-ep93xx/Kconfig +++ b/arch/arm/mach-ep93xx/Kconfig @@ -2,6 +2,11 @@ if ARCH_EP93XX menu "Cirrus EP93xx Implementation Options" +config CRUNCH + bool "Support for MaverickCrunch" + help + Enable kernel support for MaverickCrunch. + comment "EP93xx Platforms" config MACH_GESBC9312 diff --git a/include/asm-arm/fpstate.h b/include/asm-arm/fpstate.h index 132c3c5628b..6af4e6bd129 100644 --- a/include/asm-arm/fpstate.h +++ b/include/asm-arm/fpstate.h @@ -72,6 +72,14 @@ union fp_state { #define FP_SIZE (sizeof(union fp_state) / sizeof(int)) +struct crunch_state { + unsigned int mvdx[16][2]; + unsigned int mvax[4][3]; + unsigned int dspsc[2]; +}; + +#define CRUNCH_SIZE sizeof(struct crunch_state) + #endif #endif diff --git a/include/asm-arm/thread_info.h b/include/asm-arm/thread_info.h index cfbccb63c67..c46b5c84275 100644 --- a/include/asm-arm/thread_info.h +++ b/include/asm-arm/thread_info.h @@ -59,6 +59,7 @@ struct thread_info { struct cpu_context_save cpu_context; /* cpu context */ __u8 used_cp[16]; /* thread used copro */ unsigned long tp_value; + struct crunch_state crunchstate; union fp_state fpstate __attribute__((aligned(8))); union vfp_state vfpstate; struct restart_block restart_block; @@ -101,6 +102,11 @@ extern void free_thread_info(struct thread_info *); #define thread_saved_fp(tsk) \ ((unsigned long)(task_thread_info(tsk)->cpu_context.fp)) +extern void crunch_task_disable(struct thread_info *); +extern void crunch_task_copy(struct thread_info *, void *); +extern void crunch_task_restore(struct thread_info *, void *); +extern void crunch_task_release(struct thread_info *); + extern void iwmmxt_task_disable(struct thread_info *); extern void iwmmxt_task_copy(struct thread_info *, void *); extern void iwmmxt_task_restore(struct thread_info *, void *); -- cgit v1.2.3-70-g09d2 From 9d56d878ae12dbb9af744cec2858ed35dffde2b2 Mon Sep 17 00:00:00 2001 From: Jack Steiner Date: Thu, 27 Apr 2006 20:01:05 -0500 Subject: [IA64-SGI] - Pass OS logical cpu number to the SN prom (bios) Pass the OS logical cpu number to the PROM. This allows PROM to log the OS logical cpu number in error records viewed thru POD. Signed-off-by: Jack Steiner Signed-off-by: Tony Luck --- arch/ia64/sn/kernel/setup.c | 10 ++++++++-- include/asm-ia64/sn/sn_sal.h | 10 ++++++++++ 2 files changed, 18 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/arch/ia64/sn/kernel/setup.c b/arch/ia64/sn/kernel/setup.c index 93577abae36..97579348a54 100644 --- a/arch/ia64/sn/kernel/setup.c +++ b/arch/ia64/sn/kernel/setup.c @@ -577,7 +577,8 @@ void __init sn_cpu_init(void) int i; static int wars_have_been_checked; - if (smp_processor_id() == 0 && IS_MEDUSA()) { + cpuid = smp_processor_id(); + if (cpuid == 0 && IS_MEDUSA()) { if (ia64_sn_is_fake_prom()) sn_prom_type = 2; else @@ -596,6 +597,12 @@ void __init sn_cpu_init(void) BUG(); sn_hub_info->as_shift = sn_hub_info->nasid_shift - 2; + /* + * Don't check status. The SAL call is not supported on all PROMs + * but a failure is harmless. + */ + (void) ia64_sn_set_cpu_number(cpuid); + /* * The boot cpu makes this call again after platform initialization is * complete. @@ -607,7 +614,6 @@ void __init sn_cpu_init(void) if (ia64_sn_get_prom_feature_set(i, &sn_prom_features[i]) != 0) break; - cpuid = smp_processor_id(); cpuphyid = get_sapicid(); if (ia64_sn_get_sapic_info(cpuphyid, &nasid, &subnode, &slice)) diff --git a/include/asm-ia64/sn/sn_sal.h b/include/asm-ia64/sn/sn_sal.h index cd490b20d59..bd4452bda35 100644 --- a/include/asm-ia64/sn/sn_sal.h +++ b/include/asm-ia64/sn/sn_sal.h @@ -85,6 +85,7 @@ #define SN_SAL_GET_PROM_FEATURE_SET 0x02000065 #define SN_SAL_SET_OS_FEATURE_SET 0x02000066 #define SN_SAL_INJECT_ERROR 0x02000067 +#define SN_SAL_SET_CPU_NUMBER 0x02000068 /* * Service-specific constants @@ -1150,4 +1151,13 @@ sn_inject_error(u64 paddr, u64 *data, u64 *ecc) local_irq_restore(irq_flags); return ret_stuff.status; } + +static inline int +ia64_sn_set_cpu_number(int cpu) +{ + struct ia64_sal_retval rv; + + SAL_CALL_NOLOCK(rv, SN_SAL_SET_CPU_NUMBER, cpu, 0, 0, 0, 0, 0, 0); + return rv.status; +} #endif /* _ASM_IA64_SN_SN_SAL_H */ -- cgit v1.2.3-70-g09d2 From 002547b4f86c27bfac5bae344b723d250857be6b Mon Sep 17 00:00:00 2001 From: Russell King Date: Tue, 20 Jun 2006 20:46:52 +0100 Subject: [ARM] nommu: adjust headers for !MMU ARM systems Majorily based on Hyok Choi's patches, this fixes up the asm-arm header files for mmuless systems. Over and above Hyok's patches: - nommu.h merged into mmu.h (it's only a structure) - nommu_context.h is essentially the same as mmu_context.h, but without the MM switching code. so there's no point having separate files. Also, in memory.h, there's no point #ifndef'ing PHYS_OFFSET and END_MEM - both CONFIG_DRAM_BASE and CONFIG_DRAM_SIZE will always be set by the configuration scripts. Other files have minor formatting changes, but are essentially the same. Hyok's original patches were signed off thusly: Signed-off-by: Hyok S. Choi Signed-off-by: Russell King --- include/asm-arm/bugs.h | 4 ++ include/asm-arm/domain.h | 7 +++ include/asm-arm/memory.h | 75 ++++++++++++++++++------ include/asm-arm/mmu.h | 16 ++++++ include/asm-arm/mmu_context.h | 2 + include/asm-arm/page-nommu.h | 51 +++++++++++++++++ include/asm-arm/page.h | 8 +++ include/asm-arm/pgalloc.h | 8 ++- include/asm-arm/pgtable-nommu.h | 123 ++++++++++++++++++++++++++++++++++++++++ include/asm-arm/pgtable.h | 10 +++- include/asm-arm/proc-fns.h | 4 ++ 11 files changed, 287 insertions(+), 21 deletions(-) create mode 100644 include/asm-arm/page-nommu.h create mode 100644 include/asm-arm/pgtable-nommu.h (limited to 'include') diff --git a/include/asm-arm/bugs.h b/include/asm-arm/bugs.h index 4c80ec519d4..ca54eb0f12d 100644 --- a/include/asm-arm/bugs.h +++ b/include/asm-arm/bugs.h @@ -10,8 +10,12 @@ #ifndef __ASM_BUGS_H #define __ASM_BUGS_H +#ifdef CONFIG_MMU extern void check_writebuffer_bugs(void); #define check_bugs() check_writebuffer_bugs() +#else +#define check_bugs() do { } while (0) +#endif #endif diff --git a/include/asm-arm/domain.h b/include/asm-arm/domain.h index f8ea2de4848..4c2885abbe6 100644 --- a/include/asm-arm/domain.h +++ b/include/asm-arm/domain.h @@ -50,6 +50,8 @@ #define domain_val(dom,type) ((type) << (2*(dom))) #ifndef __ASSEMBLY__ + +#ifdef CONFIG_MMU #define set_domain(x) \ do { \ __asm__ __volatile__( \ @@ -66,5 +68,10 @@ set_domain(thread->cpu_domain); \ } while (0) +#else +#define set_domain(x) do { } while (0) +#define modify_domain(dom,type) do { } while (0) +#endif + #endif #endif /* !__ASSEMBLY__ */ diff --git a/include/asm-arm/memory.h b/include/asm-arm/memory.h index 731e321a57d..94f973b704f 100644 --- a/include/asm-arm/memory.h +++ b/include/asm-arm/memory.h @@ -2,6 +2,7 @@ * linux/include/asm-arm/memory.h * * Copyright (C) 2000-2002 Russell King + * modification for nommu, Hyok S. Choi, 2004 * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License version 2 as @@ -26,6 +27,8 @@ #include #include +#ifdef CONFIG_MMU + #ifndef TASK_SIZE /* * TASK_SIZE - the maximum size of a user space task. @@ -47,6 +50,60 @@ #define PAGE_OFFSET UL(0xc0000000) #endif +/* + * The module space lives between the addresses given by TASK_SIZE + * and PAGE_OFFSET - it must be within 32MB of the kernel text. + */ +#define MODULE_END (PAGE_OFFSET) +#define MODULE_START (MODULE_END - 16*1048576) + +#if TASK_SIZE > MODULE_START +#error Top of user space clashes with start of module space +#endif + +/* + * The XIP kernel gets mapped at the bottom of the module vm area. + * Since we use sections to map it, this macro replaces the physical address + * with its virtual address while keeping offset from the base section. + */ +#define XIP_VIRT_ADDR(physaddr) (MODULE_START + ((physaddr) & 0x000fffff)) + +#else /* CONFIG_MMU */ + +/* + * The limitation of user task size can grow up to the end of free ram region. + * It is difficult to define and perhaps will never meet the original meaning + * of this define that was meant to. + * Fortunately, there is no reference for this in noMMU mode, for now. + */ +#ifndef TASK_SIZE +#define TASK_SIZE (CONFIG_DRAM_SIZE) +#endif + +#ifndef TASK_UNMAPPED_BASE +#define TASK_UNMAPPED_BASE UL(0x00000000) +#endif + +#ifndef PHYS_OFFSET +#define PHYS_OFFSET (CONFIG_DRAM_BASE) +#endif + +#ifndef END_MEM +#define END_MEM (CONFIG_DRAM_BASE + CONFIG_DRAM_SIZE) +#endif + +#ifndef PAGE_OFFSET +#define PAGE_OFFSET (PHYS_OFFSET) +#endif + +/* + * The module can be at any place in ram in nommu mode. + */ +#define MODULE_END (END_MEM) +#define MODULE_START (PHYS_OFFSET) + +#endif /* !CONFIG_MMU */ + /* * Size of DMA-consistent memory region. Must be multiple of 2M, * between 2MB and 14MB inclusive. @@ -71,24 +128,6 @@ #define __phys_to_pfn(paddr) ((paddr) >> PAGE_SHIFT) #define __pfn_to_phys(pfn) ((pfn) << PAGE_SHIFT) -/* - * The module space lives between the addresses given by TASK_SIZE - * and PAGE_OFFSET - it must be within 32MB of the kernel text. - */ -#define MODULE_END (PAGE_OFFSET) -#define MODULE_START (MODULE_END - 16*1048576) - -#if TASK_SIZE > MODULE_START -#error Top of user space clashes with start of module space -#endif - -/* - * The XIP kernel gets mapped at the bottom of the module vm area. - * Since we use sections to map it, this macro replaces the physical address - * with its virtual address while keeping offset from the base section. - */ -#define XIP_VIRT_ADDR(physaddr) (MODULE_START + ((physaddr) & 0x000fffff)) - #ifndef __ASSEMBLY__ /* diff --git a/include/asm-arm/mmu.h b/include/asm-arm/mmu.h index a457cb71984..23dde52e094 100644 --- a/include/asm-arm/mmu.h +++ b/include/asm-arm/mmu.h @@ -1,6 +1,8 @@ #ifndef __ARM_MMU_H #define __ARM_MMU_H +#ifdef CONFIG_MMU + typedef struct { #if __LINUX_ARM_ARCH__ >= 6 unsigned int id; @@ -13,4 +15,18 @@ typedef struct { #define ASID(mm) (0) #endif +#else + +/* + * From nommu.h: + * Copyright (C) 2002, David McCullough + * modified for 2.6 by Hyok S. Choi + */ +typedef struct { + struct vm_list_struct *vmlist; + unsigned long end_brk; +} mm_context_t; + +#endif + #endif diff --git a/include/asm-arm/mmu_context.h b/include/asm-arm/mmu_context.h index 81c59facea3..9fadb01e030 100644 --- a/include/asm-arm/mmu_context.h +++ b/include/asm-arm/mmu_context.h @@ -82,6 +82,7 @@ static inline void switch_mm(struct mm_struct *prev, struct mm_struct *next, struct task_struct *tsk) { +#ifdef CONFIG_MMU unsigned int cpu = smp_processor_id(); if (prev != next) { @@ -91,6 +92,7 @@ switch_mm(struct mm_struct *prev, struct mm_struct *next, if (cache_is_vivt()) cpu_clear(cpu, prev->cpu_vm_mask); } +#endif } #define deactivate_mm(tsk,mm) do { } while (0) diff --git a/include/asm-arm/page-nommu.h b/include/asm-arm/page-nommu.h new file mode 100644 index 00000000000..a1bcad06048 --- /dev/null +++ b/include/asm-arm/page-nommu.h @@ -0,0 +1,51 @@ +/* + * linux/include/asm-arm/page-nommu.h + * + * Copyright (C) 2004 Hyok S. Choi + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ +#ifndef _ASMARM_PAGE_NOMMU_H +#define _ASMARM_PAGE_NOMMU_H + +#if !defined(CONFIG_SMALL_TASKS) && PAGE_SHIFT < 13 +#define KTHREAD_SIZE (8192) +#else +#define KTHREAD_SIZE PAGE_SIZE +#endif + +#define get_user_page(vaddr) __get_free_page(GFP_KERNEL) +#define free_user_page(page, addr) free_page(addr) + +#define clear_page(page) memset((page), 0, PAGE_SIZE) +#define copy_page(to,from) memcpy((to), (from), PAGE_SIZE) + +#define clear_user_page(page, vaddr, pg) clear_page(page) +#define copy_user_page(to, from, vaddr, pg) copy_page(to, from) + +/* + * These are used to make use of C type-checking.. + */ +typedef unsigned long pte_t; +typedef unsigned long pmd_t; +typedef unsigned long pgd_t[2]; +typedef unsigned long pgprot_t; + +#define pte_val(x) (x) +#define pmd_val(x) (x) +#define pgd_val(x) ((x)[0]) +#define pgprot_val(x) (x) + +#define __pte(x) (x) +#define __pmd(x) (x) +#define __pgprot(x) (x) + +/* to align the pointer to the (next) page boundary */ +#define PAGE_ALIGN(addr) (((addr)+PAGE_SIZE-1)&PAGE_MASK) + +extern unsigned long memory_start; +extern unsigned long memory_end; + +#endif diff --git a/include/asm-arm/page.h b/include/asm-arm/page.h index 66cfeb5290e..63d12f0244c 100644 --- a/include/asm-arm/page.h +++ b/include/asm-arm/page.h @@ -23,6 +23,12 @@ #ifndef __ASSEMBLY__ +#ifndef CONFIG_MMU + +#include "page-nommu.h" + +#else + #include /* @@ -171,6 +177,8 @@ typedef unsigned long pgprot_t; /* the upper-most page table pointer */ extern pmd_t *top_pmd; +#endif /* CONFIG_MMU */ + #include #endif /* !__ASSEMBLY__ */ diff --git a/include/asm-arm/pgalloc.h b/include/asm-arm/pgalloc.h index c4ac2e67768..4d439455291 100644 --- a/include/asm-arm/pgalloc.h +++ b/include/asm-arm/pgalloc.h @@ -16,6 +16,10 @@ #include #include +#define check_pgt_cache() do { } while (0) + +#ifdef CONFIG_MMU + #define _PAGE_USER_TABLE (PMD_TYPE_TABLE | PMD_BIT4 | PMD_DOMAIN(DOMAIN_USER)) #define _PAGE_KERNEL_TABLE (PMD_TYPE_TABLE | PMD_BIT4 | PMD_DOMAIN(DOMAIN_KERNEL)) @@ -32,8 +36,6 @@ extern void free_pgd_slow(pgd_t *pgd); #define pgd_alloc(mm) get_pgd_slow(mm) #define pgd_free(pgd) free_pgd_slow(pgd) -#define check_pgt_cache() do { } while (0) - /* * Allocate one PTE table. * @@ -126,4 +128,6 @@ pmd_populate(struct mm_struct *mm, pmd_t *pmdp, struct page *ptep) __pmd_populate(pmdp, page_to_pfn(ptep) << PAGE_SHIFT | _PAGE_USER_TABLE); } +#endif /* CONFIG_MMU */ + #endif diff --git a/include/asm-arm/pgtable-nommu.h b/include/asm-arm/pgtable-nommu.h new file mode 100644 index 00000000000..b13322dccf4 --- /dev/null +++ b/include/asm-arm/pgtable-nommu.h @@ -0,0 +1,123 @@ +/* + * linux/include/asm-arm/pgtable-nommu.h + * + * Copyright (C) 1995-2002 Russell King + * Copyright (C) 2004 Hyok S. Choi + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ +#ifndef _ASMARM_PGTABLE_NOMMU_H +#define _ASMARM_PGTABLE_NOMMU_H + +#ifndef __ASSEMBLY__ + +#include +#include +#include +#include +#include + +/* + * Trivial page table functions. + */ +#define pgd_present(pgd) (1) +#define pgd_none(pgd) (0) +#define pgd_bad(pgd) (0) +#define pgd_clear(pgdp) +#define kern_addr_valid(addr) (1) +#define pmd_offset(a, b) ((void *)0) +/* FIXME */ +/* + * PMD_SHIFT determines the size of the area a second-level page table can map + * PGDIR_SHIFT determines what a third-level page table entry can map + */ +#define PGDIR_SHIFT 21 + +#define PGDIR_SIZE (1UL << PGDIR_SHIFT) +#define PGDIR_MASK (~(PGDIR_SIZE-1)) +/* FIXME */ + +#define PAGE_NONE __pgprot(0) +#define PAGE_SHARED __pgprot(0) +#define PAGE_COPY __pgprot(0) +#define PAGE_READONLY __pgprot(0) +#define PAGE_KERNEL __pgprot(0) + +//extern void paging_init(struct meminfo *, struct machine_desc *); +#define swapper_pg_dir ((pgd_t *) 0) + +#define __swp_type(x) (0) +#define __swp_offset(x) (0) +#define __swp_entry(typ,off) ((swp_entry_t) { ((typ) | ((off) << 7)) }) +#define __pte_to_swp_entry(pte) ((swp_entry_t) { pte_val(pte) }) +#define __swp_entry_to_pte(x) ((pte_t) { (x).val }) + + +typedef pte_t *pte_addr_t; + +static inline int pte_file(pte_t pte) { return 0; } + +/* + * ZERO_PAGE is a global shared page that is always zero: used + * for zero-mapped memory areas etc.. + */ +#define ZERO_PAGE(vaddr) (virt_to_page(0)) + +/* + * Mark the prot value as uncacheable and unbufferable. + */ +#define pgprot_noncached(prot) __pgprot(0) +#define pgprot_writecombine(prot) __pgprot(0) + + +/* + * These would be in other places but having them here reduces the diffs. + */ +extern unsigned int kobjsize(const void *objp); +extern int is_in_rom(unsigned long); + +/* + * No page table caches to initialise. + */ +#define pgtable_cache_init() do { } while (0) +#define io_remap_page_range remap_page_range +#define io_remap_pfn_range remap_pfn_range + +#define MK_IOSPACE_PFN(space, pfn) (pfn) +#define GET_IOSPACE(pfn) 0 +#define GET_PFN(pfn) (pfn) + + +/* + * All 32bit addresses are effectively valid for vmalloc... + * Sort of meaningless for non-VM targets. + */ +#define VMALLOC_START 0 +#define VMALLOC_END 0xffffffff + +#define FIRST_USER_ADDRESS (0) + +#else + +/* + * dummy tlb and user structures. + */ +#define v3_tlb_fns (0) +#define v4_tlb_fns (0) +#define v4wb_tlb_fns (0) +#define v4wbi_tlb_fns (0) +#define v6_tlb_fns (0) + +#define v3_user_fns (0) +#define v4_user_fns (0) +#define v4_mc_user_fns (0) +#define v4wb_user_fns (0) +#define v4wt_user_fns (0) +#define v6_user_fns (0) +#define xscale_mc_user_fns (0) + +#endif /*__ASSEMBLY__*/ + +#endif /* _ASMARM_PGTABLE_H */ diff --git a/include/asm-arm/pgtable.h b/include/asm-arm/pgtable.h index e85c08d78dd..8d3919c6458 100644 --- a/include/asm-arm/pgtable.h +++ b/include/asm-arm/pgtable.h @@ -11,9 +11,15 @@ #define _ASMARM_PGTABLE_H #include +#include + +#ifndef CONFIG_MMU + +#include "pgtable-nommu.h" + +#else #include -#include #include /* @@ -378,4 +384,6 @@ extern pgd_t swapper_pg_dir[PTRS_PER_PGD]; #endif /* !__ASSEMBLY__ */ +#endif /* CONFIG_MMU */ + #endif /* _ASMARM_PGTABLE_H */ diff --git a/include/asm-arm/proc-fns.h b/include/asm-arm/proc-fns.h index e9310895e79..1bde92cdaeb 100644 --- a/include/asm-arm/proc-fns.h +++ b/include/asm-arm/proc-fns.h @@ -165,6 +165,8 @@ #include +#ifdef CONFIG_MMU + #define cpu_switch_mm(pgd,mm) cpu_do_switch_mm(virt_to_phys(pgd),mm) #define cpu_get_pgd() \ @@ -176,6 +178,8 @@ (pgd_t *)phys_to_virt(pg); \ }) +#endif + #endif /* __ASSEMBLY__ */ #endif /* __KERNEL__ */ #endif /* __ASM_PROCFNS_H */ -- cgit v1.2.3-70-g09d2 From 9641c7cc5a7f6d5c9dc9b43eea4e5f8c3c08c94e Mon Sep 17 00:00:00 2001 From: Russell King Date: Wed, 21 Jun 2006 20:38:17 +0100 Subject: [ARM] nommu: uaccess tweaks MMUless systems have only one address space for all threads, so both the usual access_ok() checks, and the exception handling do not make much sense. Hence, discard the fixup and exception tables at link time, use memcpy/memset for the user copy/clearing functions, and define the permission check macros to be constants. Some of this patch was derived from the equivalent patch by Hyok S. Choi. Signed-off-by: Hyok S. Choi Signed-off-by: Russell King --- arch/arm/kernel/armksyms.c | 7 ++- arch/arm/kernel/vmlinux.lds.S | 8 +++ arch/arm/lib/Makefile | 13 ++-- include/asm-arm/uaccess.h | 139 ++++++++++++++++++++++++++---------------- 4 files changed, 108 insertions(+), 59 deletions(-) (limited to 'include') diff --git a/arch/arm/kernel/armksyms.c b/arch/arm/kernel/armksyms.c index f8bb7abd3e9..da69e660574 100644 --- a/arch/arm/kernel/armksyms.c +++ b/arch/arm/kernel/armksyms.c @@ -109,11 +109,13 @@ EXPORT_SYMBOL(memchr); EXPORT_SYMBOL(__memzero); /* user mem (segment) */ +EXPORT_SYMBOL(__strnlen_user); +EXPORT_SYMBOL(__strncpy_from_user); + +#ifdef CONFIG_MMU EXPORT_SYMBOL(__copy_from_user); EXPORT_SYMBOL(__copy_to_user); EXPORT_SYMBOL(__clear_user); -EXPORT_SYMBOL(__strnlen_user); -EXPORT_SYMBOL(__strncpy_from_user); EXPORT_SYMBOL(__get_user_1); EXPORT_SYMBOL(__get_user_2); @@ -123,6 +125,7 @@ EXPORT_SYMBOL(__put_user_1); EXPORT_SYMBOL(__put_user_2); EXPORT_SYMBOL(__put_user_4); EXPORT_SYMBOL(__put_user_8); +#endif /* crypto hash */ EXPORT_SYMBOL(sha_transform); diff --git a/arch/arm/kernel/vmlinux.lds.S b/arch/arm/kernel/vmlinux.lds.S index 2b254e88595..2df9688a702 100644 --- a/arch/arm/kernel/vmlinux.lds.S +++ b/arch/arm/kernel/vmlinux.lds.S @@ -80,6 +80,10 @@ SECTIONS *(.exit.text) *(.exit.data) *(.exitcall.exit) +#ifndef CONFIG_MMU + *(.fixup) + *(__ex_table) +#endif } .text : { /* Real text segment */ @@ -87,7 +91,9 @@ SECTIONS *(.text) SCHED_TEXT LOCK_TEXT +#ifdef CONFIG_MMU *(.fixup) +#endif *(.gnu.warning) *(.rodata) *(.rodata.*) @@ -142,7 +148,9 @@ SECTIONS */ . = ALIGN(32); __start___ex_table = .; +#ifdef CONFIG_MMU *(__ex_table) +#endif __stop___ex_table = .; /* diff --git a/arch/arm/lib/Makefile b/arch/arm/lib/Makefile index 7b726b627ea..30351cd4560 100644 --- a/arch/arm/lib/Makefile +++ b/arch/arm/lib/Makefile @@ -6,28 +6,31 @@ lib-y := backtrace.o changebit.o csumipv6.o csumpartial.o \ csumpartialcopy.o csumpartialcopyuser.o clearbit.o \ - copy_page.o delay.o findbit.o memchr.o memcpy.o \ + delay.o findbit.o memchr.o memcpy.o \ memmove.o memset.o memzero.o setbit.o \ strncpy_from_user.o strnlen_user.o \ strchr.o strrchr.o \ testchangebit.o testclearbit.o testsetbit.o \ - getuser.o putuser.o clear_user.o \ ashldi3.o ashrdi3.o lshrdi3.o muldi3.o \ ucmpdi2.o lib1funcs.o div64.o sha1.o \ io-readsb.o io-writesb.o io-readsl.o io-writesl.o +mmu-y := clear_user.o copy_page.o getuser.o putuser.o + # the code in uaccess.S is not preemption safe and # probably faster on ARMv3 only ifeq ($(CONFIG_PREEMPT),y) - lib-y += copy_from_user.o copy_to_user.o + mmu-y += copy_from_user.o copy_to_user.o else ifneq ($(CONFIG_CPU_32v3),y) - lib-y += copy_from_user.o copy_to_user.o + mmu-y += copy_from_user.o copy_to_user.o else - lib-y += uaccess.o + mmu-y += uaccess.o endif endif +lib-$(CONFIG_MMU) += $(mmu-y) + ifeq ($(CONFIG_CPU_32v3),y) lib-y += io-readsw-armv3.o io-writesw-armv3.o else diff --git a/include/asm-arm/uaccess.h b/include/asm-arm/uaccess.h index f909dc75301..87aba57a66c 100644 --- a/include/asm-arm/uaccess.h +++ b/include/asm-arm/uaccess.h @@ -40,16 +40,25 @@ struct exception_table_entry extern int fixup_exception(struct pt_regs *regs); +/* + * These two are intentionally not defined anywhere - if the kernel + * code generates any references to them, that's a bug. + */ +extern int __get_user_bad(void); +extern int __put_user_bad(void); + /* * Note that this is actually 0x1,0000,0000 */ #define KERNEL_DS 0x00000000 -#define USER_DS TASK_SIZE - #define get_ds() (KERNEL_DS) + +#ifdef CONFIG_MMU + +#define USER_DS TASK_SIZE #define get_fs() (current_thread_info()->addr_limit) -static inline void set_fs (mm_segment_t fs) +static inline void set_fs(mm_segment_t fs) { current_thread_info()->addr_limit = fs; modify_domain(DOMAIN_KERNEL, fs ? DOMAIN_CLIENT : DOMAIN_MANAGER); @@ -75,8 +84,6 @@ static inline void set_fs (mm_segment_t fs) : "cc"); \ flag; }) -#define access_ok(type,addr,size) (__range_ok(addr,size) == 0) - /* * Single-value transfer routines. They automatically use the right * size if we just have the right pointer type. Note that the functions @@ -87,20 +94,10 @@ static inline void set_fs (mm_segment_t fs) * fixup code, but there are a few places where it intrudes on the * main code path. When we only write to user space, there is no * problem. - * - * The "__xxx" versions of the user access functions do not verify the - * address space - it must have been done previously with a separate - * "access_ok()" call. - * - * The "xxx_error" versions set the third argument to EFAULT if an - * error occurs, and leave it unchanged on success. Note that these - * versions are void (ie, don't return a value as such). */ - extern int __get_user_1(void *); extern int __get_user_2(void *); extern int __get_user_4(void *); -extern int __get_user_bad(void); #define __get_user_x(__r2,__p,__e,__s,__i...) \ __asm__ __volatile__ ( \ @@ -131,6 +128,74 @@ extern int __get_user_bad(void); __e; \ }) +extern int __put_user_1(void *, unsigned int); +extern int __put_user_2(void *, unsigned int); +extern int __put_user_4(void *, unsigned int); +extern int __put_user_8(void *, unsigned long long); + +#define __put_user_x(__r2,__p,__e,__s) \ + __asm__ __volatile__ ( \ + __asmeq("%0", "r0") __asmeq("%2", "r2") \ + "bl __put_user_" #__s \ + : "=&r" (__e) \ + : "0" (__p), "r" (__r2) \ + : "ip", "lr", "cc") + +#define put_user(x,p) \ + ({ \ + const register typeof(*(p)) __r2 asm("r2") = (x); \ + const register typeof(*(p)) __user *__p asm("r0") = (p);\ + register int __e asm("r0"); \ + switch (sizeof(*(__p))) { \ + case 1: \ + __put_user_x(__r2, __p, __e, 1); \ + break; \ + case 2: \ + __put_user_x(__r2, __p, __e, 2); \ + break; \ + case 4: \ + __put_user_x(__r2, __p, __e, 4); \ + break; \ + case 8: \ + __put_user_x(__r2, __p, __e, 8); \ + break; \ + default: __e = __put_user_bad(); break; \ + } \ + __e; \ + }) + +#else /* CONFIG_MMU */ + +/* + * uClinux has only one addr space, so has simplified address limits. + */ +#define USER_DS KERNEL_DS + +#define segment_eq(a,b) (1) +#define __addr_ok(addr) (1) +#define __range_ok(addr,size) (0) +#define get_fs() (KERNEL_DS) + +static inline void set_fs(mm_segment_t fs) +{ +} + +#define get_user(x,p) __get_user(x,p) +#define put_user(x,p) __put_user(x,p) + +#endif /* CONFIG_MMU */ + +#define access_ok(type,addr,size) (__range_ok(addr,size) == 0) + +/* + * The "__xxx" versions of the user access functions do not verify the + * address space - it must have been done previously with a separate + * "access_ok()" call. + * + * The "xxx_error" versions set the third argument to EFAULT if an + * error occurs, and leave it unchanged on success. Note that these + * versions are void (ie, don't return a value as such). + */ #define __get_user(x,ptr) \ ({ \ long __gu_err = 0; \ @@ -212,43 +277,6 @@ do { \ : "r" (addr), "i" (-EFAULT) \ : "cc") -extern int __put_user_1(void *, unsigned int); -extern int __put_user_2(void *, unsigned int); -extern int __put_user_4(void *, unsigned int); -extern int __put_user_8(void *, unsigned long long); -extern int __put_user_bad(void); - -#define __put_user_x(__r2,__p,__e,__s) \ - __asm__ __volatile__ ( \ - __asmeq("%0", "r0") __asmeq("%2", "r2") \ - "bl __put_user_" #__s \ - : "=&r" (__e) \ - : "0" (__p), "r" (__r2) \ - : "ip", "lr", "cc") - -#define put_user(x,p) \ - ({ \ - const register typeof(*(p)) __r2 asm("r2") = (x); \ - const register typeof(*(p)) __user *__p asm("r0") = (p);\ - register int __e asm("r0"); \ - switch (sizeof(*(__p))) { \ - case 1: \ - __put_user_x(__r2, __p, __e, 1); \ - break; \ - case 2: \ - __put_user_x(__r2, __p, __e, 2); \ - break; \ - case 4: \ - __put_user_x(__r2, __p, __e, 4); \ - break; \ - case 8: \ - __put_user_x(__r2, __p, __e, 8); \ - break; \ - default: __e = __put_user_bad(); break; \ - } \ - __e; \ - }) - #define __put_user(x,ptr) \ ({ \ long __pu_err = 0; \ @@ -354,9 +382,16 @@ do { \ : "cc") +#ifdef CONFIG_MMU extern unsigned long __copy_from_user(void *to, const void __user *from, unsigned long n); extern unsigned long __copy_to_user(void __user *to, const void *from, unsigned long n); extern unsigned long __clear_user(void __user *addr, unsigned long n); +#else +#define __copy_from_user(to,from,n) (memcpy(to, (void __force *)from, n), 0) +#define __copy_to_user(to,from,n) (memcpy((void __force *)to, from, n), 0) +#define __clear_user(addr,n) (memset((void __force *)addr, 0, n), 0) +#endif + extern unsigned long __strncpy_from_user(char *to, const char __user *from, unsigned long count); extern unsigned long __strnlen_user(const char __user *s, long n); -- cgit v1.2.3-70-g09d2 From 888e7bf166a0059480da137f3bd28dcd51175f3d Mon Sep 17 00:00:00 2001 From: Russell King Date: Sat, 24 Jun 2006 17:20:13 +0100 Subject: [ARM] Remove TABLE_SIZE, and several unused function prototypes TABLE_SIZE is never used in arch/arm/mm/init.c. create_memmap_holes(), memtable_init, and setup_io_desc() no longer exist in the kernel. Signed-off-by: Russell King --- arch/arm/mm/init.c | 2 -- include/asm-arm/mach/map.h | 5 ----- 2 files changed, 7 deletions(-) (limited to 'include') diff --git a/arch/arm/mm/init.c b/arch/arm/mm/init.c index 9ea1f87a707..989fd681c82 100644 --- a/arch/arm/mm/init.c +++ b/arch/arm/mm/init.c @@ -26,8 +26,6 @@ #include #include -#define TABLE_SIZE (2 * PTRS_PER_PTE * sizeof(pte_t)) - DEFINE_PER_CPU(struct mmu_gather, mmu_gathers); extern pgd_t swapper_pg_dir[PTRS_PER_PGD]; diff --git a/include/asm-arm/mach/map.h b/include/asm-arm/mach/map.h index e8ea67c97c7..c3929fc74af 100644 --- a/include/asm-arm/mach/map.h +++ b/include/asm-arm/mach/map.h @@ -16,8 +16,6 @@ struct map_desc { unsigned int type; }; -struct meminfo; - #define MT_DEVICE 0 #define MT_CACHECLEAN 1 #define MT_MINICLEAN 2 @@ -28,7 +26,4 @@ struct meminfo; #define MT_IXP2000_DEVICE 7 #define MT_NONSHARED_DEVICE 8 -extern void create_memmap_holes(struct meminfo *); -extern void memtable_init(struct meminfo *); extern void iotable_init(struct map_desc *, int); -extern void setup_io_desc(void); -- cgit v1.2.3-70-g09d2 From 092c1952e18fec06f3a951113fe1d87492d6e840 Mon Sep 17 00:00:00 2001 From: Russell King Date: Sat, 24 Jun 2006 17:34:50 +0100 Subject: [ARM] nommu: remove fault-armv, mmap and mm-armv files from nommu build Remove fault-armv.o, mmap.o and mm-armv.o from uclinux builds - these are concerned with MMU-ful operations, and as such are redundant for uclinux. Since this also removes iotable_init() and iotable_init() is used extensively in the platform support files, just make it a no-op. Based upon a couple of patches by Hyok. Signed-off-by: Hyok S. Choi Signed-off-by: Russell King --- arch/arm/mm/Makefile | 8 ++++---- include/asm-arm/mach/map.h | 4 ++++ 2 files changed, 8 insertions(+), 4 deletions(-) (limited to 'include') diff --git a/arch/arm/mm/Makefile b/arch/arm/mm/Makefile index ddce0387898..21a2770226e 100644 --- a/arch/arm/mm/Makefile +++ b/arch/arm/mm/Makefile @@ -2,11 +2,11 @@ # Makefile for the linux arm-specific parts of the memory manager. # -obj-y := consistent.o extable.o fault-armv.o \ - fault.o init.o iomap.o mmap.o \ - mm-armv.o +obj-y := consistent.o extable.o fault.o init.o \ + iomap.o -obj-$(CONFIG_MMU) += flush.o ioremap.o +obj-$(CONFIG_MMU) += fault-armv.o flush.o ioremap.o mmap.o \ + mm-armv.o ifneq ($(CONFIG_MMU),y) obj-y += nommu.o diff --git a/include/asm-arm/mach/map.h b/include/asm-arm/mach/map.h index c3929fc74af..cef5364ed5f 100644 --- a/include/asm-arm/mach/map.h +++ b/include/asm-arm/mach/map.h @@ -26,4 +26,8 @@ struct map_desc { #define MT_IXP2000_DEVICE 7 #define MT_NONSHARED_DEVICE 8 +#ifdef CONFIG_MMU extern void iotable_init(struct map_desc *, int); +#else +#define iotable_init(map,num) do { } while (0) +#endif -- cgit v1.2.3-70-g09d2 From 02ada9c112eb00ecb9d2c04b6132f48ade534dbf Mon Sep 17 00:00:00 2001 From: Takashi Iwai Date: Fri, 23 Jun 2006 15:13:57 +0200 Subject: [ALSA] Suppress irq handler mismatch messages in ALSA ISA drivers Suppress 'irq handler mismatch' messages at auto-probing of irqs in ALSA ISA drivers. Signed-off-by: Takashi Iwai Signed-off-by: Jaroslav Kysela --- include/sound/initval.h | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'include') diff --git a/include/sound/initval.h b/include/sound/initval.h index d29e3d31d14..d45170b9e0b 100644 --- a/include/sound/initval.h +++ b/include/sound/initval.h @@ -62,7 +62,8 @@ static int snd_legacy_find_free_irq(int *irq_table) { while (*irq_table != -1) { if (!request_irq(*irq_table, snd_legacy_empty_irq_handler, - SA_INTERRUPT, "ALSA Test IRQ", (void *) irq_table)) { + SA_INTERRUPT | SA_PROBEIRQ, "ALSA Test IRQ", + (void *) irq_table)) { free_irq(*irq_table, (void *) irq_table); return *irq_table; } -- cgit v1.2.3-70-g09d2 From 02856b5684677b74095069c3be4774c2992e4fdc Mon Sep 17 00:00:00 2001 From: Jaya Kumar Date: Fri, 23 Jun 2006 15:18:41 +0200 Subject: [ALSA] AD1888 mixer controls for DC mode This patch adds two mixer controls. The V_REFOUT enable is a documented register that couples the microphone input lines to the V_REFOUT DC source. The High Pass Filter enable in the AC97_AD_TEST2 (0x5c) is an undocumented register provided by Miller Puckette via Analog Devices that enables the AD codec to apply a high pass filter to the input. Signed-off-by: Jaya Kumar Signed-off-by: Takashi Iwai Signed-off-by: Jaroslav Kysela --- include/sound/ac97_codec.h | 1 + sound/pci/ac97/ac97_patch.c | 2 ++ 2 files changed, 3 insertions(+) (limited to 'include') diff --git a/include/sound/ac97_codec.h b/include/sound/ac97_codec.h index 446afc3ea27..758f8bf133c 100644 --- a/include/sound/ac97_codec.h +++ b/include/sound/ac97_codec.h @@ -265,6 +265,7 @@ /* specific - Analog Devices */ #define AC97_AD_TEST 0x5a /* test register */ +#define AC97_AD_TEST2 0x5c /* undocumented test register 2 */ #define AC97_AD_CODEC_CFG 0x70 /* codec configuration */ #define AC97_AD_JACK_SPDIF 0x72 /* Jack Sense & S/PDIF */ #define AC97_AD_SERIAL_CFG 0x74 /* Serial Configuration */ diff --git a/sound/pci/ac97/ac97_patch.c b/sound/pci/ac97/ac97_patch.c index 7f197c78081..094cfc1f3a1 100644 --- a/sound/pci/ac97/ac97_patch.c +++ b/sound/pci/ac97/ac97_patch.c @@ -1824,6 +1824,8 @@ static const struct snd_kcontrol_new snd_ac97_ad1888_controls[] = { .get = snd_ac97_ad1888_lohpsel_get, .put = snd_ac97_ad1888_lohpsel_put }, + AC97_SINGLE("V_REFOUT Enable", AC97_AD_MISC, 2, 1, 1), + AC97_SINGLE("High Pass Filter Enable", AC97_AD_TEST2, 12, 1, 1), AC97_SINGLE("Spread Front to Surround and Center/LFE", AC97_AD_MISC, 7, 1, 0), { .iface = SNDRV_CTL_ELEM_IFACE_MIXER, -- cgit v1.2.3-70-g09d2 From c83c0c470565a0aed2f6fcbaa6c80a98ef250586 Mon Sep 17 00:00:00 2001 From: Jani Alinikula Date: Tue, 27 Jun 2006 15:00:55 +0200 Subject: [ALSA] Stereo controls for M-Audio Revolution cards This patch adds stereo controls to revo cards by making the ak4xxx driver mixers configurable from the card driver. Signed-off-by: Jani Alinikula Signed-off-by: Takashi Iwai Signed-off-by: Jaroslav Kysela --- include/sound/ak4xxx-adda.h | 2 + sound/i2c/other/ak4xxx-adda.c | 89 ++++++++++++++++++++++++++++++++++++++++--- sound/pci/ice1712/revo.c | 23 +++++++++-- 3 files changed, 105 insertions(+), 9 deletions(-) (limited to 'include') diff --git a/include/sound/ak4xxx-adda.h b/include/sound/ak4xxx-adda.h index 3bf5911fe82..acf8cc1292e 100644 --- a/include/sound/ak4xxx-adda.h +++ b/include/sound/ak4xxx-adda.h @@ -53,6 +53,8 @@ struct snd_akm4xxx { SND_AK4524, SND_AK4528, SND_AK4529, SND_AK4355, SND_AK4358, SND_AK4381 } type; + unsigned int *num_stereo; /* array of combined counts for the mixer */ + char **channel_names; /* array of mixer channel names */ struct snd_ak4xxx_ops ops; }; diff --git a/sound/i2c/other/ak4xxx-adda.c b/sound/i2c/other/ak4xxx-adda.c index 045e32a311e..f68bd740e1a 100644 --- a/sound/i2c/other/ak4xxx-adda.c +++ b/sound/i2c/other/ak4xxx-adda.c @@ -292,6 +292,64 @@ static int snd_akm4xxx_volume_put(struct snd_kcontrol *kcontrol, return change; } +static int snd_akm4xxx_stereo_volume_info(struct snd_kcontrol *kcontrol, + struct snd_ctl_elem_info *uinfo) +{ + unsigned int mask = AK_GET_MASK(kcontrol->private_value); + + uinfo->type = SNDRV_CTL_ELEM_TYPE_INTEGER; + uinfo->count = 2; + uinfo->value.integer.min = 0; + uinfo->value.integer.max = mask; + return 0; +} + +static int snd_akm4xxx_stereo_volume_get(struct snd_kcontrol *kcontrol, + struct snd_ctl_elem_value *ucontrol) +{ + struct snd_akm4xxx *ak = snd_kcontrol_chip(kcontrol); + int chip = AK_GET_CHIP(kcontrol->private_value); + int addr = AK_GET_ADDR(kcontrol->private_value); + int invert = AK_GET_INVERT(kcontrol->private_value); + unsigned int mask = AK_GET_MASK(kcontrol->private_value); + unsigned char val = snd_akm4xxx_get(ak, chip, addr); + + ucontrol->value.integer.value[0] = invert ? mask - val : val; + + val = snd_akm4xxx_get(ak, chip, addr+1); + ucontrol->value.integer.value[1] = invert ? mask - val : val; + + return 0; +} + +static int snd_akm4xxx_stereo_volume_put(struct snd_kcontrol *kcontrol, + struct snd_ctl_elem_value *ucontrol) +{ + struct snd_akm4xxx *ak = snd_kcontrol_chip(kcontrol); + int chip = AK_GET_CHIP(kcontrol->private_value); + int addr = AK_GET_ADDR(kcontrol->private_value); + int invert = AK_GET_INVERT(kcontrol->private_value); + unsigned int mask = AK_GET_MASK(kcontrol->private_value); + unsigned char nval = ucontrol->value.integer.value[0] % (mask+1); + int change0, change1; + + if (invert) + nval = mask - nval; + change0 = snd_akm4xxx_get(ak, chip, addr) != nval; + if (change0) + snd_akm4xxx_write(ak, chip, addr, nval); + + nval = ucontrol->value.integer.value[1] % (mask+1); + if (invert) + nval = mask - nval; + change1 = snd_akm4xxx_get(ak, chip, addr+1) != nval; + if (change1) + snd_akm4xxx_write(ak, chip, addr+1, nval); + + + return change0 || change1; +} + static int snd_akm4xxx_ipga_gain_info(struct snd_kcontrol *kcontrol, struct snd_ctl_elem_info *uinfo) { @@ -377,20 +435,35 @@ int snd_akm4xxx_build_controls(struct snd_akm4xxx *ak) unsigned int idx, num_emphs; struct snd_kcontrol *ctl; int err; + int mixer_ch = 0; + int num_stereo; ctl = kmalloc(sizeof(*ctl), GFP_KERNEL); if (! ctl) return -ENOMEM; - for (idx = 0; idx < ak->num_dacs; ++idx) { + for (idx = 0; idx < ak->num_dacs; ) { memset(ctl, 0, sizeof(*ctl)); - strcpy(ctl->id.name, "DAC Volume"); - ctl->id.index = idx + ak->idx_offset * 2; + if (ak->channel_names == NULL) { + strcpy(ctl->id.name, "DAC Volume"); + num_stereo = 1; + ctl->id.index = mixer_ch + ak->idx_offset * 2; + } else { + strcpy(ctl->id.name, ak->channel_names[mixer_ch]); + num_stereo = ak->num_stereo[mixer_ch]; + ctl->id.index = 0; //mixer_ch + ak->idx_offset * 2; + } ctl->id.iface = SNDRV_CTL_ELEM_IFACE_MIXER; ctl->count = 1; - ctl->info = snd_akm4xxx_volume_info; - ctl->get = snd_akm4xxx_volume_get; - ctl->put = snd_akm4xxx_volume_put; + if (num_stereo == 2) { + ctl->info = snd_akm4xxx_stereo_volume_info; + ctl->get = snd_akm4xxx_stereo_volume_get; + ctl->put = snd_akm4xxx_stereo_volume_put; + } else { + ctl->info = snd_akm4xxx_volume_info; + ctl->get = snd_akm4xxx_volume_get; + ctl->put = snd_akm4xxx_volume_put; + } switch (ak->type) { case SND_AK4524: ctl->private_value = AK_COMPOSE(idx/2, (idx%2) + 6, 0, 127); /* register 6 & 7 */ @@ -419,9 +492,13 @@ int snd_akm4xxx_build_controls(struct snd_akm4xxx *ak) err = -EINVAL; goto __error; } + ctl->private_data = ak; if ((err = snd_ctl_add(ak->card, snd_ctl_new(ctl, SNDRV_CTL_ELEM_ACCESS_READ|SNDRV_CTL_ELEM_ACCESS_WRITE))) < 0) goto __error; + + idx += num_stereo; + mixer_ch++; } for (idx = 0; idx < ak->num_adcs && ak->type == SND_AK4524; ++idx) { memset(ctl, 0, sizeof(*ctl)); diff --git a/sound/pci/ice1712/revo.c b/sound/pci/ice1712/revo.c index b5754b32b80..fec9440cb31 100644 --- a/sound/pci/ice1712/revo.c +++ b/sound/pci/ice1712/revo.c @@ -87,12 +87,25 @@ static void revo_set_rate_val(struct snd_akm4xxx *ak, unsigned int rate) * initialize the chips on M-Audio Revolution cards */ +static unsigned int revo71_num_stereo_front[] = {2}; +static char *revo71_channel_names_front[] = {"PCM Playback Volume"}; + +static unsigned int revo71_num_stereo_surround[] = {1, 1, 2, 2}; +static char *revo71_channel_names_surround[] = {"PCM Center Playback Volume", "PCM LFE Playback Volume", + "PCM Side Playback Volume", "PCM Rear Playback Volume"}; + +static unsigned int revo51_num_stereo[] = {2, 1, 1, 2}; +static char *revo51_channel_names[] = {"PCM Playback Volume", "PCM Center Playback Volume", + "PCM LFE Playback Volume", "PCM Rear Playback Volume"}; + static struct snd_akm4xxx akm_revo_front __devinitdata = { .type = SND_AK4381, .num_dacs = 2, .ops = { .set_rate_val = revo_set_rate_val - } + }, + .num_stereo = revo71_num_stereo_front, + .channel_names = revo71_channel_names_front }; static struct snd_ak4xxx_private akm_revo_front_priv __devinitdata = { @@ -113,7 +126,9 @@ static struct snd_akm4xxx akm_revo_surround __devinitdata = { .num_dacs = 6, .ops = { .set_rate_val = revo_set_rate_val - } + }, + .num_stereo = revo71_num_stereo_surround, + .channel_names = revo71_channel_names_surround }; static struct snd_ak4xxx_private akm_revo_surround_priv __devinitdata = { @@ -133,7 +148,9 @@ static struct snd_akm4xxx akm_revo51 __devinitdata = { .num_dacs = 6, .ops = { .set_rate_val = revo_set_rate_val - } + }, + .num_stereo = revo51_num_stereo, + .channel_names = revo51_channel_names }; static struct snd_ak4xxx_private akm_revo51_priv __devinitdata = { -- cgit v1.2.3-70-g09d2 From cb9d24e4349013628259b5fee97e692173731b07 Mon Sep 17 00:00:00 2001 From: Takashi Iwai Date: Tue, 27 Jun 2006 17:49:12 +0200 Subject: [ALSA] ak4xxx-adda - Code clean-up Fix spaces, fold lines to fit 80 columns in ak4xxx-adda driver codes. Split a long reset function to each codec routine just for better readability. Signed-off-by: Takashi Iwai Signed-off-by: Jaroslav Kysela --- include/sound/ak4xxx-adda.h | 37 +++++--- sound/i2c/other/ak4xxx-adda.c | 203 ++++++++++++++++++++++++++++-------------- 2 files changed, 161 insertions(+), 79 deletions(-) (limited to 'include') diff --git a/include/sound/ak4xxx-adda.h b/include/sound/ak4xxx-adda.h index acf8cc1292e..3d988849202 100644 --- a/include/sound/ak4xxx-adda.h +++ b/include/sound/ak4xxx-adda.h @@ -32,8 +32,8 @@ struct snd_akm4xxx; struct snd_ak4xxx_ops { void (*lock)(struct snd_akm4xxx *ak, int chip); void (*unlock)(struct snd_akm4xxx *ak, int chip); - void (*write)(struct snd_akm4xxx *ak, int chip, unsigned char reg, unsigned char val); - // unsigned char (*read)(struct snd_akm4xxx *ak, int chip, unsigned char reg); + void (*write)(struct snd_akm4xxx *ak, int chip, unsigned char reg, + unsigned char val); void (*set_rate_val)(struct snd_akm4xxx *ak, unsigned int rate); }; @@ -41,31 +41,40 @@ struct snd_ak4xxx_ops { struct snd_akm4xxx { struct snd_card *card; - unsigned int num_adcs; /* AK4524 or AK4528 ADCs */ - unsigned int num_dacs; /* AK4524 or AK4528 DACs */ - unsigned char images[AK4XXX_IMAGE_SIZE]; /* saved register image */ - unsigned char ipga_gain[AK4XXX_MAX_CHIPS][2]; /* saved register image for IPGA (AK4528) */ + unsigned int num_adcs; /* AK4524 or AK4528 ADCs */ + unsigned int num_dacs; /* AK4524 or AK4528 DACs */ + unsigned char images[AK4XXX_IMAGE_SIZE]; /* saved register image */ + unsigned char ipga_gain[AK4XXX_MAX_CHIPS][2]; /* saved register image + * for IPGA (AK4528) + */ unsigned long private_value[AK4XXX_MAX_CHIPS]; /* helper for driver */ void *private_data[AK4XXX_MAX_CHIPS]; /* helper for driver */ /* template should fill the following fields */ - unsigned int idx_offset; /* control index offset */ + unsigned int idx_offset; /* control index offset */ enum { SND_AK4524, SND_AK4528, SND_AK4529, SND_AK4355, SND_AK4358, SND_AK4381 } type; - unsigned int *num_stereo; /* array of combined counts for the mixer */ - char **channel_names; /* array of mixer channel names */ + unsigned int *num_stereo; /* array of combined counts + * for the mixer + */ + char **channel_names; /* array of mixer channel names */ struct snd_ak4xxx_ops ops; }; -void snd_akm4xxx_write(struct snd_akm4xxx *ak, int chip, unsigned char reg, unsigned char val); +void snd_akm4xxx_write(struct snd_akm4xxx *ak, int chip, unsigned char reg, + unsigned char val); void snd_akm4xxx_reset(struct snd_akm4xxx *ak, int state); void snd_akm4xxx_init(struct snd_akm4xxx *ak); int snd_akm4xxx_build_controls(struct snd_akm4xxx *ak); -#define snd_akm4xxx_get(ak,chip,reg) (ak)->images[(chip) * 16 + (reg)] -#define snd_akm4xxx_set(ak,chip,reg,val) ((ak)->images[(chip) * 16 + (reg)] = (val)) -#define snd_akm4xxx_get_ipga(ak,chip,reg) (ak)->ipga_gain[chip][(reg)-4] -#define snd_akm4xxx_set_ipga(ak,chip,reg,val) ((ak)->ipga_gain[chip][(reg)-4] = (val)) +#define snd_akm4xxx_get(ak,chip,reg) \ + (ak)->images[(chip) * 16 + (reg)] +#define snd_akm4xxx_set(ak,chip,reg,val) \ + ((ak)->images[(chip) * 16 + (reg)] = (val)) +#define snd_akm4xxx_get_ipga(ak,chip,reg) \ + (ak)->ipga_gain[chip][(reg)-4] +#define snd_akm4xxx_set_ipga(ak,chip,reg,val) \ + ((ak)->ipga_gain[chip][(reg)-4] = (val)) #endif /* __SOUND_AK4XXX_ADDA_H */ diff --git a/sound/i2c/other/ak4xxx-adda.c b/sound/i2c/other/ak4xxx-adda.c index f68bd740e1a..dc7cc2001b7 100644 --- a/sound/i2c/other/ak4xxx-adda.c +++ b/sound/i2c/other/ak4xxx-adda.c @@ -34,7 +34,8 @@ MODULE_AUTHOR("Jaroslav Kysela , Takashi Iwai "); MODULE_DESCRIPTION("Routines for control of AK452x / AK43xx AD/DA converters"); MODULE_LICENSE("GPL"); -void snd_akm4xxx_write(struct snd_akm4xxx *ak, int chip, unsigned char reg, unsigned char val) +void snd_akm4xxx_write(struct snd_akm4xxx *ak, int chip, unsigned char reg, + unsigned char val) { ak->ops.lock(ak, chip); ak->ops.write(ak, chip, reg, val); @@ -52,6 +53,67 @@ void snd_akm4xxx_write(struct snd_akm4xxx *ak, int chip, unsigned char reg, unsi ak->ops.unlock(ak, chip); } +EXPORT_SYMBOL(snd_akm4xxx_write); + +/* reset procedure for AK4524 and AK4528 */ +static void ak4524_reset(struct snd_akm4xxx *ak, int state) +{ + unsigned int chip; + unsigned char reg, maxreg; + + if (ak->type == SND_AK4528) + maxreg = 0x06; + else + maxreg = 0x08; + for (chip = 0; chip < ak->num_dacs/2; chip++) { + snd_akm4xxx_write(ak, chip, 0x01, state ? 0x00 : 0x03); + if (state) + continue; + /* DAC volumes */ + for (reg = 0x04; reg < maxreg; reg++) + snd_akm4xxx_write(ak, chip, reg, + snd_akm4xxx_get(ak, chip, reg)); + if (ak->type == SND_AK4528) + continue; + /* IPGA */ + for (reg = 0x04; reg < 0x06; reg++) + snd_akm4xxx_write(ak, chip, reg, + snd_akm4xxx_get_ipga(ak, chip, reg)); + } +} + +/* reset procedure for AK4355 and AK4358 */ +static void ak4355_reset(struct snd_akm4xxx *ak, int state) +{ + unsigned char reg; + + if (state) { + snd_akm4xxx_write(ak, 0, 0x01, 0x02); /* reset and soft-mute */ + return; + } + for (reg = 0x00; reg < 0x0b; reg++) + if (reg != 0x01) + snd_akm4xxx_write(ak, 0, reg, + snd_akm4xxx_get(ak, 0, reg)); + snd_akm4xxx_write(ak, 0, 0x01, 0x01); /* un-reset, unmute */ +} + +/* reset procedure for AK4381 */ +static void ak4381_reset(struct snd_akm4xxx *ak, int state) +{ + unsigned int chip; + unsigned char reg; + + for (chip = 0; chip < ak->num_dacs/2; chip++) { + snd_akm4xxx_write(ak, chip, 0x00, state ? 0x0c : 0x0f); + if (state) + continue; + for (reg = 0x01; reg < 0x05; reg++) + snd_akm4xxx_write(ak, chip, reg, + snd_akm4xxx_get(ak, chip, reg)); + } +} + /* * reset the AKM codecs * @state: 1 = reset codec, 0 = restore the registers @@ -60,52 +122,26 @@ void snd_akm4xxx_write(struct snd_akm4xxx *ak, int chip, unsigned char reg, unsi */ void snd_akm4xxx_reset(struct snd_akm4xxx *ak, int state) { - unsigned int chip; - unsigned char reg; - switch (ak->type) { case SND_AK4524: case SND_AK4528: - for (chip = 0; chip < ak->num_dacs/2; chip++) { - snd_akm4xxx_write(ak, chip, 0x01, state ? 0x00 : 0x03); - if (state) - continue; - /* DAC volumes */ - for (reg = 0x04; reg < (ak->type == SND_AK4528 ? 0x06 : 0x08); reg++) - snd_akm4xxx_write(ak, chip, reg, snd_akm4xxx_get(ak, chip, reg)); - if (ak->type == SND_AK4528) - continue; - /* IPGA */ - for (reg = 0x04; reg < 0x06; reg++) - snd_akm4xxx_write(ak, chip, reg, snd_akm4xxx_get_ipga(ak, chip, reg)); - } + ak4524_reset(ak, state); break; case SND_AK4529: /* FIXME: needed for ak4529? */ break; case SND_AK4355: case SND_AK4358: - if (state) { - snd_akm4xxx_write(ak, 0, 0x01, 0x02); /* reset and soft-mute */ - return; - } - for (reg = 0x00; reg < 0x0b; reg++) - if (reg != 0x01) - snd_akm4xxx_write(ak, 0, reg, snd_akm4xxx_get(ak, 0, reg)); - snd_akm4xxx_write(ak, 0, 0x01, 0x01); /* un-reset, unmute */ + ak4355_reset(ak, state); break; case SND_AK4381: - for (chip = 0; chip < ak->num_dacs/2; chip++) { - snd_akm4xxx_write(ak, chip, 0x00, state ? 0x0c : 0x0f); - if (state) - continue; - for (reg = 0x01; reg < 0x05; reg++) - snd_akm4xxx_write(ak, chip, reg, snd_akm4xxx_get(ak, chip, reg)); - } + ak4381_reset(ak, state); break; } } +EXPORT_SYMBOL(snd_akm4xxx_reset); + /* * initialize all the ak4xxx chips */ @@ -153,7 +189,8 @@ void snd_akm4xxx_init(struct snd_akm4xxx *ak) }; static unsigned char inits_ak4355[] = { 0x01, 0x02, /* 1: reset and soft-mute */ - 0x00, 0x06, /* 0: mode3(i2s), disable auto-clock detect, disable DZF, sharp roll-off, RSTN#=0 */ + 0x00, 0x06, /* 0: mode3(i2s), disable auto-clock detect, + * disable DZF, sharp roll-off, RSTN#=0 */ 0x02, 0x0e, /* 2: DA's power up, normal speed, RSTN#=0 */ // 0x02, 0x2e, /* quad speed */ 0x03, 0x01, /* 3: de-emphasis off */ @@ -169,7 +206,8 @@ void snd_akm4xxx_init(struct snd_akm4xxx *ak) }; static unsigned char inits_ak4358[] = { 0x01, 0x02, /* 1: reset and soft-mute */ - 0x00, 0x06, /* 0: mode3(i2s), disable auto-clock detect, disable DZF, sharp roll-off, RSTN#=0 */ + 0x00, 0x06, /* 0: mode3(i2s), disable auto-clock detect, + * disable DZF, sharp roll-off, RSTN#=0 */ 0x02, 0x0e, /* 2: DA's power up, normal speed, RSTN#=0 */ // 0x02, 0x2e, /* quad speed */ 0x03, 0x01, /* 3: de-emphasis off */ @@ -187,7 +225,8 @@ void snd_akm4xxx_init(struct snd_akm4xxx *ak) }; static unsigned char inits_ak4381[] = { 0x00, 0x0c, /* 0: mode3(i2s), disable auto-clock detect */ - 0x01, 0x02, /* 1: de-emphasis off, normal speed, sharp roll-off, DZF off */ + 0x01, 0x02, /* 1: de-emphasis off, normal speed, + * sharp roll-off, DZF off */ // 0x01, 0x12, /* quad speed */ 0x02, 0x00, /* 2: DZF disabled */ 0x03, 0x00, /* 3: LATT 0 */ @@ -239,12 +278,15 @@ void snd_akm4xxx_init(struct snd_akm4xxx *ak) } } +EXPORT_SYMBOL(snd_akm4xxx_init); + #define AK_GET_CHIP(val) (((val) >> 8) & 0xff) #define AK_GET_ADDR(val) ((val) & 0xff) #define AK_GET_SHIFT(val) (((val) >> 16) & 0x7f) #define AK_GET_INVERT(val) (((val) >> 23) & 1) #define AK_GET_MASK(val) (((val) >> 24) & 0xff) -#define AK_COMPOSE(chip,addr,shift,mask) (((chip) << 8) | (addr) | ((shift) << 16) | ((mask) << 24)) +#define AK_COMPOSE(chip,addr,shift,mask) \ + (((chip) << 8) | (addr) | ((shift) << 16) | ((mask) << 24)) #define AK_INVERT (1<<23) static int snd_akm4xxx_volume_info(struct snd_kcontrol *kcontrol, @@ -293,7 +335,7 @@ static int snd_akm4xxx_volume_put(struct snd_kcontrol *kcontrol, } static int snd_akm4xxx_stereo_volume_info(struct snd_kcontrol *kcontrol, - struct snd_ctl_elem_info *uinfo) + struct snd_ctl_elem_info *uinfo) { unsigned int mask = AK_GET_MASK(kcontrol->private_value); @@ -305,7 +347,7 @@ static int snd_akm4xxx_stereo_volume_info(struct snd_kcontrol *kcontrol, } static int snd_akm4xxx_stereo_volume_get(struct snd_kcontrol *kcontrol, - struct snd_ctl_elem_value *ucontrol) + struct snd_ctl_elem_value *ucontrol) { struct snd_akm4xxx *ak = snd_kcontrol_chip(kcontrol); int chip = AK_GET_CHIP(kcontrol->private_value); @@ -323,7 +365,7 @@ static int snd_akm4xxx_stereo_volume_get(struct snd_kcontrol *kcontrol, } static int snd_akm4xxx_stereo_volume_put(struct snd_kcontrol *kcontrol, - struct snd_ctl_elem_value *ucontrol) + struct snd_ctl_elem_value *ucontrol) { struct snd_akm4xxx *ak = snd_kcontrol_chip(kcontrol); int chip = AK_GET_CHIP(kcontrol->private_value); @@ -366,7 +408,8 @@ static int snd_akm4xxx_ipga_gain_get(struct snd_kcontrol *kcontrol, struct snd_akm4xxx *ak = snd_kcontrol_chip(kcontrol); int chip = AK_GET_CHIP(kcontrol->private_value); int addr = AK_GET_ADDR(kcontrol->private_value); - ucontrol->value.integer.value[0] = snd_akm4xxx_get_ipga(ak, chip, addr) & 0x7f; + ucontrol->value.integer.value[0] = + snd_akm4xxx_get_ipga(ak, chip, addr) & 0x7f; return 0; } @@ -394,7 +437,8 @@ static int snd_akm4xxx_deemphasis_info(struct snd_kcontrol *kcontrol, uinfo->value.enumerated.items = 4; if (uinfo->value.enumerated.item >= 4) uinfo->value.enumerated.item = 3; - strcpy(uinfo->value.enumerated.name, texts[uinfo->value.enumerated.item]); + strcpy(uinfo->value.enumerated.name, + texts[uinfo->value.enumerated.item]); return 0; } @@ -405,7 +449,8 @@ static int snd_akm4xxx_deemphasis_get(struct snd_kcontrol *kcontrol, int chip = AK_GET_CHIP(kcontrol->private_value); int addr = AK_GET_ADDR(kcontrol->private_value); int shift = AK_GET_SHIFT(kcontrol->private_value); - ucontrol->value.enumerated.item[0] = (snd_akm4xxx_get(ak, chip, addr) >> shift) & 3; + ucontrol->value.enumerated.item[0] = + (snd_akm4xxx_get(ak, chip, addr) >> shift) & 3; return 0; } @@ -419,7 +464,8 @@ static int snd_akm4xxx_deemphasis_put(struct snd_kcontrol *kcontrol, unsigned char nval = ucontrol->value.enumerated.item[0] & 3; int change; - nval = (nval << shift) | (snd_akm4xxx_get(ak, chip, addr) & ~(3 << shift)); + nval = (nval << shift) | + (snd_akm4xxx_get(ak, chip, addr) & ~(3 << shift)); change = snd_akm4xxx_get(ak, chip, addr) != nval; if (change) snd_akm4xxx_write(ak, chip, addr, nval); @@ -451,7 +497,7 @@ int snd_akm4xxx_build_controls(struct snd_akm4xxx *ak) } else { strcpy(ctl->id.name, ak->channel_names[mixer_ch]); num_stereo = ak->num_stereo[mixer_ch]; - ctl->id.index = 0; //mixer_ch + ak->idx_offset * 2; + ctl->id.index = 0; } ctl->id.iface = SNDRV_CTL_ELEM_IFACE_MIXER; ctl->count = 1; @@ -466,27 +512,40 @@ int snd_akm4xxx_build_controls(struct snd_akm4xxx *ak) } switch (ak->type) { case SND_AK4524: - ctl->private_value = AK_COMPOSE(idx/2, (idx%2) + 6, 0, 127); /* register 6 & 7 */ + /* register 6 & 7 */ + ctl->private_value = + AK_COMPOSE(idx/2, (idx%2) + 6, 0, 127); break; case SND_AK4528: - ctl->private_value = AK_COMPOSE(idx/2, (idx%2) + 4, 0, 127); /* register 4 & 5 */ + /* register 4 & 5 */ + ctl->private_value = + AK_COMPOSE(idx/2, (idx%2) + 4, 0, 127); break; case SND_AK4529: { - int val = idx < 6 ? idx + 2 : (idx - 6) + 0xb; /* registers 2-7 and b,c */ - ctl->private_value = AK_COMPOSE(0, val, 0, 255) | AK_INVERT; + /* registers 2-7 and b,c */ + int val = idx < 6 ? idx + 2 : (idx - 6) + 0xb; + ctl->private_value = + AK_COMPOSE(0, val, 0, 255) | AK_INVERT; break; } case SND_AK4355: - ctl->private_value = AK_COMPOSE(0, idx + 4, 0, 255); /* register 4-9, chip #0 only */ + /* register 4-9, chip #0 only */ + ctl->private_value = AK_COMPOSE(0, idx + 4, 0, 255); break; case SND_AK4358: if (idx >= 6) - ctl->private_value = AK_COMPOSE(0, idx + 5, 0, 255); /* register 4-9, chip #0 only */ + /* register 4-9, chip #0 only */ + ctl->private_value = + AK_COMPOSE(0, idx + 5, 0, 255); else - ctl->private_value = AK_COMPOSE(0, idx + 4, 0, 255); /* register 4-9, chip #0 only */ + /* register 4-9, chip #0 only */ + ctl->private_value = + AK_COMPOSE(0, idx + 4, 0, 255); break; case SND_AK4381: - ctl->private_value = AK_COMPOSE(idx/2, (idx%2) + 3, 0, 255); /* register 3 & 4 */ + /* register 3 & 4 */ + ctl->private_value = + AK_COMPOSE(idx/2, (idx%2) + 3, 0, 255); break; default: err = -EINVAL; @@ -494,7 +553,10 @@ int snd_akm4xxx_build_controls(struct snd_akm4xxx *ak) } ctl->private_data = ak; - if ((err = snd_ctl_add(ak->card, snd_ctl_new(ctl, SNDRV_CTL_ELEM_ACCESS_READ|SNDRV_CTL_ELEM_ACCESS_WRITE))) < 0) + err = snd_ctl_add(ak->card, + snd_ctl_new(ctl, SNDRV_CTL_ELEM_ACCESS_READ| + SNDRV_CTL_ELEM_ACCESS_WRITE)); + if (err < 0) goto __error; idx += num_stereo; @@ -509,9 +571,14 @@ int snd_akm4xxx_build_controls(struct snd_akm4xxx *ak) ctl->info = snd_akm4xxx_volume_info; ctl->get = snd_akm4xxx_volume_get; ctl->put = snd_akm4xxx_volume_put; - ctl->private_value = AK_COMPOSE(idx/2, (idx%2) + 4, 0, 127); /* register 4 & 5 */ + /* register 4 & 5 */ + ctl->private_value = + AK_COMPOSE(idx/2, (idx%2) + 4, 0, 127); ctl->private_data = ak; - if ((err = snd_ctl_add(ak->card, snd_ctl_new(ctl, SNDRV_CTL_ELEM_ACCESS_READ|SNDRV_CTL_ELEM_ACCESS_WRITE))) < 0) + err = snd_ctl_add(ak->card, + snd_ctl_new(ctl, SNDRV_CTL_ELEM_ACCESS_READ| + SNDRV_CTL_ELEM_ACCESS_WRITE)); + if (err < 0) goto __error; memset(ctl, 0, sizeof(*ctl)); @@ -522,9 +589,13 @@ int snd_akm4xxx_build_controls(struct snd_akm4xxx *ak) ctl->info = snd_akm4xxx_ipga_gain_info; ctl->get = snd_akm4xxx_ipga_gain_get; ctl->put = snd_akm4xxx_ipga_gain_put; - ctl->private_value = AK_COMPOSE(idx/2, (idx%2) + 4, 0, 0); /* register 4 & 5 */ + /* register 4 & 5 */ + ctl->private_value = AK_COMPOSE(idx/2, (idx%2) + 4, 0, 0); ctl->private_data = ak; - if ((err = snd_ctl_add(ak->card, snd_ctl_new(ctl, SNDRV_CTL_ELEM_ACCESS_READ|SNDRV_CTL_ELEM_ACCESS_WRITE))) < 0) + err = snd_ctl_add(ak->card, + snd_ctl_new(ctl, SNDRV_CTL_ELEM_ACCESS_READ| + SNDRV_CTL_ELEM_ACCESS_WRITE)); + if (err < 0) goto __error; } if (ak->type == SND_AK4355 || ak->type == SND_AK4358) @@ -543,11 +614,13 @@ int snd_akm4xxx_build_controls(struct snd_akm4xxx *ak) switch (ak->type) { case SND_AK4524: case SND_AK4528: - ctl->private_value = AK_COMPOSE(idx, 3, 0, 0); /* register 3 */ + /* register 3 */ + ctl->private_value = AK_COMPOSE(idx, 3, 0, 0); break; case SND_AK4529: { int shift = idx == 3 ? 6 : (2 - idx) * 2; - ctl->private_value = AK_COMPOSE(0, 8, shift, 0); /* register 8 with shift */ + /* register 8 with shift */ + ctl->private_value = AK_COMPOSE(0, 8, shift, 0); break; } case SND_AK4355: @@ -559,7 +632,10 @@ int snd_akm4xxx_build_controls(struct snd_akm4xxx *ak) break; } ctl->private_data = ak; - if ((err = snd_ctl_add(ak->card, snd_ctl_new(ctl, SNDRV_CTL_ELEM_ACCESS_READ|SNDRV_CTL_ELEM_ACCESS_WRITE))) < 0) + err = snd_ctl_add(ak->card, + snd_ctl_new(ctl, SNDRV_CTL_ELEM_ACCESS_READ| + SNDRV_CTL_ELEM_ACCESS_WRITE)); + if (err < 0) goto __error; } err = 0; @@ -569,6 +645,8 @@ int snd_akm4xxx_build_controls(struct snd_akm4xxx *ak) return err; } +EXPORT_SYMBOL(snd_akm4xxx_build_controls); + static int __init alsa_akm4xxx_module_init(void) { return 0; @@ -580,8 +658,3 @@ static void __exit alsa_akm4xxx_module_exit(void) module_init(alsa_akm4xxx_module_init) module_exit(alsa_akm4xxx_module_exit) - -EXPORT_SYMBOL(snd_akm4xxx_write); -EXPORT_SYMBOL(snd_akm4xxx_reset); -EXPORT_SYMBOL(snd_akm4xxx_init); -EXPORT_SYMBOL(snd_akm4xxx_build_controls); -- cgit v1.2.3-70-g09d2 From f5e54d6e53a20cef45af7499e86164f0e0d16bb2 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Wed, 28 Jun 2006 04:26:44 -0700 Subject: [PATCH] mark address_space_operations const Same as with already do with the file operations: keep them in .rodata and prevents people from doing runtime patching. Signed-off-by: Christoph Hellwig Cc: Steven French Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/block/loop.c | 4 ++-- drivers/block/rd.c | 2 +- fs/9p/v9fs_vfs.h | 2 +- fs/9p/vfs_addr.c | 2 +- fs/adfs/inode.c | 2 +- fs/affs/affs.h | 6 +++--- fs/affs/file.c | 4 ++-- fs/affs/symlink.c | 2 +- fs/afs/file.c | 2 +- fs/afs/internal.h | 2 +- fs/befs/linuxvfs.c | 2 +- fs/bfs/bfs.h | 2 +- fs/bfs/file.c | 2 +- fs/block_dev.c | 2 +- fs/buffer.c | 2 +- fs/cifs/cifsfs.h | 4 ++-- fs/cifs/file.c | 4 ++-- fs/coda/symlink.c | 2 +- fs/configfs/inode.c | 2 +- fs/cramfs/inode.c | 4 ++-- fs/efs/inode.c | 2 +- fs/efs/symlink.c | 2 +- fs/ext2/ext2.h | 6 +++--- fs/ext2/inode.c | 6 +++--- fs/ext3/inode.c | 6 +++--- fs/fat/inode.c | 2 +- fs/freevxfs/vxfs_immed.c | 2 +- fs/freevxfs/vxfs_inode.c | 6 +++--- fs/freevxfs/vxfs_subr.c | 2 +- fs/fuse/file.c | 2 +- fs/hfs/hfs_fs.h | 4 ++-- fs/hfs/inode.c | 4 ++-- fs/hfsplus/hfsplus_fs.h | 4 ++-- fs/hfsplus/inode.c | 4 ++-- fs/hostfs/hostfs_kern.c | 6 +++--- fs/hpfs/file.c | 2 +- fs/hpfs/hpfs_fn.h | 4 ++-- fs/hpfs/namei.c | 2 +- fs/hugetlbfs/inode.c | 4 ++-- fs/inode.c | 2 +- fs/isofs/compress.c | 2 +- fs/isofs/inode.c | 2 +- fs/isofs/isofs.h | 2 +- fs/isofs/rock.c | 2 +- fs/isofs/zisofs.h | 2 +- fs/jffs/inode-v23.c | 4 ++-- fs/jffs2/file.c | 2 +- fs/jffs2/os-linux.h | 2 +- fs/jfs/inode.c | 2 +- fs/jfs/jfs_inode.h | 2 +- fs/jfs/jfs_metapage.c | 2 +- fs/jfs/jfs_metapage.h | 2 +- fs/minix/inode.c | 2 +- fs/ncpfs/inode.c | 2 +- fs/ncpfs/symlink.c | 2 +- fs/nfs/file.c | 2 +- fs/ntfs/aops.c | 4 ++-- fs/ntfs/ntfs.h | 4 ++-- fs/ocfs2/aops.c | 2 +- fs/ocfs2/inode.h | 2 +- fs/qnx4/inode.c | 2 +- fs/ramfs/file-mmu.c | 2 +- fs/ramfs/file-nommu.c | 2 +- fs/ramfs/internal.h | 2 +- fs/reiserfs/inode.c | 2 +- fs/romfs/inode.c | 2 +- fs/smbfs/file.c | 2 +- fs/smbfs/proto.h | 2 +- fs/sysfs/inode.c | 2 +- fs/sysv/itree.c | 2 +- fs/sysv/sysv.h | 2 +- fs/udf/file.c | 2 +- fs/udf/inode.c | 2 +- fs/udf/symlink.c | 2 +- fs/udf/udfdecl.h | 6 +++--- fs/ufs/inode.c | 2 +- fs/xfs/linux-2.6/xfs_aops.c | 2 +- fs/xfs/linux-2.6/xfs_aops.h | 2 +- fs/xfs/linux-2.6/xfs_buf.c | 2 +- include/linux/coda_linux.h | 4 ++-- include/linux/efs_fs.h | 2 +- include/linux/fs.h | 4 ++-- include/linux/nfs_fs.h | 2 +- include/linux/reiserfs_fs.h | 2 +- include/linux/ufs_fs.h | 2 +- mm/filemap.c | 4 ++-- mm/filemap_xip.c | 2 +- mm/shmem.c | 4 ++-- mm/swap_state.c | 2 +- 89 files changed, 121 insertions(+), 121 deletions(-) (limited to 'include') diff --git a/drivers/block/loop.c b/drivers/block/loop.c index 3c74ea729fc..18dd026f470 100644 --- a/drivers/block/loop.c +++ b/drivers/block/loop.c @@ -210,7 +210,7 @@ static int do_lo_send_aops(struct loop_device *lo, struct bio_vec *bvec, { struct file *file = lo->lo_backing_file; /* kudos to NFsckingS */ struct address_space *mapping = file->f_mapping; - struct address_space_operations *aops = mapping->a_ops; + const struct address_space_operations *aops = mapping->a_ops; pgoff_t index; unsigned offset, bv_offs; int len, ret; @@ -784,7 +784,7 @@ static int loop_set_fd(struct loop_device *lo, struct file *lo_file, error = -EINVAL; if (S_ISREG(inode->i_mode) || S_ISBLK(inode->i_mode)) { - struct address_space_operations *aops = mapping->a_ops; + const struct address_space_operations *aops = mapping->a_ops; /* * If we can't read - sorry. If we only can't write - well, * it's going to be read-only. diff --git a/drivers/block/rd.c b/drivers/block/rd.c index 940bfd7951e..0378da04cfa 100644 --- a/drivers/block/rd.c +++ b/drivers/block/rd.c @@ -191,7 +191,7 @@ static int ramdisk_set_page_dirty(struct page *page) return 0; } -static struct address_space_operations ramdisk_aops = { +static const struct address_space_operations ramdisk_aops = { .readpage = ramdisk_readpage, .prepare_write = ramdisk_prepare_write, .commit_write = ramdisk_commit_write, diff --git a/fs/9p/v9fs_vfs.h b/fs/9p/v9fs_vfs.h index f867b8d3e97..450b0c1b385 100644 --- a/fs/9p/v9fs_vfs.h +++ b/fs/9p/v9fs_vfs.h @@ -38,7 +38,7 @@ */ extern struct file_system_type v9fs_fs_type; -extern struct address_space_operations v9fs_addr_operations; +extern const struct address_space_operations v9fs_addr_operations; extern const struct file_operations v9fs_file_operations; extern const struct file_operations v9fs_dir_operations; extern struct dentry_operations v9fs_dentry_operations; diff --git a/fs/9p/vfs_addr.c b/fs/9p/vfs_addr.c index efda46fb64d..d4f0aa3c87f 100644 --- a/fs/9p/vfs_addr.c +++ b/fs/9p/vfs_addr.c @@ -103,6 +103,6 @@ UnmapAndUnlock: return retval; } -struct address_space_operations v9fs_addr_operations = { +const struct address_space_operations v9fs_addr_operations = { .readpage = v9fs_vfs_readpage, }; diff --git a/fs/adfs/inode.c b/fs/adfs/inode.c index a02802a3079..534f3eecc98 100644 --- a/fs/adfs/inode.c +++ b/fs/adfs/inode.c @@ -72,7 +72,7 @@ static sector_t _adfs_bmap(struct address_space *mapping, sector_t block) return generic_block_bmap(mapping, block, adfs_get_block); } -static struct address_space_operations adfs_aops = { +static const struct address_space_operations adfs_aops = { .readpage = adfs_readpage, .writepage = adfs_writepage, .sync_page = block_sync_page, diff --git a/fs/affs/affs.h b/fs/affs/affs.h index a43a876742b..0ddd4cc0d1a 100644 --- a/fs/affs/affs.h +++ b/fs/affs/affs.h @@ -195,9 +195,9 @@ extern struct inode_operations affs_symlink_inode_operations; extern const struct file_operations affs_file_operations; extern const struct file_operations affs_file_operations_ofs; extern const struct file_operations affs_dir_operations; -extern struct address_space_operations affs_symlink_aops; -extern struct address_space_operations affs_aops; -extern struct address_space_operations affs_aops_ofs; +extern const struct address_space_operations affs_symlink_aops; +extern const struct address_space_operations affs_aops; +extern const struct address_space_operations affs_aops_ofs; extern struct dentry_operations affs_dentry_operations; extern struct dentry_operations affs_dentry_operations_intl; diff --git a/fs/affs/file.c b/fs/affs/file.c index 7076262af39..3de8590e4f6 100644 --- a/fs/affs/file.c +++ b/fs/affs/file.c @@ -406,7 +406,7 @@ static sector_t _affs_bmap(struct address_space *mapping, sector_t block) { return generic_block_bmap(mapping,block,affs_get_block); } -struct address_space_operations affs_aops = { +const struct address_space_operations affs_aops = { .readpage = affs_readpage, .writepage = affs_writepage, .sync_page = block_sync_page, @@ -759,7 +759,7 @@ out: goto done; } -struct address_space_operations affs_aops_ofs = { +const struct address_space_operations affs_aops_ofs = { .readpage = affs_readpage_ofs, //.writepage = affs_writepage_ofs, //.sync_page = affs_sync_page_ofs, diff --git a/fs/affs/symlink.c b/fs/affs/symlink.c index 426f0f094f2..f802256a593 100644 --- a/fs/affs/symlink.c +++ b/fs/affs/symlink.c @@ -66,7 +66,7 @@ fail: return err; } -struct address_space_operations affs_symlink_aops = { +const struct address_space_operations affs_symlink_aops = { .readpage = affs_symlink_readpage, }; diff --git a/fs/afs/file.c b/fs/afs/file.c index 7bb716887e2..67d6634101f 100644 --- a/fs/afs/file.c +++ b/fs/afs/file.c @@ -35,7 +35,7 @@ struct inode_operations afs_file_inode_operations = { .getattr = afs_inode_getattr, }; -struct address_space_operations afs_fs_aops = { +const struct address_space_operations afs_fs_aops = { .readpage = afs_file_readpage, .sync_page = block_sync_page, .set_page_dirty = __set_page_dirty_nobuffers, diff --git a/fs/afs/internal.h b/fs/afs/internal.h index 72febdf9a35..e88b3b65ae4 100644 --- a/fs/afs/internal.h +++ b/fs/afs/internal.h @@ -69,7 +69,7 @@ extern const struct file_operations afs_dir_file_operations; /* * file.c */ -extern struct address_space_operations afs_fs_aops; +extern const struct address_space_operations afs_fs_aops; extern struct inode_operations afs_file_inode_operations; #ifdef AFS_CACHING_SUPPORT diff --git a/fs/befs/linuxvfs.c b/fs/befs/linuxvfs.c index 08201fab26c..a83e889a97c 100644 --- a/fs/befs/linuxvfs.c +++ b/fs/befs/linuxvfs.c @@ -73,7 +73,7 @@ static struct inode_operations befs_dir_inode_operations = { .lookup = befs_lookup, }; -static struct address_space_operations befs_aops = { +static const struct address_space_operations befs_aops = { .readpage = befs_readpage, .sync_page = block_sync_page, .bmap = befs_bmap, diff --git a/fs/bfs/bfs.h b/fs/bfs/bfs.h index 9d791004b21..31973bbbf05 100644 --- a/fs/bfs/bfs.h +++ b/fs/bfs/bfs.h @@ -50,7 +50,7 @@ static inline struct bfs_inode_info *BFS_I(struct inode *inode) /* file.c */ extern struct inode_operations bfs_file_inops; extern const struct file_operations bfs_file_operations; -extern struct address_space_operations bfs_aops; +extern const struct address_space_operations bfs_aops; /* dir.c */ extern struct inode_operations bfs_dir_inops; diff --git a/fs/bfs/file.c b/fs/bfs/file.c index d83cd74a2e4..3d5aca28a0a 100644 --- a/fs/bfs/file.c +++ b/fs/bfs/file.c @@ -153,7 +153,7 @@ static sector_t bfs_bmap(struct address_space *mapping, sector_t block) return generic_block_bmap(mapping, block, bfs_get_block); } -struct address_space_operations bfs_aops = { +const struct address_space_operations bfs_aops = { .readpage = bfs_readpage, .writepage = bfs_writepage, .sync_page = block_sync_page, diff --git a/fs/block_dev.c b/fs/block_dev.c index 028d9fb9c2d..7f7600e2381 100644 --- a/fs/block_dev.c +++ b/fs/block_dev.c @@ -1095,7 +1095,7 @@ static long block_ioctl(struct file *file, unsigned cmd, unsigned long arg) return blkdev_ioctl(file->f_mapping->host, file, cmd, arg); } -struct address_space_operations def_blk_aops = { +const struct address_space_operations def_blk_aops = { .readpage = blkdev_readpage, .writepage = blkdev_writepage, .sync_page = block_sync_page, diff --git a/fs/buffer.c b/fs/buffer.c index f23bb647db4..e9994722f4a 100644 --- a/fs/buffer.c +++ b/fs/buffer.c @@ -2598,7 +2598,7 @@ int nobh_truncate_page(struct address_space *mapping, loff_t from) unsigned offset = from & (PAGE_CACHE_SIZE-1); unsigned to; struct page *page; - struct address_space_operations *a_ops = mapping->a_ops; + const struct address_space_operations *a_ops = mapping->a_ops; char *kaddr; int ret = 0; diff --git a/fs/cifs/cifsfs.h b/fs/cifs/cifsfs.h index a6384d83fde..8f75c6f2470 100644 --- a/fs/cifs/cifsfs.h +++ b/fs/cifs/cifsfs.h @@ -32,8 +32,8 @@ #define TRUE 1 #endif -extern struct address_space_operations cifs_addr_ops; -extern struct address_space_operations cifs_addr_ops_smallbuf; +extern const struct address_space_operations cifs_addr_ops; +extern const struct address_space_operations cifs_addr_ops_smallbuf; /* Functions related to super block operations */ extern struct super_operations cifs_super_ops; diff --git a/fs/cifs/file.c b/fs/cifs/file.c index e9c1573f6aa..5861eb42e62 100644 --- a/fs/cifs/file.c +++ b/fs/cifs/file.c @@ -1942,7 +1942,7 @@ static int cifs_prepare_write(struct file *file, struct page *page, return 0; } -struct address_space_operations cifs_addr_ops = { +const struct address_space_operations cifs_addr_ops = { .readpage = cifs_readpage, .readpages = cifs_readpages, .writepage = cifs_writepage, @@ -1959,7 +1959,7 @@ struct address_space_operations cifs_addr_ops = { * contain the header plus one complete page of data. Otherwise, we need * to leave cifs_readpages out of the address space operations. */ -struct address_space_operations cifs_addr_ops_smallbuf = { +const struct address_space_operations cifs_addr_ops_smallbuf = { .readpage = cifs_readpage, .writepage = cifs_writepage, .writepages = cifs_writepages, diff --git a/fs/coda/symlink.c b/fs/coda/symlink.c index b35e5bbd9c9..76e00a65a75 100644 --- a/fs/coda/symlink.c +++ b/fs/coda/symlink.c @@ -50,6 +50,6 @@ fail: return error; } -struct address_space_operations coda_symlink_aops = { +const struct address_space_operations coda_symlink_aops = { .readpage = coda_symlink_filler, }; diff --git a/fs/configfs/inode.c b/fs/configfs/inode.c index c153bd9534c..e14488ca641 100644 --- a/fs/configfs/inode.c +++ b/fs/configfs/inode.c @@ -38,7 +38,7 @@ extern struct super_block * configfs_sb; -static struct address_space_operations configfs_aops = { +static const struct address_space_operations configfs_aops = { .readpage = simple_readpage, .prepare_write = simple_prepare_write, .commit_write = simple_commit_write diff --git a/fs/cramfs/inode.c b/fs/cramfs/inode.c index c45d7386080..223c0431042 100644 --- a/fs/cramfs/inode.c +++ b/fs/cramfs/inode.c @@ -30,7 +30,7 @@ static struct super_operations cramfs_ops; static struct inode_operations cramfs_dir_inode_operations; static const struct file_operations cramfs_directory_operations; -static struct address_space_operations cramfs_aops; +static const struct address_space_operations cramfs_aops; static DEFINE_MUTEX(read_mutex); @@ -501,7 +501,7 @@ static int cramfs_readpage(struct file *file, struct page * page) return 0; } -static struct address_space_operations cramfs_aops = { +static const struct address_space_operations cramfs_aops = { .readpage = cramfs_readpage }; diff --git a/fs/efs/inode.c b/fs/efs/inode.c index 180607f9314..174696f9bf1 100644 --- a/fs/efs/inode.c +++ b/fs/efs/inode.c @@ -21,7 +21,7 @@ static sector_t _efs_bmap(struct address_space *mapping, sector_t block) { return generic_block_bmap(mapping,block,efs_get_block); } -static struct address_space_operations efs_aops = { +static const struct address_space_operations efs_aops = { .readpage = efs_readpage, .sync_page = block_sync_page, .bmap = _efs_bmap diff --git a/fs/efs/symlink.c b/fs/efs/symlink.c index 3d9a350e3e7..e249cf733a6 100644 --- a/fs/efs/symlink.c +++ b/fs/efs/symlink.c @@ -53,6 +53,6 @@ fail: return err; } -struct address_space_operations efs_symlink_aops = { +const struct address_space_operations efs_symlink_aops = { .readpage = efs_symlink_readpage }; diff --git a/fs/ext2/ext2.h b/fs/ext2/ext2.h index 9f74a62be55..e65a019fc7a 100644 --- a/fs/ext2/ext2.h +++ b/fs/ext2/ext2.h @@ -162,9 +162,9 @@ extern const struct file_operations ext2_file_operations; extern const struct file_operations ext2_xip_file_operations; /* inode.c */ -extern struct address_space_operations ext2_aops; -extern struct address_space_operations ext2_aops_xip; -extern struct address_space_operations ext2_nobh_aops; +extern const struct address_space_operations ext2_aops; +extern const struct address_space_operations ext2_aops_xip; +extern const struct address_space_operations ext2_nobh_aops; /* namei.c */ extern struct inode_operations ext2_dir_inode_operations; diff --git a/fs/ext2/inode.c b/fs/ext2/inode.c index 04af9c45dce..fb4d3220eb8 100644 --- a/fs/ext2/inode.c +++ b/fs/ext2/inode.c @@ -684,7 +684,7 @@ ext2_writepages(struct address_space *mapping, struct writeback_control *wbc) return mpage_writepages(mapping, wbc, ext2_get_block); } -struct address_space_operations ext2_aops = { +const struct address_space_operations ext2_aops = { .readpage = ext2_readpage, .readpages = ext2_readpages, .writepage = ext2_writepage, @@ -697,12 +697,12 @@ struct address_space_operations ext2_aops = { .migratepage = buffer_migrate_page, }; -struct address_space_operations ext2_aops_xip = { +const struct address_space_operations ext2_aops_xip = { .bmap = ext2_bmap, .get_xip_page = ext2_get_xip_page, }; -struct address_space_operations ext2_nobh_aops = { +const struct address_space_operations ext2_nobh_aops = { .readpage = ext2_readpage, .readpages = ext2_readpages, .writepage = ext2_nobh_writepage, diff --git a/fs/ext3/inode.c b/fs/ext3/inode.c index 0321e1b9034..f804d5e9d60 100644 --- a/fs/ext3/inode.c +++ b/fs/ext3/inode.c @@ -1698,7 +1698,7 @@ static int ext3_journalled_set_page_dirty(struct page *page) return __set_page_dirty_nobuffers(page); } -static struct address_space_operations ext3_ordered_aops = { +static const struct address_space_operations ext3_ordered_aops = { .readpage = ext3_readpage, .readpages = ext3_readpages, .writepage = ext3_ordered_writepage, @@ -1712,7 +1712,7 @@ static struct address_space_operations ext3_ordered_aops = { .migratepage = buffer_migrate_page, }; -static struct address_space_operations ext3_writeback_aops = { +static const struct address_space_operations ext3_writeback_aops = { .readpage = ext3_readpage, .readpages = ext3_readpages, .writepage = ext3_writeback_writepage, @@ -1726,7 +1726,7 @@ static struct address_space_operations ext3_writeback_aops = { .migratepage = buffer_migrate_page, }; -static struct address_space_operations ext3_journalled_aops = { +static const struct address_space_operations ext3_journalled_aops = { .readpage = ext3_readpage, .readpages = ext3_readpages, .writepage = ext3_journalled_writepage, diff --git a/fs/fat/inode.c b/fs/fat/inode.c index 7c35d582ec1..31b7174176b 100644 --- a/fs/fat/inode.c +++ b/fs/fat/inode.c @@ -196,7 +196,7 @@ static sector_t _fat_bmap(struct address_space *mapping, sector_t block) return generic_block_bmap(mapping, block, fat_get_block); } -static struct address_space_operations fat_aops = { +static const struct address_space_operations fat_aops = { .readpage = fat_readpage, .readpages = fat_readpages, .writepage = fat_writepage, diff --git a/fs/freevxfs/vxfs_immed.c b/fs/freevxfs/vxfs_immed.c index 6f5df1700e9..4e25f3fbed8 100644 --- a/fs/freevxfs/vxfs_immed.c +++ b/fs/freevxfs/vxfs_immed.c @@ -56,7 +56,7 @@ struct inode_operations vxfs_immed_symlink_iops = { /* * Adress space operations for immed files and directories. */ -struct address_space_operations vxfs_immed_aops = { +const struct address_space_operations vxfs_immed_aops = { .readpage = vxfs_immed_readpage, }; diff --git a/fs/freevxfs/vxfs_inode.c b/fs/freevxfs/vxfs_inode.c index f544aae9169..ca6a3971477 100644 --- a/fs/freevxfs/vxfs_inode.c +++ b/fs/freevxfs/vxfs_inode.c @@ -41,8 +41,8 @@ #include "vxfs_extern.h" -extern struct address_space_operations vxfs_aops; -extern struct address_space_operations vxfs_immed_aops; +extern const struct address_space_operations vxfs_aops; +extern const struct address_space_operations vxfs_immed_aops; extern struct inode_operations vxfs_immed_symlink_iops; @@ -295,7 +295,7 @@ vxfs_read_inode(struct inode *ip) { struct super_block *sbp = ip->i_sb; struct vxfs_inode_info *vip; - struct address_space_operations *aops; + const struct address_space_operations *aops; ino_t ino = ip->i_ino; if (!(vip = __vxfs_iget(ino, VXFS_SBI(sbp)->vsi_ilist))) diff --git a/fs/freevxfs/vxfs_subr.c b/fs/freevxfs/vxfs_subr.c index c1be118fc06..decac62efe5 100644 --- a/fs/freevxfs/vxfs_subr.c +++ b/fs/freevxfs/vxfs_subr.c @@ -42,7 +42,7 @@ static int vxfs_readpage(struct file *, struct page *); static sector_t vxfs_bmap(struct address_space *, sector_t); -struct address_space_operations vxfs_aops = { +const struct address_space_operations vxfs_aops = { .readpage = vxfs_readpage, .bmap = vxfs_bmap, .sync_page = block_sync_page, diff --git a/fs/fuse/file.c b/fs/fuse/file.c index 28aa81eae2c..63614ed1633 100644 --- a/fs/fuse/file.c +++ b/fs/fuse/file.c @@ -770,7 +770,7 @@ static const struct file_operations fuse_direct_io_file_operations = { /* no mmap and sendfile */ }; -static struct address_space_operations fuse_file_aops = { +static const struct address_space_operations fuse_file_aops = { .readpage = fuse_readpage, .prepare_write = fuse_prepare_write, .commit_write = fuse_commit_write, diff --git a/fs/hfs/hfs_fs.h b/fs/hfs/hfs_fs.h index 3ed8663a8db..735332dfd1b 100644 --- a/fs/hfs/hfs_fs.h +++ b/fs/hfs/hfs_fs.h @@ -182,8 +182,8 @@ extern void hfs_file_truncate(struct inode *); extern int hfs_get_block(struct inode *, sector_t, struct buffer_head *, int); /* inode.c */ -extern struct address_space_operations hfs_aops; -extern struct address_space_operations hfs_btree_aops; +extern const struct address_space_operations hfs_aops; +extern const struct address_space_operations hfs_btree_aops; extern struct inode *hfs_new_inode(struct inode *, struct qstr *, int); extern void hfs_inode_write_fork(struct inode *, struct hfs_extent *, __be32 *, __be32 *); diff --git a/fs/hfs/inode.c b/fs/hfs/inode.c index 2d4ced22201..315cf44a90b 100644 --- a/fs/hfs/inode.c +++ b/fs/hfs/inode.c @@ -114,7 +114,7 @@ static int hfs_writepages(struct address_space *mapping, return mpage_writepages(mapping, wbc, hfs_get_block); } -struct address_space_operations hfs_btree_aops = { +const struct address_space_operations hfs_btree_aops = { .readpage = hfs_readpage, .writepage = hfs_writepage, .sync_page = block_sync_page, @@ -124,7 +124,7 @@ struct address_space_operations hfs_btree_aops = { .releasepage = hfs_releasepage, }; -struct address_space_operations hfs_aops = { +const struct address_space_operations hfs_aops = { .readpage = hfs_readpage, .writepage = hfs_writepage, .sync_page = block_sync_page, diff --git a/fs/hfsplus/hfsplus_fs.h b/fs/hfsplus/hfsplus_fs.h index 7ae393637a0..8a1ca5ef7ad 100644 --- a/fs/hfsplus/hfsplus_fs.h +++ b/fs/hfsplus/hfsplus_fs.h @@ -323,8 +323,8 @@ int hfsplus_file_extend(struct inode *); void hfsplus_file_truncate(struct inode *); /* inode.c */ -extern struct address_space_operations hfsplus_aops; -extern struct address_space_operations hfsplus_btree_aops; +extern const struct address_space_operations hfsplus_aops; +extern const struct address_space_operations hfsplus_btree_aops; void hfsplus_inode_read_fork(struct inode *, struct hfsplus_fork_raw *); void hfsplus_inode_write_fork(struct inode *, struct hfsplus_fork_raw *); diff --git a/fs/hfsplus/inode.c b/fs/hfsplus/inode.c index acf66dba3e0..924ecdef809 100644 --- a/fs/hfsplus/inode.c +++ b/fs/hfsplus/inode.c @@ -109,7 +109,7 @@ static int hfsplus_writepages(struct address_space *mapping, return mpage_writepages(mapping, wbc, hfsplus_get_block); } -struct address_space_operations hfsplus_btree_aops = { +const struct address_space_operations hfsplus_btree_aops = { .readpage = hfsplus_readpage, .writepage = hfsplus_writepage, .sync_page = block_sync_page, @@ -119,7 +119,7 @@ struct address_space_operations hfsplus_btree_aops = { .releasepage = hfsplus_releasepage, }; -struct address_space_operations hfsplus_aops = { +const struct address_space_operations hfsplus_aops = { .readpage = hfsplus_readpage, .writepage = hfsplus_writepage, .sync_page = block_sync_page, diff --git a/fs/hostfs/hostfs_kern.c b/fs/hostfs/hostfs_kern.c index 8e0d37743e7..b82e3d9c879 100644 --- a/fs/hostfs/hostfs_kern.c +++ b/fs/hostfs/hostfs_kern.c @@ -54,7 +54,7 @@ static int append = 0; static struct inode_operations hostfs_iops; static struct inode_operations hostfs_dir_iops; -static struct address_space_operations hostfs_link_aops; +static const struct address_space_operations hostfs_link_aops; #ifndef MODULE static int __init hostfs_args(char *options, int *add) @@ -518,7 +518,7 @@ int hostfs_commit_write(struct file *file, struct page *page, unsigned from, return(err); } -static struct address_space_operations hostfs_aops = { +static const struct address_space_operations hostfs_aops = { .writepage = hostfs_writepage, .readpage = hostfs_readpage, .set_page_dirty = __set_page_dirty_nobuffers, @@ -935,7 +935,7 @@ int hostfs_link_readpage(struct file *file, struct page *page) return(err); } -static struct address_space_operations hostfs_link_aops = { +static const struct address_space_operations hostfs_link_aops = { .readpage = hostfs_link_readpage, }; diff --git a/fs/hpfs/file.c b/fs/hpfs/file.c index d3b9fffe45a..d9eb19b7b8a 100644 --- a/fs/hpfs/file.c +++ b/fs/hpfs/file.c @@ -99,7 +99,7 @@ static sector_t _hpfs_bmap(struct address_space *mapping, sector_t block) { return generic_block_bmap(mapping,block,hpfs_get_block); } -struct address_space_operations hpfs_aops = { +const struct address_space_operations hpfs_aops = { .readpage = hpfs_readpage, .writepage = hpfs_writepage, .sync_page = block_sync_page, diff --git a/fs/hpfs/hpfs_fn.h b/fs/hpfs/hpfs_fn.h index 29b7a3e5517..f687d54ed44 100644 --- a/fs/hpfs/hpfs_fn.h +++ b/fs/hpfs/hpfs_fn.h @@ -268,7 +268,7 @@ void hpfs_set_ea(struct inode *, struct fnode *, char *, char *, int); int hpfs_file_fsync(struct file *, struct dentry *, int); extern const struct file_operations hpfs_file_ops; extern struct inode_operations hpfs_file_iops; -extern struct address_space_operations hpfs_aops; +extern const struct address_space_operations hpfs_aops; /* inode.c */ @@ -304,7 +304,7 @@ void hpfs_decide_conv(struct inode *, unsigned char *, unsigned); /* namei.c */ extern struct inode_operations hpfs_dir_iops; -extern struct address_space_operations hpfs_symlink_aops; +extern const struct address_space_operations hpfs_symlink_aops; static inline struct hpfs_inode_info *hpfs_i(struct inode *inode) { diff --git a/fs/hpfs/namei.c b/fs/hpfs/namei.c index a03abb12c61..59e7dc182a0 100644 --- a/fs/hpfs/namei.c +++ b/fs/hpfs/namei.c @@ -538,7 +538,7 @@ fail: return err; } -struct address_space_operations hpfs_symlink_aops = { +const struct address_space_operations hpfs_symlink_aops = { .readpage = hpfs_symlink_readpage }; diff --git a/fs/hugetlbfs/inode.c b/fs/hugetlbfs/inode.c index e6410d8edd0..6449cb69796 100644 --- a/fs/hugetlbfs/inode.c +++ b/fs/hugetlbfs/inode.c @@ -34,7 +34,7 @@ #define HUGETLBFS_MAGIC 0x958458f6 static struct super_operations hugetlbfs_ops; -static struct address_space_operations hugetlbfs_aops; +static const struct address_space_operations hugetlbfs_aops; const struct file_operations hugetlbfs_file_operations; static struct inode_operations hugetlbfs_dir_inode_operations; static struct inode_operations hugetlbfs_inode_operations; @@ -547,7 +547,7 @@ static void hugetlbfs_destroy_inode(struct inode *inode) kmem_cache_free(hugetlbfs_inode_cachep, HUGETLBFS_I(inode)); } -static struct address_space_operations hugetlbfs_aops = { +static const struct address_space_operations hugetlbfs_aops = { .readpage = hugetlbfs_readpage, .prepare_write = hugetlbfs_prepare_write, .commit_write = hugetlbfs_commit_write, diff --git a/fs/inode.c b/fs/inode.c index 3a2446a27d2..f42961eb983 100644 --- a/fs/inode.c +++ b/fs/inode.c @@ -102,7 +102,7 @@ static kmem_cache_t * inode_cachep __read_mostly; static struct inode *alloc_inode(struct super_block *sb) { - static struct address_space_operations empty_aops; + static const struct address_space_operations empty_aops; static struct inode_operations empty_iops; static const struct file_operations empty_fops; struct inode *inode; diff --git a/fs/isofs/compress.c b/fs/isofs/compress.c index 4917315db73..3a39158cca9 100644 --- a/fs/isofs/compress.c +++ b/fs/isofs/compress.c @@ -312,7 +312,7 @@ eio: return err; } -struct address_space_operations zisofs_aops = { +const struct address_space_operations zisofs_aops = { .readpage = zisofs_readpage, /* No sync_page operation supported? */ /* No bmap operation supported */ diff --git a/fs/isofs/inode.c b/fs/isofs/inode.c index 3f9c8ba1fa1..bb11c7fb401 100644 --- a/fs/isofs/inode.c +++ b/fs/isofs/inode.c @@ -1054,7 +1054,7 @@ static sector_t _isofs_bmap(struct address_space *mapping, sector_t block) return generic_block_bmap(mapping,block,isofs_get_block); } -static struct address_space_operations isofs_aops = { +static const struct address_space_operations isofs_aops = { .readpage = isofs_readpage, .sync_page = block_sync_page, .bmap = _isofs_bmap diff --git a/fs/isofs/isofs.h b/fs/isofs/isofs.h index b87ba066f5e..e6308c8b573 100644 --- a/fs/isofs/isofs.h +++ b/fs/isofs/isofs.h @@ -176,5 +176,5 @@ isofs_normalize_block_and_offset(struct iso_directory_record* de, extern struct inode_operations isofs_dir_inode_operations; extern const struct file_operations isofs_dir_operations; -extern struct address_space_operations isofs_symlink_aops; +extern const struct address_space_operations isofs_symlink_aops; extern struct export_operations isofs_export_ops; diff --git a/fs/isofs/rock.c b/fs/isofs/rock.c index 4326cb47f8f..f3a1db3098d 100644 --- a/fs/isofs/rock.c +++ b/fs/isofs/rock.c @@ -754,6 +754,6 @@ error: return -EIO; } -struct address_space_operations isofs_symlink_aops = { +const struct address_space_operations isofs_symlink_aops = { .readpage = rock_ridge_symlink_readpage }; diff --git a/fs/isofs/zisofs.h b/fs/isofs/zisofs.h index d78485d101c..27379570915 100644 --- a/fs/isofs/zisofs.h +++ b/fs/isofs/zisofs.h @@ -15,7 +15,7 @@ */ #ifdef CONFIG_ZISOFS -extern struct address_space_operations zisofs_aops; +extern const struct address_space_operations zisofs_aops; extern int __init zisofs_init(void); extern void zisofs_cleanup(void); #endif diff --git a/fs/jffs/inode-v23.c b/fs/jffs/inode-v23.c index 9e46ea6da75..93068697a9b 100644 --- a/fs/jffs/inode-v23.c +++ b/fs/jffs/inode-v23.c @@ -59,7 +59,7 @@ static const struct file_operations jffs_file_operations; static struct inode_operations jffs_file_inode_operations; static const struct file_operations jffs_dir_operations; static struct inode_operations jffs_dir_inode_operations; -static struct address_space_operations jffs_address_operations; +static const struct address_space_operations jffs_address_operations; kmem_cache_t *node_cache = NULL; kmem_cache_t *fm_cache = NULL; @@ -1614,7 +1614,7 @@ jffs_ioctl(struct inode *inode, struct file *filp, unsigned int cmd, } /* jffs_ioctl() */ -static struct address_space_operations jffs_address_operations = { +static const struct address_space_operations jffs_address_operations = { .readpage = jffs_readpage, .prepare_write = jffs_prepare_write, .commit_write = jffs_commit_write, diff --git a/fs/jffs2/file.c b/fs/jffs2/file.c index bb8844f40e4..3ed6e3e120b 100644 --- a/fs/jffs2/file.c +++ b/fs/jffs2/file.c @@ -62,7 +62,7 @@ struct inode_operations jffs2_file_inode_operations = .removexattr = jffs2_removexattr }; -struct address_space_operations jffs2_file_address_operations = +const struct address_space_operations jffs2_file_address_operations = { .readpage = jffs2_readpage, .prepare_write =jffs2_prepare_write, diff --git a/fs/jffs2/os-linux.h b/fs/jffs2/os-linux.h index 6b522356540..9f41fc01a37 100644 --- a/fs/jffs2/os-linux.h +++ b/fs/jffs2/os-linux.h @@ -158,7 +158,7 @@ extern struct inode_operations jffs2_dir_inode_operations; /* file.c */ extern const struct file_operations jffs2_file_operations; extern struct inode_operations jffs2_file_inode_operations; -extern struct address_space_operations jffs2_file_address_operations; +extern const struct address_space_operations jffs2_file_address_operations; int jffs2_fsync(struct file *, struct dentry *, int); int jffs2_do_readpage_unlock (struct inode *inode, struct page *pg); diff --git a/fs/jfs/inode.c b/fs/jfs/inode.c index 04eb78f1252..43e3f566aad 100644 --- a/fs/jfs/inode.c +++ b/fs/jfs/inode.c @@ -305,7 +305,7 @@ static ssize_t jfs_direct_IO(int rw, struct kiocb *iocb, offset, nr_segs, jfs_get_block, NULL); } -struct address_space_operations jfs_aops = { +const struct address_space_operations jfs_aops = { .readpage = jfs_readpage, .readpages = jfs_readpages, .writepage = jfs_writepage, diff --git a/fs/jfs/jfs_inode.h b/fs/jfs/jfs_inode.h index c3007267446..b5c7da6190d 100644 --- a/fs/jfs/jfs_inode.h +++ b/fs/jfs/jfs_inode.h @@ -33,7 +33,7 @@ extern void jfs_free_zero_link(struct inode *); extern struct dentry *jfs_get_parent(struct dentry *dentry); extern void jfs_set_inode_flags(struct inode *); -extern struct address_space_operations jfs_aops; +extern const struct address_space_operations jfs_aops; extern struct inode_operations jfs_dir_inode_operations; extern const struct file_operations jfs_dir_operations; extern struct inode_operations jfs_file_inode_operations; diff --git a/fs/jfs/jfs_metapage.c b/fs/jfs/jfs_metapage.c index 7f6e8803970..e1e0a6e6ebd 100644 --- a/fs/jfs/jfs_metapage.c +++ b/fs/jfs/jfs_metapage.c @@ -577,7 +577,7 @@ static void metapage_invalidatepage(struct page *page, unsigned long offset) metapage_releasepage(page, 0); } -struct address_space_operations jfs_metapage_aops = { +const struct address_space_operations jfs_metapage_aops = { .readpage = metapage_readpage, .writepage = metapage_writepage, .sync_page = block_sync_page, diff --git a/fs/jfs/jfs_metapage.h b/fs/jfs/jfs_metapage.h index f0b7d3282b0..d17a3290f5a 100644 --- a/fs/jfs/jfs_metapage.h +++ b/fs/jfs/jfs_metapage.h @@ -139,7 +139,7 @@ static inline void metapage_homeok(struct metapage *mp) put_metapage(mp); } -extern struct address_space_operations jfs_metapage_aops; +extern const struct address_space_operations jfs_metapage_aops; /* * This routines invalidate all pages for an extent. diff --git a/fs/minix/inode.c b/fs/minix/inode.c index a6fb509b734..9ea91c5eeb7 100644 --- a/fs/minix/inode.c +++ b/fs/minix/inode.c @@ -335,7 +335,7 @@ static sector_t minix_bmap(struct address_space *mapping, sector_t block) { return generic_block_bmap(mapping,block,minix_get_block); } -static struct address_space_operations minix_aops = { +static const struct address_space_operations minix_aops = { .readpage = minix_readpage, .writepage = minix_writepage, .sync_page = block_sync_page, diff --git a/fs/ncpfs/inode.c b/fs/ncpfs/inode.c index 90d2ea28f33..6c51c119846 100644 --- a/fs/ncpfs/inode.c +++ b/fs/ncpfs/inode.c @@ -105,7 +105,7 @@ static struct super_operations ncp_sops = extern struct dentry_operations ncp_root_dentry_operations; #if defined(CONFIG_NCPFS_EXTRAS) || defined(CONFIG_NCPFS_NFS_NS) -extern struct address_space_operations ncp_symlink_aops; +extern const struct address_space_operations ncp_symlink_aops; extern int ncp_symlink(struct inode*, struct dentry*, const char*); #endif diff --git a/fs/ncpfs/symlink.c b/fs/ncpfs/symlink.c index e935f1b34bc..f76b1392a01 100644 --- a/fs/ncpfs/symlink.c +++ b/fs/ncpfs/symlink.c @@ -99,7 +99,7 @@ fail: /* * symlinks can't do much... */ -struct address_space_operations ncp_symlink_aops = { +const struct address_space_operations ncp_symlink_aops = { .readpage = ncp_symlink_readpage, }; diff --git a/fs/nfs/file.c b/fs/nfs/file.c index add28913883..cc2b874ad5a 100644 --- a/fs/nfs/file.c +++ b/fs/nfs/file.c @@ -315,7 +315,7 @@ static int nfs_release_page(struct page *page, gfp_t gfp) return !nfs_wb_page(page->mapping->host, page); } -struct address_space_operations nfs_file_aops = { +const struct address_space_operations nfs_file_aops = { .readpage = nfs_readpage, .readpages = nfs_readpages, .set_page_dirty = __set_page_dirty_nobuffers, diff --git a/fs/ntfs/aops.c b/fs/ntfs/aops.c index 580412d330c..bc579bfdfbd 100644 --- a/fs/ntfs/aops.c +++ b/fs/ntfs/aops.c @@ -1544,7 +1544,7 @@ err_out: /** * ntfs_aops - general address space operations for inodes and attributes */ -struct address_space_operations ntfs_aops = { +const struct address_space_operations ntfs_aops = { .readpage = ntfs_readpage, /* Fill page with data. */ .sync_page = block_sync_page, /* Currently, just unplugs the disk request queue. */ @@ -1560,7 +1560,7 @@ struct address_space_operations ntfs_aops = { * ntfs_mst_aops - general address space operations for mst protecteed inodes * and attributes */ -struct address_space_operations ntfs_mst_aops = { +const struct address_space_operations ntfs_mst_aops = { .readpage = ntfs_readpage, /* Fill page with data. */ .sync_page = block_sync_page, /* Currently, just unplugs the disk request queue. */ diff --git a/fs/ntfs/ntfs.h b/fs/ntfs/ntfs.h index bf7b3d7c093..ddd3d503097 100644 --- a/fs/ntfs/ntfs.h +++ b/fs/ntfs/ntfs.h @@ -57,8 +57,8 @@ extern struct kmem_cache *ntfs_attr_ctx_cache; extern struct kmem_cache *ntfs_index_ctx_cache; /* The various operations structs defined throughout the driver files. */ -extern struct address_space_operations ntfs_aops; -extern struct address_space_operations ntfs_mst_aops; +extern const struct address_space_operations ntfs_aops; +extern const struct address_space_operations ntfs_mst_aops; extern const struct file_operations ntfs_file_ops; extern struct inode_operations ntfs_file_inode_ops; diff --git a/fs/ocfs2/aops.c b/fs/ocfs2/aops.c index 47152bf9a7f..cca71317b6d 100644 --- a/fs/ocfs2/aops.c +++ b/fs/ocfs2/aops.c @@ -666,7 +666,7 @@ out: return ret; } -struct address_space_operations ocfs2_aops = { +const struct address_space_operations ocfs2_aops = { .readpage = ocfs2_readpage, .writepage = ocfs2_writepage, .prepare_write = ocfs2_prepare_write, diff --git a/fs/ocfs2/inode.h b/fs/ocfs2/inode.h index 84c50796128..35140f6cf84 100644 --- a/fs/ocfs2/inode.h +++ b/fs/ocfs2/inode.h @@ -114,7 +114,7 @@ static inline struct ocfs2_inode_info *OCFS2_I(struct inode *inode) extern kmem_cache_t *ocfs2_inode_cache; -extern struct address_space_operations ocfs2_aops; +extern const struct address_space_operations ocfs2_aops; struct buffer_head *ocfs2_bread(struct inode *inode, int block, int *err, int reada); diff --git a/fs/qnx4/inode.c b/fs/qnx4/inode.c index 2f24c46f72a..8bc182a8874 100644 --- a/fs/qnx4/inode.c +++ b/fs/qnx4/inode.c @@ -450,7 +450,7 @@ static sector_t qnx4_bmap(struct address_space *mapping, sector_t block) { return generic_block_bmap(mapping,block,qnx4_get_block); } -static struct address_space_operations qnx4_aops = { +static const struct address_space_operations qnx4_aops = { .readpage = qnx4_readpage, .writepage = qnx4_writepage, .sync_page = block_sync_page, diff --git a/fs/ramfs/file-mmu.c b/fs/ramfs/file-mmu.c index 00a933eb820..86f14cacf64 100644 --- a/fs/ramfs/file-mmu.c +++ b/fs/ramfs/file-mmu.c @@ -26,7 +26,7 @@ #include -struct address_space_operations ramfs_aops = { +const struct address_space_operations ramfs_aops = { .readpage = simple_readpage, .prepare_write = simple_prepare_write, .commit_write = simple_commit_write diff --git a/fs/ramfs/file-nommu.c b/fs/ramfs/file-nommu.c index f443a84b98a..99fffc9e1bf 100644 --- a/fs/ramfs/file-nommu.c +++ b/fs/ramfs/file-nommu.c @@ -27,7 +27,7 @@ static int ramfs_nommu_setattr(struct dentry *, struct iattr *); -struct address_space_operations ramfs_aops = { +const struct address_space_operations ramfs_aops = { .readpage = simple_readpage, .prepare_write = simple_prepare_write, .commit_write = simple_commit_write diff --git a/fs/ramfs/internal.h b/fs/ramfs/internal.h index 313237631b4..c2bb58e7465 100644 --- a/fs/ramfs/internal.h +++ b/fs/ramfs/internal.h @@ -10,6 +10,6 @@ */ -extern struct address_space_operations ramfs_aops; +extern const struct address_space_operations ramfs_aops; extern const struct file_operations ramfs_file_operations; extern struct inode_operations ramfs_file_inode_operations; diff --git a/fs/reiserfs/inode.c b/fs/reiserfs/inode.c index 9857e50f85e..a24858a632f 100644 --- a/fs/reiserfs/inode.c +++ b/fs/reiserfs/inode.c @@ -2996,7 +2996,7 @@ int reiserfs_setattr(struct dentry *dentry, struct iattr *attr) return error; } -struct address_space_operations reiserfs_address_space_operations = { +const struct address_space_operations reiserfs_address_space_operations = { .writepage = reiserfs_writepage, .readpage = reiserfs_readpage, .readpages = reiserfs_readpages, diff --git a/fs/romfs/inode.c b/fs/romfs/inode.c index 283fbc6b8ee..22eed61ebf6 100644 --- a/fs/romfs/inode.c +++ b/fs/romfs/inode.c @@ -459,7 +459,7 @@ err_out: /* Mapping from our types to the kernel */ -static struct address_space_operations romfs_aops = { +static const struct address_space_operations romfs_aops = { .readpage = romfs_readpage }; diff --git a/fs/smbfs/file.c b/fs/smbfs/file.c index ed9a24d19d7..dae67048bab 100644 --- a/fs/smbfs/file.c +++ b/fs/smbfs/file.c @@ -306,7 +306,7 @@ static int smb_commit_write(struct file *file, struct page *page, return status; } -struct address_space_operations smb_file_aops = { +const struct address_space_operations smb_file_aops = { .readpage = smb_readpage, .writepage = smb_writepage, .prepare_write = smb_prepare_write, diff --git a/fs/smbfs/proto.h b/fs/smbfs/proto.h index 972ed7dad38..34fb462b237 100644 --- a/fs/smbfs/proto.h +++ b/fs/smbfs/proto.h @@ -63,7 +63,7 @@ extern int smb_revalidate_inode(struct dentry *dentry); extern int smb_getattr(struct vfsmount *mnt, struct dentry *dentry, struct kstat *stat); extern int smb_notify_change(struct dentry *dentry, struct iattr *attr); /* file.c */ -extern struct address_space_operations smb_file_aops; +extern const struct address_space_operations smb_file_aops; extern const struct file_operations smb_file_operations; extern struct inode_operations smb_file_inode_operations; /* ioctl.c */ diff --git a/fs/sysfs/inode.c b/fs/sysfs/inode.c index f0b347bd12c..5e0e31cc46f 100644 --- a/fs/sysfs/inode.c +++ b/fs/sysfs/inode.c @@ -16,7 +16,7 @@ extern struct super_block * sysfs_sb; -static struct address_space_operations sysfs_aops = { +static const struct address_space_operations sysfs_aops = { .readpage = simple_readpage, .prepare_write = simple_prepare_write, .commit_write = simple_commit_write diff --git a/fs/sysv/itree.c b/fs/sysv/itree.c index 86f5f8d43d0..f2bcccd1d6f 100644 --- a/fs/sysv/itree.c +++ b/fs/sysv/itree.c @@ -465,7 +465,7 @@ static sector_t sysv_bmap(struct address_space *mapping, sector_t block) { return generic_block_bmap(mapping,block,get_block); } -struct address_space_operations sysv_aops = { +const struct address_space_operations sysv_aops = { .readpage = sysv_readpage, .writepage = sysv_writepage, .sync_page = block_sync_page, diff --git a/fs/sysv/sysv.h b/fs/sysv/sysv.h index 393a480e4de..9dcc8212093 100644 --- a/fs/sysv/sysv.h +++ b/fs/sysv/sysv.h @@ -161,7 +161,7 @@ extern struct inode_operations sysv_dir_inode_operations; extern struct inode_operations sysv_fast_symlink_inode_operations; extern const struct file_operations sysv_file_operations; extern const struct file_operations sysv_dir_operations; -extern struct address_space_operations sysv_aops; +extern const struct address_space_operations sysv_aops; extern struct super_operations sysv_sops; extern struct dentry_operations sysv_dentry_operations; diff --git a/fs/udf/file.c b/fs/udf/file.c index e34b00e303f..a59e5f33daf 100644 --- a/fs/udf/file.c +++ b/fs/udf/file.c @@ -95,7 +95,7 @@ static int udf_adinicb_commit_write(struct file *file, struct page *page, unsign return 0; } -struct address_space_operations udf_adinicb_aops = { +const struct address_space_operations udf_adinicb_aops = { .readpage = udf_adinicb_readpage, .writepage = udf_adinicb_writepage, .sync_page = block_sync_page, diff --git a/fs/udf/inode.c b/fs/udf/inode.c index 2983afd5e7f..605f5111b6d 100644 --- a/fs/udf/inode.c +++ b/fs/udf/inode.c @@ -132,7 +132,7 @@ static sector_t udf_bmap(struct address_space *mapping, sector_t block) return generic_block_bmap(mapping,block,udf_get_block); } -struct address_space_operations udf_aops = { +const struct address_space_operations udf_aops = { .readpage = udf_readpage, .writepage = udf_writepage, .sync_page = block_sync_page, diff --git a/fs/udf/symlink.c b/fs/udf/symlink.c index 674bb40edc8..ba068a78656 100644 --- a/fs/udf/symlink.c +++ b/fs/udf/symlink.c @@ -113,6 +113,6 @@ out: /* * symlinks can't do much... */ -struct address_space_operations udf_symlink_aops = { +const struct address_space_operations udf_symlink_aops = { .readpage = udf_symlink_filler, }; diff --git a/fs/udf/udfdecl.h b/fs/udf/udfdecl.h index 023e19ba5a2..2f992387cc9 100644 --- a/fs/udf/udfdecl.h +++ b/fs/udf/udfdecl.h @@ -47,9 +47,9 @@ extern struct inode_operations udf_dir_inode_operations; extern const struct file_operations udf_dir_operations; extern struct inode_operations udf_file_inode_operations; extern const struct file_operations udf_file_operations; -extern struct address_space_operations udf_aops; -extern struct address_space_operations udf_adinicb_aops; -extern struct address_space_operations udf_symlink_aops; +extern const struct address_space_operations udf_aops; +extern const struct address_space_operations udf_adinicb_aops; +extern const struct address_space_operations udf_symlink_aops; struct udf_fileident_bh { diff --git a/fs/ufs/inode.c b/fs/ufs/inode.c index 259bd196099..8e1f90e4204 100644 --- a/fs/ufs/inode.c +++ b/fs/ufs/inode.c @@ -574,7 +574,7 @@ static sector_t ufs_bmap(struct address_space *mapping, sector_t block) { return generic_block_bmap(mapping,block,ufs_getfrag_block); } -struct address_space_operations ufs_aops = { +const struct address_space_operations ufs_aops = { .readpage = ufs_readpage, .writepage = ufs_writepage, .sync_page = block_sync_page, diff --git a/fs/xfs/linux-2.6/xfs_aops.c b/fs/xfs/linux-2.6/xfs_aops.c index 3e807b828e2..c40f81ba9b1 100644 --- a/fs/xfs/linux-2.6/xfs_aops.c +++ b/fs/xfs/linux-2.6/xfs_aops.c @@ -1454,7 +1454,7 @@ xfs_vm_invalidatepage( block_invalidatepage(page, offset); } -struct address_space_operations xfs_address_space_operations = { +const struct address_space_operations xfs_address_space_operations = { .readpage = xfs_vm_readpage, .readpages = xfs_vm_readpages, .writepage = xfs_vm_writepage, diff --git a/fs/xfs/linux-2.6/xfs_aops.h b/fs/xfs/linux-2.6/xfs_aops.h index 706d8c781b8..2244e516b66 100644 --- a/fs/xfs/linux-2.6/xfs_aops.h +++ b/fs/xfs/linux-2.6/xfs_aops.h @@ -40,7 +40,7 @@ typedef struct xfs_ioend { struct work_struct io_work; /* xfsdatad work queue */ } xfs_ioend_t; -extern struct address_space_operations xfs_address_space_operations; +extern const struct address_space_operations xfs_address_space_operations; extern int xfs_get_blocks(struct inode *, sector_t, struct buffer_head *, int); #endif /* __XFS_AOPS_H__ */ diff --git a/fs/xfs/linux-2.6/xfs_buf.c b/fs/xfs/linux-2.6/xfs_buf.c index 26fed0756f0..2af528dcfb0 100644 --- a/fs/xfs/linux-2.6/xfs_buf.c +++ b/fs/xfs/linux-2.6/xfs_buf.c @@ -1520,7 +1520,7 @@ xfs_mapping_buftarg( struct backing_dev_info *bdi; struct inode *inode; struct address_space *mapping; - static struct address_space_operations mapping_aops = { + static const struct address_space_operations mapping_aops = { .sync_page = block_sync_page, .migratepage = fail_migrate_page, }; diff --git a/include/linux/coda_linux.h b/include/linux/coda_linux.h index 7b5c5df5cb6..be512cc9879 100644 --- a/include/linux/coda_linux.h +++ b/include/linux/coda_linux.h @@ -27,8 +27,8 @@ extern struct inode_operations coda_dir_inode_operations; extern struct inode_operations coda_file_inode_operations; extern struct inode_operations coda_ioctl_inode_operations; -extern struct address_space_operations coda_file_aops; -extern struct address_space_operations coda_symlink_aops; +extern const struct address_space_operations coda_file_aops; +extern const struct address_space_operations coda_symlink_aops; extern const struct file_operations coda_dir_operations; extern const struct file_operations coda_file_operations; diff --git a/include/linux/efs_fs.h b/include/linux/efs_fs.h index fbfa6b52e2f..278ef449581 100644 --- a/include/linux/efs_fs.h +++ b/include/linux/efs_fs.h @@ -38,7 +38,7 @@ struct statfs; extern struct inode_operations efs_dir_inode_operations; extern const struct file_operations efs_dir_operations; -extern struct address_space_operations efs_symlink_aops; +extern const struct address_space_operations efs_symlink_aops; extern void efs_read_inode(struct inode *); extern efs_block_t efs_map_block(struct inode *, efs_block_t); diff --git a/include/linux/fs.h b/include/linux/fs.h index 2d8b348c119..e04a5cfe874 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -392,7 +392,7 @@ struct address_space { unsigned int truncate_count; /* Cover race condition with truncate */ unsigned long nrpages; /* number of total pages */ pgoff_t writeback_index;/* writeback starts here */ - struct address_space_operations *a_ops; /* methods */ + const struct address_space_operations *a_ops; /* methods */ unsigned long flags; /* error bits/gfp mask */ struct backing_dev_info *backing_dev_info; /* device readahead, etc */ spinlock_t private_lock; /* for use by the address_space */ @@ -1405,7 +1405,7 @@ extern void bd_forget(struct inode *inode); extern void bdput(struct block_device *); extern struct block_device *open_by_devnum(dev_t, unsigned); extern const struct file_operations def_blk_fops; -extern struct address_space_operations def_blk_aops; +extern const struct address_space_operations def_blk_aops; extern const struct file_operations def_chr_fops; extern const struct file_operations bad_sock_fops; extern const struct file_operations def_fifo_fops; diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h index 0a1740b2532..d90b1bb3756 100644 --- a/include/linux/nfs_fs.h +++ b/include/linux/nfs_fs.h @@ -335,7 +335,7 @@ extern struct inode_operations nfs_file_inode_operations; extern struct inode_operations nfs3_file_inode_operations; #endif /* CONFIG_NFS_V3 */ extern const struct file_operations nfs_file_operations; -extern struct address_space_operations nfs_file_aops; +extern const struct address_space_operations nfs_file_aops; static inline struct rpc_cred *nfs_file_cred(struct file *file) { diff --git a/include/linux/reiserfs_fs.h b/include/linux/reiserfs_fs.h index 5676c4210e2..daa2d83cefe 100644 --- a/include/linux/reiserfs_fs.h +++ b/include/linux/reiserfs_fs.h @@ -1973,7 +1973,7 @@ void reiserfs_unmap_buffer(struct buffer_head *); /* file.c */ extern struct inode_operations reiserfs_file_inode_operations; extern const struct file_operations reiserfs_file_operations; -extern struct address_space_operations reiserfs_address_space_operations; +extern const struct address_space_operations reiserfs_address_space_operations; /* fix_nodes.c */ diff --git a/include/linux/ufs_fs.h b/include/linux/ufs_fs.h index 914f911325b..e39b7cc4339 100644 --- a/include/linux/ufs_fs.h +++ b/include/linux/ufs_fs.h @@ -966,7 +966,7 @@ extern void ufs_set_link(struct inode *dir, struct ufs_dir_entry *de, extern struct inode_operations ufs_file_inode_operations; extern const struct file_operations ufs_file_operations; -extern struct address_space_operations ufs_aops; +extern const struct address_space_operations ufs_aops; /* ialloc.c */ extern void ufs_free_inode (struct inode *inode); diff --git a/mm/filemap.c b/mm/filemap.c index d504d6e9888..4082b3b3cea 100644 --- a/mm/filemap.c +++ b/mm/filemap.c @@ -2069,7 +2069,7 @@ generic_file_buffered_write(struct kiocb *iocb, const struct iovec *iov, { struct file *file = iocb->ki_filp; struct address_space * mapping = file->f_mapping; - struct address_space_operations *a_ops = mapping->a_ops; + const struct address_space_operations *a_ops = mapping->a_ops; struct inode *inode = mapping->host; long status = 0; struct page *page; @@ -2219,7 +2219,7 @@ __generic_file_aio_write_nolock(struct kiocb *iocb, const struct iovec *iov, unsigned long nr_segs, loff_t *ppos) { struct file *file = iocb->ki_filp; - struct address_space * mapping = file->f_mapping; + const struct address_space * mapping = file->f_mapping; size_t ocount; /* original count */ size_t count; /* after file limit checks */ struct inode *inode = mapping->host; diff --git a/mm/filemap_xip.c b/mm/filemap_xip.c index b960ac8e591..b4fd0d7c9bf 100644 --- a/mm/filemap_xip.c +++ b/mm/filemap_xip.c @@ -273,7 +273,7 @@ __xip_file_write(struct file *filp, const char __user *buf, size_t count, loff_t pos, loff_t *ppos) { struct address_space * mapping = filp->f_mapping; - struct address_space_operations *a_ops = mapping->a_ops; + const struct address_space_operations *a_ops = mapping->a_ops; struct inode *inode = mapping->host; long status = 0; struct page *page; diff --git a/mm/shmem.c b/mm/shmem.c index 38bc3334f26..ea64c07cbe7 100644 --- a/mm/shmem.c +++ b/mm/shmem.c @@ -174,7 +174,7 @@ static inline void shmem_unacct_blocks(unsigned long flags, long pages) } static struct super_operations shmem_ops; -static struct address_space_operations shmem_aops; +static const struct address_space_operations shmem_aops; static struct file_operations shmem_file_operations; static struct inode_operations shmem_inode_operations; static struct inode_operations shmem_dir_inode_operations; @@ -2162,7 +2162,7 @@ static void destroy_inodecache(void) printk(KERN_INFO "shmem_inode_cache: not all structures were freed\n"); } -static struct address_space_operations shmem_aops = { +static const struct address_space_operations shmem_aops = { .writepage = shmem_writepage, .set_page_dirty = __set_page_dirty_nobuffers, #ifdef CONFIG_TMPFS diff --git a/mm/swap_state.c b/mm/swap_state.c index e0e1583f32c..7535211bb49 100644 --- a/mm/swap_state.c +++ b/mm/swap_state.c @@ -24,7 +24,7 @@ * vmscan's shrink_list, to make sync_page look nicer, and to allow * future use of radix_tree tags in the swap cache. */ -static struct address_space_operations swap_aops = { +static const struct address_space_operations swap_aops = { .writepage = swap_writepage, .sync_page = block_sync_page, .set_page_dirty = __set_page_dirty_nobuffers, -- cgit v1.2.3-70-g09d2 From f71d20e961474dde77e6558396efb93d6ac80a4b Mon Sep 17 00:00:00 2001 From: Arjan van de Ven Date: Wed, 28 Jun 2006 04:26:45 -0700 Subject: [PATCH] Add EXPORT_UNUSED_SYMBOL and EXPORT_UNUSED_SYMBOL_GPL Temporarily add EXPORT_UNUSED_SYMBOL and EXPORT_UNUSED_SYMBOL_GPL. These will be used as a transition measure for symbols that aren't used in the kernel and are on the way out. When a module uses such a symbol, a warning is printk'd at modprobe time. The main reason for removing unused exports is size: eacho export takes roughly between 100 and 150 bytes of kernel space in the binary. This patch gives users the option to immediately get this size gain via a config option. Signed-off-by: Arjan van de Ven Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- Documentation/feature-removal-schedule.txt | 10 ++++ include/asm-generic/vmlinux.lds.h | 28 +++++++++++ include/linux/module.h | 20 ++++++++ kernel/module.c | 78 +++++++++++++++++++++++++++++- lib/Kconfig.debug | 16 ++++++ 5 files changed, 150 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/Documentation/feature-removal-schedule.txt b/Documentation/feature-removal-schedule.txt index 027285d0c26..033ac91da07 100644 --- a/Documentation/feature-removal-schedule.txt +++ b/Documentation/feature-removal-schedule.txt @@ -177,6 +177,16 @@ Who: Jean Delvare --------------------------- +What: Unused EXPORT_SYMBOL/EXPORT_SYMBOL_GPL exports + (temporary transition config option provided until then) + The transition config option will also be removed at the same time. +When: before 2.6.19 +Why: Unused symbols are both increasing the size of the kernel binary + and are often a sign of "wrong API" +Who: Arjan van de Ven + +--------------------------- + What: remove EXPORT_SYMBOL(tasklist_lock) When: August 2006 Files: kernel/fork.c diff --git a/include/asm-generic/vmlinux.lds.h b/include/asm-generic/vmlinux.lds.h index 9d11550b481..db5a3732f10 100644 --- a/include/asm-generic/vmlinux.lds.h +++ b/include/asm-generic/vmlinux.lds.h @@ -58,6 +58,20 @@ VMLINUX_SYMBOL(__stop___ksymtab_gpl) = .; \ } \ \ + /* Kernel symbol table: Normal unused symbols */ \ + __ksymtab_unused : AT(ADDR(__ksymtab_unused) - LOAD_OFFSET) { \ + VMLINUX_SYMBOL(__start___ksymtab_unused) = .; \ + *(__ksymtab_unused) \ + VMLINUX_SYMBOL(__stop___ksymtab_unused) = .; \ + } \ + \ + /* Kernel symbol table: GPL-only unused symbols */ \ + __ksymtab_unused_gpl : AT(ADDR(__ksymtab_unused_gpl) - LOAD_OFFSET) { \ + VMLINUX_SYMBOL(__start___ksymtab_unused_gpl) = .; \ + *(__ksymtab_unused_gpl) \ + VMLINUX_SYMBOL(__stop___ksymtab_unused_gpl) = .; \ + } \ + \ /* Kernel symbol table: GPL-future-only symbols */ \ __ksymtab_gpl_future : AT(ADDR(__ksymtab_gpl_future) - LOAD_OFFSET) { \ VMLINUX_SYMBOL(__start___ksymtab_gpl_future) = .; \ @@ -79,6 +93,20 @@ VMLINUX_SYMBOL(__stop___kcrctab_gpl) = .; \ } \ \ + /* Kernel symbol table: Normal unused symbols */ \ + __kcrctab_unused : AT(ADDR(__kcrctab_unused) - LOAD_OFFSET) { \ + VMLINUX_SYMBOL(__start___kcrctab_unused) = .; \ + *(__kcrctab_unused) \ + VMLINUX_SYMBOL(__stop___kcrctab_unused) = .; \ + } \ + \ + /* Kernel symbol table: GPL-only unused symbols */ \ + __kcrctab_unused_gpl : AT(ADDR(__kcrctab_unused_gpl) - LOAD_OFFSET) { \ + VMLINUX_SYMBOL(__start___kcrctab_unused_gpl) = .; \ + *(__kcrctab_unused_gpl) \ + VMLINUX_SYMBOL(__stop___kcrctab_unused_gpl) = .; \ + } \ + \ /* Kernel symbol table: GPL-future-only symbols */ \ __kcrctab_gpl_future : AT(ADDR(__kcrctab_gpl_future) - LOAD_OFFSET) { \ VMLINUX_SYMBOL(__start___kcrctab_gpl_future) = .; \ diff --git a/include/linux/module.h b/include/linux/module.h index 9ebbb74b7b7..9e9dc7c24d9 100644 --- a/include/linux/module.h +++ b/include/linux/module.h @@ -203,6 +203,15 @@ void *__symbol_get_gpl(const char *symbol); #define EXPORT_SYMBOL_GPL_FUTURE(sym) \ __EXPORT_SYMBOL(sym, "_gpl_future") + +#ifdef CONFIG_UNUSED_SYMBOLS +#define EXPORT_UNUSED_SYMBOL(sym) __EXPORT_SYMBOL(sym, "_unused") +#define EXPORT_UNUSED_SYMBOL_GPL(sym) __EXPORT_SYMBOL(sym, "_unused_gpl") +#else +#define EXPORT_UNUSED_SYMBOL(sym) +#define EXPORT_UNUSED_SYMBOL_GPL(sym) +#endif + #endif struct module_ref @@ -261,6 +270,15 @@ struct module unsigned int num_gpl_syms; const unsigned long *gpl_crcs; + /* unused exported symbols. */ + const struct kernel_symbol *unused_syms; + unsigned int num_unused_syms; + const unsigned long *unused_crcs; + /* GPL-only, unused exported symbols. */ + const struct kernel_symbol *unused_gpl_syms; + unsigned int num_unused_gpl_syms; + const unsigned long *unused_gpl_crcs; + /* symbols that will be GPL-only in the near future. */ const struct kernel_symbol *gpl_future_syms; unsigned int num_gpl_future_syms; @@ -456,6 +474,8 @@ void module_remove_driver(struct device_driver *); #define EXPORT_SYMBOL(sym) #define EXPORT_SYMBOL_GPL(sym) #define EXPORT_SYMBOL_GPL_FUTURE(sym) +#define EXPORT_UNUSED_SYMBOL(sym) +#define EXPORT_UNUSED_SYMBOL_GPL(sym) /* Given an address, look for it in the exception tables. */ static inline const struct exception_table_entry * diff --git a/kernel/module.c b/kernel/module.c index 10e5b872adf..03b738172a8 100644 --- a/kernel/module.c +++ b/kernel/module.c @@ -1,4 +1,4 @@ -/* Rewritten by Rusty Russell, on the backs of many others... +/* Copyright (C) 2002 Richard Henderson Copyright (C) 2001 Rusty Russell, 2002 Rusty Russell IBM. @@ -122,9 +122,17 @@ extern const struct kernel_symbol __start___ksymtab_gpl[]; extern const struct kernel_symbol __stop___ksymtab_gpl[]; extern const struct kernel_symbol __start___ksymtab_gpl_future[]; extern const struct kernel_symbol __stop___ksymtab_gpl_future[]; +extern const struct kernel_symbol __start___ksymtab_unused[]; +extern const struct kernel_symbol __stop___ksymtab_unused[]; +extern const struct kernel_symbol __start___ksymtab_unused_gpl[]; +extern const struct kernel_symbol __stop___ksymtab_unused_gpl[]; +extern const struct kernel_symbol __start___ksymtab_gpl_future[]; +extern const struct kernel_symbol __stop___ksymtab_gpl_future[]; extern const unsigned long __start___kcrctab[]; extern const unsigned long __start___kcrctab_gpl[]; extern const unsigned long __start___kcrctab_gpl_future[]; +extern const unsigned long __start___kcrctab_unused[]; +extern const unsigned long __start___kcrctab_unused_gpl[]; #ifndef CONFIG_MODVERSIONS #define symversion(base, idx) NULL @@ -144,6 +152,17 @@ static const struct kernel_symbol *lookup_symbol(const char *name, return NULL; } +static void printk_unused_warning(const char *name) +{ + printk(KERN_WARNING "Symbol %s is marked as UNUSED, " + "however this module is using it.\n", name); + printk(KERN_WARNING "This symbol will go away in the future.\n"); + printk(KERN_WARNING "Please evalute if this is the right api to use, " + "and if it really is, submit a report the linux kernel " + "mailinglist together with submitting your code for " + "inclusion.\n"); +} + /* Find a symbol, return value, crc and module which owns it */ static unsigned long __find_symbol(const char *name, struct module **owner, @@ -186,6 +205,25 @@ static unsigned long __find_symbol(const char *name, return ks->value; } + ks = lookup_symbol(name, __start___ksymtab_unused, + __stop___ksymtab_unused); + if (ks) { + printk_unused_warning(name); + *crc = symversion(__start___kcrctab_unused, + (ks - __start___ksymtab_unused)); + return ks->value; + } + + if (gplok) + ks = lookup_symbol(name, __start___ksymtab_unused_gpl, + __stop___ksymtab_unused_gpl); + if (ks) { + printk_unused_warning(name); + *crc = symversion(__start___kcrctab_unused_gpl, + (ks - __start___ksymtab_unused_gpl)); + return ks->value; + } + /* Now try modules. */ list_for_each_entry(mod, &modules, list) { *owner = mod; @@ -204,6 +242,23 @@ static unsigned long __find_symbol(const char *name, return ks->value; } } + ks = lookup_symbol(name, mod->unused_syms, mod->unused_syms + mod->num_unused_syms); + if (ks) { + printk_unused_warning(name); + *crc = symversion(mod->unused_crcs, (ks - mod->unused_syms)); + return ks->value; + } + + if (gplok) { + ks = lookup_symbol(name, mod->unused_gpl_syms, + mod->unused_gpl_syms + mod->num_unused_gpl_syms); + if (ks) { + printk_unused_warning(name); + *crc = symversion(mod->unused_gpl_crcs, + (ks - mod->unused_gpl_syms)); + return ks->value; + } + } ks = lookup_symbol(name, mod->gpl_future_syms, (mod->gpl_future_syms + mod->num_gpl_future_syms)); @@ -1407,6 +1462,8 @@ static struct module *load_module(void __user *umod, exportindex, modindex, obsparmindex, infoindex, gplindex, crcindex, gplcrcindex, versindex, pcpuindex, gplfutureindex, gplfuturecrcindex, unwindex = 0; + unsigned int unusedindex, unusedcrcindex, unusedgplindex, + unusedgplcrcindex; struct module *mod; long err = 0; void *percpu = NULL, *ptr = NULL; /* Stops spurious gcc warning */ @@ -1487,9 +1544,13 @@ static struct module *load_module(void __user *umod, exportindex = find_sec(hdr, sechdrs, secstrings, "__ksymtab"); gplindex = find_sec(hdr, sechdrs, secstrings, "__ksymtab_gpl"); gplfutureindex = find_sec(hdr, sechdrs, secstrings, "__ksymtab_gpl_future"); + unusedindex = find_sec(hdr, sechdrs, secstrings, "__ksymtab_unused"); + unusedgplindex = find_sec(hdr, sechdrs, secstrings, "__ksymtab_unused_gpl"); crcindex = find_sec(hdr, sechdrs, secstrings, "__kcrctab"); gplcrcindex = find_sec(hdr, sechdrs, secstrings, "__kcrctab_gpl"); gplfuturecrcindex = find_sec(hdr, sechdrs, secstrings, "__kcrctab_gpl_future"); + unusedcrcindex = find_sec(hdr, sechdrs, secstrings, "__kcrctab_unused"); + unusedgplcrcindex = find_sec(hdr, sechdrs, secstrings, "__kcrctab_unused_gpl"); setupindex = find_sec(hdr, sechdrs, secstrings, "__param"); exindex = find_sec(hdr, sechdrs, secstrings, "__ex_table"); obsparmindex = find_sec(hdr, sechdrs, secstrings, "__obsparm"); @@ -1638,14 +1699,27 @@ static struct module *load_module(void __user *umod, mod->gpl_crcs = (void *)sechdrs[gplcrcindex].sh_addr; mod->num_gpl_future_syms = sechdrs[gplfutureindex].sh_size / sizeof(*mod->gpl_future_syms); + mod->num_unused_syms = sechdrs[unusedindex].sh_size / + sizeof(*mod->unused_syms); + mod->num_unused_gpl_syms = sechdrs[unusedgplindex].sh_size / + sizeof(*mod->unused_gpl_syms); mod->gpl_future_syms = (void *)sechdrs[gplfutureindex].sh_addr; if (gplfuturecrcindex) mod->gpl_future_crcs = (void *)sechdrs[gplfuturecrcindex].sh_addr; + mod->unused_syms = (void *)sechdrs[unusedindex].sh_addr; + if (unusedcrcindex) + mod->unused_crcs = (void *)sechdrs[unusedcrcindex].sh_addr; + mod->unused_gpl_syms = (void *)sechdrs[unusedgplindex].sh_addr; + if (unusedgplcrcindex) + mod->unused_crcs = (void *)sechdrs[unusedgplcrcindex].sh_addr; + #ifdef CONFIG_MODVERSIONS if ((mod->num_syms && !crcindex) || (mod->num_gpl_syms && !gplcrcindex) || - (mod->num_gpl_future_syms && !gplfuturecrcindex)) { + (mod->num_gpl_future_syms && !gplfuturecrcindex) || + (mod->num_unused_syms && !unusedcrcindex) || + (mod->num_unused_gpl_syms && !unusedgplcrcindex)) { printk(KERN_WARNING "%s: No versions for exported symbols." " Tainting kernel.\n", mod->name); add_taint(TAINT_FORCED_MODULE); diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug index 5330911ebd3..e4fcbd12cf6 100644 --- a/lib/Kconfig.debug +++ b/lib/Kconfig.debug @@ -23,6 +23,22 @@ config MAGIC_SYSRQ keys are documented in . Don't say Y unless you really know what this hack does. +config UNUSED_SYMBOLS + bool "Enable unused/obsolete exported symbols" + default y if X86 + help + Unused but exported symbols make the kernel needlessly bigger. For + that reason most of these unused exports will soon be removed. This + option is provided temporarily to provide a transition period in case + some external kernel module needs one of these symbols anyway. If you + encounter such a case in your module, consider if you are actually + using the right API. (rationale: since nobody in the kernel is using + this in a module, there is a pretty good chance it's actually the + wrong interface to use). If you really need the symbol, please send a + mail to the linux kernel mailing list mentioning the symbol and why + you really need it, and what the merge plan to the mainline kernel for + your module is. + config DEBUG_KERNEL bool "Kernel debugging" help -- cgit v1.2.3-70-g09d2 From 817d6d3bceaf34c99f5343820f9b9e6021f0655c Mon Sep 17 00:00:00 2001 From: Paul Fulghum Date: Wed, 28 Jun 2006 04:26:47 -0700 Subject: [PATCH] remove TTY_DONT_FLIP Remove TTY_DONT_FLIP tty flag. This flag was introduced in 2.1.X kernels to prevent the N_TTY line discipline functions read_chan() and n_tty_receive_buf() from running at the same time. 2.2.15 introduced tty->read_lock to protect access to the N_TTY read buffer, which is the only state requiring protection between these two functions. The current TTY_DONT_FLIP implementation is broken for SMP, and is not universally honored by drivers that send data directly to the line discipline receive_buf function. Because TTY_DONT_FLIP is not necessary, is broken in implementation, and is not universally honored, it is removed. Signed-off-by: Paul Fulghum Cc: Alan Cox Cc: Theodore Ts'o Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/char/mxser.c | 1 - drivers/char/n_tty.c | 6 +----- drivers/char/pty.c | 2 +- drivers/char/tty_io.c | 15 ++------------- drivers/serial/crisv10.c | 6 ------ drivers/serial/jsm/jsm_tty.c | 7 ------- drivers/usb/serial/ir-usb.c | 3 +-- include/linux/tty.h | 1 - net/bluetooth/rfcomm/tty.c | 8 ++------ 9 files changed, 7 insertions(+), 42 deletions(-) (limited to 'include') diff --git a/drivers/char/mxser.c b/drivers/char/mxser.c index 645d9d713ae..72cfd09091e 100644 --- a/drivers/char/mxser.c +++ b/drivers/char/mxser.c @@ -996,7 +996,6 @@ static int mxser_open(struct tty_struct *tty, struct file *filp) info->session = current->signal->session; info->pgrp = process_group(current); - clear_bit(TTY_DONT_FLIP, &tty->flags); /* status = mxser_get_msr(info->base, 0, info->port); diff --git a/drivers/char/n_tty.c b/drivers/char/n_tty.c index b9371d5bf79..603b9ade5eb 100644 --- a/drivers/char/n_tty.c +++ b/drivers/char/n_tty.c @@ -1132,7 +1132,7 @@ static inline int input_available_p(struct tty_struct *tty, int amt) * buffer, and once to drain the space from the (physical) beginning of * the buffer to head pointer. * - * Called under the tty->atomic_read_lock sem and with TTY_DONT_FLIP set + * Called under the tty->atomic_read_lock sem * */ @@ -1271,7 +1271,6 @@ do_it_again: } add_wait_queue(&tty->read_wait, &wait); - set_bit(TTY_DONT_FLIP, &tty->flags); while (nr) { /* First test for status change. */ if (tty->packet && tty->link->ctrl_status) { @@ -1315,9 +1314,7 @@ do_it_again: break; } n_tty_set_room(tty); - clear_bit(TTY_DONT_FLIP, &tty->flags); timeout = schedule_timeout(timeout); - set_bit(TTY_DONT_FLIP, &tty->flags); continue; } __set_current_state(TASK_RUNNING); @@ -1394,7 +1391,6 @@ do_it_again: if (time) timeout = time; } - clear_bit(TTY_DONT_FLIP, &tty->flags); mutex_unlock(&tty->atomic_read_lock); remove_wait_queue(&tty->read_wait, &wait); diff --git a/drivers/char/pty.c b/drivers/char/pty.c index 9b5a2c0e700..0c17f61549b 100644 --- a/drivers/char/pty.c +++ b/drivers/char/pty.c @@ -101,7 +101,7 @@ static void pty_unthrottle(struct tty_struct * tty) * * FIXME: Our pty_write method is called with our ldisc lock held but * not our partners. We can't just take the other one blindly without - * risking deadlocks. There is also the small matter of TTY_DONT_FLIP + * risking deadlocks. */ static int pty_write(struct tty_struct * tty, const unsigned char *buf, int count) { diff --git a/drivers/char/tty_io.c b/drivers/char/tty_io.c index bd74e82d8a7..b846d87f2b5 100644 --- a/drivers/char/tty_io.c +++ b/drivers/char/tty_io.c @@ -784,11 +784,8 @@ restart: } clear_bit(TTY_LDISC, &tty->flags); - clear_bit(TTY_DONT_FLIP, &tty->flags); - if (o_tty) { + if (o_tty) clear_bit(TTY_LDISC, &o_tty->flags); - clear_bit(TTY_DONT_FLIP, &o_tty->flags); - } spin_unlock_irqrestore(&tty_ldisc_lock, flags); /* @@ -1955,7 +1952,6 @@ static void release_dev(struct file * filp) * race with the set_ldisc code path. */ clear_bit(TTY_LDISC, &tty->flags); - clear_bit(TTY_DONT_FLIP, &tty->flags); cancel_delayed_work(&tty->buf.work); /* @@ -2784,13 +2780,6 @@ static void flush_to_ldisc(void *private_) if (disc == NULL) /* !TTY_LDISC */ return; - if (test_bit(TTY_DONT_FLIP, &tty->flags)) { - /* - * Do it after the next timer tick: - */ - schedule_delayed_work(&tty->buf.work, 1); - goto out; - } spin_lock_irqsave(&tty->buf.lock, flags); while((tbuf = tty->buf.head) != NULL) { while ((count = tbuf->commit - tbuf->read) != 0) { @@ -2809,7 +2798,7 @@ static void flush_to_ldisc(void *private_) tty_buffer_free(tty, tbuf); } spin_unlock_irqrestore(&tty->buf.lock, flags); -out: + tty_ldisc_deref(disc); } diff --git a/drivers/serial/crisv10.c b/drivers/serial/crisv10.c index 89700141f87..5cacc5e74a9 100644 --- a/drivers/serial/crisv10.c +++ b/drivers/serial/crisv10.c @@ -2573,12 +2573,6 @@ static void flush_to_flip_buffer(struct e100_serial *info) DFLIP( if (1) { - - if (test_bit(TTY_DONT_FLIP, &tty->flags)) { - DEBUG_LOG(info->line, "*** TTY_DONT_FLIP set flip.count %i ***\n", tty->flip.count); - DEBUG_LOG(info->line, "*** recv_cnt %i\n", info->recv_cnt); - } else { - } DEBUG_LOG(info->line, "*** rxtot %i\n", info->icount.rx); DEBUG_LOG(info->line, "ldisc %lu\n", tty->ldisc.chars_in_buffer(tty)); DEBUG_LOG(info->line, "room %lu\n", tty->ldisc.receive_room(tty)); diff --git a/drivers/serial/jsm/jsm_tty.c b/drivers/serial/jsm/jsm_tty.c index 7d823705193..f8262e6ad8d 100644 --- a/drivers/serial/jsm/jsm_tty.c +++ b/drivers/serial/jsm/jsm_tty.c @@ -588,13 +588,6 @@ void jsm_input(struct jsm_channel *ch) len = min(len, (N_TTY_BUF_SIZE - 1) - tp->read_cnt); ld = tty_ldisc_ref(tp); - /* - * If the DONT_FLIP flag is on, don't flush our buffer, and act - * like the ld doesn't have any space to put the data right now. - */ - if (test_bit(TTY_DONT_FLIP, &tp->flags)) - len = 0; - /* * If we were unable to get a reference to the ld, * don't flush our buffer, and act like the ld doesn't diff --git a/drivers/usb/serial/ir-usb.c b/drivers/usb/serial/ir-usb.c index 9432c730227..d7f3f736a69 100644 --- a/drivers/usb/serial/ir-usb.c +++ b/drivers/usb/serial/ir-usb.c @@ -453,8 +453,7 @@ static void ir_read_bulk_callback (struct urb *urb, struct pt_regs *regs) tty = port->tty; /* - * FIXME: must not do this in IRQ context, - * must honour TTY_DONT_FLIP + * FIXME: must not do this in IRQ context */ tty->ldisc.receive_buf( tty, diff --git a/include/linux/tty.h b/include/linux/tty.h index cb35ca50a0a..341cc4552c0 100644 --- a/include/linux/tty.h +++ b/include/linux/tty.h @@ -259,7 +259,6 @@ struct tty_struct { #define TTY_DO_WRITE_WAKEUP 5 /* Call write_wakeup after queuing new */ #define TTY_PUSH 6 /* n_tty private */ #define TTY_CLOSING 7 /* ->close() in progress */ -#define TTY_DONT_FLIP 8 /* Defer buffer flip */ #define TTY_LDISC 9 /* Line discipline attached */ #define TTY_HW_COOK_OUT 14 /* Hardware can do output cooking */ #define TTY_HW_COOK_IN 15 /* Hardware can do input cooking */ diff --git a/net/bluetooth/rfcomm/tty.c b/net/bluetooth/rfcomm/tty.c index 74368f79ee5..b81fad89332 100644 --- a/net/bluetooth/rfcomm/tty.c +++ b/net/bluetooth/rfcomm/tty.c @@ -480,12 +480,8 @@ static void rfcomm_dev_data_ready(struct rfcomm_dlc *dlc, struct sk_buff *skb) BT_DBG("dlc %p tty %p len %d", dlc, tty, skb->len); - if (test_bit(TTY_DONT_FLIP, &tty->flags)) { - tty_buffer_request_room(tty, skb->len); - tty_insert_flip_string(tty, skb->data, skb->len); - tty_flip_buffer_push(tty); - } else - tty->ldisc.receive_buf(tty, skb->data, NULL, skb->len); + tty_insert_flip_string(tty, skb->data, skb->len); + tty_flip_buffer_push(tty); kfree_skb(skb); } -- cgit v1.2.3-70-g09d2 From 33b37a33c242542fac2980b8ccd90977388b7a8d Mon Sep 17 00:00:00 2001 From: Paul Fulghum Date: Wed, 28 Jun 2006 04:26:49 -0700 Subject: [PATCH] remove active field from tty buffer structure Remove 'active' field from tty buffer structure. This was added in 2.6.16 as part of a patch to make the new tty buffering SMP safe. This field is unnecessary with the more intelligently written flush_to_ldisc that adds receive_room handling. Removing this field reverts to simpler logic where the tail buffer is always the 'active' buffer, which should not be freed by flush_to_ldisc. (active == buffer being filled with new data) The result is simpler, smaller, and faster tty buffer code. Signed-off-by: Paul Fulghum Cc: Alan Cox Cc: Theodore Ts'o Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/char/tty_io.c | 16 ++++------------ include/linux/kbd_kern.h | 4 +--- include/linux/tty.h | 1 - include/linux/tty_flip.h | 2 +- 4 files changed, 6 insertions(+), 17 deletions(-) (limited to 'include') diff --git a/drivers/char/tty_io.c b/drivers/char/tty_io.c index 1f03ebf165d..8d19f7281f0 100644 --- a/drivers/char/tty_io.c +++ b/drivers/char/tty_io.c @@ -267,7 +267,6 @@ static struct tty_buffer *tty_buffer_alloc(size_t size) p->used = 0; p->size = size; p->next = NULL; - p->active = 0; p->commit = 0; p->read = 0; p->char_buf_ptr = (char *)(p->data); @@ -327,10 +326,9 @@ int tty_buffer_request_room(struct tty_struct *tty, size_t size) /* OPTIMISATION: We could keep a per tty "zero" sized buffer to remove this conditional if its worth it. This would be invisible to the callers */ - if ((b = tty->buf.tail) != NULL) { + if ((b = tty->buf.tail) != NULL) left = b->size - b->used; - b->active = 1; - } else + else left = 0; if (left < size) { @@ -338,12 +336,10 @@ int tty_buffer_request_room(struct tty_struct *tty, size_t size) if ((n = tty_buffer_find(tty, size)) != NULL) { if (b != NULL) { b->next = n; - b->active = 0; b->commit = b->used; } else tty->buf.head = n; tty->buf.tail = n; - n->active = 1; } else size = left; } @@ -404,10 +400,8 @@ void tty_schedule_flip(struct tty_struct *tty) { unsigned long flags; spin_lock_irqsave(&tty->buf.lock, flags); - if (tty->buf.tail != NULL) { - tty->buf.tail->active = 0; + if (tty->buf.tail != NULL) tty->buf.tail->commit = tty->buf.tail->used; - } spin_unlock_irqrestore(&tty->buf.lock, flags); schedule_delayed_work(&tty->buf.work, 1); } @@ -2902,10 +2896,8 @@ void tty_flip_buffer_push(struct tty_struct *tty) { unsigned long flags; spin_lock_irqsave(&tty->buf.lock, flags); - if (tty->buf.tail != NULL) { - tty->buf.tail->active = 0; + if (tty->buf.tail != NULL) tty->buf.tail->commit = tty->buf.tail->used; - } spin_unlock_irqrestore(&tty->buf.lock, flags); if (tty->low_latency) diff --git a/include/linux/kbd_kern.h b/include/linux/kbd_kern.h index 4eb851ece08..efe0ee4cc80 100644 --- a/include/linux/kbd_kern.h +++ b/include/linux/kbd_kern.h @@ -155,10 +155,8 @@ static inline void con_schedule_flip(struct tty_struct *t) { unsigned long flags; spin_lock_irqsave(&t->buf.lock, flags); - if (t->buf.tail != NULL) { - t->buf.tail->active = 0; + if (t->buf.tail != NULL) t->buf.tail->commit = t->buf.tail->used; - } spin_unlock_irqrestore(&t->buf.lock, flags); schedule_work(&t->buf.work); } diff --git a/include/linux/tty.h b/include/linux/tty.h index 341cc4552c0..b3b807e4b05 100644 --- a/include/linux/tty.h +++ b/include/linux/tty.h @@ -57,7 +57,6 @@ struct tty_buffer { unsigned char *flag_buf_ptr; int used; int size; - int active; int commit; int read; /* Data points here */ diff --git a/include/linux/tty_flip.h b/include/linux/tty_flip.h index 31548303ee3..eb677cf5610 100644 --- a/include/linux/tty_flip.h +++ b/include/linux/tty_flip.h @@ -12,7 +12,7 @@ static inline int tty_insert_flip_char(struct tty_struct *tty, unsigned char ch, char flag) { struct tty_buffer *tb = tty->buf.tail; - if (tb && tb->active && tb->used < tb->size) { + if (tb && tb->used < tb->size) { tb->flag_buf_ptr[tb->used] = flag; tb->char_buf_ptr[tb->used++] = ch; return 1; -- cgit v1.2.3-70-g09d2 From be6990e7473fcd11becda747c24b94a478413245 Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Wed, 28 Jun 2006 04:26:57 -0700 Subject: [PATCH] ac97_codec: make bitfield unsigned Make a 1-bit bitfield unsigned (no space for sign bit). Removes 24 sparse warnings from this one file: include/linux/ac97_codec.h:262:13: error: dubious one-bit signed bitfield Signed-off-by: Randy Dunlap Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/ac97_codec.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include') diff --git a/include/linux/ac97_codec.h b/include/linux/ac97_codec.h index c35833824e1..2ed2fd85513 100644 --- a/include/linux/ac97_codec.h +++ b/include/linux/ac97_codec.h @@ -259,7 +259,7 @@ struct ac97_codec { int type; u32 model; - int modem:1; + unsigned int modem:1; struct ac97_ops *codec_ops; -- cgit v1.2.3-70-g09d2 From da574af755bcb1d604e01feadf2a8c31b364447c Mon Sep 17 00:00:00 2001 From: Alan Cox Date: Wed, 28 Jun 2006 04:27:01 -0700 Subject: [PATCH] ide: fix error handling for drives which clear the FIFO on error If the controller FIFO cleared automatically on error we must not try and drain it as this will hang some chips. Based in concept on a broken patch from -mm some while back Signed-off-by: Alan Cox Cc: Sergei Shtylyov Cc: Bartlomiej Zolnierkiewicz Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/ide/ide-io.c | 2 +- include/linux/ide.h | 1 + 2 files changed, 2 insertions(+), 1 deletion(-) (limited to 'include') diff --git a/drivers/ide/ide-io.c b/drivers/ide/ide-io.c index 935cb258377..26ceab1e90b 100644 --- a/drivers/ide/ide-io.c +++ b/drivers/ide/ide-io.c @@ -505,7 +505,7 @@ static ide_startstop_t ide_ata_error(ide_drive_t *drive, struct request *rq, u8 } } - if ((stat & DRQ_STAT) && rq_data_dir(rq) == READ) + if ((stat & DRQ_STAT) && rq_data_dir(rq) == READ && hwif->err_stops_fifo == 0) try_to_flush_leftover_data(drive); if (hwif->INB(IDE_STATUS_REG) & (BUSY_STAT|DRQ_STAT)) diff --git a/include/linux/ide.h b/include/linux/ide.h index ef7bef207f4..0c100168c0c 100644 --- a/include/linux/ide.h +++ b/include/linux/ide.h @@ -793,6 +793,7 @@ typedef struct hwif_s { unsigned auto_poll : 1; /* supports nop auto-poll */ unsigned sg_mapped : 1; /* sg_table and sg_nents are ready */ unsigned no_io_32bit : 1; /* 1 = can not do 32-bit IO ops */ + unsigned err_stops_fifo : 1; /* 1=data FIFO is cleared by an error */ struct device gendev; struct completion gendev_rel_comp; /* To deal with device release() */ -- cgit v1.2.3-70-g09d2 From 0aee77b221a68bf8ef5401712bb55c89e36b461c Mon Sep 17 00:00:00 2001 From: Greg Ungerer Date: Wed, 28 Jun 2006 16:12:45 +1000 Subject: [PATCH] m68knommu: fix clobber list in uCdimm/uCsimm helper asm Fix clobber list in uCsimm/uCdimm boot load helper asm. Signed-off-by: Greg Ungerer Signed-off-by: Linus Torvalds --- include/asm-m68knommu/bootstd.h | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) (limited to 'include') diff --git a/include/asm-m68knommu/bootstd.h b/include/asm-m68knommu/bootstd.h index 3fdc79f06d5..bdc1a4ac4fe 100644 --- a/include/asm-m68knommu/bootstd.h +++ b/include/asm-m68knommu/bootstd.h @@ -52,7 +52,7 @@ type name(void) \ __asm__ __volatile__ ("trap #2" \ : "=g" (__res) \ : "0" (__res) \ - : "%d0"); \ + ); \ __bsc_return(type,__res); \ } @@ -64,7 +64,7 @@ type name(atype a) \ __asm__ __volatile__ ("trap #2" \ : "=g" (__res) \ : "0" (__res), "d" (__a) \ - : "%d0"); \ + ); \ __bsc_return(type,__res); \ } @@ -77,7 +77,7 @@ type name(atype a, btype b) \ __asm__ __volatile__ ("trap #2" \ : "=g" (__res) \ : "0" (__res), "d" (__a), "d" (__b) \ - : "%d0"); \ + ); \ __bsc_return(type,__res); \ } @@ -92,7 +92,7 @@ type name(atype a, btype b, ctype c) \ : "=g" (__res) \ : "0" (__res), "d" (__a), "d" (__b), \ "d" (__c) \ - : "%d0"); \ + ); \ __bsc_return(type,__res); \ } @@ -108,7 +108,7 @@ type name(atype a, btype b, ctype c, dtype d) \ : "=g" (__res) \ : "0" (__res), "d" (__a), "d" (__b), \ "d" (__c), "d" (__d) \ - : "%d0"); \ + ); \ __bsc_return(type,__res); \ } @@ -125,7 +125,7 @@ type name(atype a, btype b, ctype c, dtype d, etype e) \ : "=g" (__res) \ : "0" (__res), "d" (__a), "d" (__b), \ "d" (__c), "d" (__d), "d" (__e) \ - : "%d0"); \ + ); \ __bsc_return(type,__res); \ } -- cgit v1.2.3-70-g09d2 From 980a01c9bfb090cb8a991e39e56ac379c30c61b8 Mon Sep 17 00:00:00 2001 From: David Brownell Date: Wed, 28 Jun 2006 07:47:15 -0700 Subject: [PATCH] SPI: infrastructure to initialize spi_device.mode early This patch adds earlier initialization of spi_device.mode, as needed on boards using nondefault chipselect polarity. An example would be ones using the RS5C348 RTC without an external signal inverter between the RTC chipselect and the SPI controller. Without this mechanism, the first setup() call for that chip would wrongly enable chips, corrupting transfers to/from other chips sharing that SPI bus. Signed-off-by: David Brownell Signed-off-by: Linus Torvalds --- drivers/spi/spi.c | 1 + include/linux/spi/spi.h | 6 +++++- 2 files changed, 6 insertions(+), 1 deletion(-) (limited to 'include') diff --git a/drivers/spi/spi.c b/drivers/spi/spi.c index 1cea4a6799f..ed1cdf6ac8f 100644 --- a/drivers/spi/spi.c +++ b/drivers/spi/spi.c @@ -210,6 +210,7 @@ spi_new_device(struct spi_master *master, struct spi_board_info *chip) proxy->master = master; proxy->chip_select = chip->chip_select; proxy->max_speed_hz = chip->max_speed_hz; + proxy->mode = chip->mode; proxy->irq = chip->irq; proxy->modalias = chip->modalias; diff --git a/include/linux/spi/spi.h b/include/linux/spi/spi.h index e928c0dcc29..c8bb68099eb 100644 --- a/include/linux/spi/spi.h +++ b/include/linux/spi/spi.h @@ -642,10 +642,14 @@ struct spi_board_info { u16 bus_num; u16 chip_select; + /* mode becomes spi_device.mode, and is essential for chips + * where the default of SPI_CS_HIGH = 0 is wrong. + */ + u8 mode; + /* ... may need additional spi_device chip config data here. * avoid stuff protocol drivers can set; but include stuff * needed to behave without being bound to a driver: - * - chipselect polarity * - quirks like clock rate mattering when not selected */ }; -- cgit v1.2.3-70-g09d2 From b44597906e03d5e2b467c17a3b73585596c0d7be Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Wed, 28 Jun 2006 17:14:07 +0200 Subject: [PATCH] Fix plist include dependency plist.h uses container_of, which is defined in kernel.h. Include kernel.h in plist.h as the kernel.h include does not longer happen automatically on all architectures. Signed-off-by: Thomas Gleixner Signed-off-by: Linus Torvalds --- include/linux/plist.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include') diff --git a/include/linux/plist.h b/include/linux/plist.h index 3404faef542..b95818a037a 100644 --- a/include/linux/plist.h +++ b/include/linux/plist.h @@ -73,6 +73,7 @@ #ifndef _LINUX_PLIST_H_ #define _LINUX_PLIST_H_ +#include #include #include -- cgit v1.2.3-70-g09d2 From 1729dc7833014eaba3520091a673dbdc9242d9f9 Mon Sep 17 00:00:00 2001 From: Paul Mackerras Date: Thu, 29 Jun 2006 16:16:15 +1000 Subject: [POWERPC] Correct the MAX_CONTEXT definition When we increased the address space per process to 2^44 bytes, the number of contexts that we could actually use reduced, but we forgot to decrease the MAX_CONTEXT definition. (Fortunately this would only cause problems if we actually had more than 512k user processes running.) This patch corrects the definition. Signed-off-by: Paul Mackerras --- include/asm-powerpc/mmu_context.h | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) (limited to 'include') diff --git a/include/asm-powerpc/mmu_context.h b/include/asm-powerpc/mmu_context.h index 8c6b1a6d944..083ac917bd2 100644 --- a/include/asm-powerpc/mmu_context.h +++ b/include/asm-powerpc/mmu_context.h @@ -25,8 +25,13 @@ static inline void enter_lazy_tlb(struct mm_struct *mm, { } +/* + * The proto-VSID space has 2^35 - 1 segments available for user mappings. + * Each segment contains 2^28 bytes. Each context maps 2^44 bytes, + * so we can support 2^19-1 contexts (19 == 35 + 28 - 44). + */ #define NO_CONTEXT 0 -#define MAX_CONTEXT (0x100000-1) +#define MAX_CONTEXT ((1UL << 19) - 1) extern int init_new_context(struct task_struct *tsk, struct mm_struct *mm); extern void destroy_context(struct mm_struct *mm); -- cgit v1.2.3-70-g09d2 From 2b9d7467a6dbc41872c605511e105cbde1eda100 Mon Sep 17 00:00:00 2001 From: Zang Roy-r61911 Date: Tue, 13 Jun 2006 15:07:23 +0800 Subject: [POWERPC] Add tsi108 pci and platform device data register function Add Tundra Semiconductor tsi108 pci and platform device data register function support. Signed-off-by: Alexandre Bounine Signed-off-by: Roy Zang --- Signed-off-by: Paul Mackerras --- arch/powerpc/sysdev/Makefile | 1 + arch/powerpc/sysdev/tsi108_dev.c | 145 ++++++++++++++ arch/powerpc/sysdev/tsi108_pci.c | 412 +++++++++++++++++++++++++++++++++++++++ include/asm-powerpc/tsi108.h | 109 +++++++++++ 4 files changed, 667 insertions(+) create mode 100644 arch/powerpc/sysdev/tsi108_dev.c create mode 100644 arch/powerpc/sysdev/tsi108_pci.c create mode 100644 include/asm-powerpc/tsi108.h (limited to 'include') diff --git a/arch/powerpc/sysdev/Makefile b/arch/powerpc/sysdev/Makefile index cef95b02373..63e81e74b45 100644 --- a/arch/powerpc/sysdev/Makefile +++ b/arch/powerpc/sysdev/Makefile @@ -12,3 +12,4 @@ obj-$(CONFIG_U3_DART) += dart_iommu.o obj-$(CONFIG_MMIO_NVRAM) += mmio_nvram.o obj-$(CONFIG_PPC_83xx) += ipic.o obj-$(CONFIG_FSL_SOC) += fsl_soc.o +obj-$(CONFIG_TSI108_BRIDGE) += tsi108_pci.o tsi108_dev.o diff --git a/arch/powerpc/sysdev/tsi108_dev.c b/arch/powerpc/sysdev/tsi108_dev.c new file mode 100644 index 00000000000..26a0cc820cd --- /dev/null +++ b/arch/powerpc/sysdev/tsi108_dev.c @@ -0,0 +1,145 @@ +/* + * tsi108/109 device setup code + * + * Maintained by Roy Zang < tie-fei.zang@freescale.com > + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License as published by the + * Free Software Foundation; either version 2 of the License, or (at your + * option) any later version. + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include +#include +#include +#include +#include +#include + +#undef DEBUG + +#ifdef DEBUG +#define DBG(fmt...) do { printk(fmt); } while(0) +#else +#define DBG(fmt...) do { } while(0) +#endif + +static phys_addr_t tsi108_csr_base = -1; + +phys_addr_t get_csrbase(void) +{ + struct device_node *tsi; + + if (tsi108_csr_base != -1) + return tsi108_csr_base; + + tsi = of_find_node_by_type(NULL, "tsi-bridge"); + if (tsi) { + unsigned int size; + void *prop = get_property(tsi, "reg", &size); + tsi108_csr_base = of_translate_address(tsi, prop); + of_node_put(tsi); + }; + return tsi108_csr_base; +} + +u32 get_vir_csrbase(void) +{ + return (u32) (ioremap(get_csrbase(), 0x10000)); +} + +EXPORT_SYMBOL(get_csrbase); +EXPORT_SYMBOL(get_vir_csrbase); + +static int __init tsi108_eth_of_init(void) +{ + struct device_node *np; + unsigned int i; + struct platform_device *tsi_eth_dev; + struct resource res; + int ret; + + for (np = NULL, i = 0; + (np = of_find_compatible_node(np, "network", "tsi-ethernet")) != NULL; + i++) { + struct resource r[2]; + struct device_node *phy; + hw_info tsi_eth_data; + unsigned int *id; + unsigned int *phy_id; + void *mac_addr; + phandle *ph; + + memset(r, 0, sizeof(r)); + memset(&tsi_eth_data, 0, sizeof(tsi_eth_data)); + + ret = of_address_to_resource(np, 0, &r[0]); + DBG("%s: name:start->end = %s:0x%lx-> 0x%lx\n", + __FUNCTION__,r[0].name, r[0].start, r[0].end); + if (ret) + goto err; + + r[1].name = "tx"; + r[1].start = np->intrs[0].line; + r[1].end = np->intrs[0].line; + r[1].flags = IORESOURCE_IRQ; + + tsi_eth_dev = + platform_device_register_simple("tsi-ethernet", i, &r[0], + np->n_intrs + 1); + + if (IS_ERR(tsi_eth_dev)) { + ret = PTR_ERR(tsi_eth_dev); + goto err; + } + + mac_addr = get_property(np, "address", NULL); + memcpy(tsi_eth_data.mac_addr, mac_addr, 6); + + ph = (phandle *) get_property(np, "phy-handle", NULL); + phy = of_find_node_by_phandle(*ph); + + if (phy == NULL) { + ret = -ENODEV; + goto unreg; + } + + id = (u32 *) get_property(phy, "reg", NULL); + phy_id = (u32 *) get_property(phy, "phy-id", NULL); + ret = of_address_to_resource(phy, 0, &res); + if (ret) { + of_node_put(phy); + goto unreg; + } + tsi_eth_data.regs = r[0].start; + tsi_eth_data.phyregs = res.start; + tsi_eth_data.phy = *phy_id; + tsi_eth_data.irq_num = np->intrs[0].line; + of_node_put(phy); + ret = + platform_device_add_data(tsi_eth_dev, &tsi_eth_data, + sizeof(hw_info)); + if (ret) + goto unreg; + } + return 0; +unreg: + platform_device_unregister(tsi_eth_dev); +err: + return ret; +} + +arch_initcall(tsi108_eth_of_init); diff --git a/arch/powerpc/sysdev/tsi108_pci.c b/arch/powerpc/sysdev/tsi108_pci.c new file mode 100644 index 00000000000..3265d54c82e --- /dev/null +++ b/arch/powerpc/sysdev/tsi108_pci.c @@ -0,0 +1,412 @@ +/* + * Common routines for Tundra Semiconductor TSI108 host bridge. + * + * 2004-2005 (c) Tundra Semiconductor Corp. + * Author: Alex Bounine (alexandreb@tundra.com) + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License as published by the Free + * Software Foundation; either version 2 of the License, or (at your option) + * any later version. + * + * This program is distributed in the hope that it will be useful, but WITHOUT + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for + * more details. + * + * You should have received a copy of the GNU General Public License along with + * this program; if not, write to the Free Software Foundation, Inc., 59 + * Temple Place - Suite 330, Boston, MA 02111-1307, USA. + */ + +#include +#include +#include +#include +#include +#include + + +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#undef DEBUG +#ifdef DEBUG +#define DBG(x...) printk(x) +#else +#define DBG(x...) +#endif + +#define tsi_mk_config_addr(bus, devfunc, offset) \ + ((((bus)<<16) | ((devfunc)<<8) | (offset & 0xfc)) + tsi108_pci_cfg_base) + +u32 tsi108_pci_cfg_base; +u32 tsi108_csr_vir_base; + +extern u32 get_vir_csrbase(void); +extern u32 tsi108_read_reg(u32 reg_offset); +extern void tsi108_write_reg(u32 reg_offset, u32 val); + +int +tsi108_direct_write_config(struct pci_bus *bus, unsigned int devfunc, + int offset, int len, u32 val) +{ + volatile unsigned char *cfg_addr; + + if (ppc_md.pci_exclude_device) + if (ppc_md.pci_exclude_device(bus->number, devfunc)) + return PCIBIOS_DEVICE_NOT_FOUND; + + cfg_addr = (unsigned char *)(tsi_mk_config_addr(bus->number, + devfunc, offset) | + (offset & 0x03)); + +#ifdef DEBUG + printk("PCI CFG write : "); + printk("%d:0x%x:0x%x ", bus->number, devfunc, offset); + printk("%d ADDR=0x%08x ", len, (uint) cfg_addr); + printk("data = 0x%08x\n", val); +#endif + + switch (len) { + case 1: + out_8((u8 *) cfg_addr, val); + break; + case 2: + out_le16((u16 *) cfg_addr, val); + break; + default: + out_le32((u32 *) cfg_addr, val); + break; + } + + return PCIBIOS_SUCCESSFUL; +} + +void tsi108_clear_pci_error(u32 pci_cfg_base) +{ + u32 err_stat, err_addr, pci_stat; + + /* + * Quietly clear PB and PCI error flags set as result + * of PCI/X configuration read requests. + */ + + /* Read PB Error Log Registers */ + + err_stat = tsi108_read_reg(TSI108_PB_OFFSET + TSI108_PB_ERRCS); + err_addr = tsi108_read_reg(TSI108_PB_OFFSET + TSI108_PB_AERR); + + if (err_stat & TSI108_PB_ERRCS_ES) { + /* Clear error flag */ + tsi108_write_reg(TSI108_PB_OFFSET + TSI108_PB_ERRCS, + TSI108_PB_ERRCS_ES); + + /* Clear read error reported in PB_ISR */ + tsi108_write_reg(TSI108_PB_OFFSET + TSI108_PB_ISR, + TSI108_PB_ISR_PBS_RD_ERR); + + /* Clear PCI/X bus cfg errors if applicable */ + if ((err_addr & 0xFF000000) == pci_cfg_base) { + pci_stat = + tsi108_read_reg(TSI108_PCI_OFFSET + TSI108_PCI_CSR); + tsi108_write_reg(TSI108_PCI_OFFSET + TSI108_PCI_CSR, + pci_stat); + } + } + + return; +} + +#define __tsi108_read_pci_config(x, addr, op) \ + __asm__ __volatile__( \ + " "op" %0,0,%1\n" \ + "1: eieio\n" \ + "2:\n" \ + ".section .fixup,\"ax\"\n" \ + "3: li %0,-1\n" \ + " b 2b\n" \ + ".section __ex_table,\"a\"\n" \ + " .align 2\n" \ + " .long 1b,3b\n" \ + ".text" \ + : "=r"(x) : "r"(addr)) + +int +tsi108_direct_read_config(struct pci_bus *bus, unsigned int devfn, int offset, + int len, u32 * val) +{ + volatile unsigned char *cfg_addr; + u32 temp; + + if (ppc_md.pci_exclude_device) + if (ppc_md.pci_exclude_device(bus->number, devfn)) + return PCIBIOS_DEVICE_NOT_FOUND; + + cfg_addr = (unsigned char *)(tsi_mk_config_addr(bus->number, + devfn, + offset) | (offset & + 0x03)); + + switch (len) { + case 1: + __tsi108_read_pci_config(temp, cfg_addr, "lbzx"); + break; + case 2: + __tsi108_read_pci_config(temp, cfg_addr, "lhbrx"); + break; + default: + __tsi108_read_pci_config(temp, cfg_addr, "lwbrx"); + break; + } + + *val = temp; + +#ifdef DEBUG + if ((0xFFFFFFFF != temp) && (0xFFFF != temp) && (0xFF != temp)) { + printk("PCI CFG read : "); + printk("%d:0x%x:0x%x ", bus->number, devfn, offset); + printk("%d ADDR=0x%08x ", len, (uint) cfg_addr); + printk("data = 0x%x\n", *val); + } +#endif + return PCIBIOS_SUCCESSFUL; +} + +void tsi108_clear_pci_cfg_error(void) +{ + tsi108_clear_pci_error(TSI108_PCI_CFG_BASE_PHYS); +} + +static struct pci_ops tsi108_direct_pci_ops = { + tsi108_direct_read_config, + tsi108_direct_write_config +}; + +int __init tsi108_setup_pci(struct device_node *dev) +{ + int len; + struct pci_controller *hose; + struct resource rsrc; + int *bus_range; + int primary = 0, has_address = 0; + + /* PCI Config mapping */ + tsi108_pci_cfg_base = (u32)ioremap(TSI108_PCI_CFG_BASE_PHYS, + TSI108_PCI_CFG_SIZE); + DBG("TSI_PCI: %s tsi108_pci_cfg_base=0x%x\n", __FUNCTION__, + tsi108_pci_cfg_base); + + /* Fetch host bridge registers address */ + has_address = (of_address_to_resource(dev, 0, &rsrc) == 0); + + /* Get bus range if any */ + bus_range = (int *)get_property(dev, "bus-range", &len); + if (bus_range == NULL || len < 2 * sizeof(int)) { + printk(KERN_WARNING "Can't get bus-range for %s, assume" + " bus 0\n", dev->full_name); + } + + hose = pcibios_alloc_controller(); + + if (!hose) { + printk("PCI Host bridge init failed\n"); + return -ENOMEM; + } + hose->arch_data = dev; + hose->set_cfg_type = 1; + + hose->first_busno = bus_range ? bus_range[0] : 0; + hose->last_busno = bus_range ? bus_range[1] : 0xff; + + (hose)->ops = &tsi108_direct_pci_ops; + + printk(KERN_INFO "Found tsi108 PCI host bridge at 0x%08lx. " + "Firmware bus number: %d->%d\n", + rsrc.start, hose->first_busno, hose->last_busno); + + /* Interpret the "ranges" property */ + /* This also maps the I/O region and sets isa_io/mem_base */ + pci_process_bridge_OF_ranges(hose, dev, primary); + return 0; +} + +/* + * Low level utility functions + */ + +static void tsi108_pci_int_mask(u_int irq) +{ + u_int irp_cfg; + int int_line = (irq - IRQ_PCI_INTAD_BASE); + + irp_cfg = tsi108_read_reg(TSI108_PCI_OFFSET + TSI108_PCI_IRP_CFG_CTL); + mb(); + irp_cfg |= (1 << int_line); /* INTx_DIR = output */ + irp_cfg &= ~(3 << (8 + (int_line * 2))); /* INTx_TYPE = unused */ + tsi108_write_reg(TSI108_PCI_OFFSET + TSI108_PCI_IRP_CFG_CTL, irp_cfg); + mb(); + irp_cfg = tsi108_read_reg(TSI108_PCI_OFFSET + TSI108_PCI_IRP_CFG_CTL); +} + +static void tsi108_pci_int_unmask(u_int irq) +{ + u_int irp_cfg; + int int_line = (irq - IRQ_PCI_INTAD_BASE); + + irp_cfg = tsi108_read_reg(TSI108_PCI_OFFSET + TSI108_PCI_IRP_CFG_CTL); + mb(); + irp_cfg &= ~(1 << int_line); + irp_cfg |= (3 << (8 + (int_line * 2))); + tsi108_write_reg(TSI108_PCI_OFFSET + TSI108_PCI_IRP_CFG_CTL, irp_cfg); + mb(); +} + +static void init_pci_source(void) +{ + tsi108_write_reg(TSI108_PCI_OFFSET + TSI108_PCI_IRP_CFG_CTL, + 0x0000ff00); + tsi108_write_reg(TSI108_PCI_OFFSET + TSI108_PCI_IRP_ENABLE, + TSI108_PCI_IRP_ENABLE_P_INT); + mb(); +} + +static inline int get_pci_source(void) +{ + u_int temp = 0; + int irq = -1; + int i; + u_int pci_irp_stat; + static int mask = 0; + + /* Read PCI/X block interrupt status register */ + pci_irp_stat = tsi108_read_reg(TSI108_PCI_OFFSET + TSI108_PCI_IRP_STAT); + mb(); + + if (pci_irp_stat & TSI108_PCI_IRP_STAT_P_INT) { + /* Process Interrupt from PCI bus INTA# - INTD# lines */ + temp = + tsi108_read_reg(TSI108_PCI_OFFSET + + TSI108_PCI_IRP_INTAD) & 0xf; + mb(); + for (i = 0; i < 4; i++, mask++) { + if (temp & (1 << mask % 4)) { + irq = IRQ_PCI_INTA + mask % 4; + mask++; + break; + } + } + + /* Disable interrupts from PCI block */ + temp = tsi108_read_reg(TSI108_PCI_OFFSET + TSI108_PCI_IRP_ENABLE); + tsi108_write_reg(TSI108_PCI_OFFSET + TSI108_PCI_IRP_ENABLE, + temp & ~TSI108_PCI_IRP_ENABLE_P_INT); + mb(); + (void)tsi108_read_reg(TSI108_PCI_OFFSET + TSI108_PCI_IRP_ENABLE); + mb(); + } +#ifdef DEBUG + else { + printk("TSI108_PIC: error in TSI108_PCI_IRP_STAT\n"); + pci_irp_stat = + tsi108_read_reg(TSI108_PCI_OFFSET + TSI108_PCI_IRP_STAT); + temp = + tsi108_read_reg(TSI108_PCI_OFFSET + TSI108_PCI_IRP_INTAD); + mb(); + printk(">> stat=0x%08x intad=0x%08x ", pci_irp_stat, temp); + temp = + tsi108_read_reg(TSI108_PCI_OFFSET + TSI108_PCI_IRP_CFG_CTL); + mb(); + printk("cfg_ctl=0x%08x ", temp); + temp = + tsi108_read_reg(TSI108_PCI_OFFSET + TSI108_PCI_IRP_ENABLE); + mb(); + printk("irp_enable=0x%08x\n", temp); + } +#endif /* end of DEBUG */ + + return irq; +} + + +/* + * Linux descriptor level callbacks + */ + +static void tsi108_pci_irq_enable(u_int irq) +{ + tsi108_pci_int_unmask(irq); +} + +static void tsi108_pci_irq_disable(u_int irq) +{ + tsi108_pci_int_mask(irq); +} + +static void tsi108_pci_irq_ack(u_int irq) +{ + tsi108_pci_int_mask(irq); +} + +static void tsi108_pci_irq_end(u_int irq) +{ + tsi108_pci_int_unmask(irq); + + /* Enable interrupts from PCI block */ + tsi108_write_reg(TSI108_PCI_OFFSET + TSI108_PCI_IRP_ENABLE, + tsi108_read_reg(TSI108_PCI_OFFSET + + TSI108_PCI_IRP_ENABLE) | + TSI108_PCI_IRP_ENABLE_P_INT); + mb(); +} + +/* + * Interrupt controller descriptor for cascaded PCI interrupt controller. + */ + +struct hw_interrupt_type tsi108_pci_irq = { + .typename = "tsi108_PCI_int", + .enable = tsi108_pci_irq_enable, + .disable = tsi108_pci_irq_disable, + .ack = tsi108_pci_irq_ack, + .end = tsi108_pci_irq_end, +}; + +/* + * Exported functions + */ + +/* + * The Tsi108 PCI interrupts initialization routine. + * + * The INTA# - INTD# interrupts on the PCI bus are reported by the PCI block + * to the MPIC using single interrupt source (IRQ_TSI108_PCI). Therefore the + * PCI block has to be treated as a cascaded interrupt controller connected + * to the MPIC. + */ + +void __init tsi108_pci_int_init(void) +{ + u_int i; + + DBG("Tsi108_pci_int_init: initializing PCI interrupts\n"); + + for (i = 0; i < NUM_PCI_IRQS; i++) { + irq_desc[i + IRQ_PCI_INTAD_BASE].handler = &tsi108_pci_irq; + irq_desc[i + IRQ_PCI_INTAD_BASE].status |= IRQ_LEVEL; + } + + init_pci_source(); +} + +int tsi108_irq_cascade(struct pt_regs *regs, void *unused) +{ + return get_pci_source(); +} diff --git a/include/asm-powerpc/tsi108.h b/include/asm-powerpc/tsi108.h new file mode 100644 index 00000000000..c4c278d72f7 --- /dev/null +++ b/include/asm-powerpc/tsi108.h @@ -0,0 +1,109 @@ +/* + * include/asm-ppc/tsi108.h + * + * common routine and memory layout for Tundra TSI108(Grendel) host bridge + * memory controller. + * + * Author: Jacob Pan (jacob.pan@freescale.com) + * Alex Bounine (alexandreb@tundra.com) + * 2004 (c) Freescale Semiconductor Inc. This file is licensed under + * the terms of the GNU General Public License version 2. This program + * is licensed "as is" without any warranty of any kind, whether express + * or implied. + */ +#ifndef __PPC_KERNEL_TSI108_H +#define __PPC_KERNEL_TSI108_H + +#include + +/* Size of entire register space */ +#define TSI108_REG_SIZE (0x10000) + +/* Sizes of register spaces for individual blocks */ +#define TSI108_HLP_SIZE 0x1000 +#define TSI108_PCI_SIZE 0x1000 +#define TSI108_CLK_SIZE 0x1000 +#define TSI108_PB_SIZE 0x1000 +#define TSI108_SD_SIZE 0x1000 +#define TSI108_DMA_SIZE 0x1000 +#define TSI108_ETH_SIZE 0x1000 +#define TSI108_I2C_SIZE 0x400 +#define TSI108_MPIC_SIZE 0x400 +#define TSI108_UART0_SIZE 0x200 +#define TSI108_GPIO_SIZE 0x200 +#define TSI108_UART1_SIZE 0x200 + +/* Offsets within Tsi108(A) CSR space for individual blocks */ +#define TSI108_HLP_OFFSET 0x0000 +#define TSI108_PCI_OFFSET 0x1000 +#define TSI108_CLK_OFFSET 0x2000 +#define TSI108_PB_OFFSET 0x3000 +#define TSI108_SD_OFFSET 0x4000 +#define TSI108_DMA_OFFSET 0x5000 +#define TSI108_ETH_OFFSET 0x6000 +#define TSI108_I2C_OFFSET 0x7000 +#define TSI108_MPIC_OFFSET 0x7400 +#define TSI108_UART0_OFFSET 0x7800 +#define TSI108_GPIO_OFFSET 0x7A00 +#define TSI108_UART1_OFFSET 0x7C00 + +/* Tsi108 registers used by common code components */ +#define TSI108_PCI_CSR (0x004) +#define TSI108_PCI_IRP_CFG_CTL (0x180) +#define TSI108_PCI_IRP_STAT (0x184) +#define TSI108_PCI_IRP_ENABLE (0x188) +#define TSI108_PCI_IRP_INTAD (0x18C) + +#define TSI108_PCI_IRP_STAT_P_INT (0x00400000) +#define TSI108_PCI_IRP_ENABLE_P_INT (0x00400000) + +#define TSI108_CG_PWRUP_STATUS (0x234) + +#define TSI108_PB_ISR (0x00C) +#define TSI108_PB_ERRCS (0x404) +#define TSI108_PB_AERR (0x408) + +#define TSI108_PB_ERRCS_ES (1 << 1) +#define TSI108_PB_ISR_PBS_RD_ERR (1 << 8) + +#define TSI108_PCI_CFG_BASE_PHYS (0xfb000000) +#define TSI108_PCI_CFG_SIZE (0x01000000) +/* Global variables */ + +extern u32 tsi108_pci_cfg_base; +/* Exported functions */ + +extern int tsi108_bridge_init(struct pci_controller *hose, uint phys_csr_base); +extern unsigned long tsi108_get_mem_size(void); +extern unsigned long tsi108_get_cpu_clk(void); +extern unsigned long tsi108_get_sdc_clk(void); +extern int tsi108_direct_write_config(struct pci_bus *bus, unsigned int devfn, + int offset, int len, u32 val); +extern int tsi108_direct_read_config(struct pci_bus *bus, unsigned int devfn, + int offset, int len, u32 * val); +extern void tsi108_clear_pci_error(u32 pci_cfg_base); + +extern phys_addr_t get_csrbase(void); + +typedef struct { + u32 regs; /* hw registers base address */ + u32 phyregs; /* phy registers base address */ + u16 phy; /* phy address */ + u16 irq_num; /* irq number */ + u8 mac_addr[6]; /* phy mac address */ +} hw_info; + +extern u32 get_vir_csrbase(void); +extern u32 tsi108_csr_vir_base; + +extern inline u32 tsi108_read_reg(u32 reg_offset) +{ + return in_be32((volatile u32 *)(tsi108_csr_vir_base + reg_offset)); +} + +extern inline void tsi108_write_reg(u32 reg_offset, u32 val) +{ + out_be32((volatile u32 *)(tsi108_csr_vir_base + reg_offset), val); +} + +#endif /* __PPC_KERNEL_TSI108_H */ -- cgit v1.2.3-70-g09d2 From c220153654ede57b41900159eb8d1f6029d85642 Mon Sep 17 00:00:00 2001 From: "Mark A. Greer" Date: Wed, 28 Jun 2006 14:13:20 -0700 Subject: [POWERPC] todc: add support for Time-Of-Day-Clock This is a resubmit with a proper subject and with all comments addressed. Applies cleanly to powerpc.git 649e85797259162f7fdc696420e7492f20226f2d Mark -- The todc code from arch/ppc supports many todc/rtc chips and is needed in arch/powerpc. This patch adds the todc code to arch/powerpc. Signed-off-by: Mark A. Greer -- arch/powerpc/Kconfig | 7 arch/powerpc/sysdev/Makefile | 1 arch/powerpc/sysdev/todc.c | 392 ++++++++++++++++++++++++++++++++++ include/asm-powerpc/todc.h | 487 +++++++++++++++++++++++++++++++++++++++++++ 4 files changed, 887 insertions(+) -- Signed-off-by: Paul Mackerras --- arch/powerpc/Kconfig | 7 + arch/powerpc/sysdev/Makefile | 1 + arch/powerpc/sysdev/todc.c | 392 ++++++++++++++++++++++++++++++++++ include/asm-powerpc/todc.h | 487 +++++++++++++++++++++++++++++++++++++++++++ 4 files changed, 887 insertions(+) create mode 100644 arch/powerpc/sysdev/todc.c create mode 100644 include/asm-powerpc/todc.h (limited to 'include') diff --git a/arch/powerpc/Kconfig b/arch/powerpc/Kconfig index 0b5e225c2d5..e898da2381a 100644 --- a/arch/powerpc/Kconfig +++ b/arch/powerpc/Kconfig @@ -563,6 +563,13 @@ config TAU_AVERAGE /proc/cpuinfo. If in doubt, say N here. + +config PPC_TODC + depends on EMBEDDED6xx + bool "Generic Time-of-day Clock (TODC) support" + ---help--- + This adds support for many TODC/RTC chips. + endmenu source arch/powerpc/platforms/embedded6xx/Kconfig diff --git a/arch/powerpc/sysdev/Makefile b/arch/powerpc/sysdev/Makefile index 63e81e74b45..054bd8b41ef 100644 --- a/arch/powerpc/sysdev/Makefile +++ b/arch/powerpc/sysdev/Makefile @@ -12,4 +12,5 @@ obj-$(CONFIG_U3_DART) += dart_iommu.o obj-$(CONFIG_MMIO_NVRAM) += mmio_nvram.o obj-$(CONFIG_PPC_83xx) += ipic.o obj-$(CONFIG_FSL_SOC) += fsl_soc.o +obj-$(CONFIG_PPC_TODC) += todc.o obj-$(CONFIG_TSI108_BRIDGE) += tsi108_pci.o tsi108_dev.o diff --git a/arch/powerpc/sysdev/todc.c b/arch/powerpc/sysdev/todc.c new file mode 100644 index 00000000000..0a65980efb5 --- /dev/null +++ b/arch/powerpc/sysdev/todc.c @@ -0,0 +1,392 @@ +/* + * Time of Day Clock support for the M48T35, M48T37, M48T59, and MC146818 + * Real Time Clocks/Timekeepers. + * + * Author: Mark A. Greer + * + * 2001-2004 (c) MontaVista, Software, Inc. This file is licensed under + * the terms of the GNU General Public License version 2. This program + * is licensed "as is" without any warranty of any kind, whether express + * or implied. + */ +#include +#include +#include +#include +#include +#include +#include + +#include +#include +#include +#include + +/* + * Depending on the hardware on your board and your board design, the + * RTC/NVRAM may be accessed either directly (like normal memory) or via + * address/data registers. If your board uses the direct method, set + * 'nvram_data' to the base address of your nvram and leave 'nvram_as0' and + * 'nvram_as1' NULL. If your board uses address/data regs to access nvram, + * set 'nvram_as0' to the address of the lower byte, set 'nvram_as1' to the + * address of the upper byte (leave NULL if using mc146818), and set + * 'nvram_data' to the address of the 8-bit data register. + * + * Note: Even though the documentation for the various RTC chips say that it + * take up to a second before it starts updating once the 'R' bit is + * cleared, they always seem to update even though we bang on it many + * times a second. This is true, except for the Dallas Semi 1746/1747 + * (possibly others). Those chips seem to have a real problem whenever + * we set the 'R' bit before reading them, they basically stop counting. + * --MAG + */ + +/* + * 'todc_info' should be initialized in your *_setup.c file to + * point to a fully initialized 'todc_info_t' structure. + * This structure holds all the register offsets for your particular + * TODC/RTC chip. + * TODC_ALLOC()/TODC_INIT() will allocate and initialize this table for you. + */ + +#ifdef RTC_FREQ_SELECT +#undef RTC_FREQ_SELECT +#define RTC_FREQ_SELECT control_b /* Register A */ +#endif + +#ifdef RTC_CONTROL +#undef RTC_CONTROL +#define RTC_CONTROL control_a /* Register B */ +#endif + +#ifdef RTC_INTR_FLAGS +#undef RTC_INTR_FLAGS +#define RTC_INTR_FLAGS watchdog /* Register C */ +#endif + +#ifdef RTC_VALID +#undef RTC_VALID +#define RTC_VALID interrupts /* Register D */ +#endif + +/* Access routines when RTC accessed directly (like normal memory) */ +u_char +todc_direct_read_val(int addr) +{ + return readb((void __iomem *)(todc_info->nvram_data + addr)); +} + +void +todc_direct_write_val(int addr, unsigned char val) +{ + writeb(val, (void __iomem *)(todc_info->nvram_data + addr)); + return; +} + +/* Access routines for accessing m48txx type chips via addr/data regs */ +u_char +todc_m48txx_read_val(int addr) +{ + outb(addr, todc_info->nvram_as0); + outb(addr>>todc_info->as0_bits, todc_info->nvram_as1); + return inb(todc_info->nvram_data); +} + +void +todc_m48txx_write_val(int addr, unsigned char val) +{ + outb(addr, todc_info->nvram_as0); + outb(addr>>todc_info->as0_bits, todc_info->nvram_as1); + outb(val, todc_info->nvram_data); + return; +} + +/* Access routines for accessing mc146818 type chips via addr/data regs */ +u_char +todc_mc146818_read_val(int addr) +{ + outb_p(addr, todc_info->nvram_as0); + return inb_p(todc_info->nvram_data); +} + +void +todc_mc146818_write_val(int addr, unsigned char val) +{ + outb_p(addr, todc_info->nvram_as0); + outb_p(val, todc_info->nvram_data); +} + + +/* + * Routines to make RTC chips with NVRAM buried behind an addr/data pair + * have the NVRAM and clock regs appear at the same level. + * The NVRAM will appear to start at addr 0 and the clock regs will appear + * to start immediately after the NVRAM (actually, start at offset + * todc_info->nvram_size). + */ +static inline u_char +todc_read_val(int addr) +{ + u_char val; + + if (todc_info->sw_flags & TODC_FLAG_2_LEVEL_NVRAM) { + if (addr < todc_info->nvram_size) { /* NVRAM */ + ppc_md.rtc_write_val(todc_info->nvram_addr_reg, addr); + val = ppc_md.rtc_read_val(todc_info->nvram_data_reg); + } else { /* Clock Reg */ + addr -= todc_info->nvram_size; + val = ppc_md.rtc_read_val(addr); + } + } else + val = ppc_md.rtc_read_val(addr); + + return val; +} + +static inline void +todc_write_val(int addr, u_char val) +{ + if (todc_info->sw_flags & TODC_FLAG_2_LEVEL_NVRAM) { + if (addr < todc_info->nvram_size) { /* NVRAM */ + ppc_md.rtc_write_val(todc_info->nvram_addr_reg, addr); + ppc_md.rtc_write_val(todc_info->nvram_data_reg, val); + } else { /* Clock Reg */ + addr -= todc_info->nvram_size; + ppc_md.rtc_write_val(addr, val); + } + } else + ppc_md.rtc_write_val(addr, val); +} + +/* + * TODC routines + * + * There is some ugly stuff in that there are assumptions for the mc146818. + * + * Assumptions: + * - todc_info->control_a has the offset as mc146818 Register B reg + * - todc_info->control_b has the offset as mc146818 Register A reg + * - m48txx control reg's write enable or 'W' bit is same as + * mc146818 Register B 'SET' bit (i.e., 0x80) + * + * These assumptions were made to make the code simpler. + */ +long __init +todc_time_init(void) +{ + u_char cntl_b; + + if (!ppc_md.rtc_read_val) + ppc_md.rtc_read_val = ppc_md.nvram_read_val; + if (!ppc_md.rtc_write_val) + ppc_md.rtc_write_val = ppc_md.nvram_write_val; + + cntl_b = todc_read_val(todc_info->control_b); + + if (todc_info->rtc_type == TODC_TYPE_MC146818) { + if ((cntl_b & 0x70) != 0x20) { + printk(KERN_INFO "TODC real-time-clock was stopped." + " Now starting..."); + cntl_b &= ~0x70; + cntl_b |= 0x20; + } + + todc_write_val(todc_info->control_b, cntl_b); + } else if (todc_info->rtc_type == TODC_TYPE_DS17285) { + u_char mode; + + mode = todc_read_val(TODC_TYPE_DS17285_CNTL_A); + /* Make sure countdown clear is not set */ + mode &= ~0x40; + /* Enable oscillator, extended register set */ + mode |= 0x30; + todc_write_val(TODC_TYPE_DS17285_CNTL_A, mode); + + } else if (todc_info->rtc_type == TODC_TYPE_DS1501) { + u_char month; + + todc_info->enable_read = TODC_DS1501_CNTL_B_TE; + todc_info->enable_write = TODC_DS1501_CNTL_B_TE; + + month = todc_read_val(todc_info->month); + + if ((month & 0x80) == 0x80) { + printk(KERN_INFO "TODC %s %s\n", + "real-time-clock was stopped.", + "Now starting..."); + month &= ~0x80; + todc_write_val(todc_info->month, month); + } + + cntl_b &= ~TODC_DS1501_CNTL_B_TE; + todc_write_val(todc_info->control_b, cntl_b); + } else { /* must be a m48txx type */ + u_char cntl_a; + + todc_info->enable_read = TODC_MK48TXX_CNTL_A_R; + todc_info->enable_write = TODC_MK48TXX_CNTL_A_W; + + cntl_a = todc_read_val(todc_info->control_a); + + /* Check & clear STOP bit in control B register */ + if (cntl_b & TODC_MK48TXX_DAY_CB) { + printk(KERN_INFO "TODC %s %s\n", + "real-time-clock was stopped.", + "Now starting..."); + + cntl_a |= todc_info->enable_write; + cntl_b &= ~TODC_MK48TXX_DAY_CB;/* Start Oscil */ + + todc_write_val(todc_info->control_a, cntl_a); + todc_write_val(todc_info->control_b, cntl_b); + } + + /* Make sure READ & WRITE bits are cleared. */ + cntl_a &= ~(todc_info->enable_write | todc_info->enable_read); + todc_write_val(todc_info->control_a, cntl_a); + } + + return 0; +} + +/* + * There is some ugly stuff in that there are assumptions that for a mc146818, + * the todc_info->control_a has the offset of the mc146818 Register B reg and + * that the register'ss 'SET' bit is the same as the m48txx's write enable + * bit in the control register of the m48txx (i.e., 0x80). + * + * It was done to make the code look simpler. + */ +void +todc_get_rtc_time(struct rtc_time *tm) +{ + uint year = 0, mon = 0, mday = 0, hour = 0, min = 0, sec = 0; + uint limit, i; + u_char save_control, uip = 0; + extern void GregorianDay(struct rtc_time *); + + spin_lock(&rtc_lock); + save_control = todc_read_val(todc_info->control_a); + + if (todc_info->rtc_type != TODC_TYPE_MC146818) { + limit = 1; + + switch (todc_info->rtc_type) { + case TODC_TYPE_DS1553: + case TODC_TYPE_DS1557: + case TODC_TYPE_DS1743: + case TODC_TYPE_DS1746: /* XXXX BAD HACK -> FIX */ + case TODC_TYPE_DS1747: + case TODC_TYPE_DS17285: + break; + default: + todc_write_val(todc_info->control_a, + (save_control | todc_info->enable_read)); + } + } else + limit = 100000000; + + for (i=0; irtc_type == TODC_TYPE_MC146818) + uip = todc_read_val(todc_info->RTC_FREQ_SELECT); + + sec = todc_read_val(todc_info->seconds) & 0x7f; + min = todc_read_val(todc_info->minutes) & 0x7f; + hour = todc_read_val(todc_info->hours) & 0x3f; + mday = todc_read_val(todc_info->day_of_month) & 0x3f; + mon = todc_read_val(todc_info->month) & 0x1f; + year = todc_read_val(todc_info->year) & 0xff; + + if (todc_info->rtc_type == TODC_TYPE_MC146818) { + uip |= todc_read_val(todc_info->RTC_FREQ_SELECT); + if ((uip & RTC_UIP) == 0) + break; + } + } + + if (todc_info->rtc_type != TODC_TYPE_MC146818) { + switch (todc_info->rtc_type) { + case TODC_TYPE_DS1553: + case TODC_TYPE_DS1557: + case TODC_TYPE_DS1743: + case TODC_TYPE_DS1746: /* XXXX BAD HACK -> FIX */ + case TODC_TYPE_DS1747: + case TODC_TYPE_DS17285: + break; + default: + save_control &= ~(todc_info->enable_read); + todc_write_val(todc_info->control_a, save_control); + } + } + spin_unlock(&rtc_lock); + + if ((todc_info->rtc_type != TODC_TYPE_MC146818) + || ((save_control & RTC_DM_BINARY) == 0) + || RTC_ALWAYS_BCD) { + BCD_TO_BIN(sec); + BCD_TO_BIN(min); + BCD_TO_BIN(hour); + BCD_TO_BIN(mday); + BCD_TO_BIN(mon); + BCD_TO_BIN(year); + } + + if ((year + 1900) < 1970) { + year += 100; + } + + tm->tm_sec = sec; + tm->tm_min = min; + tm->tm_hour = hour; + tm->tm_mday = mday; + tm->tm_mon = mon; + tm->tm_year = year; + + GregorianDay(tm); +} + +int +todc_set_rtc_time(struct rtc_time *tm) +{ + u_char save_control, save_freq_select = 0; + + spin_lock(&rtc_lock); + save_control = todc_read_val(todc_info->control_a); + + /* Assuming MK48T59_RTC_CA_WRITE & RTC_SET are equal */ + todc_write_val(todc_info->control_a, + (save_control | todc_info->enable_write)); + save_control &= ~(todc_info->enable_write); /* in case it was set */ + + if (todc_info->rtc_type == TODC_TYPE_MC146818) { + save_freq_select = todc_read_val(todc_info->RTC_FREQ_SELECT); + todc_write_val(todc_info->RTC_FREQ_SELECT, + save_freq_select | RTC_DIV_RESET2); + } + + if ((todc_info->rtc_type != TODC_TYPE_MC146818) + || ((save_control & RTC_DM_BINARY) == 0) + || RTC_ALWAYS_BCD) { + BIN_TO_BCD(tm->tm_sec); + BIN_TO_BCD(tm->tm_min); + BIN_TO_BCD(tm->tm_hour); + BIN_TO_BCD(tm->tm_mon); + BIN_TO_BCD(tm->tm_mday); + BIN_TO_BCD(tm->tm_year); + } + + todc_write_val(todc_info->seconds, tm->tm_sec); + todc_write_val(todc_info->minutes, tm->tm_min); + todc_write_val(todc_info->hours, tm->tm_hour); + todc_write_val(todc_info->month, tm->tm_mon); + todc_write_val(todc_info->day_of_month, tm->tm_mday); + todc_write_val(todc_info->year, tm->tm_year); + + todc_write_val(todc_info->control_a, save_control); + + if (todc_info->rtc_type == TODC_TYPE_MC146818) + todc_write_val(todc_info->RTC_FREQ_SELECT, save_freq_select); + + spin_unlock(&rtc_lock); + return 0; +} diff --git a/include/asm-powerpc/todc.h b/include/asm-powerpc/todc.h new file mode 100644 index 00000000000..60a8c39b8c1 --- /dev/null +++ b/include/asm-powerpc/todc.h @@ -0,0 +1,487 @@ +/* + * Definitions for the M48Txx and mc146818 series of Time of day/Real Time + * Clock chips. + * + * Author: Mark A. Greer + * + * 2001 (c) MontaVista, Software, Inc. This file is licensed under + * the terms of the GNU General Public License version 2. This program + * is licensed "as is" without any warranty of any kind, whether express + * or implied. + */ + +/* + * Support for the M48T37/M48T59/.../mc146818 Real Time Clock chips. + * Purpose is to make one generic file that handles all of these chips instead + * of every platform implementing the same code over & over again. + */ + +#ifndef __PPC_KERNEL_TODC_H +#define __PPC_KERNEL_TODC_H + +typedef struct { + uint rtc_type; /* your particular chip */ + + /* + * Following are the addresses of the AS0, AS1, and DATA registers + * of these chips. Note that these are board-specific. + */ + unsigned int nvram_as0; + unsigned int nvram_as1; + unsigned int nvram_data; + + /* + * Define bits to stop external set of regs from changing so + * the chip can be read/written reliably. + */ + unsigned char enable_read; + unsigned char enable_write; + + /* + * Following is the number of AS0 address bits. This is normally + * 8 but some bad hardware routes address lines incorrectly. + */ + int as0_bits; + + int nvram_size; /* Size of NVRAM on chip */ + int sw_flags; /* Software control flags */ + + /* Following are the register offsets for the particular chip */ + int year; + int month; + int day_of_month; + int day_of_week; + int hours; + int minutes; + int seconds; + int control_b; + int control_a; + int watchdog; + int interrupts; + int alarm_date; + int alarm_hour; + int alarm_minutes; + int alarm_seconds; + int century; + int flags; + + /* + * Some RTC chips have their NVRAM buried behind a addr/data pair of + * regs on the first level/clock registers. The following fields + * are the addresses for those addr/data regs. + */ + int nvram_addr_reg; + int nvram_data_reg; +} todc_info_t; + +/* + * Define the types of TODC/RTC variants that are supported in + * arch/ppc/kernel/todc_time.c + * Make a new one of these for any chip somehow differs from what's already + * defined. That way, if you ever need to put in code to touch those + * bits/registers in todc_time.c, you can put it inside an + * 'if (todc_info->rtc_type == TODC_TYPE_XXX)' so you won't break + * anyone else. + */ +#define TODC_TYPE_MK48T35 1 +#define TODC_TYPE_MK48T37 2 +#define TODC_TYPE_MK48T59 3 +#define TODC_TYPE_DS1693 4 /* Dallas DS1693 RTC */ +#define TODC_TYPE_DS1743 5 /* Dallas DS1743 RTC */ +#define TODC_TYPE_DS1746 6 /* Dallas DS1746 RTC */ +#define TODC_TYPE_DS1747 7 /* Dallas DS1747 RTC */ +#define TODC_TYPE_DS1501 8 /* Dallas DS1501 RTC */ +#define TODC_TYPE_DS1643 9 /* Dallas DS1643 RTC */ +#define TODC_TYPE_PC97307 10 /* PC97307 internal RTC */ +#define TODC_TYPE_DS1557 11 /* Dallas DS1557 RTC */ +#define TODC_TYPE_DS17285 12 /* Dallas DS17285 RTC */ +#define TODC_TYPE_DS1553 13 /* Dallas DS1553 RTC */ +#define TODC_TYPE_MC146818 100 /* Leave room for m48txx's */ + +/* + * Bit to clear/set to enable reads/writes to the chip + */ +#define TODC_MK48TXX_CNTL_A_R 0x40 +#define TODC_MK48TXX_CNTL_A_W 0x80 +#define TODC_MK48TXX_DAY_CB 0x80 + +#define TODC_DS1501_CNTL_B_TE 0x80 + +/* + * Define flag bits used by todc routines. + */ +#define TODC_FLAG_2_LEVEL_NVRAM 0x00000001 + +/* + * Define the values for the various RTC's that should to into the todc_info + * table. + * Note: The XXX_NVRAM_SIZE, XXX_NVRAM_ADDR_REG, and XXX_NVRAM_DATA_REG only + * matter if XXX_SW_FLAGS has TODC_FLAG_2_LEVEL_NVRAM set. + */ +#define TODC_TYPE_MK48T35_NVRAM_SIZE 0x7ff8 +#define TODC_TYPE_MK48T35_SW_FLAGS 0 +#define TODC_TYPE_MK48T35_YEAR 0x7fff +#define TODC_TYPE_MK48T35_MONTH 0x7ffe +#define TODC_TYPE_MK48T35_DOM 0x7ffd /* Day of Month */ +#define TODC_TYPE_MK48T35_DOW 0x7ffc /* Day of Week */ +#define TODC_TYPE_MK48T35_HOURS 0x7ffb +#define TODC_TYPE_MK48T35_MINUTES 0x7ffa +#define TODC_TYPE_MK48T35_SECONDS 0x7ff9 +#define TODC_TYPE_MK48T35_CNTL_B 0x7ff9 +#define TODC_TYPE_MK48T35_CNTL_A 0x7ff8 +#define TODC_TYPE_MK48T35_WATCHDOG 0x0000 +#define TODC_TYPE_MK48T35_INTERRUPTS 0x0000 +#define TODC_TYPE_MK48T35_ALARM_DATE 0x0000 +#define TODC_TYPE_MK48T35_ALARM_HOUR 0x0000 +#define TODC_TYPE_MK48T35_ALARM_MINUTES 0x0000 +#define TODC_TYPE_MK48T35_ALARM_SECONDS 0x0000 +#define TODC_TYPE_MK48T35_CENTURY 0x0000 +#define TODC_TYPE_MK48T35_FLAGS 0x0000 +#define TODC_TYPE_MK48T35_NVRAM_ADDR_REG 0 +#define TODC_TYPE_MK48T35_NVRAM_DATA_REG 0 + +#define TODC_TYPE_MK48T37_NVRAM_SIZE 0x7ff0 +#define TODC_TYPE_MK48T37_SW_FLAGS 0 +#define TODC_TYPE_MK48T37_YEAR 0x7fff +#define TODC_TYPE_MK48T37_MONTH 0x7ffe +#define TODC_TYPE_MK48T37_DOM 0x7ffd /* Day of Month */ +#define TODC_TYPE_MK48T37_DOW 0x7ffc /* Day of Week */ +#define TODC_TYPE_MK48T37_HOURS 0x7ffb +#define TODC_TYPE_MK48T37_MINUTES 0x7ffa +#define TODC_TYPE_MK48T37_SECONDS 0x7ff9 +#define TODC_TYPE_MK48T37_CNTL_B 0x7ff9 +#define TODC_TYPE_MK48T37_CNTL_A 0x7ff8 +#define TODC_TYPE_MK48T37_WATCHDOG 0x7ff7 +#define TODC_TYPE_MK48T37_INTERRUPTS 0x7ff6 +#define TODC_TYPE_MK48T37_ALARM_DATE 0x7ff5 +#define TODC_TYPE_MK48T37_ALARM_HOUR 0x7ff4 +#define TODC_TYPE_MK48T37_ALARM_MINUTES 0x7ff3 +#define TODC_TYPE_MK48T37_ALARM_SECONDS 0x7ff2 +#define TODC_TYPE_MK48T37_CENTURY 0x7ff1 +#define TODC_TYPE_MK48T37_FLAGS 0x7ff0 +#define TODC_TYPE_MK48T37_NVRAM_ADDR_REG 0 +#define TODC_TYPE_MK48T37_NVRAM_DATA_REG 0 + +#define TODC_TYPE_MK48T59_NVRAM_SIZE 0x1ff0 +#define TODC_TYPE_MK48T59_SW_FLAGS 0 +#define TODC_TYPE_MK48T59_YEAR 0x1fff +#define TODC_TYPE_MK48T59_MONTH 0x1ffe +#define TODC_TYPE_MK48T59_DOM 0x1ffd /* Day of Month */ +#define TODC_TYPE_MK48T59_DOW 0x1ffc /* Day of Week */ +#define TODC_TYPE_MK48T59_HOURS 0x1ffb +#define TODC_TYPE_MK48T59_MINUTES 0x1ffa +#define TODC_TYPE_MK48T59_SECONDS 0x1ff9 +#define TODC_TYPE_MK48T59_CNTL_B 0x1ff9 +#define TODC_TYPE_MK48T59_CNTL_A 0x1ff8 +#define TODC_TYPE_MK48T59_WATCHDOG 0x1fff +#define TODC_TYPE_MK48T59_INTERRUPTS 0x1fff +#define TODC_TYPE_MK48T59_ALARM_DATE 0x1fff +#define TODC_TYPE_MK48T59_ALARM_HOUR 0x1fff +#define TODC_TYPE_MK48T59_ALARM_MINUTES 0x1fff +#define TODC_TYPE_MK48T59_ALARM_SECONDS 0x1fff +#define TODC_TYPE_MK48T59_CENTURY 0x1fff +#define TODC_TYPE_MK48T59_FLAGS 0x1fff +#define TODC_TYPE_MK48T59_NVRAM_ADDR_REG 0 +#define TODC_TYPE_MK48T59_NVRAM_DATA_REG 0 + +#define TODC_TYPE_DS1501_NVRAM_SIZE 0x100 +#define TODC_TYPE_DS1501_SW_FLAGS TODC_FLAG_2_LEVEL_NVRAM +#define TODC_TYPE_DS1501_YEAR (TODC_TYPE_DS1501_NVRAM_SIZE + 0x06) +#define TODC_TYPE_DS1501_MONTH (TODC_TYPE_DS1501_NVRAM_SIZE + 0x05) +#define TODC_TYPE_DS1501_DOM (TODC_TYPE_DS1501_NVRAM_SIZE + 0x04) +#define TODC_TYPE_DS1501_DOW (TODC_TYPE_DS1501_NVRAM_SIZE + 0x03) +#define TODC_TYPE_DS1501_HOURS (TODC_TYPE_DS1501_NVRAM_SIZE + 0x02) +#define TODC_TYPE_DS1501_MINUTES (TODC_TYPE_DS1501_NVRAM_SIZE + 0x01) +#define TODC_TYPE_DS1501_SECONDS (TODC_TYPE_DS1501_NVRAM_SIZE + 0x00) +#define TODC_TYPE_DS1501_CNTL_B (TODC_TYPE_DS1501_NVRAM_SIZE + 0x0f) +#define TODC_TYPE_DS1501_CNTL_A (TODC_TYPE_DS1501_NVRAM_SIZE + 0x0f) +#define TODC_TYPE_DS1501_WATCHDOG (TODC_TYPE_DS1501_NVRAM_SIZE + 0xff) +#define TODC_TYPE_DS1501_INTERRUPTS (TODC_TYPE_DS1501_NVRAM_SIZE + 0xff) +#define TODC_TYPE_DS1501_ALARM_DATE (TODC_TYPE_DS1501_NVRAM_SIZE + 0x0b) +#define TODC_TYPE_DS1501_ALARM_HOUR (TODC_TYPE_DS1501_NVRAM_SIZE + 0x0a) +#define TODC_TYPE_DS1501_ALARM_MINUTES (TODC_TYPE_DS1501_NVRAM_SIZE + 0x09) +#define TODC_TYPE_DS1501_ALARM_SECONDS (TODC_TYPE_DS1501_NVRAM_SIZE + 0x08) +#define TODC_TYPE_DS1501_CENTURY (TODC_TYPE_DS1501_NVRAM_SIZE + 0x07) +#define TODC_TYPE_DS1501_FLAGS (TODC_TYPE_DS1501_NVRAM_SIZE + 0xff) +#define TODC_TYPE_DS1501_NVRAM_ADDR_REG 0x10 +#define TODC_TYPE_DS1501_NVRAM_DATA_REG 0x13 + +#define TODC_TYPE_DS1553_NVRAM_SIZE 0x1ff0 +#define TODC_TYPE_DS1553_SW_FLAGS 0 +#define TODC_TYPE_DS1553_YEAR 0x1fff +#define TODC_TYPE_DS1553_MONTH 0x1ffe +#define TODC_TYPE_DS1553_DOM 0x1ffd /* Day of Month */ +#define TODC_TYPE_DS1553_DOW 0x1ffc /* Day of Week */ +#define TODC_TYPE_DS1553_HOURS 0x1ffb +#define TODC_TYPE_DS1553_MINUTES 0x1ffa +#define TODC_TYPE_DS1553_SECONDS 0x1ff9 +#define TODC_TYPE_DS1553_CNTL_B 0x1ff9 +#define TODC_TYPE_DS1553_CNTL_A 0x1ff8 /* control_a R/W regs */ +#define TODC_TYPE_DS1553_WATCHDOG 0x1ff7 +#define TODC_TYPE_DS1553_INTERRUPTS 0x1ff6 +#define TODC_TYPE_DS1553_ALARM_DATE 0x1ff5 +#define TODC_TYPE_DS1553_ALARM_HOUR 0x1ff4 +#define TODC_TYPE_DS1553_ALARM_MINUTES 0x1ff3 +#define TODC_TYPE_DS1553_ALARM_SECONDS 0x1ff2 +#define TODC_TYPE_DS1553_CENTURY 0x1ff8 +#define TODC_TYPE_DS1553_FLAGS 0x1ff0 +#define TODC_TYPE_DS1553_NVRAM_ADDR_REG 0 +#define TODC_TYPE_DS1553_NVRAM_DATA_REG 0 + +#define TODC_TYPE_DS1557_NVRAM_SIZE 0x7fff0 +#define TODC_TYPE_DS1557_SW_FLAGS 0 +#define TODC_TYPE_DS1557_YEAR 0x7ffff +#define TODC_TYPE_DS1557_MONTH 0x7fffe +#define TODC_TYPE_DS1557_DOM 0x7fffd /* Day of Month */ +#define TODC_TYPE_DS1557_DOW 0x7fffc /* Day of Week */ +#define TODC_TYPE_DS1557_HOURS 0x7fffb +#define TODC_TYPE_DS1557_MINUTES 0x7fffa +#define TODC_TYPE_DS1557_SECONDS 0x7fff9 +#define TODC_TYPE_DS1557_CNTL_B 0x7fff9 +#define TODC_TYPE_DS1557_CNTL_A 0x7fff8 /* control_a R/W regs */ +#define TODC_TYPE_DS1557_WATCHDOG 0x7fff7 +#define TODC_TYPE_DS1557_INTERRUPTS 0x7fff6 +#define TODC_TYPE_DS1557_ALARM_DATE 0x7fff5 +#define TODC_TYPE_DS1557_ALARM_HOUR 0x7fff4 +#define TODC_TYPE_DS1557_ALARM_MINUTES 0x7fff3 +#define TODC_TYPE_DS1557_ALARM_SECONDS 0x7fff2 +#define TODC_TYPE_DS1557_CENTURY 0x7fff8 +#define TODC_TYPE_DS1557_FLAGS 0x7fff0 +#define TODC_TYPE_DS1557_NVRAM_ADDR_REG 0 +#define TODC_TYPE_DS1557_NVRAM_DATA_REG 0 + +#define TODC_TYPE_DS1643_NVRAM_SIZE 0x1ff8 +#define TODC_TYPE_DS1643_SW_FLAGS 0 +#define TODC_TYPE_DS1643_YEAR 0x1fff +#define TODC_TYPE_DS1643_MONTH 0x1ffe +#define TODC_TYPE_DS1643_DOM 0x1ffd /* Day of Month */ +#define TODC_TYPE_DS1643_DOW 0x1ffc /* Day of Week */ +#define TODC_TYPE_DS1643_HOURS 0x1ffb +#define TODC_TYPE_DS1643_MINUTES 0x1ffa +#define TODC_TYPE_DS1643_SECONDS 0x1ff9 +#define TODC_TYPE_DS1643_CNTL_B 0x1ff9 +#define TODC_TYPE_DS1643_CNTL_A 0x1ff8 /* control_a R/W regs */ +#define TODC_TYPE_DS1643_WATCHDOG 0x1fff +#define TODC_TYPE_DS1643_INTERRUPTS 0x1fff +#define TODC_TYPE_DS1643_ALARM_DATE 0x1fff +#define TODC_TYPE_DS1643_ALARM_HOUR 0x1fff +#define TODC_TYPE_DS1643_ALARM_MINUTES 0x1fff +#define TODC_TYPE_DS1643_ALARM_SECONDS 0x1fff +#define TODC_TYPE_DS1643_CENTURY 0x1ff8 +#define TODC_TYPE_DS1643_FLAGS 0x1fff +#define TODC_TYPE_DS1643_NVRAM_ADDR_REG 0 +#define TODC_TYPE_DS1643_NVRAM_DATA_REG 0 + +#define TODC_TYPE_DS1693_NVRAM_SIZE 0 /* Not handled yet */ +#define TODC_TYPE_DS1693_SW_FLAGS 0 +#define TODC_TYPE_DS1693_YEAR 0x09 +#define TODC_TYPE_DS1693_MONTH 0x08 +#define TODC_TYPE_DS1693_DOM 0x07 /* Day of Month */ +#define TODC_TYPE_DS1693_DOW 0x06 /* Day of Week */ +#define TODC_TYPE_DS1693_HOURS 0x04 +#define TODC_TYPE_DS1693_MINUTES 0x02 +#define TODC_TYPE_DS1693_SECONDS 0x00 +#define TODC_TYPE_DS1693_CNTL_B 0x0b +#define TODC_TYPE_DS1693_CNTL_A 0x0a +#define TODC_TYPE_DS1693_WATCHDOG 0xff +#define TODC_TYPE_DS1693_INTERRUPTS 0xff +#define TODC_TYPE_DS1693_ALARM_DATE 0x49 +#define TODC_TYPE_DS1693_ALARM_HOUR 0x05 +#define TODC_TYPE_DS1693_ALARM_MINUTES 0x03 +#define TODC_TYPE_DS1693_ALARM_SECONDS 0x01 +#define TODC_TYPE_DS1693_CENTURY 0x48 +#define TODC_TYPE_DS1693_FLAGS 0xff +#define TODC_TYPE_DS1693_NVRAM_ADDR_REG 0 +#define TODC_TYPE_DS1693_NVRAM_DATA_REG 0 + +#define TODC_TYPE_DS1743_NVRAM_SIZE 0x1ff8 +#define TODC_TYPE_DS1743_SW_FLAGS 0 +#define TODC_TYPE_DS1743_YEAR 0x1fff +#define TODC_TYPE_DS1743_MONTH 0x1ffe +#define TODC_TYPE_DS1743_DOM 0x1ffd /* Day of Month */ +#define TODC_TYPE_DS1743_DOW 0x1ffc /* Day of Week */ +#define TODC_TYPE_DS1743_HOURS 0x1ffb +#define TODC_TYPE_DS1743_MINUTES 0x1ffa +#define TODC_TYPE_DS1743_SECONDS 0x1ff9 +#define TODC_TYPE_DS1743_CNTL_B 0x1ff9 +#define TODC_TYPE_DS1743_CNTL_A 0x1ff8 /* control_a R/W regs */ +#define TODC_TYPE_DS1743_WATCHDOG 0x1fff +#define TODC_TYPE_DS1743_INTERRUPTS 0x1fff +#define TODC_TYPE_DS1743_ALARM_DATE 0x1fff +#define TODC_TYPE_DS1743_ALARM_HOUR 0x1fff +#define TODC_TYPE_DS1743_ALARM_MINUTES 0x1fff +#define TODC_TYPE_DS1743_ALARM_SECONDS 0x1fff +#define TODC_TYPE_DS1743_CENTURY 0x1ff8 +#define TODC_TYPE_DS1743_FLAGS 0x1fff +#define TODC_TYPE_DS1743_NVRAM_ADDR_REG 0 +#define TODC_TYPE_DS1743_NVRAM_DATA_REG 0 + +#define TODC_TYPE_DS1746_NVRAM_SIZE 0x1fff8 +#define TODC_TYPE_DS1746_SW_FLAGS 0 +#define TODC_TYPE_DS1746_YEAR 0x1ffff +#define TODC_TYPE_DS1746_MONTH 0x1fffe +#define TODC_TYPE_DS1746_DOM 0x1fffd /* Day of Month */ +#define TODC_TYPE_DS1746_DOW 0x1fffc /* Day of Week */ +#define TODC_TYPE_DS1746_HOURS 0x1fffb +#define TODC_TYPE_DS1746_MINUTES 0x1fffa +#define TODC_TYPE_DS1746_SECONDS 0x1fff9 +#define TODC_TYPE_DS1746_CNTL_B 0x1fff9 +#define TODC_TYPE_DS1746_CNTL_A 0x1fff8 /* control_a R/W regs */ +#define TODC_TYPE_DS1746_WATCHDOG 0x00000 +#define TODC_TYPE_DS1746_INTERRUPTS 0x00000 +#define TODC_TYPE_DS1746_ALARM_DATE 0x00000 +#define TODC_TYPE_DS1746_ALARM_HOUR 0x00000 +#define TODC_TYPE_DS1746_ALARM_MINUTES 0x00000 +#define TODC_TYPE_DS1746_ALARM_SECONDS 0x00000 +#define TODC_TYPE_DS1746_CENTURY 0x00000 +#define TODC_TYPE_DS1746_FLAGS 0x00000 +#define TODC_TYPE_DS1746_NVRAM_ADDR_REG 0 +#define TODC_TYPE_DS1746_NVRAM_DATA_REG 0 + +#define TODC_TYPE_DS1747_NVRAM_SIZE 0x7fff8 +#define TODC_TYPE_DS1747_SW_FLAGS 0 +#define TODC_TYPE_DS1747_YEAR 0x7ffff +#define TODC_TYPE_DS1747_MONTH 0x7fffe +#define TODC_TYPE_DS1747_DOM 0x7fffd /* Day of Month */ +#define TODC_TYPE_DS1747_DOW 0x7fffc /* Day of Week */ +#define TODC_TYPE_DS1747_HOURS 0x7fffb +#define TODC_TYPE_DS1747_MINUTES 0x7fffa +#define TODC_TYPE_DS1747_SECONDS 0x7fff9 +#define TODC_TYPE_DS1747_CNTL_B 0x7fff9 +#define TODC_TYPE_DS1747_CNTL_A 0x7fff8 /* control_a R/W regs */ +#define TODC_TYPE_DS1747_WATCHDOG 0x00000 +#define TODC_TYPE_DS1747_INTERRUPTS 0x00000 +#define TODC_TYPE_DS1747_ALARM_DATE 0x00000 +#define TODC_TYPE_DS1747_ALARM_HOUR 0x00000 +#define TODC_TYPE_DS1747_ALARM_MINUTES 0x00000 +#define TODC_TYPE_DS1747_ALARM_SECONDS 0x00000 +#define TODC_TYPE_DS1747_CENTURY 0x00000 +#define TODC_TYPE_DS1747_FLAGS 0x00000 +#define TODC_TYPE_DS1747_NVRAM_ADDR_REG 0 +#define TODC_TYPE_DS1747_NVRAM_DATA_REG 0 + +#define TODC_TYPE_DS17285_NVRAM_SIZE (0x1000-0x80) /* 4Kx8 NVRAM (minus RTC regs) */ +#define TODC_TYPE_DS17285_SW_FLAGS TODC_FLAG_2_LEVEL_NVRAM +#define TODC_TYPE_DS17285_SECONDS (TODC_TYPE_DS17285_NVRAM_SIZE + 0x00) +#define TODC_TYPE_DS17285_ALARM_SECONDS (TODC_TYPE_DS17285_NVRAM_SIZE + 0x01) +#define TODC_TYPE_DS17285_MINUTES (TODC_TYPE_DS17285_NVRAM_SIZE + 0x02) +#define TODC_TYPE_DS17285_ALARM_MINUTES (TODC_TYPE_DS17285_NVRAM_SIZE + 0x03) +#define TODC_TYPE_DS17285_HOURS (TODC_TYPE_DS17285_NVRAM_SIZE + 0x04) +#define TODC_TYPE_DS17285_ALARM_HOUR (TODC_TYPE_DS17285_NVRAM_SIZE + 0x05) +#define TODC_TYPE_DS17285_DOW (TODC_TYPE_DS17285_NVRAM_SIZE + 0x06) +#define TODC_TYPE_DS17285_DOM (TODC_TYPE_DS17285_NVRAM_SIZE + 0x07) +#define TODC_TYPE_DS17285_MONTH (TODC_TYPE_DS17285_NVRAM_SIZE + 0x08) +#define TODC_TYPE_DS17285_YEAR (TODC_TYPE_DS17285_NVRAM_SIZE + 0x09) +#define TODC_TYPE_DS17285_CNTL_A (TODC_TYPE_DS17285_NVRAM_SIZE + 0x0A) +#define TODC_TYPE_DS17285_CNTL_B (TODC_TYPE_DS17285_NVRAM_SIZE + 0x0B) +#define TODC_TYPE_DS17285_CNTL_C (TODC_TYPE_DS17285_NVRAM_SIZE + 0x0C) +#define TODC_TYPE_DS17285_CNTL_D (TODC_TYPE_DS17285_NVRAM_SIZE + 0x0D) +#define TODC_TYPE_DS17285_WATCHDOG 0 +#define TODC_TYPE_DS17285_INTERRUPTS 0 +#define TODC_TYPE_DS17285_ALARM_DATE 0 +#define TODC_TYPE_DS17285_CENTURY 0 +#define TODC_TYPE_DS17285_FLAGS 0 +#define TODC_TYPE_DS17285_NVRAM_ADDR_REG 0x50 +#define TODC_TYPE_DS17285_NVRAM_DATA_REG 0x53 + +#define TODC_TYPE_MC146818_NVRAM_SIZE 0 /* XXXX */ +#define TODC_TYPE_MC146818_SW_FLAGS 0 +#define TODC_TYPE_MC146818_YEAR 0x09 +#define TODC_TYPE_MC146818_MONTH 0x08 +#define TODC_TYPE_MC146818_DOM 0x07 /* Day of Month */ +#define TODC_TYPE_MC146818_DOW 0x06 /* Day of Week */ +#define TODC_TYPE_MC146818_HOURS 0x04 +#define TODC_TYPE_MC146818_MINUTES 0x02 +#define TODC_TYPE_MC146818_SECONDS 0x00 +#define TODC_TYPE_MC146818_CNTL_B 0x0a +#define TODC_TYPE_MC146818_CNTL_A 0x0b /* control_a R/W regs */ +#define TODC_TYPE_MC146818_WATCHDOG 0 +#define TODC_TYPE_MC146818_INTERRUPTS 0x0c +#define TODC_TYPE_MC146818_ALARM_DATE 0xff +#define TODC_TYPE_MC146818_ALARM_HOUR 0x05 +#define TODC_TYPE_MC146818_ALARM_MINUTES 0x03 +#define TODC_TYPE_MC146818_ALARM_SECONDS 0x01 +#define TODC_TYPE_MC146818_CENTURY 0xff +#define TODC_TYPE_MC146818_FLAGS 0xff +#define TODC_TYPE_MC146818_NVRAM_ADDR_REG 0 +#define TODC_TYPE_MC146818_NVRAM_DATA_REG 0 + +#define TODC_TYPE_PC97307_NVRAM_SIZE 0 /* No NVRAM? */ +#define TODC_TYPE_PC97307_SW_FLAGS 0 +#define TODC_TYPE_PC97307_YEAR 0x09 +#define TODC_TYPE_PC97307_MONTH 0x08 +#define TODC_TYPE_PC97307_DOM 0x07 /* Day of Month */ +#define TODC_TYPE_PC97307_DOW 0x06 /* Day of Week */ +#define TODC_TYPE_PC97307_HOURS 0x04 +#define TODC_TYPE_PC97307_MINUTES 0x02 +#define TODC_TYPE_PC97307_SECONDS 0x00 +#define TODC_TYPE_PC97307_CNTL_B 0x0a +#define TODC_TYPE_PC97307_CNTL_A 0x0b /* control_a R/W regs */ +#define TODC_TYPE_PC97307_WATCHDOG 0x0c +#define TODC_TYPE_PC97307_INTERRUPTS 0x0d +#define TODC_TYPE_PC97307_ALARM_DATE 0xff +#define TODC_TYPE_PC97307_ALARM_HOUR 0x05 +#define TODC_TYPE_PC97307_ALARM_MINUTES 0x03 +#define TODC_TYPE_PC97307_ALARM_SECONDS 0x01 +#define TODC_TYPE_PC97307_CENTURY 0xff +#define TODC_TYPE_PC97307_FLAGS 0xff +#define TODC_TYPE_PC97307_NVRAM_ADDR_REG 0 +#define TODC_TYPE_PC97307_NVRAM_DATA_REG 0 + +/* + * Define macros to allocate and init the todc_info_t table that will + * be used by the todc_time.c routines. + */ +#define TODC_ALLOC() \ + static todc_info_t todc_info_alloc; \ + todc_info_t *todc_info = &todc_info_alloc; + +#define TODC_INIT(clock_type, as0, as1, data, bits) { \ + todc_info->rtc_type = clock_type; \ + \ + todc_info->nvram_as0 = (unsigned int)(as0); \ + todc_info->nvram_as1 = (unsigned int)(as1); \ + todc_info->nvram_data = (unsigned int)(data); \ + \ + todc_info->as0_bits = (bits); \ + \ + todc_info->nvram_size = clock_type ##_NVRAM_SIZE; \ + todc_info->sw_flags = clock_type ##_SW_FLAGS; \ + \ + todc_info->year = clock_type ##_YEAR; \ + todc_info->month = clock_type ##_MONTH; \ + todc_info->day_of_month = clock_type ##_DOM; \ + todc_info->day_of_week = clock_type ##_DOW; \ + todc_info->hours = clock_type ##_HOURS; \ + todc_info->minutes = clock_type ##_MINUTES; \ + todc_info->seconds = clock_type ##_SECONDS; \ + todc_info->control_b = clock_type ##_CNTL_B; \ + todc_info->control_a = clock_type ##_CNTL_A; \ + todc_info->watchdog = clock_type ##_WATCHDOG; \ + todc_info->interrupts = clock_type ##_INTERRUPTS; \ + todc_info->alarm_date = clock_type ##_ALARM_DATE; \ + todc_info->alarm_hour = clock_type ##_ALARM_HOUR; \ + todc_info->alarm_minutes = clock_type ##_ALARM_MINUTES; \ + todc_info->alarm_seconds = clock_type ##_ALARM_SECONDS; \ + todc_info->century = clock_type ##_CENTURY; \ + todc_info->flags = clock_type ##_FLAGS; \ + \ + todc_info->nvram_addr_reg = clock_type ##_NVRAM_ADDR_REG; \ + todc_info->nvram_data_reg = clock_type ##_NVRAM_DATA_REG; \ +} + +extern todc_info_t *todc_info; + +unsigned char todc_direct_read_val(int addr); +void todc_direct_write_val(int addr, unsigned char val); +unsigned char todc_m48txx_read_val(int addr); +void todc_m48txx_write_val(int addr, unsigned char val); +unsigned char todc_mc146818_read_val(int addr); +void todc_mc146818_write_val(int addr, unsigned char val); + +long todc_time_init(void); +void todc_get_rtc_time(struct rtc_time *); +int todc_set_rtc_time(struct rtc_time *); +void todc_calibrate_decr(void); + +#endif /* __PPC_KERNEL_TODC_H */ -- cgit v1.2.3-70-g09d2 From c406abd3a6d0cf5ce8db4db155a729a28fb98c4f Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Thu, 29 Jun 2006 14:56:13 +0200 Subject: [S390] cleanup bitops.h. Encapsulate complete bitops.h with #ifdef __KERNEL__ and remove the now superfluous ALIGN_CS define and its users. This patch is needed for compiling klibc. Signed-off-by: Heiko Carstens Signed-off-by: Martin Schwidefsky --- include/asm-s390/bitops.h | 42 +++--------------------------------------- 1 file changed, 3 insertions(+), 39 deletions(-) (limited to 'include') diff --git a/include/asm-s390/bitops.h b/include/asm-s390/bitops.h index 4d2b126ba15..0ddcdba79e4 100644 --- a/include/asm-s390/bitops.h +++ b/include/asm-s390/bitops.h @@ -12,6 +12,9 @@ * Copyright (C) 1992, Linus Torvalds * */ + +#ifdef __KERNEL__ + #include /* @@ -50,19 +53,6 @@ * with operation of the form "set_bit(bitnr, flags)". */ -/* set ALIGN_CS to 1 if the SMP safe bit operations should - * align the address to 4 byte boundary. It seems to work - * without the alignment. - */ -#ifdef __KERNEL__ -#define ALIGN_CS 0 -#else -#define ALIGN_CS 1 -#ifndef CONFIG_SMP -#error "bitops won't work without CONFIG_SMP" -#endif -#endif - /* bitmap tables from arch/S390/kernel/bitmap.S */ extern const char _oi_bitmap[]; extern const char _ni_bitmap[]; @@ -121,10 +111,6 @@ static inline void set_bit_cs(unsigned long nr, volatile unsigned long *ptr) unsigned long addr, old, new, mask; addr = (unsigned long) ptr; -#if ALIGN_CS == 1 - nr += (addr & __BITOPS_ALIGN) << 3; /* add alignment to bit number */ - addr ^= addr & __BITOPS_ALIGN; /* align address to 8 */ -#endif /* calculate address for CS */ addr += (nr ^ (nr & (__BITOPS_WORDSIZE - 1))) >> 3; /* make OR mask */ @@ -141,10 +127,6 @@ static inline void clear_bit_cs(unsigned long nr, volatile unsigned long *ptr) unsigned long addr, old, new, mask; addr = (unsigned long) ptr; -#if ALIGN_CS == 1 - nr += (addr & __BITOPS_ALIGN) << 3; /* add alignment to bit number */ - addr ^= addr & __BITOPS_ALIGN; /* align address to 8 */ -#endif /* calculate address for CS */ addr += (nr ^ (nr & (__BITOPS_WORDSIZE - 1))) >> 3; /* make AND mask */ @@ -161,10 +143,6 @@ static inline void change_bit_cs(unsigned long nr, volatile unsigned long *ptr) unsigned long addr, old, new, mask; addr = (unsigned long) ptr; -#if ALIGN_CS == 1 - nr += (addr & __BITOPS_ALIGN) << 3; /* add alignment to bit number */ - addr ^= addr & __BITOPS_ALIGN; /* align address to 8 */ -#endif /* calculate address for CS */ addr += (nr ^ (nr & (__BITOPS_WORDSIZE - 1))) >> 3; /* make XOR mask */ @@ -182,10 +160,6 @@ test_and_set_bit_cs(unsigned long nr, volatile unsigned long *ptr) unsigned long addr, old, new, mask; addr = (unsigned long) ptr; -#if ALIGN_CS == 1 - nr += (addr & __BITOPS_ALIGN) << 3; /* add alignment to bit number */ - addr ^= addr & __BITOPS_ALIGN; /* align address to 8 */ -#endif /* calculate address for CS */ addr += (nr ^ (nr & (__BITOPS_WORDSIZE - 1))) >> 3; /* make OR/test mask */ @@ -205,10 +179,6 @@ test_and_clear_bit_cs(unsigned long nr, volatile unsigned long *ptr) unsigned long addr, old, new, mask; addr = (unsigned long) ptr; -#if ALIGN_CS == 1 - nr += (addr & __BITOPS_ALIGN) << 3; /* add alignment to bit number */ - addr ^= addr & __BITOPS_ALIGN; /* align address to 8 */ -#endif /* calculate address for CS */ addr += (nr ^ (nr & (__BITOPS_WORDSIZE - 1))) >> 3; /* make AND/test mask */ @@ -228,10 +198,6 @@ test_and_change_bit_cs(unsigned long nr, volatile unsigned long *ptr) unsigned long addr, old, new, mask; addr = (unsigned long) ptr; -#if ALIGN_CS == 1 - nr += (addr & __BITOPS_ALIGN) << 3; /* add alignment to bit number */ - addr ^= addr & __BITOPS_ALIGN; /* align address to 8 */ -#endif /* calculate address for CS */ addr += (nr ^ (nr & (__BITOPS_WORDSIZE - 1))) >> 3; /* make XOR/test mask */ @@ -834,8 +800,6 @@ static inline int sched_find_first_bit(unsigned long *b) #include -#ifdef __KERNEL__ - /* * ATTENTION: intel byte ordering convention for ext2 and minix !! * bit 0 is the LSB of addr; bit 31 is the MSB of addr; -- cgit v1.2.3-70-g09d2 From 40154b824331cd9c81c06545761338f3d80a36e2 Mon Sep 17 00:00:00 2001 From: Peter Oberparleiter Date: Thu, 29 Jun 2006 14:57:03 +0200 Subject: [S390] cio async subchannel reprobe. Changes in the DASD driver require an asynchronous implementation of the subchannel reprobe loop. This loop was so far only used by the blacklisting mechanism but is now available to all CCW device drivers. Signed-off-by: Peter Oberparleiter Signed-off-by: Martin Schwidefsky --- drivers/s390/cio/blacklist.c | 35 +----------------------- drivers/s390/cio/css.c | 63 ++++++++++++++++++++++++++++++++++++++++++++ drivers/s390/cio/device.c | 4 +-- drivers/s390/cio/device.h | 6 +++++ include/asm-s390/cio.h | 2 ++ 5 files changed, 74 insertions(+), 36 deletions(-) (limited to 'include') diff --git a/drivers/s390/cio/blacklist.c b/drivers/s390/cio/blacklist.c index 0960bef7b19..15b895496a4 100644 --- a/drivers/s390/cio/blacklist.c +++ b/drivers/s390/cio/blacklist.c @@ -224,39 +224,6 @@ is_blacklisted (int ssid, int devno) } #ifdef CONFIG_PROC_FS -static int -__s390_redo_validation(struct subchannel_id schid, void *data) -{ - int ret; - struct subchannel *sch; - - sch = get_subchannel_by_schid(schid); - if (sch) { - /* Already known. */ - put_device(&sch->dev); - return 0; - } - ret = css_probe_device(schid); - if (ret == -ENXIO) - return ret; /* We're through. */ - if (ret == -ENOMEM) - /* Stop validation for now. Bad, but no need for a panic. */ - return ret; - return 0; -} - -/* - * Function: s390_redo_validation - * Look for no longer blacklisted devices - * FIXME: there must be a better way to do this */ -static inline void -s390_redo_validation (void) -{ - CIO_TRACE_EVENT (0, "redoval"); - - for_each_subchannel(__s390_redo_validation, NULL); -} - /* * Function: blacklist_parse_proc_parameters * parse the stuff which is piped to /proc/cio_ignore @@ -281,7 +248,7 @@ blacklist_parse_proc_parameters (char *buf) return; } - s390_redo_validation (); + css_schedule_reprobe(); } /* Iterator struct for all devices. */ diff --git a/drivers/s390/cio/css.c b/drivers/s390/cio/css.c index 74ea8aac4b7..1d3be80797f 100644 --- a/drivers/s390/cio/css.c +++ b/drivers/s390/cio/css.c @@ -19,9 +19,11 @@ #include "cio_debug.h" #include "ioasm.h" #include "chsc.h" +#include "device.h" int need_rescan = 0; int css_init_done = 0; +static int need_reprobe = 0; static int max_ssid = 0; struct channel_subsystem *css[__MAX_CSSID + 1]; @@ -339,6 +341,67 @@ typedef void (*workfunc)(void *); DECLARE_WORK(slow_path_work, (workfunc)css_trigger_slow_path, NULL); struct workqueue_struct *slow_path_wq; +/* Reprobe subchannel if unregistered. */ +static int reprobe_subchannel(struct subchannel_id schid, void *data) +{ + struct subchannel *sch; + int ret; + + CIO_DEBUG(KERN_INFO, 6, "cio: reprobe 0.%x.%04x\n", + schid.ssid, schid.sch_no); + if (need_reprobe) + return -EAGAIN; + + sch = get_subchannel_by_schid(schid); + if (sch) { + /* Already known. */ + put_device(&sch->dev); + return 0; + } + + ret = css_probe_device(schid); + switch (ret) { + case 0: + break; + case -ENXIO: + case -ENOMEM: + /* These should abort looping */ + break; + default: + ret = 0; + } + + return ret; +} + +/* Work function used to reprobe all unregistered subchannels. */ +static void reprobe_all(void *data) +{ + int ret; + + CIO_MSG_EVENT(2, "reprobe start\n"); + + need_reprobe = 0; + /* Make sure initial subchannel scan is done. */ + wait_event(ccw_device_init_wq, + atomic_read(&ccw_device_init_count) == 0); + ret = for_each_subchannel(reprobe_subchannel, NULL); + + CIO_MSG_EVENT(2, "reprobe done (rc=%d, need_reprobe=%d)\n", ret, + need_reprobe); +} + +DECLARE_WORK(css_reprobe_work, reprobe_all, NULL); + +/* Schedule reprobing of all unregistered subchannels. */ +void css_schedule_reprobe(void) +{ + need_reprobe = 1; + queue_work(ccw_device_work, &css_reprobe_work); +} + +EXPORT_SYMBOL_GPL(css_schedule_reprobe); + /* * Rescan for new devices. FIXME: This is slow. * This function is called when we have lost CRWs due to overflows and we have diff --git a/drivers/s390/cio/device.c b/drivers/s390/cio/device.c index 8e3053c2a45..eafde43e841 100644 --- a/drivers/s390/cio/device.c +++ b/drivers/s390/cio/device.c @@ -133,8 +133,8 @@ struct css_driver io_subchannel_driver = { struct workqueue_struct *ccw_device_work; struct workqueue_struct *ccw_device_notify_work; -static wait_queue_head_t ccw_device_init_wq; -static atomic_t ccw_device_init_count; +wait_queue_head_t ccw_device_init_wq; +atomic_t ccw_device_init_count; static int __init init_ccw_bus_type (void) diff --git a/drivers/s390/cio/device.h b/drivers/s390/cio/device.h index 11587ebb728..8e0d1db3dd4 100644 --- a/drivers/s390/cio/device.h +++ b/drivers/s390/cio/device.h @@ -1,6 +1,10 @@ #ifndef S390_DEVICE_H #define S390_DEVICE_H +#include +#include +#include + /* * states of the device statemachine */ @@ -67,6 +71,8 @@ dev_fsm_final_state(struct ccw_device *cdev) extern struct workqueue_struct *ccw_device_work; extern struct workqueue_struct *ccw_device_notify_work; +extern wait_queue_head_t ccw_device_init_wq; +extern atomic_t ccw_device_init_count; void io_subchannel_recog_done(struct ccw_device *cdev); diff --git a/include/asm-s390/cio.h b/include/asm-s390/cio.h index 089cf567c31..2b161930635 100644 --- a/include/asm-s390/cio.h +++ b/include/asm-s390/cio.h @@ -276,6 +276,8 @@ extern void wait_cons_dev(void); extern void clear_all_subchannels(void); +extern void css_schedule_reprobe(void); + #endif #endif -- cgit v1.2.3-70-g09d2 From 4980082db1a8aa3ec45aa22cd4a10021955e22ed Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Thu, 29 Jun 2006 14:57:25 +0200 Subject: [S390] __syscall_return error check. Fix __syscall_return macro: valid error numbers are in the range of -1..-4095. Signed-off-by: Heiko Carstens Signed-off-by: Martin Schwidefsky --- include/asm-s390/unistd.h | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) (limited to 'include') diff --git a/include/asm-s390/unistd.h b/include/asm-s390/unistd.h index e21443d3ea1..aa7a243862e 100644 --- a/include/asm-s390/unistd.h +++ b/include/asm-s390/unistd.h @@ -394,11 +394,9 @@ #ifdef __KERNEL__ -/* user-visible error numbers are in the range -1 - -122: see */ - #define __syscall_return(type, res) \ do { \ - if ((unsigned long)(res) >= (unsigned long)(-125)) { \ + if ((unsigned long)(res) >= (unsigned long)(-4095)) {\ errno = -(res); \ res = -1; \ } \ -- cgit v1.2.3-70-g09d2 From cc43ec1ab0ae67c9cd5b3c2483a731afc98cbc2c Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Thu, 29 Jun 2006 15:02:47 +0200 Subject: [S390] preempt_count initialization. The preempt_count in the thread_info structure must be initialized to 1. Signed-off-by: Heiko Carstens Signed-off-by: Martin Schwidefsky --- include/asm-s390/thread_info.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include') diff --git a/include/asm-s390/thread_info.h b/include/asm-s390/thread_info.h index 8e0c7ed73d0..0a518915bf9 100644 --- a/include/asm-s390/thread_info.h +++ b/include/asm-s390/thread_info.h @@ -63,6 +63,7 @@ struct thread_info { .exec_domain = &default_exec_domain, \ .flags = 0, \ .cpu = 0, \ + .preempt_count = 1, \ .restart_block = { \ .fn = do_no_restart_syscall, \ }, \ -- cgit v1.2.3-70-g09d2 From 405455734e1cdec09c37233216f9240cb1a058e5 Mon Sep 17 00:00:00 2001 From: Horst Hummel Date: Thu, 29 Jun 2006 15:08:18 +0200 Subject: [S390] add PAV support to the dasd driver. Add support for parallel-access-volumes to the dasd driver. This allows concurrent access to dasd devices with multiple channel programs. Signed-off-by: Horst Hummel Signed-off-by: Martin Schwidefsky --- drivers/s390/block/dasd.c | 51 +++++++------- drivers/s390/block/dasd_devmap.c | 80 +++++++++++++++++++--- drivers/s390/block/dasd_eckd.c | 140 ++++++++++++++++++++++++++++++--------- drivers/s390/block/dasd_eckd.h | 16 ++++- drivers/s390/block/dasd_fba.c | 20 +----- drivers/s390/block/dasd_int.h | 2 +- include/asm-s390/dasd.h | 8 ++- 7 files changed, 227 insertions(+), 90 deletions(-) (limited to 'include') diff --git a/drivers/s390/block/dasd.c b/drivers/s390/block/dasd.c index 7e9978ad144..bafcd2f20ae 100644 --- a/drivers/s390/block/dasd.c +++ b/drivers/s390/block/dasd.c @@ -1855,15 +1855,34 @@ dasd_generic_probe (struct ccw_device *cdev, { int ret; + ret = ccw_device_set_options(cdev, CCWDEV_DO_PATHGROUP); + if (ret) { + printk(KERN_WARNING + "dasd_generic_probe: could not set ccw-device options " + "for %s\n", cdev->dev.bus_id); + return ret; + } ret = dasd_add_sysfs_files(cdev); if (ret) { printk(KERN_WARNING "dasd_generic_probe: could not add sysfs entries " "for %s\n", cdev->dev.bus_id); - } else { - cdev->handler = &dasd_int_handler; + return ret; } + cdev->handler = &dasd_int_handler; + /* + * Automatically online either all dasd devices (dasd_autodetect) + * or all devices specified with dasd= parameters during + * initial probe. + */ + if ((dasd_get_feature(cdev, DASD_FEATURE_INITIAL_ONLINE) > 0 ) || + (dasd_autodetect && dasd_busid_known(cdev->dev.bus_id) != 0)) + ret = ccw_device_set_online(cdev); + if (ret) + printk(KERN_WARNING + "dasd_generic_probe: could not initially online " + "ccw-device %s\n", cdev->dev.bus_id); return ret; } @@ -1911,6 +1930,8 @@ dasd_generic_set_online (struct ccw_device *cdev, struct dasd_device *device; int rc; + /* first online clears initial online feature flag */ + dasd_set_feature(cdev, DASD_FEATURE_INITIAL_ONLINE, 0); device = dasd_create_device(cdev); if (IS_ERR(device)) return PTR_ERR(device); @@ -2065,31 +2086,6 @@ dasd_generic_notify(struct ccw_device *cdev, int event) return ret; } -/* - * Automatically online either all dasd devices (dasd_autodetect) or - * all devices specified with dasd= parameters. - */ -static int -__dasd_auto_online(struct device *dev, void *data) -{ - struct ccw_device *cdev; - - cdev = to_ccwdev(dev); - if (dasd_autodetect || dasd_busid_known(cdev->dev.bus_id) == 0) - ccw_device_set_online(cdev); - return 0; -} - -void -dasd_generic_auto_online (struct ccw_driver *dasd_discipline_driver) -{ - struct device_driver *drv; - - drv = get_driver(&dasd_discipline_driver->driver); - driver_for_each_device(drv, NULL, NULL, __dasd_auto_online); - put_driver(drv); -} - static int __init dasd_init(void) @@ -2170,5 +2166,4 @@ EXPORT_SYMBOL_GPL(dasd_generic_remove); EXPORT_SYMBOL_GPL(dasd_generic_notify); EXPORT_SYMBOL_GPL(dasd_generic_set_online); EXPORT_SYMBOL_GPL(dasd_generic_set_offline); -EXPORT_SYMBOL_GPL(dasd_generic_auto_online); diff --git a/drivers/s390/block/dasd_devmap.c b/drivers/s390/block/dasd_devmap.c index 672e50314b1..9e9ae717960 100644 --- a/drivers/s390/block/dasd_devmap.c +++ b/drivers/s390/block/dasd_devmap.c @@ -27,7 +27,7 @@ #include "dasd_int.h" kmem_cache_t *dasd_page_cache; -EXPORT_SYMBOL(dasd_page_cache); +EXPORT_SYMBOL_GPL(dasd_page_cache); /* * dasd_devmap_t is used to store the features and the relation @@ -48,6 +48,20 @@ struct dasd_devmap { struct dasd_uid uid; }; +/* + * dasd_servermap is used to store the server_id of all storage servers + * accessed by DASD device driver. + */ +struct dasd_servermap { + struct list_head list; + struct server_id { + char vendor[4]; + char serial[15]; + } sid; +}; + +static struct list_head dasd_serverlist; + /* * Parameter parsing functions for dasd= parameter. The syntax is: * : (0x)?[0-9a-fA-F]+ @@ -64,6 +78,8 @@ struct dasd_devmap { int dasd_probeonly = 0; /* is true, when probeonly mode is active */ int dasd_autodetect = 0; /* is true, when autodetection is active */ +int dasd_nopav = 0; /* is true, when PAV is disabled */ +EXPORT_SYMBOL_GPL(dasd_nopav); /* * char *dasd[] is intended to hold the ranges supplied by the dasd= statement @@ -228,19 +244,24 @@ dasd_parse_keyword( char *parsestring ) { length = strlen(parsestring); residual_str = parsestring + length; } - if (strncmp ("autodetect", parsestring, length) == 0) { + if (strncmp("autodetect", parsestring, length) == 0) { dasd_autodetect = 1; MESSAGE (KERN_INFO, "%s", "turning to autodetection mode"); return residual_str; } - if (strncmp ("probeonly", parsestring, length) == 0) { + if (strncmp("probeonly", parsestring, length) == 0) { dasd_probeonly = 1; MESSAGE(KERN_INFO, "%s", "turning to probeonly mode"); return residual_str; } - if (strncmp ("fixedbuffers", parsestring, length) == 0) { + if (strncmp("nopav", parsestring, length) == 0) { + dasd_nopav = 1; + MESSAGE(KERN_INFO, "%s", "disable PAV mode"); + return residual_str; + } + if (strncmp("fixedbuffers", parsestring, length) == 0) { if (dasd_page_cache) return residual_str; dasd_page_cache = @@ -294,6 +315,8 @@ dasd_parse_range( char *parsestring ) { features = dasd_feature_list(str, &str); if (features < 0) return ERR_PTR(-EINVAL); + /* each device in dasd= parameter should be set initially online */ + features |= DASD_FEATURE_INITIAL_ONLINE; while (from <= to) { sprintf(bus_id, "%01x.%01x.%04x", from_id0, from_id1, from++); @@ -836,6 +859,38 @@ static struct attribute_group dasd_attr_group = { .attrs = dasd_attrs, }; +/* + * Check if the related storage server is already contained in the + * dasd_serverlist. If server is not contained, create new entry. + * Return 0 if server was already in serverlist, + * 1 if the server was added successfully + * <0 in case of error. + */ +static int +dasd_add_server(struct dasd_uid *uid) +{ + struct dasd_servermap *new, *tmp; + + /* check if server is already contained */ + list_for_each_entry(tmp, &dasd_serverlist, list) + // normale cmp? + if (strncmp(tmp->sid.vendor, uid->vendor, + sizeof(tmp->sid.vendor)) == 0 + && strncmp(tmp->sid.serial, uid->serial, + sizeof(tmp->sid.serial)) == 0) + return 0; + + new = (struct dasd_servermap *) + kzalloc(sizeof(struct dasd_servermap), GFP_KERNEL); + if (!new) + return -ENOMEM; + + strncpy(new->sid.vendor, uid->vendor, sizeof(new->sid.vendor)); + strncpy(new->sid.serial, uid->serial, sizeof(new->sid.serial)); + list_add(&new->list, &dasd_serverlist); + return 1; +} + /* * Return copy of the device unique identifier. @@ -856,21 +911,26 @@ dasd_get_uid(struct ccw_device *cdev, struct dasd_uid *uid) /* * Register the given device unique identifier into devmap struct. + * Return 0 if server was already in serverlist, + * 1 if the server was added successful + * <0 in case of error. */ int dasd_set_uid(struct ccw_device *cdev, struct dasd_uid *uid) { struct dasd_devmap *devmap; + int rc; devmap = dasd_find_busid(cdev->dev.bus_id); if (IS_ERR(devmap)) return PTR_ERR(devmap); spin_lock(&dasd_devmap_lock); devmap->uid = *uid; + rc = dasd_add_server(uid); spin_unlock(&dasd_devmap_lock); - return 0; + return rc; } -EXPORT_SYMBOL(dasd_set_uid); +EXPORT_SYMBOL_GPL(dasd_set_uid); /* * Return value of the specified feature. @@ -882,7 +942,7 @@ dasd_get_feature(struct ccw_device *cdev, int feature) devmap = dasd_find_busid(cdev->dev.bus_id); if (IS_ERR(devmap)) - return (int) PTR_ERR(devmap); + return PTR_ERR(devmap); return ((devmap->features & feature) != 0); } @@ -898,7 +958,7 @@ dasd_set_feature(struct ccw_device *cdev, int feature, int flag) devmap = dasd_find_busid(cdev->dev.bus_id); if (IS_ERR(devmap)) - return (int) PTR_ERR(devmap); + return PTR_ERR(devmap); spin_lock(&dasd_devmap_lock); if (flag) @@ -934,8 +994,10 @@ dasd_devmap_init(void) dasd_max_devindex = 0; for (i = 0; i < 256; i++) INIT_LIST_HEAD(&dasd_hashlists[i]); - return 0; + /* Initialize servermap structure. */ + INIT_LIST_HEAD(&dasd_serverlist); + return 0; } void diff --git a/drivers/s390/block/dasd_eckd.c b/drivers/s390/block/dasd_eckd.c index 12257776e79..0dfab30e808 100644 --- a/drivers/s390/block/dasd_eckd.c +++ b/drivers/s390/block/dasd_eckd.c @@ -24,6 +24,7 @@ #include #include #include +#include #include #include "dasd_int.h" @@ -89,17 +90,22 @@ dasd_eckd_probe (struct ccw_device *cdev) { int ret; - ret = dasd_generic_probe (cdev, &dasd_eckd_discipline); - if (ret) + /* set ECKD specific ccw-device options */ + ret = ccw_device_set_options(cdev, CCWDEV_ALLOW_FORCE); + if (ret) { + printk(KERN_WARNING + "dasd_eckd_probe: could not set ccw-device options " + "for %s\n", cdev->dev.bus_id); return ret; - ccw_device_set_options(cdev, CCWDEV_DO_PATHGROUP | CCWDEV_ALLOW_FORCE); - return 0; + } + ret = dasd_generic_probe(cdev, &dasd_eckd_discipline); + return ret; } static int dasd_eckd_set_online(struct ccw_device *cdev) { - return dasd_generic_set_online (cdev, &dasd_eckd_discipline); + return dasd_generic_set_online(cdev, &dasd_eckd_discipline); } static struct ccw_driver dasd_eckd_driver = { @@ -540,6 +546,86 @@ dasd_eckd_read_conf(struct dasd_device *device) return 0; } +/* + * Build CP for Perform Subsystem Function - SSC. + */ +struct dasd_ccw_req * +dasd_eckd_build_psf_ssc(struct dasd_device *device) +{ + struct dasd_ccw_req *cqr; + struct dasd_psf_ssc_data *psf_ssc_data; + struct ccw1 *ccw; + + cqr = dasd_smalloc_request("ECKD", 1 /* PSF */ , + sizeof(struct dasd_psf_ssc_data), + device); + + if (IS_ERR(cqr)) { + DEV_MESSAGE(KERN_WARNING, device, "%s", + "Could not allocate PSF-SSC request"); + return cqr; + } + psf_ssc_data = (struct dasd_psf_ssc_data *)cqr->data; + psf_ssc_data->order = PSF_ORDER_SSC; + psf_ssc_data->suborder = 0x08; + + ccw = cqr->cpaddr; + ccw->cmd_code = DASD_ECKD_CCW_PSF; + ccw->cda = (__u32)(addr_t)psf_ssc_data; + ccw->count = 66; + + cqr->device = device; + cqr->expires = 10*HZ; + cqr->buildclk = get_clock(); + cqr->status = DASD_CQR_FILLED; + return cqr; +} + +/* + * Perform Subsystem Function. + * It is necessary to trigger CIO for channel revalidation since this + * call might change behaviour of DASD devices. + */ +static int +dasd_eckd_psf_ssc(struct dasd_device *device) +{ + struct dasd_ccw_req *cqr; + int rc; + + cqr = dasd_eckd_build_psf_ssc(device); + if (IS_ERR(cqr)) + return PTR_ERR(cqr); + + rc = dasd_sleep_on(cqr); + if (!rc) + /* trigger CIO to reprobe devices */ + css_schedule_reprobe(); + dasd_sfree_request(cqr, cqr->device); + return rc; +} + +/* + * Valide storage server of current device. + */ +static int +dasd_eckd_validate_server(struct dasd_device *device) +{ + int rc; + + /* Currently PAV is the only reason to 'validate' server on LPAR */ + if (dasd_nopav || MACHINE_IS_VM) + return 0; + + rc = dasd_eckd_psf_ssc(device); + if (rc) + /* may be requested feature is not available on server, + * therefore just report error and go ahead */ + DEV_MESSAGE(KERN_INFO, device, + "Perform Subsystem Function returned rc=%d", rc); + /* RE-Read Configuration Data */ + return dasd_eckd_read_conf(device); +} + /* * Check device characteristics. * If the device is accessible using ECKD discipline, the device is enabled. @@ -570,16 +656,29 @@ dasd_eckd_check_characteristics(struct dasd_device *device) private->attrib.operation = DASD_NORMAL_CACHE; private->attrib.nr_cyl = 0; + /* Read Configuration Data */ + rc = dasd_eckd_read_conf(device); + if (rc) + return rc; + + /* Generate device unique id and register in devmap */ + rc = dasd_eckd_generate_uid(device, &uid); + if (rc) + return rc; + rc = dasd_set_uid(device->cdev, &uid); + if (rc == 1) /* new server found */ + rc = dasd_eckd_validate_server(device); + if (rc) + return rc; + /* Read Device Characteristics */ rdc_data = (void *) &(private->rdc_data); memset(rdc_data, 0, sizeof(rdc_data)); rc = read_dev_chars(device->cdev, &rdc_data, 64); - if (rc) { + if (rc) DEV_MESSAGE(KERN_WARNING, device, - "Read device characteristics returned error %d", - rc); - return rc; - } + "Read device characteristics returned " + "rc=%d", rc); DEV_MESSAGE(KERN_INFO, device, "%04X/%02X(CU:%04X/%02X) Cyl:%d Head:%d Sec:%d", @@ -590,19 +689,6 @@ dasd_eckd_check_characteristics(struct dasd_device *device) private->rdc_data.no_cyl, private->rdc_data.trk_per_cyl, private->rdc_data.sec_per_trk); - - /* Read Configuration Data */ - rc = dasd_eckd_read_conf (device); - if (rc) - return rc; - - /* Generate device unique id and register in devmap */ - rc = dasd_eckd_generate_uid(device, &uid); - if (rc) - return rc; - - rc = dasd_set_uid(device->cdev, &uid); - return rc; } @@ -1687,14 +1773,8 @@ static struct dasd_discipline dasd_eckd_discipline = { static int __init dasd_eckd_init(void) { - int ret; - ASCEBC(dasd_eckd_discipline.ebcname, 4); - - ret = ccw_driver_register(&dasd_eckd_driver); - if (!ret) - dasd_generic_auto_online(&dasd_eckd_driver); - return ret; + return ccw_driver_register(&dasd_eckd_driver); } static void __exit diff --git a/drivers/s390/block/dasd_eckd.h b/drivers/s390/block/dasd_eckd.h index 9d91c863256..712ff165013 100644 --- a/drivers/s390/block/dasd_eckd.h +++ b/drivers/s390/block/dasd_eckd.h @@ -41,9 +41,10 @@ #define DASD_ECKD_CCW_RESERVE 0xB4 /* - *Perform Subsystem Function / Sub-Orders + * Perform Subsystem Function / Sub-Orders */ -#define PSF_ORDER_PRSSD 0x18 +#define PSF_ORDER_PRSSD 0x18 +#define PSF_ORDER_SSC 0x1D /***************************************************************************** * SECTION: Type Definitions @@ -353,4 +354,15 @@ struct dasd_psf_prssd_data { unsigned char varies[9]; } __attribute__ ((packed)); +/* + * Perform Subsystem Function - Set Subsystem Characteristics + */ +struct dasd_psf_ssc_data { + unsigned char order; + unsigned char flags; + unsigned char cu_type[4]; + unsigned char suborder; + unsigned char reserved[59]; +} __attribute__((packed)); + #endif /* DASD_ECKD_H */ diff --git a/drivers/s390/block/dasd_fba.c b/drivers/s390/block/dasd_fba.c index d331c6e22c5..bb7755b9b19 100644 --- a/drivers/s390/block/dasd_fba.c +++ b/drivers/s390/block/dasd_fba.c @@ -56,19 +56,13 @@ static struct ccw_driver dasd_fba_driver; /* see below */ static int dasd_fba_probe(struct ccw_device *cdev) { - int ret; - - ret = dasd_generic_probe (cdev, &dasd_fba_discipline); - if (ret) - return ret; - ccw_device_set_options(cdev, CCWDEV_DO_PATHGROUP); - return 0; + return dasd_generic_probe(cdev, &dasd_fba_discipline); } static int dasd_fba_set_online(struct ccw_device *cdev) { - return dasd_generic_set_online (cdev, &dasd_fba_discipline); + return dasd_generic_set_online(cdev, &dasd_fba_discipline); } static struct ccw_driver dasd_fba_driver = { @@ -569,16 +563,8 @@ static struct dasd_discipline dasd_fba_discipline = { static int __init dasd_fba_init(void) { - int ret; - ASCEBC(dasd_fba_discipline.ebcname, 4); - - ret = ccw_driver_register(&dasd_fba_driver); - if (ret) - return ret; - - dasd_generic_auto_online(&dasd_fba_driver); - return 0; + return ccw_driver_register(&dasd_fba_driver); } static void __exit diff --git a/drivers/s390/block/dasd_int.h b/drivers/s390/block/dasd_int.h index 2c1fa85a3da..03a83efc34c 100644 --- a/drivers/s390/block/dasd_int.h +++ b/drivers/s390/block/dasd_int.h @@ -513,12 +513,12 @@ void dasd_generic_remove (struct ccw_device *cdev); int dasd_generic_set_online(struct ccw_device *, struct dasd_discipline *); int dasd_generic_set_offline (struct ccw_device *cdev); int dasd_generic_notify(struct ccw_device *, int); -void dasd_generic_auto_online (struct ccw_driver *); /* externals in dasd_devmap.c */ extern int dasd_max_devindex; extern int dasd_probeonly; extern int dasd_autodetect; +extern int dasd_nopav; int dasd_devmap_init(void); void dasd_devmap_exit(void); diff --git a/include/asm-s390/dasd.h b/include/asm-s390/dasd.h index 1630c26e8f4..c042f957808 100644 --- a/include/asm-s390/dasd.h +++ b/include/asm-s390/dasd.h @@ -68,10 +68,12 @@ typedef struct dasd_information2_t { * 0x00: default features * 0x01: readonly (ro) * 0x02: use diag discipline (diag) + * 0x04: set the device initially online (internal use only) */ -#define DASD_FEATURE_DEFAULT 0 -#define DASD_FEATURE_READONLY 1 -#define DASD_FEATURE_USEDIAG 2 +#define DASD_FEATURE_DEFAULT 0x00 +#define DASD_FEATURE_READONLY 0x01 +#define DASD_FEATURE_USEDIAG 0x02 +#define DASD_FEATURE_INITIAL_ONLINE 0x04 #define DASD_PARTN_BITS 2 -- cgit v1.2.3-70-g09d2 From 94bb063312d872d9269deb2e5c0c7c6d5b0318e1 Mon Sep 17 00:00:00 2001 From: Cornelia Huck Date: Thu, 29 Jun 2006 15:08:41 +0200 Subject: [S390] rework of channel measurement facility. Fixes for several channel measurement facility bugs: * Blocks copied from the hardware might not be consistent. Solve this by moving the copying into idle state and repeating the copying. * avg_sample_interval changed with every read, even though no new block was available. Solve this by storing a timestamp when the last new block was received. * Several locking issues. * Measurements were not reenabled after a disconnected device became available again. * Remove #defines for ioctls that were never implemented. Signed-off-by: Cornelia Huck Signed-off-by: Martin Schwidefsky --- drivers/s390/cio/cmf.c | 623 ++++++++++++++++++++++++++++++------------ drivers/s390/cio/device.h | 4 + drivers/s390/cio/device_fsm.c | 18 +- include/asm-s390/cmb.h | 4 - 4 files changed, 468 insertions(+), 181 deletions(-) (limited to 'include') diff --git a/drivers/s390/cio/cmf.c b/drivers/s390/cio/cmf.c index 07ef3f640f4..1c3e8e9012b 100644 --- a/drivers/s390/cio/cmf.c +++ b/drivers/s390/cio/cmf.c @@ -3,9 +3,10 @@ * * Linux on zSeries Channel Measurement Facility support * - * Copyright 2000,2003 IBM Corporation + * Copyright 2000,2006 IBM Corporation * - * Author: Arnd Bergmann + * Authors: Arnd Bergmann + * Cornelia Huck * * original idea from Natarajan Krishnaswami * @@ -96,9 +97,9 @@ module_param(format, bool, 0444); /** * struct cmb_operations - functions to use depending on cmb_format * - * all these functions operate on a struct cmf_device. There is only - * one instance of struct cmb_operations because all cmf_device - * objects are guaranteed to be of the same type. + * Most of these functions operate on a struct ccw_device. There is only + * one instance of struct cmb_operations because the format of the measurement + * data is guaranteed to be the same for every ccw_device. * * @alloc: allocate memory for a channel measurement block, * either with the help of a special pool or with kmalloc @@ -107,6 +108,7 @@ module_param(format, bool, 0444); * @readall: read a measurement block in a common format * @reset: clear the data in the associated measurement block and * reset its time stamp + * @align: align an allocated block so that the hardware can use it */ struct cmb_operations { int (*alloc) (struct ccw_device*); @@ -115,11 +117,19 @@ struct cmb_operations { u64 (*read) (struct ccw_device*, int); int (*readall)(struct ccw_device*, struct cmbdata *); void (*reset) (struct ccw_device*); + void * (*align) (void *); struct attribute_group *attr_group; }; static struct cmb_operations *cmbops; +struct cmb_data { + void *hw_block; /* Pointer to block updated by hardware */ + void *last_block; /* Last changed block copied from hardware block */ + int size; /* Size of hw_block and last_block */ + unsigned long long last_update; /* when last_block was updated */ +}; + /* our user interface is designed in terms of nanoseconds, * while the hardware measures total times in its own * unit.*/ @@ -226,63 +236,229 @@ struct set_schib_struct { unsigned long address; wait_queue_head_t wait; int ret; + struct kref kref; }; +static void cmf_set_schib_release(struct kref *kref) +{ + struct set_schib_struct *set_data; + + set_data = container_of(kref, struct set_schib_struct, kref); + kfree(set_data); +} + +#define CMF_PENDING 1 + static int set_schib_wait(struct ccw_device *cdev, u32 mme, int mbfc, unsigned long address) { - struct set_schib_struct s = { - .mme = mme, - .mbfc = mbfc, - .address = address, - .wait = __WAIT_QUEUE_HEAD_INITIALIZER(s.wait), - }; + struct set_schib_struct *set_data; + int ret; spin_lock_irq(cdev->ccwlock); - s.ret = set_schib(cdev, mme, mbfc, address); - if (s.ret != -EBUSY) { - goto out_nowait; + if (!cdev->private->cmb) { + ret = -ENODEV; + goto out; } + set_data = kzalloc(sizeof(struct set_schib_struct), GFP_ATOMIC); + if (!set_data) { + ret = -ENOMEM; + goto out; + } + init_waitqueue_head(&set_data->wait); + kref_init(&set_data->kref); + set_data->mme = mme; + set_data->mbfc = mbfc; + set_data->address = address; + + ret = set_schib(cdev, mme, mbfc, address); + if (ret != -EBUSY) + goto out_put; if (cdev->private->state != DEV_STATE_ONLINE) { - s.ret = -EBUSY; /* if the device is not online, don't even try again */ - goto out_nowait; + ret = -EBUSY; + goto out_put; } + cdev->private->state = DEV_STATE_CMFCHANGE; - cdev->private->cmb_wait = &s; - s.ret = 1; + set_data->ret = CMF_PENDING; + cdev->private->cmb_wait = set_data; spin_unlock_irq(cdev->ccwlock); - if (wait_event_interruptible(s.wait, s.ret != 1)) { + if (wait_event_interruptible(set_data->wait, + set_data->ret != CMF_PENDING)) { spin_lock_irq(cdev->ccwlock); - if (s.ret == 1) { - s.ret = -ERESTARTSYS; - cdev->private->cmb_wait = 0; + if (set_data->ret == CMF_PENDING) { + set_data->ret = -ERESTARTSYS; if (cdev->private->state == DEV_STATE_CMFCHANGE) cdev->private->state = DEV_STATE_ONLINE; } spin_unlock_irq(cdev->ccwlock); } - return s.ret; - -out_nowait: + spin_lock_irq(cdev->ccwlock); + cdev->private->cmb_wait = NULL; + ret = set_data->ret; +out_put: + kref_put(&set_data->kref, cmf_set_schib_release); +out: spin_unlock_irq(cdev->ccwlock); - return s.ret; + return ret; } void retry_set_schib(struct ccw_device *cdev) { - struct set_schib_struct *s; + struct set_schib_struct *set_data; + + set_data = cdev->private->cmb_wait; + if (!set_data) { + WARN_ON(1); + return; + } + kref_get(&set_data->kref); + set_data->ret = set_schib(cdev, set_data->mme, set_data->mbfc, + set_data->address); + wake_up(&set_data->wait); + kref_put(&set_data->kref, cmf_set_schib_release); +} + +static int cmf_copy_block(struct ccw_device *cdev) +{ + struct subchannel *sch; + void *reference_buf; + void *hw_block; + struct cmb_data *cmb_data; + + sch = to_subchannel(cdev->dev.parent); + + if (stsch(sch->schid, &sch->schib)) + return -ENODEV; + + if (sch->schib.scsw.fctl & SCSW_FCTL_START_FUNC) { + /* Don't copy if a start function is in progress. */ + if ((!sch->schib.scsw.actl & SCSW_ACTL_SUSPENDED) && + (sch->schib.scsw.actl & + (SCSW_ACTL_DEVACT | SCSW_ACTL_SCHACT)) && + (!sch->schib.scsw.stctl & SCSW_STCTL_SEC_STATUS)) + return -EBUSY; + } + cmb_data = cdev->private->cmb; + hw_block = cmbops->align(cmb_data->hw_block); + if (!memcmp(cmb_data->last_block, hw_block, cmb_data->size)) + /* No need to copy. */ + return 0; + reference_buf = kzalloc(cmb_data->size, GFP_ATOMIC); + if (!reference_buf) + return -ENOMEM; + /* Ensure consistency of block copied from hardware. */ + do { + memcpy(cmb_data->last_block, hw_block, cmb_data->size); + memcpy(reference_buf, hw_block, cmb_data->size); + } while (memcmp(cmb_data->last_block, reference_buf, cmb_data->size)); + cmb_data->last_update = get_clock(); + kfree(reference_buf); + return 0; +} + +struct copy_block_struct { + wait_queue_head_t wait; + int ret; + struct kref kref; +}; + +static void cmf_copy_block_release(struct kref *kref) +{ + struct copy_block_struct *copy_block; + + copy_block = container_of(kref, struct copy_block_struct, kref); + kfree(copy_block); +} + +static int cmf_cmb_copy_wait(struct ccw_device *cdev) +{ + struct copy_block_struct *copy_block; + int ret; + unsigned long flags; + + spin_lock_irqsave(cdev->ccwlock, flags); + if (!cdev->private->cmb) { + ret = -ENODEV; + goto out; + } + copy_block = kzalloc(sizeof(struct copy_block_struct), GFP_ATOMIC); + if (!copy_block) { + ret = -ENOMEM; + goto out; + } + init_waitqueue_head(©_block->wait); + kref_init(©_block->kref); + + ret = cmf_copy_block(cdev); + if (ret != -EBUSY) + goto out_put; + + if (cdev->private->state != DEV_STATE_ONLINE) { + ret = -EBUSY; + goto out_put; + } + + cdev->private->state = DEV_STATE_CMFUPDATE; + copy_block->ret = CMF_PENDING; + cdev->private->cmb_wait = copy_block; + + spin_unlock_irqrestore(cdev->ccwlock, flags); + if (wait_event_interruptible(copy_block->wait, + copy_block->ret != CMF_PENDING)) { + spin_lock_irqsave(cdev->ccwlock, flags); + if (copy_block->ret == CMF_PENDING) { + copy_block->ret = -ERESTARTSYS; + if (cdev->private->state == DEV_STATE_CMFUPDATE) + cdev->private->state = DEV_STATE_ONLINE; + } + spin_unlock_irqrestore(cdev->ccwlock, flags); + } + spin_lock_irqsave(cdev->ccwlock, flags); + cdev->private->cmb_wait = NULL; + ret = copy_block->ret; +out_put: + kref_put(©_block->kref, cmf_copy_block_release); +out: + spin_unlock_irqrestore(cdev->ccwlock, flags); + return ret; +} + +void cmf_retry_copy_block(struct ccw_device *cdev) +{ + struct copy_block_struct *copy_block; - s = cdev->private->cmb_wait; - cdev->private->cmb_wait = 0; - if (!s) { + copy_block = cdev->private->cmb_wait; + if (!copy_block) { WARN_ON(1); return; } - s->ret = set_schib(cdev, s->mme, s->mbfc, s->address); - wake_up(&s->wait); + kref_get(©_block->kref); + copy_block->ret = cmf_copy_block(cdev); + wake_up(©_block->wait); + kref_put(©_block->kref, cmf_copy_block_release); +} + +static void cmf_generic_reset(struct ccw_device *cdev) +{ + struct cmb_data *cmb_data; + + spin_lock_irq(cdev->ccwlock); + cmb_data = cdev->private->cmb; + if (cmb_data) { + memset(cmb_data->last_block, 0, cmb_data->size); + /* + * Need to reset hw block as well to make the hardware start + * from 0 again. + */ + memset(cmbops->align(cmb_data->hw_block), 0, cmb_data->size); + cmb_data->last_update = 0; + } + cdev->private->cmb_start_time = get_clock(); + spin_unlock_irq(cdev->ccwlock); } /** @@ -343,8 +519,8 @@ struct cmb { /* insert a single device into the cmb_area list * called with cmb_area.lock held from alloc_cmb */ -static inline int -alloc_cmb_single (struct ccw_device *cdev) +static inline int alloc_cmb_single (struct ccw_device *cdev, + struct cmb_data *cmb_data) { struct cmb *cmb; struct ccw_device_private *node; @@ -358,10 +534,12 @@ alloc_cmb_single (struct ccw_device *cdev) /* find first unused cmb in cmb_area.mem. * this is a little tricky: cmb_area.list - * remains sorted by ->cmb pointers */ + * remains sorted by ->cmb->hw_data pointers */ cmb = cmb_area.mem; list_for_each_entry(node, &cmb_area.list, cmb_list) { - if ((struct cmb*)node->cmb > cmb) + struct cmb_data *data; + data = node->cmb; + if ((struct cmb*)data->hw_block > cmb) break; cmb++; } @@ -372,7 +550,8 @@ alloc_cmb_single (struct ccw_device *cdev) /* insert new cmb */ list_add_tail(&cdev->private->cmb_list, &node->cmb_list); - cdev->private->cmb = cmb; + cmb_data->hw_block = cmb; + cdev->private->cmb = cmb_data; ret = 0; out: spin_unlock_irq(cdev->ccwlock); @@ -385,7 +564,19 @@ alloc_cmb (struct ccw_device *cdev) int ret; struct cmb *mem; ssize_t size; + struct cmb_data *cmb_data; + + /* Allocate private cmb_data. */ + cmb_data = kzalloc(sizeof(struct cmb_data), GFP_KERNEL); + if (!cmb_data) + return -ENOMEM; + cmb_data->last_block = kzalloc(sizeof(struct cmb), GFP_KERNEL); + if (!cmb_data->last_block) { + kfree(cmb_data); + return -ENOMEM; + } + cmb_data->size = sizeof(struct cmb); spin_lock(&cmb_area.lock); if (!cmb_area.mem) { @@ -414,29 +605,36 @@ alloc_cmb (struct ccw_device *cdev) } /* do the actual allocation */ - ret = alloc_cmb_single(cdev); + ret = alloc_cmb_single(cdev, cmb_data); out: spin_unlock(&cmb_area.lock); - + if (ret) { + kfree(cmb_data->last_block); + kfree(cmb_data); + } return ret; } -static void -free_cmb(struct ccw_device *cdev) +static void free_cmb(struct ccw_device *cdev) { struct ccw_device_private *priv; - - priv = cdev->private; + struct cmb_data *cmb_data; spin_lock(&cmb_area.lock); spin_lock_irq(cdev->ccwlock); + priv = cdev->private; + if (list_empty(&priv->cmb_list)) { /* already freed */ goto out; } + cmb_data = priv->cmb; priv->cmb = NULL; + if (cmb_data) + kfree(cmb_data->last_block); + kfree(cmb_data); list_del_init(&priv->cmb_list); if (list_empty(&cmb_area.list)) { @@ -451,83 +649,97 @@ out: spin_unlock(&cmb_area.lock); } -static int -set_cmb(struct ccw_device *cdev, u32 mme) +static int set_cmb(struct ccw_device *cdev, u32 mme) { u16 offset; + struct cmb_data *cmb_data; + unsigned long flags; - if (!cdev->private->cmb) + spin_lock_irqsave(cdev->ccwlock, flags); + if (!cdev->private->cmb) { + spin_unlock_irqrestore(cdev->ccwlock, flags); return -EINVAL; - - offset = mme ? (struct cmb *)cdev->private->cmb - cmb_area.mem : 0; + } + cmb_data = cdev->private->cmb; + offset = mme ? (struct cmb *)cmb_data->hw_block - cmb_area.mem : 0; + spin_unlock_irqrestore(cdev->ccwlock, flags); return set_schib_wait(cdev, mme, 0, offset); } -static u64 -read_cmb (struct ccw_device *cdev, int index) +static u64 read_cmb (struct ccw_device *cdev, int index) { - /* yes, we have to put it on the stack - * because the cmb must only be accessed - * atomically, e.g. with mvc */ - struct cmb cmb; - unsigned long flags; + struct cmb *cmb; u32 val; + int ret; + unsigned long flags; + + ret = cmf_cmb_copy_wait(cdev); + if (ret < 0) + return 0; spin_lock_irqsave(cdev->ccwlock, flags); if (!cdev->private->cmb) { - spin_unlock_irqrestore(cdev->ccwlock, flags); - return 0; + ret = 0; + goto out; } - - cmb = *(struct cmb*)cdev->private->cmb; - spin_unlock_irqrestore(cdev->ccwlock, flags); + cmb = ((struct cmb_data *)cdev->private->cmb)->last_block; switch (index) { case cmb_ssch_rsch_count: - return cmb.ssch_rsch_count; + ret = cmb->ssch_rsch_count; + goto out; case cmb_sample_count: - return cmb.sample_count; + ret = cmb->sample_count; + goto out; case cmb_device_connect_time: - val = cmb.device_connect_time; + val = cmb->device_connect_time; break; case cmb_function_pending_time: - val = cmb.function_pending_time; + val = cmb->function_pending_time; break; case cmb_device_disconnect_time: - val = cmb.device_disconnect_time; + val = cmb->device_disconnect_time; break; case cmb_control_unit_queuing_time: - val = cmb.control_unit_queuing_time; + val = cmb->control_unit_queuing_time; break; case cmb_device_active_only_time: - val = cmb.device_active_only_time; + val = cmb->device_active_only_time; break; default: - return 0; + ret = 0; + goto out; } - return time_to_avg_nsec(val, cmb.sample_count); + ret = time_to_avg_nsec(val, cmb->sample_count); +out: + spin_unlock_irqrestore(cdev->ccwlock, flags); + return ret; } -static int -readall_cmb (struct ccw_device *cdev, struct cmbdata *data) +static int readall_cmb (struct ccw_device *cdev, struct cmbdata *data) { - /* yes, we have to put it on the stack - * because the cmb must only be accessed - * atomically, e.g. with mvc */ - struct cmb cmb; - unsigned long flags; + struct cmb *cmb; + struct cmb_data *cmb_data; u64 time; + unsigned long flags; + int ret; + ret = cmf_cmb_copy_wait(cdev); + if (ret < 0) + return ret; spin_lock_irqsave(cdev->ccwlock, flags); - if (!cdev->private->cmb) { - spin_unlock_irqrestore(cdev->ccwlock, flags); - return -ENODEV; + cmb_data = cdev->private->cmb; + if (!cmb_data) { + ret = -ENODEV; + goto out; } - - cmb = *(struct cmb*)cdev->private->cmb; - time = get_clock() - cdev->private->cmb_start_time; - spin_unlock_irqrestore(cdev->ccwlock, flags); + if (cmb_data->last_update == 0) { + ret = -EAGAIN; + goto out; + } + cmb = cmb_data->last_block; + time = cmb_data->last_update - cdev->private->cmb_start_time; memset(data, 0, sizeof(struct cmbdata)); @@ -538,31 +750,32 @@ readall_cmb (struct ccw_device *cdev, struct cmbdata *data) data->elapsed_time = (time * 1000) >> 12; /* copy data to new structure */ - data->ssch_rsch_count = cmb.ssch_rsch_count; - data->sample_count = cmb.sample_count; + data->ssch_rsch_count = cmb->ssch_rsch_count; + data->sample_count = cmb->sample_count; /* time fields are converted to nanoseconds while copying */ - data->device_connect_time = time_to_nsec(cmb.device_connect_time); - data->function_pending_time = time_to_nsec(cmb.function_pending_time); - data->device_disconnect_time = time_to_nsec(cmb.device_disconnect_time); + data->device_connect_time = time_to_nsec(cmb->device_connect_time); + data->function_pending_time = time_to_nsec(cmb->function_pending_time); + data->device_disconnect_time = + time_to_nsec(cmb->device_disconnect_time); data->control_unit_queuing_time - = time_to_nsec(cmb.control_unit_queuing_time); + = time_to_nsec(cmb->control_unit_queuing_time); data->device_active_only_time - = time_to_nsec(cmb.device_active_only_time); + = time_to_nsec(cmb->device_active_only_time); + ret = 0; +out: + spin_unlock_irqrestore(cdev->ccwlock, flags); + return ret; +} - return 0; +static void reset_cmb(struct ccw_device *cdev) +{ + cmf_generic_reset(cdev); } -static void -reset_cmb(struct ccw_device *cdev) +static void * align_cmb(void *area) { - struct cmb *cmb; - spin_lock_irq(cdev->ccwlock); - cmb = cdev->private->cmb; - if (cmb) - memset (cmb, 0, sizeof (*cmb)); - cdev->private->cmb_start_time = get_clock(); - spin_unlock_irq(cdev->ccwlock); + return area; } static struct attribute_group cmf_attr_group; @@ -574,6 +787,7 @@ static struct cmb_operations cmbops_basic = { .read = read_cmb, .readall = readall_cmb, .reset = reset_cmb, + .align = align_cmb, .attr_group = &cmf_attr_group, }; @@ -610,22 +824,34 @@ static inline struct cmbe* cmbe_align(struct cmbe *c) return (struct cmbe*)addr; } -static int -alloc_cmbe (struct ccw_device *cdev) +static int alloc_cmbe (struct ccw_device *cdev) { struct cmbe *cmbe; - cmbe = kmalloc (sizeof (*cmbe) * 2, GFP_KERNEL); + struct cmb_data *cmb_data; + int ret; + + cmbe = kzalloc (sizeof (*cmbe) * 2, GFP_KERNEL); if (!cmbe) return -ENOMEM; - + cmb_data = kzalloc(sizeof(struct cmb_data), GFP_KERNEL); + if (!cmb_data) { + ret = -ENOMEM; + goto out_free; + } + cmb_data->last_block = kzalloc(sizeof(struct cmbe), GFP_KERNEL); + if (!cmb_data->last_block) { + ret = -ENOMEM; + goto out_free; + } + cmb_data->size = sizeof(struct cmbe); spin_lock_irq(cdev->ccwlock); if (cdev->private->cmb) { - kfree(cmbe); spin_unlock_irq(cdev->ccwlock); - return -EBUSY; + ret = -EBUSY; + goto out_free; } - - cdev->private->cmb = cmbe; + cmb_data->hw_block = cmbe; + cdev->private->cmb = cmb_data; spin_unlock_irq(cdev->ccwlock); /* activate global measurement if this is the first channel */ @@ -636,14 +862,24 @@ alloc_cmbe (struct ccw_device *cdev) spin_unlock(&cmb_area.lock); return 0; +out_free: + if (cmb_data) + kfree(cmb_data->last_block); + kfree(cmb_data); + kfree(cmbe); + return ret; } -static void -free_cmbe (struct ccw_device *cdev) +static void free_cmbe (struct ccw_device *cdev) { + struct cmb_data *cmb_data; + spin_lock_irq(cdev->ccwlock); - kfree(cdev->private->cmb); + cmb_data = cdev->private->cmb; cdev->private->cmb = NULL; + if (cmb_data) + kfree(cmb_data->last_block); + kfree(cmb_data); spin_unlock_irq(cdev->ccwlock); /* deactivate global measurement if this is the last channel */ @@ -654,89 +890,105 @@ free_cmbe (struct ccw_device *cdev) spin_unlock(&cmb_area.lock); } -static int -set_cmbe(struct ccw_device *cdev, u32 mme) +static int set_cmbe(struct ccw_device *cdev, u32 mme) { unsigned long mba; + struct cmb_data *cmb_data; + unsigned long flags; - if (!cdev->private->cmb) + spin_lock_irqsave(cdev->ccwlock, flags); + if (!cdev->private->cmb) { + spin_unlock_irqrestore(cdev->ccwlock, flags); return -EINVAL; - mba = mme ? (unsigned long) cmbe_align(cdev->private->cmb) : 0; + } + cmb_data = cdev->private->cmb; + mba = mme ? (unsigned long) cmbe_align(cmb_data->hw_block) : 0; + spin_unlock_irqrestore(cdev->ccwlock, flags); return set_schib_wait(cdev, mme, 1, mba); } -u64 -read_cmbe (struct ccw_device *cdev, int index) +static u64 read_cmbe (struct ccw_device *cdev, int index) { - /* yes, we have to put it on the stack - * because the cmb must only be accessed - * atomically, e.g. with mvc */ - struct cmbe cmb; - unsigned long flags; + struct cmbe *cmb; + struct cmb_data *cmb_data; u32 val; + int ret; + unsigned long flags; - spin_lock_irqsave(cdev->ccwlock, flags); - if (!cdev->private->cmb) { - spin_unlock_irqrestore(cdev->ccwlock, flags); + ret = cmf_cmb_copy_wait(cdev); + if (ret < 0) return 0; - } - cmb = *cmbe_align(cdev->private->cmb); - spin_unlock_irqrestore(cdev->ccwlock, flags); + spin_lock_irqsave(cdev->ccwlock, flags); + cmb_data = cdev->private->cmb; + if (!cmb_data) { + ret = 0; + goto out; + } + cmb = cmb_data->last_block; switch (index) { case cmb_ssch_rsch_count: - return cmb.ssch_rsch_count; + ret = cmb->ssch_rsch_count; + goto out; case cmb_sample_count: - return cmb.sample_count; + ret = cmb->sample_count; + goto out; case cmb_device_connect_time: - val = cmb.device_connect_time; + val = cmb->device_connect_time; break; case cmb_function_pending_time: - val = cmb.function_pending_time; + val = cmb->function_pending_time; break; case cmb_device_disconnect_time: - val = cmb.device_disconnect_time; + val = cmb->device_disconnect_time; break; case cmb_control_unit_queuing_time: - val = cmb.control_unit_queuing_time; + val = cmb->control_unit_queuing_time; break; case cmb_device_active_only_time: - val = cmb.device_active_only_time; + val = cmb->device_active_only_time; break; case cmb_device_busy_time: - val = cmb.device_busy_time; + val = cmb->device_busy_time; break; case cmb_initial_command_response_time: - val = cmb.initial_command_response_time; + val = cmb->initial_command_response_time; break; default: - return 0; + ret = 0; + goto out; } - return time_to_avg_nsec(val, cmb.sample_count); + ret = time_to_avg_nsec(val, cmb->sample_count); +out: + spin_unlock_irqrestore(cdev->ccwlock, flags); + return ret; } -static int -readall_cmbe (struct ccw_device *cdev, struct cmbdata *data) +static int readall_cmbe (struct ccw_device *cdev, struct cmbdata *data) { - /* yes, we have to put it on the stack - * because the cmb must only be accessed - * atomically, e.g. with mvc */ - struct cmbe cmb; - unsigned long flags; + struct cmbe *cmb; + struct cmb_data *cmb_data; u64 time; + unsigned long flags; + int ret; + ret = cmf_cmb_copy_wait(cdev); + if (ret < 0) + return ret; spin_lock_irqsave(cdev->ccwlock, flags); - if (!cdev->private->cmb) { - spin_unlock_irqrestore(cdev->ccwlock, flags); - return -ENODEV; + cmb_data = cdev->private->cmb; + if (!cmb_data) { + ret = -ENODEV; + goto out; } - - cmb = *cmbe_align(cdev->private->cmb); - time = get_clock() - cdev->private->cmb_start_time; - spin_unlock_irqrestore(cdev->ccwlock, flags); + if (cmb_data->last_update == 0) { + ret = -EAGAIN; + goto out; + } + time = cmb_data->last_update - cdev->private->cmb_start_time; memset (data, 0, sizeof(struct cmbdata)); @@ -746,35 +998,38 @@ readall_cmbe (struct ccw_device *cdev, struct cmbdata *data) /* conver to nanoseconds */ data->elapsed_time = (time * 1000) >> 12; + cmb = cmb_data->last_block; /* copy data to new structure */ - data->ssch_rsch_count = cmb.ssch_rsch_count; - data->sample_count = cmb.sample_count; + data->ssch_rsch_count = cmb->ssch_rsch_count; + data->sample_count = cmb->sample_count; /* time fields are converted to nanoseconds while copying */ - data->device_connect_time = time_to_nsec(cmb.device_connect_time); - data->function_pending_time = time_to_nsec(cmb.function_pending_time); - data->device_disconnect_time = time_to_nsec(cmb.device_disconnect_time); + data->device_connect_time = time_to_nsec(cmb->device_connect_time); + data->function_pending_time = time_to_nsec(cmb->function_pending_time); + data->device_disconnect_time = + time_to_nsec(cmb->device_disconnect_time); data->control_unit_queuing_time - = time_to_nsec(cmb.control_unit_queuing_time); + = time_to_nsec(cmb->control_unit_queuing_time); data->device_active_only_time - = time_to_nsec(cmb.device_active_only_time); - data->device_busy_time = time_to_nsec(cmb.device_busy_time); + = time_to_nsec(cmb->device_active_only_time); + data->device_busy_time = time_to_nsec(cmb->device_busy_time); data->initial_command_response_time - = time_to_nsec(cmb.initial_command_response_time); + = time_to_nsec(cmb->initial_command_response_time); - return 0; + ret = 0; +out: + spin_unlock_irqrestore(cdev->ccwlock, flags); + return ret; } -static void -reset_cmbe(struct ccw_device *cdev) +static void reset_cmbe(struct ccw_device *cdev) { - struct cmbe *cmb; - spin_lock_irq(cdev->ccwlock); - cmb = cmbe_align(cdev->private->cmb); - if (cmb) - memset (cmb, 0, sizeof (*cmb)); - cdev->private->cmb_start_time = get_clock(); - spin_unlock_irq(cdev->ccwlock); + cmf_generic_reset(cdev); +} + +static void * align_cmbe(void *area) +{ + return cmbe_align(area); } static struct attribute_group cmf_attr_group_ext; @@ -786,6 +1041,7 @@ static struct cmb_operations cmbops_extended = { .read = read_cmbe, .readall = readall_cmbe, .reset = reset_cmbe, + .align = align_cmbe, .attr_group = &cmf_attr_group_ext, }; @@ -803,14 +1059,19 @@ cmb_show_avg_sample_interval(struct device *dev, struct device_attribute *attr, struct ccw_device *cdev; long interval; unsigned long count; + struct cmb_data *cmb_data; cdev = to_ccwdev(dev); - interval = get_clock() - cdev->private->cmb_start_time; count = cmf_read(cdev, cmb_sample_count); - if (count) + spin_lock_irq(cdev->ccwlock); + cmb_data = cdev->private->cmb; + if (count) { + interval = cmb_data->last_update - + cdev->private->cmb_start_time; interval /= count; - else + } else interval = -1; + spin_unlock_irq(cdev->ccwlock); return sprintf(buf, "%ld\n", interval); } @@ -823,7 +1084,10 @@ cmb_show_avg_utilization(struct device *dev, struct device_attribute *attr, char int ret; ret = cmf_readall(to_ccwdev(dev), &data); - if (ret) + if (ret == -EAGAIN || ret == -ENODEV) + /* No data (yet/currently) available to use for calculation. */ + return sprintf(buf, "n/a\n"); + else if (ret) return ret; utilization = data.device_connect_time + @@ -982,6 +1246,13 @@ cmf_readall(struct ccw_device *cdev, struct cmbdata *data) return cmbops->readall(cdev, data); } +/* Reenable cmf when a disconnected device becomes available again. */ +int cmf_reenable(struct ccw_device *cdev) +{ + cmbops->reset(cdev); + return cmbops->set(cdev, 2); +} + static int __init init_cmf(void) { diff --git a/drivers/s390/cio/device.h b/drivers/s390/cio/device.h index 8e0d1db3dd4..00be9a5b4ac 100644 --- a/drivers/s390/cio/device.h +++ b/drivers/s390/cio/device.h @@ -27,6 +27,7 @@ enum dev_state { DEV_STATE_DISCONNECTED, DEV_STATE_DISCONNECTED_SENSE_ID, DEV_STATE_CMFCHANGE, + DEV_STATE_CMFUPDATE, /* last element! */ NR_DEV_STATES }; @@ -118,5 +119,8 @@ int ccw_device_stlck(struct ccw_device *); void ccw_device_set_timeout(struct ccw_device *, int); extern struct subchannel_id ccw_device_get_subchannel_id(struct ccw_device *); +/* Channel measurement facility related */ void retry_set_schib(struct ccw_device *cdev); +void cmf_retry_copy_block(struct ccw_device *); +int cmf_reenable(struct ccw_device *); #endif diff --git a/drivers/s390/cio/device_fsm.c b/drivers/s390/cio/device_fsm.c index 5ec8ef1df9f..7d0dd72635e 100644 --- a/drivers/s390/cio/device_fsm.c +++ b/drivers/s390/cio/device_fsm.c @@ -336,8 +336,11 @@ ccw_device_oper_notify(void *data) if (!ret) /* Driver doesn't want device back. */ ccw_device_do_unreg_rereg((void *)cdev); - else + else { + /* Reenable channel measurements, if needed. */ + cmf_reenable(cdev); wake_up(&cdev->private->wait_q); + } } /* @@ -1095,6 +1098,13 @@ ccw_device_change_cmfstate(struct ccw_device *cdev, enum dev_event dev_event) dev_fsm_event(cdev, dev_event); } +static void ccw_device_update_cmfblock(struct ccw_device *cdev, + enum dev_event dev_event) +{ + cmf_retry_copy_block(cdev); + cdev->private->state = DEV_STATE_ONLINE; + dev_fsm_event(cdev, dev_event); +} static void ccw_device_quiesce_done(struct ccw_device *cdev, enum dev_event dev_event) @@ -1249,6 +1259,12 @@ fsm_func_t *dev_jumptable[NR_DEV_STATES][NR_DEV_EVENTS] = { [DEV_EVENT_TIMEOUT] = ccw_device_change_cmfstate, [DEV_EVENT_VERIFY] = ccw_device_change_cmfstate, }, + [DEV_STATE_CMFUPDATE] = { + [DEV_EVENT_NOTOPER] = ccw_device_update_cmfblock, + [DEV_EVENT_INTERRUPT] = ccw_device_update_cmfblock, + [DEV_EVENT_TIMEOUT] = ccw_device_update_cmfblock, + [DEV_EVENT_VERIFY] = ccw_device_update_cmfblock, + }, }; /* diff --git a/include/asm-s390/cmb.h b/include/asm-s390/cmb.h index 2d09950a9c1..241756f80df 100644 --- a/include/asm-s390/cmb.h +++ b/include/asm-s390/cmb.h @@ -44,10 +44,6 @@ struct cmbdata { #define BIODASDCMFENABLE _IO(DASD_IOCTL_LETTER,32) /* enable channel measurement */ #define BIODASDCMFDISABLE _IO(DASD_IOCTL_LETTER,33) -/* reset channel measurement block */ -#define BIODASDRESETCMB _IO(DASD_IOCTL_LETTER,34) -/* read channel measurement data */ -#define BIODASDREADCMB _IOWR(DASD_IOCTL_LETTER,32,__u64) /* read channel measurement data */ #define BIODASDREADALLCMB _IOWR(DASD_IOCTL_LETTER,33,struct cmbdata) -- cgit v1.2.3-70-g09d2 From 0686cd8fbe3e5fb1441ae84b9cbc813f9297b879 Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Thu, 29 Jun 2006 02:24:25 -0700 Subject: [PATCH] fix sgivwfb compile drivers/built-in.o: In function `sgivwfb_set_par': sgivwfb.c:(.text+0x88583): undefined reference to `sgivwfb_mem_phys' sgivwfb.c:(.text+0x88596): undefined reference to `sgivwfb_mem_phys' sgivwfb.c:(.text+0x885a8): undefined reference to `sgivwfb_mem_phys' drivers/built-in.o: In function `sgivwfb_check_var': sgivwfb.c:(.text+0x88ad0): undefined reference to `sgivwfb_mem_size' drivers/built-in.o: In function `sgivwfb_mmap': sgivwfb.c:(.text+0x88c75): undefined reference to `sgivwfb_mem_size' sgivwfb.c:(.text+0x88c7f): undefined reference to `sgivwfb_mem_phys' drivers/built-in.o: In function `sgivwfb_probe': sgivwfb.c:(.init.text+0x4060): undefined reference to `sgivwfb_mem_size' sgivwfb.c:(.init.text+0x4065): undefined reference to `sgivwfb_mem_phys' sgivwfb.c:(.init.text+0x4076): undefined reference to `sgivwfb_mem_phys' sgivwfb.c:(.init.text+0x409c): undefined reference to `sgivwfb_mem_size' sgivwfb.c:(.init.text+0x410e): undefined reference to `sgivwfb_mem_size' sgivwfb.c:(.init.text+0x4113): undefined reference to `sgivwfb_mem_phys' sgivwfb.c:(.init.text+0x4162): undefined reference to `sgivwfb_mem_size' sgivwfb.c:(.init.text+0x4168): undefined reference to `sgivwfb_mem_phys' make: *** [.tmp_vmlinux1] Error 1 Signed-off-by: Adrian Bunk Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/i386/mach-visws/setup.c | 4 ++-- drivers/video/sgivwfb.c | 6 ++---- include/asm-i386/mach-visws/setup_arch.h | 3 +++ 3 files changed, 7 insertions(+), 6 deletions(-) (limited to 'include') diff --git a/arch/i386/mach-visws/setup.c b/arch/i386/mach-visws/setup.c index 8a9e1a6f745..11e83a0e1d6 100644 --- a/arch/i386/mach-visws/setup.c +++ b/arch/i386/mach-visws/setup.c @@ -140,8 +140,8 @@ void __init time_init_hook(void) #define MB (1024 * 1024) -static unsigned long sgivwfb_mem_phys; -static unsigned long sgivwfb_mem_size; +unsigned long sgivwfb_mem_phys; +unsigned long sgivwfb_mem_size; long long mem_size __initdata = 0; diff --git a/drivers/video/sgivwfb.c b/drivers/video/sgivwfb.c index 2e6df1fcb2b..c0cc5e3ba7b 100644 --- a/drivers/video/sgivwfb.c +++ b/drivers/video/sgivwfb.c @@ -23,6 +23,8 @@ #include #include +#include + #define INCLUDE_TIMING_TABLE_DATA #define DBE_REG_BASE par->regs #include