From 6c29256c5703ad7c3cf62276dbc38d802a05669e Mon Sep 17 00:00:00 2001 From: Jeff Dike Date: Mon, 27 Mar 2006 01:14:37 -0800 Subject: [PATCH] uml: allow ubd devices to be shared in a cluster This adds a 'c' option to the ubd switch which turns off host file locking so that the device can be shared, as with a cluster. There's also some whitespace cleanup while I was in this file. Signed-off-by: Jeff Dike Cc: Paolo 'Blaisorblade' Giarrusso Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/um/drivers/ubd_kern.c | 76 ++++++++++++++++++++++++++-------------------- 1 file changed, 43 insertions(+), 33 deletions(-) (limited to 'arch/um/drivers') diff --git a/arch/um/drivers/ubd_kern.c b/arch/um/drivers/ubd_kern.c index fa617e0719a..0336575d244 100644 --- a/arch/um/drivers/ubd_kern.c +++ b/arch/um/drivers/ubd_kern.c @@ -1,4 +1,4 @@ -/* +/* * Copyright (C) 2000 Jeff Dike (jdike@karaya.com) * Licensed under the GPL */ @@ -71,7 +71,7 @@ struct io_thread_req { int error; }; -extern int open_ubd_file(char *file, struct openflags *openflags, +extern int open_ubd_file(char *file, struct openflags *openflags, int shared, char **backing_file_out, int *bitmap_offset_out, unsigned long *bitmap_len_out, int *data_offset_out, int *create_cow_out); @@ -137,7 +137,7 @@ static int fake_major = MAJOR_NR; static struct gendisk *ubd_gendisk[MAX_DEV]; static struct gendisk *fake_gendisk[MAX_DEV]; - + #ifdef CONFIG_BLK_DEV_UBD_SYNC #define OPEN_FLAGS ((struct openflags) { .r = 1, .w = 1, .s = 1, .c = 0, \ .cl = 1 }) @@ -168,6 +168,7 @@ struct ubd { __u64 size; struct openflags boot_openflags; struct openflags openflags; + int shared; int no_cow; struct cow cow; struct platform_device pdev; @@ -189,6 +190,7 @@ struct ubd { .boot_openflags = OPEN_FLAGS, \ .openflags = OPEN_FLAGS, \ .no_cow = 0, \ + .shared = 0, \ .cow = DEFAULT_COW, \ } @@ -305,7 +307,7 @@ static int ubd_setup_common(char *str, int *index_out) } major = simple_strtoul(str, &end, 0); if((*end != '\0') || (end == str)){ - printk(KERN_ERR + printk(KERN_ERR "ubd_setup : didn't parse major number\n"); return(1); } @@ -316,7 +318,7 @@ static int ubd_setup_common(char *str, int *index_out) printk(KERN_ERR "Can't assign a fake major twice\n"); goto out1; } - + fake_major = major; printk(KERN_INFO "Setting extra ubd major number to %d\n", @@ -351,7 +353,7 @@ static int ubd_setup_common(char *str, int *index_out) if (index_out) *index_out = n; - for (i = 0; i < 4; i++) { + for (i = 0; i < sizeof("rscd="); i++) { switch (*str) { case 'r': flags.w = 0; @@ -362,11 +364,14 @@ static int ubd_setup_common(char *str, int *index_out) case 'd': dev->no_cow = 1; break; + case 'c': + dev->shared = 1; + break; case '=': str++; goto break_loop; default: - printk(KERN_ERR "ubd_setup : Expected '=' or flag letter (r,s or d)\n"); + printk(KERN_ERR "ubd_setup : Expected '=' or flag letter (r, s, c, or d)\n"); goto out; } str++; @@ -515,7 +520,7 @@ static void ubd_handler(void) spin_unlock(&ubd_io_lock); return; } - + ubd_finish(rq, req.error); reactivate_fd(thread_fd, UBD_IRQ); do_ubd_request(ubd_queue); @@ -532,7 +537,7 @@ static int io_pid = -1; void kill_io_thread(void) { - if(io_pid != -1) + if(io_pid != -1) os_kill_process(io_pid, 1); } @@ -567,14 +572,15 @@ static int ubd_open_dev(struct ubd *dev) create_cow = 0; create_ptr = (dev->cow.file != NULL) ? &create_cow : NULL; back_ptr = dev->no_cow ? NULL : &dev->cow.file; - dev->fd = open_ubd_file(dev->file, &dev->openflags, back_ptr, - &dev->cow.bitmap_offset, &dev->cow.bitmap_len, - &dev->cow.data_offset, create_ptr); + dev->fd = open_ubd_file(dev->file, &dev->openflags, dev->shared, + back_ptr, &dev->cow.bitmap_offset, + &dev->cow.bitmap_len, &dev->cow.data_offset, + create_ptr); if((dev->fd == -ENOENT) && create_cow){ - dev->fd = create_cow_file(dev->file, dev->cow.file, + dev->fd = create_cow_file(dev->file, dev->cow.file, dev->openflags, 1 << 9, PAGE_SIZE, - &dev->cow.bitmap_offset, + &dev->cow.bitmap_offset, &dev->cow.bitmap_len, &dev->cow.data_offset); if(dev->fd >= 0){ @@ -598,16 +604,16 @@ static int ubd_open_dev(struct ubd *dev) } flush_tlb_kernel_vm(); - err = read_cow_bitmap(dev->fd, dev->cow.bitmap, - dev->cow.bitmap_offset, + err = read_cow_bitmap(dev->fd, dev->cow.bitmap, + dev->cow.bitmap_offset, dev->cow.bitmap_len); if(err < 0) goto error; flags = dev->openflags; flags.w = 0; - err = open_ubd_file(dev->cow.file, &flags, NULL, NULL, NULL, - NULL, NULL); + err = open_ubd_file(dev->cow.file, &flags, dev->shared, NULL, + NULL, NULL, NULL, NULL); if(err < 0) goto error; dev->cow.fd = err; } @@ -685,11 +691,11 @@ static int ubd_add(int n) dev->size = ROUND_BLOCK(dev->size); err = ubd_new_disk(MAJOR_NR, dev->size, n, &ubd_gendisk[n]); - if(err) + if(err) goto out_close; - + if(fake_major != MAJOR_NR) - ubd_new_disk(fake_major, dev->size, n, + ubd_new_disk(fake_major, dev->size, n, &fake_gendisk[n]); /* perhaps this should also be under the "if (fake_major)" above */ @@ -854,7 +860,7 @@ int ubd_init(void) return -1; } platform_driver_register(&ubd_driver); - for (i = 0; i < MAX_DEV; i++) + for (i = 0; i < MAX_DEV; i++) ubd_add(i); return 0; } @@ -872,16 +878,16 @@ int ubd_driver_init(void){ * enough. So use anyway the io thread. */ } stack = alloc_stack(0, 0); - io_pid = start_io_thread(stack + PAGE_SIZE - sizeof(void *), + io_pid = start_io_thread(stack + PAGE_SIZE - sizeof(void *), &thread_fd); if(io_pid < 0){ - printk(KERN_ERR + printk(KERN_ERR "ubd : Failed to start I/O thread (errno = %d) - " "falling back to synchronous I/O\n", -io_pid); io_pid = -1; return(0); } - err = um_request_irq(UBD_IRQ, thread_fd, IRQ_READ, ubd_intr, + err = um_request_irq(UBD_IRQ, thread_fd, IRQ_READ, ubd_intr, SA_INTERRUPT, "ubd", ubd_dev); if(err != 0) printk(KERN_ERR "um_request_irq failed - errno = %d\n", -err); @@ -978,7 +984,7 @@ static void cowify_req(struct io_thread_req *req, unsigned long *bitmap, if(req->op == UBD_READ) { for(i = 0; i < req->length >> 9; i++){ if(ubd_test_bit(sector + i, (unsigned char *) bitmap)) - ubd_set_bit(i, (unsigned char *) + ubd_set_bit(i, (unsigned char *) &req->sector_mask); } } @@ -999,7 +1005,7 @@ static int prepare_request(struct request *req, struct io_thread_req *io_req) /* This should be impossible now */ if((rq_data_dir(req) == WRITE) && !dev->openflags.w){ - printk("Write attempted on readonly ubd device %s\n", + printk("Write attempted on readonly ubd device %s\n", disk->disk_name); end_request(req, 0); return(1); @@ -1182,7 +1188,7 @@ int read_cow_bitmap(int fd, void *buf, int offset, int len) return(0); } -int open_ubd_file(char *file, struct openflags *openflags, +int open_ubd_file(char *file, struct openflags *openflags, int shared, char **backing_file_out, int *bitmap_offset_out, unsigned long *bitmap_len_out, int *data_offset_out, int *create_cow_out) @@ -1206,10 +1212,14 @@ int open_ubd_file(char *file, struct openflags *openflags, return fd; } - err = os_lock_file(fd, openflags->w); - if(err < 0){ - printk("Failed to lock '%s', err = %d\n", file, -err); - goto out_close; + if(shared) + printk("Not locking \"%s\" on the host\n", file); + else { + err = os_lock_file(fd, openflags->w); + if(err < 0){ + printk("Failed to lock '%s', err = %d\n", file, -err); + goto out_close; + } } /* Succesful return case! */ @@ -1260,7 +1270,7 @@ int create_cow_file(char *cow_file, char *backing_file, struct openflags flags, int err, fd; flags.c = 1; - fd = open_ubd_file(cow_file, &flags, NULL, NULL, NULL, NULL, NULL); + fd = open_ubd_file(cow_file, &flags, 0, NULL, NULL, NULL, NULL, NULL); if(fd < 0){ err = fd; printk("Open of COW file '%s' failed, errno = %d\n", cow_file, -- cgit v1.2.3-70-g09d2 From e041c683412d5bf44dc2b109053e3b837b71742d Mon Sep 17 00:00:00 2001 From: Alan Stern Date: Mon, 27 Mar 2006 01:16:30 -0800 Subject: [PATCH] Notifier chain update: API changes The kernel's implementation of notifier chains is unsafe. There is no protection against entries being added to or removed from a chain while the chain is in use. The issues were discussed in this thread: http://marc.theaimsgroup.com/?l=linux-kernel&m=113018709002036&w=2 We noticed that notifier chains in the kernel fall into two basic usage classes: "Blocking" chains are always called from a process context and the callout routines are allowed to sleep; "Atomic" chains can be called from an atomic context and the callout routines are not allowed to sleep. We decided to codify this distinction and make it part of the API. Therefore this set of patches introduces three new, parallel APIs: one for blocking notifiers, one for atomic notifiers, and one for "raw" notifiers (which is really just the old API under a new name). New kinds of data structures are used for the heads of the chains, and new routines are defined for registration, unregistration, and calling a chain. The three APIs are explained in include/linux/notifier.h and their implementation is in kernel/sys.c. With atomic and blocking chains, the implementation guarantees that the chain links will not be corrupted and that chain callers will not get messed up by entries being added or removed. For raw chains the implementation provides no guarantees at all; users of this API must provide their own protections. (The idea was that situations may come up where the assumptions of the atomic and blocking APIs are not appropriate, so it should be possible for users to handle these things in their own way.) There are some limitations, which should not be too hard to live with. For atomic/blocking chains, registration and unregistration must always be done in a process context since the chain is protected by a mutex/rwsem. Also, a callout routine for a non-raw chain must not try to register or unregister entries on its own chain. (This did happen in a couple of places and the code had to be changed to avoid it.) Since atomic chains may be called from within an NMI handler, they cannot use spinlocks for synchronization. Instead we use RCU. The overhead falls almost entirely in the unregister routine, which is okay since unregistration is much less frequent that calling a chain. Here is the list of chains that we adjusted and their classifications. None of them use the raw API, so for the moment it is only a placeholder. ATOMIC CHAINS ------------- arch/i386/kernel/traps.c: i386die_chain arch/ia64/kernel/traps.c: ia64die_chain arch/powerpc/kernel/traps.c: powerpc_die_chain arch/sparc64/kernel/traps.c: sparc64die_chain arch/x86_64/kernel/traps.c: die_chain drivers/char/ipmi/ipmi_si_intf.c: xaction_notifier_list kernel/panic.c: panic_notifier_list kernel/profile.c: task_free_notifier net/bluetooth/hci_core.c: hci_notifier net/ipv4/netfilter/ip_conntrack_core.c: ip_conntrack_chain net/ipv4/netfilter/ip_conntrack_core.c: ip_conntrack_expect_chain net/ipv6/addrconf.c: inet6addr_chain net/netfilter/nf_conntrack_core.c: nf_conntrack_chain net/netfilter/nf_conntrack_core.c: nf_conntrack_expect_chain net/netlink/af_netlink.c: netlink_chain BLOCKING CHAINS --------------- arch/powerpc/platforms/pseries/reconfig.c: pSeries_reconfig_chain arch/s390/kernel/process.c: idle_chain arch/x86_64/kernel/process.c idle_notifier drivers/base/memory.c: memory_chain drivers/cpufreq/cpufreq.c cpufreq_policy_notifier_list drivers/cpufreq/cpufreq.c cpufreq_transition_notifier_list drivers/macintosh/adb.c: adb_client_list drivers/macintosh/via-pmu.c sleep_notifier_list drivers/macintosh/via-pmu68k.c sleep_notifier_list drivers/macintosh/windfarm_core.c wf_client_list drivers/usb/core/notify.c usb_notifier_list drivers/video/fbmem.c fb_notifier_list kernel/cpu.c cpu_chain kernel/module.c module_notify_list kernel/profile.c munmap_notifier kernel/profile.c task_exit_notifier kernel/sys.c reboot_notifier_list net/core/dev.c netdev_chain net/decnet/dn_dev.c: dnaddr_chain net/ipv4/devinet.c: inetaddr_chain It's possible that some of these classifications are wrong. If they are, please let us know or submit a patch to fix them. Note that any chain that gets called very frequently should be atomic, because the rwsem read-locking used for blocking chains is very likely to incur cache misses on SMP systems. (However, if the chain's callout routines may sleep then the chain cannot be atomic.) The patch set was written by Alan Stern and Chandra Seetharaman, incorporating material written by Keith Owens and suggestions from Paul McKenney and Andrew Morton. [jes@sgi.com: restructure the notifier chain initialization macros] Signed-off-by: Alan Stern Signed-off-by: Chandra Seetharaman Signed-off-by: Jes Sorensen Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/alpha/kernel/setup.c | 5 +- arch/arm/mach-omap1/board-netstar.c | 2 +- arch/arm/mach-omap1/board-voiceblue.c | 2 +- arch/i386/kernel/traps.c | 17 +- arch/ia64/kernel/traps.c | 6 +- arch/mips/lasat/setup.c | 3 +- arch/mips/sgi-ip22/ip22-reset.c | 2 +- arch/mips/sgi-ip32/ip32-reset.c | 2 +- arch/parisc/kernel/pdc_chassis.c | 3 +- arch/powerpc/kernel/setup_64.c | 3 +- arch/powerpc/kernel/traps.c | 16 +- arch/powerpc/platforms/pseries/reconfig.c | 10 +- arch/ppc/platforms/prep_setup.c | 2 +- arch/s390/kernel/process.c | 11 +- arch/sparc64/kernel/traps.c | 17 +- arch/um/drivers/mconsole_kern.c | 3 +- arch/um/kernel/um_arch.c | 3 +- arch/x86_64/kernel/process.c | 17 +- arch/x86_64/kernel/traps.c | 18 +- arch/xtensa/platform-iss/setup.c | 2 +- drivers/base/memory.c | 8 +- drivers/char/ipmi/ipmi_msghandler.c | 4 +- drivers/char/ipmi/ipmi_si_intf.c | 7 +- drivers/char/ipmi/ipmi_watchdog.c | 6 +- drivers/cpufreq/cpufreq.c | 61 +++--- drivers/firmware/dcdbas.c | 19 +- drivers/macintosh/adb.c | 11 +- drivers/macintosh/adbhid.c | 3 +- drivers/macintosh/via-pmu.c | 2 +- drivers/macintosh/via-pmu68k.c | 7 +- drivers/macintosh/windfarm_core.c | 8 +- drivers/misc/ibmasm/heartbeat.c | 5 +- drivers/net/bonding/bond_main.c | 2 +- drivers/parisc/led.c | 14 +- drivers/parisc/power.c | 6 +- drivers/scsi/gdth.c | 9 +- drivers/usb/core/notify.c | 65 +----- drivers/video/fbmem.c | 31 +-- include/asm-i386/kdebug.h | 10 +- include/asm-ia64/kdebug.h | 4 +- include/asm-powerpc/kdebug.h | 12 +- include/asm-sparc64/kdebug.h | 11 +- include/asm-x86_64/kdebug.h | 23 +- include/linux/adb.h | 2 +- include/linux/kernel.h | 2 +- include/linux/memory.h | 1 - include/linux/netfilter_ipv4/ip_conntrack.h | 17 +- include/linux/notifier.h | 96 +++++++- include/net/netfilter/nf_conntrack.h | 17 +- kernel/cpu.c | 29 +-- kernel/module.c | 20 +- kernel/panic.c | 4 +- kernel/profile.c | 53 ++--- kernel/softlockup.c | 2 +- kernel/sys.c | 327 ++++++++++++++++++++++------ net/bluetooth/hci_core.c | 8 +- net/core/dev.c | 42 ++-- net/decnet/dn_dev.c | 10 +- net/ipv4/devinet.c | 16 +- net/ipv4/netfilter/ip_conntrack_core.c | 6 +- net/ipv6/addrconf.c | 10 +- net/netfilter/nf_conntrack_core.c | 6 +- net/netlink/af_netlink.c | 9 +- 63 files changed, 677 insertions(+), 472 deletions(-) (limited to 'arch/um/drivers') diff --git a/arch/alpha/kernel/setup.c b/arch/alpha/kernel/setup.c index 9402624453c..dd876967059 100644 --- a/arch/alpha/kernel/setup.c +++ b/arch/alpha/kernel/setup.c @@ -43,7 +43,7 @@ #include #include -extern struct notifier_block *panic_notifier_list; +extern struct atomic_notifier_head panic_notifier_list; static int alpha_panic_event(struct notifier_block *, unsigned long, void *); static struct notifier_block alpha_panic_block = { alpha_panic_event, @@ -500,7 +500,8 @@ setup_arch(char **cmdline_p) } /* Register a call for panic conditions. */ - notifier_chain_register(&panic_notifier_list, &alpha_panic_block); + atomic_notifier_chain_register(&panic_notifier_list, + &alpha_panic_block); #ifdef CONFIG_ALPHA_GENERIC /* Assume that we've booted from SRM if we haven't booted from MILO. diff --git a/arch/arm/mach-omap1/board-netstar.c b/arch/arm/mach-omap1/board-netstar.c index 60d5f8a3339..7520e602d7a 100644 --- a/arch/arm/mach-omap1/board-netstar.c +++ b/arch/arm/mach-omap1/board-netstar.c @@ -141,7 +141,7 @@ static int __init netstar_late_init(void) /* TODO: Setup front panel switch here */ /* Setup panic notifier */ - notifier_chain_register(&panic_notifier_list, &panic_block); + atomic_notifier_chain_register(&panic_notifier_list, &panic_block); return 0; } diff --git a/arch/arm/mach-omap1/board-voiceblue.c b/arch/arm/mach-omap1/board-voiceblue.c index bfd5fdd1a87..52e4a9d6964 100644 --- a/arch/arm/mach-omap1/board-voiceblue.c +++ b/arch/arm/mach-omap1/board-voiceblue.c @@ -235,7 +235,7 @@ static struct notifier_block panic_block = { static int __init voiceblue_setup(void) { /* Setup panic notifier */ - notifier_chain_register(&panic_notifier_list, &panic_block); + atomic_notifier_chain_register(&panic_notifier_list, &panic_block); return 0; } diff --git a/arch/i386/kernel/traps.c b/arch/i386/kernel/traps.c index 4624f8ca245..6b63a5aa1e4 100644 --- a/arch/i386/kernel/traps.c +++ b/arch/i386/kernel/traps.c @@ -92,22 +92,21 @@ asmlinkage void spurious_interrupt_bug(void); asmlinkage void machine_check(void); static int kstack_depth_to_print = 24; -struct notifier_block *i386die_chain; -static DEFINE_SPINLOCK(die_notifier_lock); +ATOMIC_NOTIFIER_HEAD(i386die_chain); int register_die_notifier(struct notifier_block *nb) { - int err = 0; - unsigned long flags; - vmalloc_sync_all(); - spin_lock_irqsave(&die_notifier_lock, flags); - err = notifier_chain_register(&i386die_chain, nb); - spin_unlock_irqrestore(&die_notifier_lock, flags); - return err; + return atomic_notifier_chain_register(&i386die_chain, nb); } EXPORT_SYMBOL(register_die_notifier); +int unregister_die_notifier(struct notifier_block *nb) +{ + return atomic_notifier_chain_unregister(&i386die_chain, nb); +} +EXPORT_SYMBOL(unregister_die_notifier); + static inline int valid_stack_ptr(struct thread_info *tinfo, void *p) { return p > (void *)tinfo && diff --git a/arch/ia64/kernel/traps.c b/arch/ia64/kernel/traps.c index dabd6c32641..7c1ddc8ac44 100644 --- a/arch/ia64/kernel/traps.c +++ b/arch/ia64/kernel/traps.c @@ -30,19 +30,19 @@ extern spinlock_t timerlist_lock; fpswa_interface_t *fpswa_interface; EXPORT_SYMBOL(fpswa_interface); -struct notifier_block *ia64die_chain; +ATOMIC_NOTIFIER_HEAD(ia64die_chain); int register_die_notifier(struct notifier_block *nb) { - return notifier_chain_register(&ia64die_chain, nb); + return atomic_notifier_chain_register(&ia64die_chain, nb); } EXPORT_SYMBOL_GPL(register_die_notifier); int unregister_die_notifier(struct notifier_block *nb) { - return notifier_chain_unregister(&ia64die_chain, nb); + return atomic_notifier_chain_unregister(&ia64die_chain, nb); } EXPORT_SYMBOL_GPL(unregister_die_notifier); diff --git a/arch/mips/lasat/setup.c b/arch/mips/lasat/setup.c index 83eb08b7a07..e9e9a89c674 100644 --- a/arch/mips/lasat/setup.c +++ b/arch/mips/lasat/setup.c @@ -165,7 +165,8 @@ void __init plat_setup(void) /* Set up panic notifier */ for (i = 0; i < sizeof(lasat_panic_block) / sizeof(struct notifier_block); i++) - notifier_chain_register(&panic_notifier_list, &lasat_panic_block[i]); + atomic_notifier_chain_register(&panic_notifier_list, + &lasat_panic_block[i]); lasat_reboot_setup(); diff --git a/arch/mips/sgi-ip22/ip22-reset.c b/arch/mips/sgi-ip22/ip22-reset.c index 92a3b3c15ed..a9c58e067b5 100644 --- a/arch/mips/sgi-ip22/ip22-reset.c +++ b/arch/mips/sgi-ip22/ip22-reset.c @@ -238,7 +238,7 @@ static int __init reboot_setup(void) request_irq(SGI_PANEL_IRQ, panel_int, 0, "Front Panel", NULL); init_timer(&blink_timer); blink_timer.function = blink_timeout; - notifier_chain_register(&panic_notifier_list, &panic_block); + atomic_notifier_chain_register(&panic_notifier_list, &panic_block); return 0; } diff --git a/arch/mips/sgi-ip32/ip32-reset.c b/arch/mips/sgi-ip32/ip32-reset.c index 0c948008b02..ab9d9cef089 100644 --- a/arch/mips/sgi-ip32/ip32-reset.c +++ b/arch/mips/sgi-ip32/ip32-reset.c @@ -193,7 +193,7 @@ static __init int ip32_reboot_setup(void) init_timer(&blink_timer); blink_timer.function = blink_timeout; - notifier_chain_register(&panic_notifier_list, &panic_block); + atomic_notifier_chain_register(&panic_notifier_list, &panic_block); request_irq(MACEISA_RTC_IRQ, ip32_rtc_int, 0, "rtc", NULL); diff --git a/arch/parisc/kernel/pdc_chassis.c b/arch/parisc/kernel/pdc_chassis.c index 2a01fe1bdc9..0cea6958f42 100644 --- a/arch/parisc/kernel/pdc_chassis.c +++ b/arch/parisc/kernel/pdc_chassis.c @@ -150,7 +150,8 @@ void __init parisc_pdc_chassis_init(void) if (handle) { /* initialize panic notifier chain */ - notifier_chain_register(&panic_notifier_list, &pdc_chassis_panic_block); + atomic_notifier_chain_register(&panic_notifier_list, + &pdc_chassis_panic_block); /* initialize reboot notifier chain */ register_reboot_notifier(&pdc_chassis_reboot_block); diff --git a/arch/powerpc/kernel/setup_64.c b/arch/powerpc/kernel/setup_64.c index 2f3fdad3559..e20c1fae342 100644 --- a/arch/powerpc/kernel/setup_64.c +++ b/arch/powerpc/kernel/setup_64.c @@ -579,7 +579,8 @@ void __init setup_arch(char **cmdline_p) panic_timeout = 180; if (ppc_md.panic) - notifier_chain_register(&panic_notifier_list, &ppc64_panic_block); + atomic_notifier_chain_register(&panic_notifier_list, + &ppc64_panic_block); init_mm.start_code = PAGE_OFFSET; init_mm.end_code = (unsigned long) _etext; diff --git a/arch/powerpc/kernel/traps.c b/arch/powerpc/kernel/traps.c index 98660aedeeb..9763faab673 100644 --- a/arch/powerpc/kernel/traps.c +++ b/arch/powerpc/kernel/traps.c @@ -74,19 +74,19 @@ EXPORT_SYMBOL(__debugger_dabr_match); EXPORT_SYMBOL(__debugger_fault_handler); #endif -struct notifier_block *powerpc_die_chain; -static DEFINE_SPINLOCK(die_notifier_lock); +ATOMIC_NOTIFIER_HEAD(powerpc_die_chain); int register_die_notifier(struct notifier_block *nb) { - int err = 0; - unsigned long flags; + return atomic_notifier_chain_register(&powerpc_die_chain, nb); +} +EXPORT_SYMBOL(register_die_notifier); - spin_lock_irqsave(&die_notifier_lock, flags); - err = notifier_chain_register(&powerpc_die_chain, nb); - spin_unlock_irqrestore(&die_notifier_lock, flags); - return err; +int unregister_die_notifier(struct notifier_block *nb) +{ + return atomic_notifier_chain_unregister(&powerpc_die_chain, nb); } +EXPORT_SYMBOL(unregister_die_notifier); /* * Trap & Exception support diff --git a/arch/powerpc/platforms/pseries/reconfig.c b/arch/powerpc/platforms/pseries/reconfig.c index 86cfa6ecdcf..5ad90676567 100644 --- a/arch/powerpc/platforms/pseries/reconfig.c +++ b/arch/powerpc/platforms/pseries/reconfig.c @@ -94,16 +94,16 @@ static struct device_node *derive_parent(const char *path) return parent; } -static struct notifier_block *pSeries_reconfig_chain; +static BLOCKING_NOTIFIER_HEAD(pSeries_reconfig_chain); int pSeries_reconfig_notifier_register(struct notifier_block *nb) { - return notifier_chain_register(&pSeries_reconfig_chain, nb); + return blocking_notifier_chain_register(&pSeries_reconfig_chain, nb); } void pSeries_reconfig_notifier_unregister(struct notifier_block *nb) { - notifier_chain_unregister(&pSeries_reconfig_chain, nb); + blocking_notifier_chain_unregister(&pSeries_reconfig_chain, nb); } static int pSeries_reconfig_add_node(const char *path, struct property *proplist) @@ -131,7 +131,7 @@ static int pSeries_reconfig_add_node(const char *path, struct property *proplist goto out_err; } - err = notifier_call_chain(&pSeries_reconfig_chain, + err = blocking_notifier_call_chain(&pSeries_reconfig_chain, PSERIES_RECONFIG_ADD, np); if (err == NOTIFY_BAD) { printk(KERN_ERR "Failed to add device node %s\n", path); @@ -171,7 +171,7 @@ static int pSeries_reconfig_remove_node(struct device_node *np) remove_node_proc_entries(np); - notifier_call_chain(&pSeries_reconfig_chain, + blocking_notifier_call_chain(&pSeries_reconfig_chain, PSERIES_RECONFIG_REMOVE, np); of_detach_node(np); diff --git a/arch/ppc/platforms/prep_setup.c b/arch/ppc/platforms/prep_setup.c index a0fc628ffb1..d95c05d9824 100644 --- a/arch/ppc/platforms/prep_setup.c +++ b/arch/ppc/platforms/prep_setup.c @@ -736,7 +736,7 @@ ibm_statusled_progress(char *s, unsigned short hex) hex = 0xfff; if (!notifier_installed) { ++notifier_installed; - notifier_chain_register(&panic_notifier_list, + atomic_notifier_chain_register(&panic_notifier_list, &ibm_statusled_block); } } diff --git a/arch/s390/kernel/process.c b/arch/s390/kernel/process.c index 99182a415fe..4a0f5a1551e 100644 --- a/arch/s390/kernel/process.c +++ b/arch/s390/kernel/process.c @@ -76,17 +76,17 @@ unsigned long thread_saved_pc(struct task_struct *tsk) /* * Need to know about CPUs going idle? */ -static struct notifier_block *idle_chain; +static ATOMIC_NOTIFIER_HEAD(idle_chain); int register_idle_notifier(struct notifier_block *nb) { - return notifier_chain_register(&idle_chain, nb); + return atomic_notifier_chain_register(&idle_chain, nb); } EXPORT_SYMBOL(register_idle_notifier); int unregister_idle_notifier(struct notifier_block *nb) { - return notifier_chain_unregister(&idle_chain, nb); + return atomic_notifier_chain_unregister(&idle_chain, nb); } EXPORT_SYMBOL(unregister_idle_notifier); @@ -95,7 +95,7 @@ void do_monitor_call(struct pt_regs *regs, long interruption_code) /* disable monitor call class 0 */ __ctl_clear_bit(8, 15); - notifier_call_chain(&idle_chain, CPU_NOT_IDLE, + atomic_notifier_call_chain(&idle_chain, CPU_NOT_IDLE, (void *)(long) smp_processor_id()); } @@ -116,7 +116,8 @@ static void default_idle(void) return; } - rc = notifier_call_chain(&idle_chain, CPU_IDLE, (void *)(long) cpu); + rc = atomic_notifier_call_chain(&idle_chain, + CPU_IDLE, (void *)(long) cpu); if (rc != NOTIFY_OK && rc != NOTIFY_DONE) BUG(); if (rc != NOTIFY_OK) { diff --git a/arch/sparc64/kernel/traps.c b/arch/sparc64/kernel/traps.c index df612e4f75f..ff090bb9734 100644 --- a/arch/sparc64/kernel/traps.c +++ b/arch/sparc64/kernel/traps.c @@ -43,18 +43,19 @@ #include #endif -struct notifier_block *sparc64die_chain; -static DEFINE_SPINLOCK(die_notifier_lock); +ATOMIC_NOTIFIER_HEAD(sparc64die_chain); int register_die_notifier(struct notifier_block *nb) { - int err = 0; - unsigned long flags; - spin_lock_irqsave(&die_notifier_lock, flags); - err = notifier_chain_register(&sparc64die_chain, nb); - spin_unlock_irqrestore(&die_notifier_lock, flags); - return err; + return atomic_notifier_chain_register(&sparc64die_chain, nb); } +EXPORT_SYMBOL(register_die_notifier); + +int unregister_die_notifier(struct notifier_block *nb) +{ + return atomic_notifier_chain_unregister(&sparc64die_chain, nb); +} +EXPORT_SYMBOL(unregister_die_notifier); /* When an irrecoverable trap occurs at tl > 0, the trap entry * code logs the trap state registers at every level in the trap diff --git a/arch/um/drivers/mconsole_kern.c b/arch/um/drivers/mconsole_kern.c index 54388d10bcf..1488816588e 100644 --- a/arch/um/drivers/mconsole_kern.c +++ b/arch/um/drivers/mconsole_kern.c @@ -762,7 +762,8 @@ static struct notifier_block panic_exit_notifier = { static int add_notifier(void) { - notifier_chain_register(&panic_notifier_list, &panic_exit_notifier); + atomic_notifier_chain_register(&panic_notifier_list, + &panic_exit_notifier); return(0); } diff --git a/arch/um/kernel/um_arch.c b/arch/um/kernel/um_arch.c index bb1c87211ac..7d51dd7201c 100644 --- a/arch/um/kernel/um_arch.c +++ b/arch/um/kernel/um_arch.c @@ -477,7 +477,8 @@ static struct notifier_block panic_exit_notifier = { void __init setup_arch(char **cmdline_p) { - notifier_chain_register(&panic_notifier_list, &panic_exit_notifier); + atomic_notifier_chain_register(&panic_notifier_list, + &panic_exit_notifier); paging_init(); strlcpy(saved_command_line, command_line, COMMAND_LINE_SIZE); *cmdline_p = command_line; diff --git a/arch/x86_64/kernel/process.c b/arch/x86_64/kernel/process.c index 0370720515f..70dd8e5c688 100644 --- a/arch/x86_64/kernel/process.c +++ b/arch/x86_64/kernel/process.c @@ -66,24 +66,17 @@ EXPORT_SYMBOL(boot_option_idle_override); void (*pm_idle)(void); static DEFINE_PER_CPU(unsigned int, cpu_idle_state); -static struct notifier_block *idle_notifier; -static DEFINE_SPINLOCK(idle_notifier_lock); +static ATOMIC_NOTIFIER_HEAD(idle_notifier); void idle_notifier_register(struct notifier_block *n) { - unsigned long flags; - spin_lock_irqsave(&idle_notifier_lock, flags); - notifier_chain_register(&idle_notifier, n); - spin_unlock_irqrestore(&idle_notifier_lock, flags); + atomic_notifier_chain_register(&idle_notifier, n); } EXPORT_SYMBOL_GPL(idle_notifier_register); void idle_notifier_unregister(struct notifier_block *n) { - unsigned long flags; - spin_lock_irqsave(&idle_notifier_lock, flags); - notifier_chain_unregister(&idle_notifier, n); - spin_unlock_irqrestore(&idle_notifier_lock, flags); + atomic_notifier_chain_unregister(&idle_notifier, n); } EXPORT_SYMBOL(idle_notifier_unregister); @@ -93,13 +86,13 @@ static DEFINE_PER_CPU(enum idle_state, idle_state) = CPU_NOT_IDLE; void enter_idle(void) { __get_cpu_var(idle_state) = CPU_IDLE; - notifier_call_chain(&idle_notifier, IDLE_START, NULL); + atomic_notifier_call_chain(&idle_notifier, IDLE_START, NULL); } static void __exit_idle(void) { __get_cpu_var(idle_state) = CPU_NOT_IDLE; - notifier_call_chain(&idle_notifier, IDLE_END, NULL); + atomic_notifier_call_chain(&idle_notifier, IDLE_END, NULL); } /* Called from interrupts to signify idle end */ diff --git a/arch/x86_64/kernel/traps.c b/arch/x86_64/kernel/traps.c index 7b148309c52..edaa9fe654d 100644 --- a/arch/x86_64/kernel/traps.c +++ b/arch/x86_64/kernel/traps.c @@ -69,20 +69,20 @@ asmlinkage void alignment_check(void); asmlinkage void machine_check(void); asmlinkage void spurious_interrupt_bug(void); -struct notifier_block *die_chain; -static DEFINE_SPINLOCK(die_notifier_lock); +ATOMIC_NOTIFIER_HEAD(die_chain); int register_die_notifier(struct notifier_block *nb) { - int err = 0; - unsigned long flags; - vmalloc_sync_all(); - spin_lock_irqsave(&die_notifier_lock, flags); - err = notifier_chain_register(&die_chain, nb); - spin_unlock_irqrestore(&die_notifier_lock, flags); - return err; + return atomic_notifier_chain_register(&die_chain, nb); +} +EXPORT_SYMBOL(register_die_notifier); + +int unregister_die_notifier(struct notifier_block *nb) +{ + return atomic_notifier_chain_unregister(&die_chain, nb); } +EXPORT_SYMBOL(unregister_die_notifier); static inline void conditional_sti(struct pt_regs *regs) { diff --git a/arch/xtensa/platform-iss/setup.c b/arch/xtensa/platform-iss/setup.c index 2e6dcbf0cc0..23790a5610e 100644 --- a/arch/xtensa/platform-iss/setup.c +++ b/arch/xtensa/platform-iss/setup.c @@ -108,5 +108,5 @@ static struct notifier_block iss_panic_block = { void __init platform_setup(char **p_cmdline) { - notifier_chain_register(&panic_notifier_list, &iss_panic_block); + atomic_notifier_chain_register(&panic_notifier_list, &iss_panic_block); } diff --git a/drivers/base/memory.c b/drivers/base/memory.c index 105a0d61eb1..dd547af4681 100644 --- a/drivers/base/memory.c +++ b/drivers/base/memory.c @@ -47,16 +47,16 @@ static struct kset_uevent_ops memory_uevent_ops = { .uevent = memory_uevent, }; -static struct notifier_block *memory_chain; +static BLOCKING_NOTIFIER_HEAD(memory_chain); int register_memory_notifier(struct notifier_block *nb) { - return notifier_chain_register(&memory_chain, nb); + return blocking_notifier_chain_register(&memory_chain, nb); } void unregister_memory_notifier(struct notifier_block *nb) { - notifier_chain_unregister(&memory_chain, nb); + blocking_notifier_chain_unregister(&memory_chain, nb); } /* @@ -140,7 +140,7 @@ static ssize_t show_mem_state(struct sys_device *dev, char *buf) static inline int memory_notify(unsigned long val, void *v) { - return notifier_call_chain(&memory_chain, val, v); + return blocking_notifier_call_chain(&memory_chain, val, v); } /* diff --git a/drivers/char/ipmi/ipmi_msghandler.c b/drivers/char/ipmi/ipmi_msghandler.c index b8fb87c6c29..40eb005b9d7 100644 --- a/drivers/char/ipmi/ipmi_msghandler.c +++ b/drivers/char/ipmi/ipmi_msghandler.c @@ -3744,7 +3744,7 @@ static int ipmi_init_msghandler(void) ipmi_timer.expires = jiffies + IPMI_TIMEOUT_JIFFIES; add_timer(&ipmi_timer); - notifier_chain_register(&panic_notifier_list, &panic_block); + atomic_notifier_chain_register(&panic_notifier_list, &panic_block); initialized = 1; @@ -3764,7 +3764,7 @@ static __exit void cleanup_ipmi(void) if (!initialized) return; - notifier_chain_unregister(&panic_notifier_list, &panic_block); + atomic_notifier_chain_unregister(&panic_notifier_list, &panic_block); /* This can't be called if any interfaces exist, so no worry about shutting down the interfaces. */ diff --git a/drivers/char/ipmi/ipmi_si_intf.c b/drivers/char/ipmi/ipmi_si_intf.c index 12f858dc999..35fbd4d8ed4 100644 --- a/drivers/char/ipmi/ipmi_si_intf.c +++ b/drivers/char/ipmi/ipmi_si_intf.c @@ -237,10 +237,10 @@ struct smi_info static int try_smi_init(struct smi_info *smi); -static struct notifier_block *xaction_notifier_list; +static ATOMIC_NOTIFIER_HEAD(xaction_notifier_list); static int register_xaction_notifier(struct notifier_block * nb) { - return notifier_chain_register(&xaction_notifier_list, nb); + return atomic_notifier_chain_register(&xaction_notifier_list, nb); } static void si_restart_short_timer(struct smi_info *smi_info); @@ -302,7 +302,8 @@ static enum si_sm_result start_next_msg(struct smi_info *smi_info) do_gettimeofday(&t); printk("**Start2: %d.%9.9d\n", t.tv_sec, t.tv_usec); #endif - err = notifier_call_chain(&xaction_notifier_list, 0, smi_info); + err = atomic_notifier_call_chain(&xaction_notifier_list, + 0, smi_info); if (err & NOTIFY_STOP_MASK) { rv = SI_SM_CALL_WITHOUT_DELAY; goto out; diff --git a/drivers/char/ipmi/ipmi_watchdog.c b/drivers/char/ipmi/ipmi_watchdog.c index 616539310d9..7ece9f3c8f7 100644 --- a/drivers/char/ipmi/ipmi_watchdog.c +++ b/drivers/char/ipmi/ipmi_watchdog.c @@ -1158,7 +1158,8 @@ static int __init ipmi_wdog_init(void) } register_reboot_notifier(&wdog_reboot_notifier); - notifier_chain_register(&panic_notifier_list, &wdog_panic_notifier); + atomic_notifier_chain_register(&panic_notifier_list, + &wdog_panic_notifier); printk(KERN_INFO PFX "driver initialized\n"); @@ -1176,7 +1177,8 @@ static __exit void ipmi_unregister_watchdog(void) release_nmi(&ipmi_nmi_handler); #endif - notifier_chain_unregister(&panic_notifier_list, &wdog_panic_notifier); + atomic_notifier_chain_unregister(&panic_notifier_list, + &wdog_panic_notifier); unregister_reboot_notifier(&wdog_reboot_notifier); if (! watchdog_user) diff --git a/drivers/cpufreq/cpufreq.c b/drivers/cpufreq/cpufreq.c index aed80e6aec6..9b6ae7dc8b8 100644 --- a/drivers/cpufreq/cpufreq.c +++ b/drivers/cpufreq/cpufreq.c @@ -52,9 +52,8 @@ static void handle_update(void *data); * changes to devices when the CPU clock speed changes. * The mutex locks both lists. */ -static struct notifier_block *cpufreq_policy_notifier_list; -static struct notifier_block *cpufreq_transition_notifier_list; -static DECLARE_RWSEM (cpufreq_notifier_rwsem); +static BLOCKING_NOTIFIER_HEAD(cpufreq_policy_notifier_list); +static BLOCKING_NOTIFIER_HEAD(cpufreq_transition_notifier_list); static LIST_HEAD(cpufreq_governor_list); @@ -247,8 +246,6 @@ void cpufreq_notify_transition(struct cpufreq_freqs *freqs, unsigned int state) dprintk("notification %u of frequency transition to %u kHz\n", state, freqs->new); - down_read(&cpufreq_notifier_rwsem); - policy = cpufreq_cpu_data[freqs->cpu]; switch (state) { @@ -266,20 +263,19 @@ void cpufreq_notify_transition(struct cpufreq_freqs *freqs, unsigned int state) freqs->old = policy->cur; } } - notifier_call_chain(&cpufreq_transition_notifier_list, - CPUFREQ_PRECHANGE, freqs); + blocking_notifier_call_chain(&cpufreq_transition_notifier_list, + CPUFREQ_PRECHANGE, freqs); adjust_jiffies(CPUFREQ_PRECHANGE, freqs); break; case CPUFREQ_POSTCHANGE: adjust_jiffies(CPUFREQ_POSTCHANGE, freqs); - notifier_call_chain(&cpufreq_transition_notifier_list, - CPUFREQ_POSTCHANGE, freqs); + blocking_notifier_call_chain(&cpufreq_transition_notifier_list, + CPUFREQ_POSTCHANGE, freqs); if (likely(policy) && likely(policy->cpu == freqs->cpu)) policy->cur = freqs->new; break; } - up_read(&cpufreq_notifier_rwsem); } EXPORT_SYMBOL_GPL(cpufreq_notify_transition); @@ -1007,7 +1003,7 @@ static int cpufreq_suspend(struct sys_device * sysdev, pm_message_t pmsg) freqs.old = cpu_policy->cur; freqs.new = cur_freq; - notifier_call_chain(&cpufreq_transition_notifier_list, + blocking_notifier_call_chain(&cpufreq_transition_notifier_list, CPUFREQ_SUSPENDCHANGE, &freqs); adjust_jiffies(CPUFREQ_SUSPENDCHANGE, &freqs); @@ -1088,7 +1084,8 @@ static int cpufreq_resume(struct sys_device * sysdev) freqs.old = cpu_policy->cur; freqs.new = cur_freq; - notifier_call_chain(&cpufreq_transition_notifier_list, + blocking_notifier_call_chain( + &cpufreq_transition_notifier_list, CPUFREQ_RESUMECHANGE, &freqs); adjust_jiffies(CPUFREQ_RESUMECHANGE, &freqs); @@ -1125,24 +1122,24 @@ static struct sysdev_driver cpufreq_sysdev_driver = { * changes in cpufreq policy. * * This function may sleep, and has the same return conditions as - * notifier_chain_register. + * blocking_notifier_chain_register. */ int cpufreq_register_notifier(struct notifier_block *nb, unsigned int list) { int ret; - down_write(&cpufreq_notifier_rwsem); switch (list) { case CPUFREQ_TRANSITION_NOTIFIER: - ret = notifier_chain_register(&cpufreq_transition_notifier_list, nb); + ret = blocking_notifier_chain_register( + &cpufreq_transition_notifier_list, nb); break; case CPUFREQ_POLICY_NOTIFIER: - ret = notifier_chain_register(&cpufreq_policy_notifier_list, nb); + ret = blocking_notifier_chain_register( + &cpufreq_policy_notifier_list, nb); break; default: ret = -EINVAL; } - up_write(&cpufreq_notifier_rwsem); return ret; } @@ -1157,24 +1154,24 @@ EXPORT_SYMBOL(cpufreq_register_notifier); * Remove a driver from the CPU frequency notifier list. * * This function may sleep, and has the same return conditions as - * notifier_chain_unregister. + * blocking_notifier_chain_unregister. */ int cpufreq_unregister_notifier(struct notifier_block *nb, unsigned int list) { int ret; - down_write(&cpufreq_notifier_rwsem); switch (list) { case CPUFREQ_TRANSITION_NOTIFIER: - ret = notifier_chain_unregister(&cpufreq_transition_notifier_list, nb); + ret = blocking_notifier_chain_unregister( + &cpufreq_transition_notifier_list, nb); break; case CPUFREQ_POLICY_NOTIFIER: - ret = notifier_chain_unregister(&cpufreq_policy_notifier_list, nb); + ret = blocking_notifier_chain_unregister( + &cpufreq_policy_notifier_list, nb); break; default: ret = -EINVAL; } - up_write(&cpufreq_notifier_rwsem); return ret; } @@ -1346,29 +1343,23 @@ static int __cpufreq_set_policy(struct cpufreq_policy *data, struct cpufreq_poli if (ret) goto error_out; - down_read(&cpufreq_notifier_rwsem); - /* adjust if necessary - all reasons */ - notifier_call_chain(&cpufreq_policy_notifier_list, CPUFREQ_ADJUST, - policy); + blocking_notifier_call_chain(&cpufreq_policy_notifier_list, + CPUFREQ_ADJUST, policy); /* adjust if necessary - hardware incompatibility*/ - notifier_call_chain(&cpufreq_policy_notifier_list, CPUFREQ_INCOMPATIBLE, - policy); + blocking_notifier_call_chain(&cpufreq_policy_notifier_list, + CPUFREQ_INCOMPATIBLE, policy); /* verify the cpu speed can be set within this limit, which might be different to the first one */ ret = cpufreq_driver->verify(policy); - if (ret) { - up_read(&cpufreq_notifier_rwsem); + if (ret) goto error_out; - } /* notification of the new policy */ - notifier_call_chain(&cpufreq_policy_notifier_list, CPUFREQ_NOTIFY, - policy); - - up_read(&cpufreq_notifier_rwsem); + blocking_notifier_call_chain(&cpufreq_policy_notifier_list, + CPUFREQ_NOTIFY, policy); data->min = policy->min; data->max = policy->max; diff --git a/drivers/firmware/dcdbas.c b/drivers/firmware/dcdbas.c index d6543fc4a92..339f405ff70 100644 --- a/drivers/firmware/dcdbas.c +++ b/drivers/firmware/dcdbas.c @@ -484,26 +484,15 @@ static void dcdbas_host_control(void) static int dcdbas_reboot_notify(struct notifier_block *nb, unsigned long code, void *unused) { - static unsigned int notify_cnt = 0; - switch (code) { case SYS_DOWN: case SYS_HALT: case SYS_POWER_OFF: if (host_control_on_shutdown) { /* firmware is going to perform host control action */ - if (++notify_cnt == 2) { - printk(KERN_WARNING - "Please wait for shutdown " - "action to complete...\n"); - dcdbas_host_control(); - } - /* - * register again and initiate the host control - * action on the second notification to allow - * everyone that registered to be notified - */ - register_reboot_notifier(nb); + printk(KERN_WARNING "Please wait for shutdown " + "action to complete...\n"); + dcdbas_host_control(); } break; } @@ -514,7 +503,7 @@ static int dcdbas_reboot_notify(struct notifier_block *nb, unsigned long code, static struct notifier_block dcdbas_reboot_nb = { .notifier_call = dcdbas_reboot_notify, .next = NULL, - .priority = 0 + .priority = INT_MIN }; static DCDBAS_BIN_ATTR_RW(smi_data); diff --git a/drivers/macintosh/adb.c b/drivers/macintosh/adb.c index d2ead1776c1..34fcabac5fd 100644 --- a/drivers/macintosh/adb.c +++ b/drivers/macintosh/adb.c @@ -80,7 +80,7 @@ static struct adb_driver *adb_driver_list[] = { static struct class *adb_dev_class; struct adb_driver *adb_controller; -struct notifier_block *adb_client_list = NULL; +BLOCKING_NOTIFIER_HEAD(adb_client_list); static int adb_got_sleep; static int adb_inited; static pid_t adb_probe_task_pid; @@ -354,7 +354,8 @@ adb_notify_sleep(struct pmu_sleep_notifier *self, int when) /* Stop autopoll */ if (adb_controller->autopoll) adb_controller->autopoll(0); - ret = notifier_call_chain(&adb_client_list, ADB_MSG_POWERDOWN, NULL); + ret = blocking_notifier_call_chain(&adb_client_list, + ADB_MSG_POWERDOWN, NULL); if (ret & NOTIFY_STOP_MASK) { up(&adb_probe_mutex); return PBOOK_SLEEP_REFUSE; @@ -391,7 +392,8 @@ do_adb_reset_bus(void) if (adb_controller->autopoll) adb_controller->autopoll(0); - nret = notifier_call_chain(&adb_client_list, ADB_MSG_PRE_RESET, NULL); + nret = blocking_notifier_call_chain(&adb_client_list, + ADB_MSG_PRE_RESET, NULL); if (nret & NOTIFY_STOP_MASK) { if (adb_controller->autopoll) adb_controller->autopoll(autopoll_devs); @@ -426,7 +428,8 @@ do_adb_reset_bus(void) } up(&adb_handler_sem); - nret = notifier_call_chain(&adb_client_list, ADB_MSG_POST_RESET, NULL); + nret = blocking_notifier_call_chain(&adb_client_list, + ADB_MSG_POST_RESET, NULL); if (nret & NOTIFY_STOP_MASK) return -EBUSY; diff --git a/drivers/macintosh/adbhid.c b/drivers/macintosh/adbhid.c index c0b46bceb5d..f5779a73184 100644 --- a/drivers/macintosh/adbhid.c +++ b/drivers/macintosh/adbhid.c @@ -1214,7 +1214,8 @@ static int __init adbhid_init(void) adbhid_probe(); - notifier_chain_register(&adb_client_list, &adbhid_adb_notifier); + blocking_notifier_chain_register(&adb_client_list, + &adbhid_adb_notifier); return 0; } diff --git a/drivers/macintosh/via-pmu.c b/drivers/macintosh/via-pmu.c index 4f5f3abc9cb..0b5ff553e39 100644 --- a/drivers/macintosh/via-pmu.c +++ b/drivers/macintosh/via-pmu.c @@ -187,7 +187,7 @@ extern int disable_kernel_backlight; int __fake_sleep; int asleep; -struct notifier_block *sleep_notifier_list; +BLOCKING_NOTIFIER_HEAD(sleep_notifier_list); #ifdef CONFIG_ADB static int adb_dev_map = 0; diff --git a/drivers/macintosh/via-pmu68k.c b/drivers/macintosh/via-pmu68k.c index f08e52f2107..35b70323e7e 100644 --- a/drivers/macintosh/via-pmu68k.c +++ b/drivers/macintosh/via-pmu68k.c @@ -102,7 +102,7 @@ static int pmu_kind = PMU_UNKNOWN; static int pmu_fully_inited = 0; int asleep; -struct notifier_block *sleep_notifier_list; +BLOCKING_NOTIFIER_HEAD(sleep_notifier_list); static int pmu_probe(void); static int pmu_init(void); @@ -913,7 +913,8 @@ int powerbook_sleep(void) struct adb_request sleep_req; /* Notify device drivers */ - ret = notifier_call_chain(&sleep_notifier_list, PBOOK_SLEEP, NULL); + ret = blocking_notifier_call_chain(&sleep_notifier_list, + PBOOK_SLEEP, NULL); if (ret & NOTIFY_STOP_MASK) return -EBUSY; @@ -984,7 +985,7 @@ int powerbook_sleep(void) enable_irq(i); /* Notify drivers */ - notifier_call_chain(&sleep_notifier_list, PBOOK_WAKE, NULL); + blocking_notifier_call_chain(&sleep_notifier_list, PBOOK_WAKE, NULL); /* reenable ADB autopoll */ pmu_adb_autopoll(adb_dev_map); diff --git a/drivers/macintosh/windfarm_core.c b/drivers/macintosh/windfarm_core.c index 6c0ba04bc57..ab3faa702d5 100644 --- a/drivers/macintosh/windfarm_core.c +++ b/drivers/macintosh/windfarm_core.c @@ -52,7 +52,7 @@ static LIST_HEAD(wf_controls); static LIST_HEAD(wf_sensors); static DEFINE_MUTEX(wf_lock); -static struct notifier_block *wf_client_list; +static BLOCKING_NOTIFIER_HEAD(wf_client_list); static int wf_client_count; static unsigned int wf_overtemp; static unsigned int wf_overtemp_counter; @@ -68,7 +68,7 @@ static struct platform_device wf_platform_device = { static inline void wf_notify(int event, void *param) { - notifier_call_chain(&wf_client_list, event, param); + blocking_notifier_call_chain(&wf_client_list, event, param); } int wf_critical_overtemp(void) @@ -398,7 +398,7 @@ int wf_register_client(struct notifier_block *nb) struct wf_sensor *sr; mutex_lock(&wf_lock); - rc = notifier_chain_register(&wf_client_list, nb); + rc = blocking_notifier_chain_register(&wf_client_list, nb); if (rc != 0) goto bail; wf_client_count++; @@ -417,7 +417,7 @@ EXPORT_SYMBOL_GPL(wf_register_client); int wf_unregister_client(struct notifier_block *nb) { mutex_lock(&wf_lock); - notifier_chain_unregister(&wf_client_list, nb); + blocking_notifier_chain_unregister(&wf_client_list, nb); wf_client_count++; if (wf_client_count == 0) wf_stop_thread(); diff --git a/drivers/misc/ibmasm/heartbeat.c b/drivers/misc/ibmasm/heartbeat.c index f295401fac2..7fd7a43e38d 100644 --- a/drivers/misc/ibmasm/heartbeat.c +++ b/drivers/misc/ibmasm/heartbeat.c @@ -52,12 +52,13 @@ static struct notifier_block panic_notifier = { panic_happened, NULL, 1 }; void ibmasm_register_panic_notifier(void) { - notifier_chain_register(&panic_notifier_list, &panic_notifier); + atomic_notifier_chain_register(&panic_notifier_list, &panic_notifier); } void ibmasm_unregister_panic_notifier(void) { - notifier_chain_unregister(&panic_notifier_list, &panic_notifier); + atomic_notifier_chain_unregister(&panic_notifier_list, + &panic_notifier); } diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c index 2d0ac169a86..f13a539dc16 100644 --- a/drivers/net/bonding/bond_main.c +++ b/drivers/net/bonding/bond_main.c @@ -3159,7 +3159,7 @@ static int bond_slave_netdev_event(unsigned long event, struct net_device *slave * bond_netdev_event: handle netdev notifier chain events. * * This function receives events for the netdev chain. The caller (an - * ioctl handler calling notifier_call_chain) holds the necessary + * ioctl handler calling blocking_notifier_call_chain) holds the necessary * locks for us to safely manipulate the slave devices (RTNL lock, * dev_probe_lock). */ diff --git a/drivers/parisc/led.c b/drivers/parisc/led.c index 3627a2d7f79..298f2ddb2c1 100644 --- a/drivers/parisc/led.c +++ b/drivers/parisc/led.c @@ -499,11 +499,16 @@ static int led_halt(struct notifier_block *, unsigned long, void *); static struct notifier_block led_notifier = { .notifier_call = led_halt, }; +static int notifier_disabled = 0; static int led_halt(struct notifier_block *nb, unsigned long event, void *buf) { char *txt; - + + if (notifier_disabled) + return NOTIFY_OK; + + notifier_disabled = 1; switch (event) { case SYS_RESTART: txt = "SYSTEM RESTART"; break; @@ -527,7 +532,6 @@ static int led_halt(struct notifier_block *nb, unsigned long event, void *buf) if (led_func_ptr) led_func_ptr(0xff); /* turn all LEDs ON */ - unregister_reboot_notifier(&led_notifier); return NOTIFY_OK; } @@ -758,6 +762,12 @@ not_found: return 1; } +static void __exit led_exit(void) +{ + unregister_reboot_notifier(&led_notifier); + return; +} + #ifdef CONFIG_PROC_FS module_init(led_create_procfs) #endif diff --git a/drivers/parisc/power.c b/drivers/parisc/power.c index 54b2b7f20b9..0bcab83b408 100644 --- a/drivers/parisc/power.c +++ b/drivers/parisc/power.c @@ -251,7 +251,8 @@ static int __init power_init(void) } /* Register a call for panic conditions. */ - notifier_chain_register(&panic_notifier_list, &parisc_panic_block); + atomic_notifier_chain_register(&panic_notifier_list, + &parisc_panic_block); tasklet_enable(&power_tasklet); @@ -264,7 +265,8 @@ static void __exit power_exit(void) return; tasklet_disable(&power_tasklet); - notifier_chain_unregister(&panic_notifier_list, &parisc_panic_block); + atomic_notifier_chain_unregister(&panic_notifier_list, + &parisc_panic_block); power_tasklet.func = NULL; pdc_soft_power_button(0); } diff --git a/drivers/scsi/gdth.c b/drivers/scsi/gdth.c index 62e3cda859a..7f7013e80a8 100644 --- a/drivers/scsi/gdth.c +++ b/drivers/scsi/gdth.c @@ -671,7 +671,7 @@ static struct file_operations gdth_fops = { static struct notifier_block gdth_notifier = { gdth_halt, NULL, 0 }; - +static int notifier_disabled = 0; static void gdth_delay(int milliseconds) { @@ -4595,13 +4595,13 @@ static int __init gdth_detect(struct scsi_host_template *shtp) add_timer(&gdth_timer); #endif major = register_chrdev(0,"gdth",&gdth_fops); + notifier_disabled = 0; register_reboot_notifier(&gdth_notifier); } gdth_polling = FALSE; return gdth_ctr_vcount; } - static int gdth_release(struct Scsi_Host *shp) { int hanum; @@ -5632,10 +5632,14 @@ static int gdth_halt(struct notifier_block *nb, ulong event, void *buf) char cmnd[MAX_COMMAND_SIZE]; #endif + if (notifier_disabled) + return NOTIFY_OK; + TRACE2(("gdth_halt() event %d\n",(int)event)); if (event != SYS_RESTART && event != SYS_HALT && event != SYS_POWER_OFF) return NOTIFY_DONE; + notifier_disabled = 1; printk("GDT-HA: Flushing all host drives .. "); for (hanum = 0; hanum < gdth_ctr_count; ++hanum) { gdth_flush(hanum); @@ -5679,7 +5683,6 @@ static int gdth_halt(struct notifier_block *nb, ulong event, void *buf) #ifdef GDTH_STATISTICS del_timer(&gdth_timer); #endif - unregister_reboot_notifier(&gdth_notifier); return NOTIFY_OK; } diff --git a/drivers/usb/core/notify.c b/drivers/usb/core/notify.c index 4b55285de9a..fe0ed54fa0a 100644 --- a/drivers/usb/core/notify.c +++ b/drivers/usb/core/notify.c @@ -16,57 +16,7 @@ #include #include "usb.h" - -static struct notifier_block *usb_notifier_list; -static DEFINE_MUTEX(usb_notifier_lock); - -static void usb_notifier_chain_register(struct notifier_block **list, - struct notifier_block *n) -{ - mutex_lock(&usb_notifier_lock); - while (*list) { - if (n->priority > (*list)->priority) - break; - list = &((*list)->next); - } - n->next = *list; - *list = n; - mutex_unlock(&usb_notifier_lock); -} - -static void usb_notifier_chain_unregister(struct notifier_block **nl, - struct notifier_block *n) -{ - mutex_lock(&usb_notifier_lock); - while ((*nl)!=NULL) { - if ((*nl)==n) { - *nl = n->next; - goto exit; - } - nl=&((*nl)->next); - } -exit: - mutex_unlock(&usb_notifier_lock); -} - -static int usb_notifier_call_chain(struct notifier_block **n, - unsigned long val, void *v) -{ - int ret=NOTIFY_DONE; - struct notifier_block *nb = *n; - - mutex_lock(&usb_notifier_lock); - while (nb) { - ret = nb->notifier_call(nb,val,v); - if (ret&NOTIFY_STOP_MASK) { - goto exit; - } - nb = nb->next; - } -exit: - mutex_unlock(&usb_notifier_lock); - return ret; -} +static BLOCKING_NOTIFIER_HEAD(usb_notifier_list); /** * usb_register_notify - register a notifier callback whenever a usb change happens @@ -76,7 +26,7 @@ exit: */ void usb_register_notify(struct notifier_block *nb) { - usb_notifier_chain_register(&usb_notifier_list, nb); + blocking_notifier_chain_register(&usb_notifier_list, nb); } EXPORT_SYMBOL_GPL(usb_register_notify); @@ -89,27 +39,28 @@ EXPORT_SYMBOL_GPL(usb_register_notify); */ void usb_unregister_notify(struct notifier_block *nb) { - usb_notifier_chain_unregister(&usb_notifier_list, nb); + blocking_notifier_chain_unregister(&usb_notifier_list, nb); } EXPORT_SYMBOL_GPL(usb_unregister_notify); void usb_notify_add_device(struct usb_device *udev) { - usb_notifier_call_chain(&usb_notifier_list, USB_DEVICE_ADD, udev); + blocking_notifier_call_chain(&usb_notifier_list, USB_DEVICE_ADD, udev); } void usb_notify_remove_device(struct usb_device *udev) { - usb_notifier_call_chain(&usb_notifier_list, USB_DEVICE_REMOVE, udev); + blocking_notifier_call_chain(&usb_notifier_list, + USB_DEVICE_REMOVE, udev); } void usb_notify_add_bus(struct usb_bus *ubus) { - usb_notifier_call_chain(&usb_notifier_list, USB_BUS_ADD, ubus); + blocking_notifier_call_chain(&usb_notifier_list, USB_BUS_ADD, ubus); } void usb_notify_remove_bus(struct usb_bus *ubus) { - usb_notifier_call_chain(&usb_notifier_list, USB_BUS_REMOVE, ubus); + blocking_notifier_call_chain(&usb_notifier_list, USB_BUS_REMOVE, ubus); } diff --git a/drivers/video/fbmem.c b/drivers/video/fbmem.c index 07d882b1439..b1a8dca7643 100644 --- a/drivers/video/fbmem.c +++ b/drivers/video/fbmem.c @@ -55,7 +55,7 @@ #define FBPIXMAPSIZE (1024 * 8) -static struct notifier_block *fb_notifier_list; +static BLOCKING_NOTIFIER_HEAD(fb_notifier_list); struct fb_info *registered_fb[FB_MAX]; int num_registered_fb; @@ -784,7 +784,7 @@ fb_set_var(struct fb_info *info, struct fb_var_screeninfo *var) event.info = info; event.data = &mode1; - ret = notifier_call_chain(&fb_notifier_list, + ret = blocking_notifier_call_chain(&fb_notifier_list, FB_EVENT_MODE_DELETE, &event); } @@ -830,8 +830,8 @@ fb_set_var(struct fb_info *info, struct fb_var_screeninfo *var) info->flags &= ~FBINFO_MISC_USEREVENT; event.info = info; - notifier_call_chain(&fb_notifier_list, evnt, - &event); + blocking_notifier_call_chain(&fb_notifier_list, + evnt, &event); } } } @@ -854,7 +854,8 @@ fb_blank(struct fb_info *info, int blank) event.info = info; event.data = ␣ - notifier_call_chain(&fb_notifier_list, FB_EVENT_BLANK, &event); + blocking_notifier_call_chain(&fb_notifier_list, + FB_EVENT_BLANK, &event); } return ret; @@ -925,7 +926,7 @@ fb_ioctl(struct inode *inode, struct file *file, unsigned int cmd, con2fb.framebuffer = -1; event.info = info; event.data = &con2fb; - notifier_call_chain(&fb_notifier_list, + blocking_notifier_call_chain(&fb_notifier_list, FB_EVENT_GET_CONSOLE_MAP, &event); return copy_to_user(argp, &con2fb, sizeof(con2fb)) ? -EFAULT : 0; @@ -944,7 +945,7 @@ fb_ioctl(struct inode *inode, struct file *file, unsigned int cmd, return -EINVAL; event.info = info; event.data = &con2fb; - return notifier_call_chain(&fb_notifier_list, + return blocking_notifier_call_chain(&fb_notifier_list, FB_EVENT_SET_CONSOLE_MAP, &event); case FBIOBLANK: @@ -1324,7 +1325,7 @@ register_framebuffer(struct fb_info *fb_info) devfs_mk_cdev(MKDEV(FB_MAJOR, i), S_IFCHR | S_IRUGO | S_IWUGO, "fb/%d", i); event.info = fb_info; - notifier_call_chain(&fb_notifier_list, + blocking_notifier_call_chain(&fb_notifier_list, FB_EVENT_FB_REGISTERED, &event); return 0; } @@ -1366,7 +1367,7 @@ unregister_framebuffer(struct fb_info *fb_info) */ int fb_register_client(struct notifier_block *nb) { - return notifier_chain_register(&fb_notifier_list, nb); + return blocking_notifier_chain_register(&fb_notifier_list, nb); } /** @@ -1375,7 +1376,7 @@ int fb_register_client(struct notifier_block *nb) */ int fb_unregister_client(struct notifier_block *nb) { - return notifier_chain_unregister(&fb_notifier_list, nb); + return blocking_notifier_chain_unregister(&fb_notifier_list, nb); } /** @@ -1393,11 +1394,13 @@ void fb_set_suspend(struct fb_info *info, int state) event.info = info; if (state) { - notifier_call_chain(&fb_notifier_list, FB_EVENT_SUSPEND, &event); + blocking_notifier_call_chain(&fb_notifier_list, + FB_EVENT_SUSPEND, &event); info->state = FBINFO_STATE_SUSPENDED; } else { info->state = FBINFO_STATE_RUNNING; - notifier_call_chain(&fb_notifier_list, FB_EVENT_RESUME, &event); + blocking_notifier_call_chain(&fb_notifier_list, + FB_EVENT_RESUME, &event); } } @@ -1469,7 +1472,7 @@ int fb_new_modelist(struct fb_info *info) if (!list_empty(&info->modelist)) { event.info = info; - err = notifier_call_chain(&fb_notifier_list, + err = blocking_notifier_call_chain(&fb_notifier_list, FB_EVENT_NEW_MODELIST, &event); } @@ -1495,7 +1498,7 @@ int fb_con_duit(struct fb_info *info, int event, void *data) evnt.info = info; evnt.data = data; - return notifier_call_chain(&fb_notifier_list, event, &evnt); + return blocking_notifier_call_chain(&fb_notifier_list, event, &evnt); } EXPORT_SYMBOL(fb_con_duit); diff --git a/include/asm-i386/kdebug.h b/include/asm-i386/kdebug.h index 316138e8991..96d0828ce09 100644 --- a/include/asm-i386/kdebug.h +++ b/include/asm-i386/kdebug.h @@ -17,11 +17,9 @@ struct die_args { int signr; }; -/* Note - you should never unregister because that can race with NMIs. - If you really want to do it first unregister - then synchronize_sched - then free. - */ -int register_die_notifier(struct notifier_block *nb); -extern struct notifier_block *i386die_chain; +extern int register_die_notifier(struct notifier_block *); +extern int unregister_die_notifier(struct notifier_block *); +extern struct atomic_notifier_head i386die_chain; /* Grossly misnamed. */ @@ -51,7 +49,7 @@ static inline int notify_die(enum die_val val, const char *str, .trapnr = trap, .signr = sig }; - return notifier_call_chain(&i386die_chain, val, &args); + return atomic_notifier_call_chain(&i386die_chain, val, &args); } #endif diff --git a/include/asm-ia64/kdebug.h b/include/asm-ia64/kdebug.h index 8b01a083dde..218c458ab60 100644 --- a/include/asm-ia64/kdebug.h +++ b/include/asm-ia64/kdebug.h @@ -40,7 +40,7 @@ struct die_args { extern int register_die_notifier(struct notifier_block *); extern int unregister_die_notifier(struct notifier_block *); -extern struct notifier_block *ia64die_chain; +extern struct atomic_notifier_head ia64die_chain; enum die_val { DIE_BREAK = 1, @@ -81,7 +81,7 @@ static inline int notify_die(enum die_val val, char *str, struct pt_regs *regs, .signr = sig }; - return notifier_call_chain(&ia64die_chain, val, &args); + return atomic_notifier_call_chain(&ia64die_chain, val, &args); } #endif diff --git a/include/asm-powerpc/kdebug.h b/include/asm-powerpc/kdebug.h index 7c16265568e..c01786ab5fa 100644 --- a/include/asm-powerpc/kdebug.h +++ b/include/asm-powerpc/kdebug.h @@ -16,13 +16,9 @@ struct die_args { int signr; }; -/* - Note - you should never unregister because that can race with NMIs. - If you really want to do it first unregister - then synchronize_sched - - then free. - */ -int register_die_notifier(struct notifier_block *nb); -extern struct notifier_block *powerpc_die_chain; +extern int register_die_notifier(struct notifier_block *); +extern int unregister_die_notifier(struct notifier_block *); +extern struct atomic_notifier_head powerpc_die_chain; /* Grossly misnamed. */ enum die_val { @@ -37,7 +33,7 @@ enum die_val { static inline int notify_die(enum die_val val,char *str,struct pt_regs *regs,long err,int trap, int sig) { struct die_args args = { .regs=regs, .str=str, .err=err, .trapnr=trap,.signr=sig }; - return notifier_call_chain(&powerpc_die_chain, val, &args); + return atomic_notifier_call_chain(&powerpc_die_chain, val, &args); } #endif /* __KERNEL__ */ diff --git a/include/asm-sparc64/kdebug.h b/include/asm-sparc64/kdebug.h index 6321f5a0198..4040d127ac3 100644 --- a/include/asm-sparc64/kdebug.h +++ b/include/asm-sparc64/kdebug.h @@ -15,12 +15,9 @@ struct die_args { int signr; }; -/* Note - you should never unregister because that can race with NMIs. - * If you really want to do it first unregister - then synchronize_sched - * - then free. - */ -int register_die_notifier(struct notifier_block *nb); -extern struct notifier_block *sparc64die_chain; +extern int register_die_notifier(struct notifier_block *); +extern int unregister_die_notifier(struct notifier_block *); +extern struct atomic_notifier_head sparc64die_chain; extern void bad_trap(struct pt_regs *, long); @@ -46,7 +43,7 @@ static inline int notify_die(enum die_val val,char *str, struct pt_regs *regs, .trapnr = trap, .signr = sig }; - return notifier_call_chain(&sparc64die_chain, val, &args); + return atomic_notifier_call_chain(&sparc64die_chain, val, &args); } #endif diff --git a/include/asm-x86_64/kdebug.h b/include/asm-x86_64/kdebug.h index b9ed4c0c878..cf795631d9b 100644 --- a/include/asm-x86_64/kdebug.h +++ b/include/asm-x86_64/kdebug.h @@ -5,21 +5,20 @@ struct pt_regs; -struct die_args { +struct die_args { struct pt_regs *regs; const char *str; - long err; + long err; int trapnr; int signr; -}; +}; + +extern int register_die_notifier(struct notifier_block *); +extern int unregister_die_notifier(struct notifier_block *); +extern struct atomic_notifier_head die_chain; -/* Note - you should never unregister because that can race with NMIs. - If you really want to do it first unregister - then synchronize_sched - then free. - */ -int register_die_notifier(struct notifier_block *nb); -extern struct notifier_block *die_chain; /* Grossly misnamed. */ -enum die_val { +enum die_val { DIE_OOPS = 1, DIE_INT3, DIE_DEBUG, @@ -33,8 +32,8 @@ enum die_val { DIE_CALL, DIE_NMI_IPI, DIE_PAGE_FAULT, -}; - +}; + static inline int notify_die(enum die_val val, const char *str, struct pt_regs *regs, long err, int trap, int sig) { @@ -45,7 +44,7 @@ static inline int notify_die(enum die_val val, const char *str, .trapnr = trap, .signr = sig }; - return notifier_call_chain(&die_chain, val, &args); + return atomic_notifier_call_chain(&die_chain, val, &args); } extern int printk_address(unsigned long address); diff --git a/include/linux/adb.h b/include/linux/adb.h index e9fdc63483c..b7305b17827 100644 --- a/include/linux/adb.h +++ b/include/linux/adb.h @@ -85,7 +85,7 @@ enum adb_message { ADB_MSG_POST_RESET /* Called after resetting the bus (re-do init & register) */ }; extern struct adb_driver *adb_controller; -extern struct notifier_block *adb_client_list; +extern struct blocking_notifier_head adb_client_list; int adb_request(struct adb_request *req, void (*done)(struct adb_request *), int flags, int nbytes, ...); diff --git a/include/linux/kernel.h b/include/linux/kernel.h index 03d6cfaa5b8..a3720f973ea 100644 --- a/include/linux/kernel.h +++ b/include/linux/kernel.h @@ -87,7 +87,7 @@ extern int cond_resched(void); (__x < 0) ? -__x : __x; \ }) -extern struct notifier_block *panic_notifier_list; +extern struct atomic_notifier_head panic_notifier_list; extern long (*panic_blink)(long time); NORET_TYPE void panic(const char * fmt, ...) __attribute__ ((NORET_AND format (printf, 1, 2))); diff --git a/include/linux/memory.h b/include/linux/memory.h index e251dc43d0f..8f04143ca36 100644 --- a/include/linux/memory.h +++ b/include/linux/memory.h @@ -77,7 +77,6 @@ extern int remove_memory_block(unsigned long, struct mem_section *, int); #define CONFIG_MEM_BLOCK_SIZE (PAGES_PER_SECTION< +#include +#include -struct notifier_block -{ - int (*notifier_call)(struct notifier_block *self, unsigned long, void *); +/* + * Notifier chains are of three types: + * + * Atomic notifier chains: Chain callbacks run in interrupt/atomic + * context. Callouts are not allowed to block. + * Blocking notifier chains: Chain callbacks run in process context. + * Callouts are allowed to block. + * Raw notifier chains: There are no restrictions on callbacks, + * registration, or unregistration. All locking and protection + * must be provided by the caller. + * + * atomic_notifier_chain_register() may be called from an atomic context, + * but blocking_notifier_chain_register() must be called from a process + * context. Ditto for the corresponding _unregister() routines. + * + * atomic_notifier_chain_unregister() and blocking_notifier_chain_unregister() + * _must not_ be called from within the call chain. + */ + +struct notifier_block { + int (*notifier_call)(struct notifier_block *, unsigned long, void *); struct notifier_block *next; int priority; }; +struct atomic_notifier_head { + spinlock_t lock; + struct notifier_block *head; +}; + +struct blocking_notifier_head { + struct rw_semaphore rwsem; + struct notifier_block *head; +}; + +struct raw_notifier_head { + struct notifier_block *head; +}; + +#define ATOMIC_INIT_NOTIFIER_HEAD(name) do { \ + spin_lock_init(&(name)->lock); \ + (name)->head = NULL; \ + } while (0) +#define BLOCKING_INIT_NOTIFIER_HEAD(name) do { \ + init_rwsem(&(name)->rwsem); \ + (name)->head = NULL; \ + } while (0) +#define RAW_INIT_NOTIFIER_HEAD(name) do { \ + (name)->head = NULL; \ + } while (0) + +#define ATOMIC_NOTIFIER_INIT(name) { \ + .lock = SPIN_LOCK_UNLOCKED, \ + .head = NULL } +#define BLOCKING_NOTIFIER_INIT(name) { \ + .rwsem = __RWSEM_INITIALIZER((name).rwsem), \ + .head = NULL } +#define RAW_NOTIFIER_INIT(name) { \ + .head = NULL } + +#define ATOMIC_NOTIFIER_HEAD(name) \ + struct atomic_notifier_head name = \ + ATOMIC_NOTIFIER_INIT(name) +#define BLOCKING_NOTIFIER_HEAD(name) \ + struct blocking_notifier_head name = \ + BLOCKING_NOTIFIER_INIT(name) +#define RAW_NOTIFIER_HEAD(name) \ + struct raw_notifier_head name = \ + RAW_NOTIFIER_INIT(name) #ifdef __KERNEL__ -extern int notifier_chain_register(struct notifier_block **list, struct notifier_block *n); -extern int notifier_chain_unregister(struct notifier_block **nl, struct notifier_block *n); -extern int notifier_call_chain(struct notifier_block **n, unsigned long val, void *v); +extern int atomic_notifier_chain_register(struct atomic_notifier_head *, + struct notifier_block *); +extern int blocking_notifier_chain_register(struct blocking_notifier_head *, + struct notifier_block *); +extern int raw_notifier_chain_register(struct raw_notifier_head *, + struct notifier_block *); + +extern int atomic_notifier_chain_unregister(struct atomic_notifier_head *, + struct notifier_block *); +extern int blocking_notifier_chain_unregister(struct blocking_notifier_head *, + struct notifier_block *); +extern int raw_notifier_chain_unregister(struct raw_notifier_head *, + struct notifier_block *); + +extern int atomic_notifier_call_chain(struct atomic_notifier_head *, + unsigned long val, void *v); +extern int blocking_notifier_call_chain(struct blocking_notifier_head *, + unsigned long val, void *v); +extern int raw_notifier_call_chain(struct raw_notifier_head *, + unsigned long val, void *v); #define NOTIFY_DONE 0x0000 /* Don't care */ #define NOTIFY_OK 0x0001 /* Suits me */ #define NOTIFY_STOP_MASK 0x8000 /* Don't call further */ -#define NOTIFY_BAD (NOTIFY_STOP_MASK|0x0002) /* Bad/Veto action */ +#define NOTIFY_BAD (NOTIFY_STOP_MASK|0x0002) + /* Bad/Veto action */ /* * Clean way to return from the notifier and stop further calls. */ diff --git a/include/net/netfilter/nf_conntrack.h b/include/net/netfilter/nf_conntrack.h index b6f0905a4ee..916013ca4a5 100644 --- a/include/net/netfilter/nf_conntrack.h +++ b/include/net/netfilter/nf_conntrack.h @@ -300,29 +300,30 @@ DECLARE_PER_CPU(struct nf_conntrack_ecache, nf_conntrack_ecache); #define CONNTRACK_ECACHE(x) (__get_cpu_var(nf_conntrack_ecache).x) -extern struct notifier_block *nf_conntrack_chain; -extern struct notifier_block *nf_conntrack_expect_chain; +extern struct atomic_notifier_head nf_conntrack_chain; +extern struct atomic_notifier_head nf_conntrack_expect_chain; static inline int nf_conntrack_register_notifier(struct notifier_block *nb) { - return notifier_chain_register(&nf_conntrack_chain, nb); + return atomic_notifier_chain_register(&nf_conntrack_chain, nb); } static inline int nf_conntrack_unregister_notifier(struct notifier_block *nb) { - return notifier_chain_unregister(&nf_conntrack_chain, nb); + return atomic_notifier_chain_unregister(&nf_conntrack_chain, nb); } static inline int nf_conntrack_expect_register_notifier(struct notifier_block *nb) { - return notifier_chain_register(&nf_conntrack_expect_chain, nb); + return atomic_notifier_chain_register(&nf_conntrack_expect_chain, nb); } static inline int nf_conntrack_expect_unregister_notifier(struct notifier_block *nb) { - return notifier_chain_unregister(&nf_conntrack_expect_chain, nb); + return atomic_notifier_chain_unregister(&nf_conntrack_expect_chain, + nb); } extern void nf_ct_deliver_cached_events(const struct nf_conn *ct); @@ -347,14 +348,14 @@ static inline void nf_conntrack_event(enum ip_conntrack_events event, struct nf_conn *ct) { if (nf_ct_is_confirmed(ct) && !nf_ct_is_dying(ct)) - notifier_call_chain(&nf_conntrack_chain, event, ct); + atomic_notifier_call_chain(&nf_conntrack_chain, event, ct); } static inline void nf_conntrack_expect_event(enum ip_conntrack_expect_events event, struct nf_conntrack_expect *exp) { - notifier_call_chain(&nf_conntrack_expect_chain, event, exp); + atomic_notifier_call_chain(&nf_conntrack_expect_chain, event, exp); } #else /* CONFIG_NF_CONNTRACK_EVENTS */ static inline void nf_conntrack_event_cache(enum ip_conntrack_events event, diff --git a/kernel/cpu.c b/kernel/cpu.c index 8be22bd8093..fe2b8d0bfe4 100644 --- a/kernel/cpu.c +++ b/kernel/cpu.c @@ -18,7 +18,7 @@ /* This protects CPUs going up and down... */ static DECLARE_MUTEX(cpucontrol); -static struct notifier_block *cpu_chain; +static BLOCKING_NOTIFIER_HEAD(cpu_chain); #ifdef CONFIG_HOTPLUG_CPU static struct task_struct *lock_cpu_hotplug_owner; @@ -71,21 +71,13 @@ EXPORT_SYMBOL_GPL(lock_cpu_hotplug_interruptible); /* Need to know about CPUs going up/down? */ int register_cpu_notifier(struct notifier_block *nb) { - int ret; - - if ((ret = lock_cpu_hotplug_interruptible()) != 0) - return ret; - ret = notifier_chain_register(&cpu_chain, nb); - unlock_cpu_hotplug(); - return ret; + return blocking_notifier_chain_register(&cpu_chain, nb); } EXPORT_SYMBOL(register_cpu_notifier); void unregister_cpu_notifier(struct notifier_block *nb) { - lock_cpu_hotplug(); - notifier_chain_unregister(&cpu_chain, nb); - unlock_cpu_hotplug(); + blocking_notifier_chain_unregister(&cpu_chain, nb); } EXPORT_SYMBOL(unregister_cpu_notifier); @@ -141,7 +133,7 @@ int cpu_down(unsigned int cpu) goto out; } - err = notifier_call_chain(&cpu_chain, CPU_DOWN_PREPARE, + err = blocking_notifier_call_chain(&cpu_chain, CPU_DOWN_PREPARE, (void *)(long)cpu); if (err == NOTIFY_BAD) { printk("%s: attempt to take down CPU %u failed\n", @@ -159,7 +151,7 @@ int cpu_down(unsigned int cpu) p = __stop_machine_run(take_cpu_down, NULL, cpu); if (IS_ERR(p)) { /* CPU didn't die: tell everyone. Can't complain. */ - if (notifier_call_chain(&cpu_chain, CPU_DOWN_FAILED, + if (blocking_notifier_call_chain(&cpu_chain, CPU_DOWN_FAILED, (void *)(long)cpu) == NOTIFY_BAD) BUG(); @@ -182,8 +174,8 @@ int cpu_down(unsigned int cpu) put_cpu(); /* CPU is completely dead: tell everyone. Too late to complain. */ - if (notifier_call_chain(&cpu_chain, CPU_DEAD, (void *)(long)cpu) - == NOTIFY_BAD) + if (blocking_notifier_call_chain(&cpu_chain, CPU_DEAD, + (void *)(long)cpu) == NOTIFY_BAD) BUG(); check_for_tasks(cpu); @@ -211,7 +203,7 @@ int __devinit cpu_up(unsigned int cpu) goto out; } - ret = notifier_call_chain(&cpu_chain, CPU_UP_PREPARE, hcpu); + ret = blocking_notifier_call_chain(&cpu_chain, CPU_UP_PREPARE, hcpu); if (ret == NOTIFY_BAD) { printk("%s: attempt to bring up CPU %u failed\n", __FUNCTION__, cpu); @@ -226,11 +218,12 @@ int __devinit cpu_up(unsigned int cpu) BUG_ON(!cpu_online(cpu)); /* Now call notifier in preparation. */ - notifier_call_chain(&cpu_chain, CPU_ONLINE, hcpu); + blocking_notifier_call_chain(&cpu_chain, CPU_ONLINE, hcpu); out_notify: if (ret != 0) - notifier_call_chain(&cpu_chain, CPU_UP_CANCELED, hcpu); + blocking_notifier_call_chain(&cpu_chain, + CPU_UP_CANCELED, hcpu); out: unlock_cpu_hotplug(); return ret; diff --git a/kernel/module.c b/kernel/module.c index ddfe45ac2fd..4fafd58038a 100644 --- a/kernel/module.c +++ b/kernel/module.c @@ -64,26 +64,17 @@ static DEFINE_SPINLOCK(modlist_lock); static DEFINE_MUTEX(module_mutex); static LIST_HEAD(modules); -static DEFINE_MUTEX(notify_mutex); -static struct notifier_block * module_notify_list; +static BLOCKING_NOTIFIER_HEAD(module_notify_list); int register_module_notifier(struct notifier_block * nb) { - int err; - mutex_lock(¬ify_mutex); - err = notifier_chain_register(&module_notify_list, nb); - mutex_unlock(¬ify_mutex); - return err; + return blocking_notifier_chain_register(&module_notify_list, nb); } EXPORT_SYMBOL(register_module_notifier); int unregister_module_notifier(struct notifier_block * nb) { - int err; - mutex_lock(¬ify_mutex); - err = notifier_chain_unregister(&module_notify_list, nb); - mutex_unlock(¬ify_mutex); - return err; + return blocking_notifier_chain_unregister(&module_notify_list, nb); } EXPORT_SYMBOL(unregister_module_notifier); @@ -1816,9 +1807,8 @@ sys_init_module(void __user *umod, /* Drop lock so they can recurse */ mutex_unlock(&module_mutex); - mutex_lock(¬ify_mutex); - notifier_call_chain(&module_notify_list, MODULE_STATE_COMING, mod); - mutex_unlock(¬ify_mutex); + blocking_notifier_call_chain(&module_notify_list, + MODULE_STATE_COMING, mod); /* Start the module */ if (mod->init != NULL) diff --git a/kernel/panic.c b/kernel/panic.c index acd95adddb9..f895c7c01d5 100644 --- a/kernel/panic.c +++ b/kernel/panic.c @@ -29,7 +29,7 @@ static DEFINE_SPINLOCK(pause_on_oops_lock); int panic_timeout; EXPORT_SYMBOL(panic_timeout); -struct notifier_block *panic_notifier_list; +ATOMIC_NOTIFIER_HEAD(panic_notifier_list); EXPORT_SYMBOL(panic_notifier_list); @@ -97,7 +97,7 @@ NORET_TYPE void panic(const char * fmt, ...) smp_send_stop(); #endif - notifier_call_chain(&panic_notifier_list, 0, buf); + atomic_notifier_call_chain(&panic_notifier_list, 0, buf); if (!panic_blink) panic_blink = no_blink; diff --git a/kernel/profile.c b/kernel/profile.c index ad81f799a9b..5a730fdb1a2 100644 --- a/kernel/profile.c +++ b/kernel/profile.c @@ -87,72 +87,52 @@ void __init profile_init(void) #ifdef CONFIG_PROFILING -static DECLARE_RWSEM(profile_rwsem); -static DEFINE_RWLOCK(handoff_lock); -static struct notifier_block * task_exit_notifier; -static struct notifier_block * task_free_notifier; -static struct notifier_block * munmap_notifier; +static BLOCKING_NOTIFIER_HEAD(task_exit_notifier); +static ATOMIC_NOTIFIER_HEAD(task_free_notifier); +static BLOCKING_NOTIFIER_HEAD(munmap_notifier); void profile_task_exit(struct task_struct * task) { - down_read(&profile_rwsem); - notifier_call_chain(&task_exit_notifier, 0, task); - up_read(&profile_rwsem); + blocking_notifier_call_chain(&task_exit_notifier, 0, task); } int profile_handoff_task(struct task_struct * task) { int ret; - read_lock(&handoff_lock); - ret = notifier_call_chain(&task_free_notifier, 0, task); - read_unlock(&handoff_lock); + ret = atomic_notifier_call_chain(&task_free_notifier, 0, task); return (ret == NOTIFY_OK) ? 1 : 0; } void profile_munmap(unsigned long addr) { - down_read(&profile_rwsem); - notifier_call_chain(&munmap_notifier, 0, (void *)addr); - up_read(&profile_rwsem); + blocking_notifier_call_chain(&munmap_notifier, 0, (void *)addr); } int task_handoff_register(struct notifier_block * n) { - int err = -EINVAL; - - write_lock(&handoff_lock); - err = notifier_chain_register(&task_free_notifier, n); - write_unlock(&handoff_lock); - return err; + return atomic_notifier_chain_register(&task_free_notifier, n); } int task_handoff_unregister(struct notifier_block * n) { - int err = -EINVAL; - - write_lock(&handoff_lock); - err = notifier_chain_unregister(&task_free_notifier, n); - write_unlock(&handoff_lock); - return err; + return atomic_notifier_chain_unregister(&task_free_notifier, n); } int profile_event_register(enum profile_type type, struct notifier_block * n) { int err = -EINVAL; - down_write(&profile_rwsem); - switch (type) { case PROFILE_TASK_EXIT: - err = notifier_chain_register(&task_exit_notifier, n); + err = blocking_notifier_chain_register( + &task_exit_notifier, n); break; case PROFILE_MUNMAP: - err = notifier_chain_register(&munmap_notifier, n); + err = blocking_notifier_chain_register( + &munmap_notifier, n); break; } - up_write(&profile_rwsem); - return err; } @@ -161,18 +141,17 @@ int profile_event_unregister(enum profile_type type, struct notifier_block * n) { int err = -EINVAL; - down_write(&profile_rwsem); - switch (type) { case PROFILE_TASK_EXIT: - err = notifier_chain_unregister(&task_exit_notifier, n); + err = blocking_notifier_chain_unregister( + &task_exit_notifier, n); break; case PROFILE_MUNMAP: - err = notifier_chain_unregister(&munmap_notifier, n); + err = blocking_notifier_chain_unregister( + &munmap_notifier, n); break; } - up_write(&profile_rwsem); return err; } diff --git a/kernel/softlockup.c b/kernel/softlockup.c index d9b3d5847ed..ced91e1ff56 100644 --- a/kernel/softlockup.c +++ b/kernel/softlockup.c @@ -152,5 +152,5 @@ __init void spawn_softlockup_task(void) cpu_callback(&cpu_nfb, CPU_ONLINE, cpu); register_cpu_notifier(&cpu_nfb); - notifier_chain_register(&panic_notifier_list, &panic_block); + atomic_notifier_chain_register(&panic_notifier_list, &panic_block); } diff --git a/kernel/sys.c b/kernel/sys.c index 38bc73ede2b..c93d37f71ae 100644 --- a/kernel/sys.c +++ b/kernel/sys.c @@ -95,99 +95,304 @@ int cad_pid = 1; * and the like. */ -static struct notifier_block *reboot_notifier_list; -static DEFINE_RWLOCK(notifier_lock); +static BLOCKING_NOTIFIER_HEAD(reboot_notifier_list); + +/* + * Notifier chain core routines. The exported routines below + * are layered on top of these, with appropriate locking added. + */ + +static int notifier_chain_register(struct notifier_block **nl, + struct notifier_block *n) +{ + while ((*nl) != NULL) { + if (n->priority > (*nl)->priority) + break; + nl = &((*nl)->next); + } + n->next = *nl; + rcu_assign_pointer(*nl, n); + return 0; +} + +static int notifier_chain_unregister(struct notifier_block **nl, + struct notifier_block *n) +{ + while ((*nl) != NULL) { + if ((*nl) == n) { + rcu_assign_pointer(*nl, n->next); + return 0; + } + nl = &((*nl)->next); + } + return -ENOENT; +} + +static int __kprobes notifier_call_chain(struct notifier_block **nl, + unsigned long val, void *v) +{ + int ret = NOTIFY_DONE; + struct notifier_block *nb; + + nb = rcu_dereference(*nl); + while (nb) { + ret = nb->notifier_call(nb, val, v); + if ((ret & NOTIFY_STOP_MASK) == NOTIFY_STOP_MASK) + break; + nb = rcu_dereference(nb->next); + } + return ret; +} + +/* + * Atomic notifier chain routines. Registration and unregistration + * use a mutex, and call_chain is synchronized by RCU (no locks). + */ /** - * notifier_chain_register - Add notifier to a notifier chain - * @list: Pointer to root list pointer + * atomic_notifier_chain_register - Add notifier to an atomic notifier chain + * @nh: Pointer to head of the atomic notifier chain * @n: New entry in notifier chain * - * Adds a notifier to a notifier chain. + * Adds a notifier to an atomic notifier chain. * * Currently always returns zero. */ + +int atomic_notifier_chain_register(struct atomic_notifier_head *nh, + struct notifier_block *n) +{ + unsigned long flags; + int ret; + + spin_lock_irqsave(&nh->lock, flags); + ret = notifier_chain_register(&nh->head, n); + spin_unlock_irqrestore(&nh->lock, flags); + return ret; +} + +EXPORT_SYMBOL_GPL(atomic_notifier_chain_register); + +/** + * atomic_notifier_chain_unregister - Remove notifier from an atomic notifier chain + * @nh: Pointer to head of the atomic notifier chain + * @n: Entry to remove from notifier chain + * + * Removes a notifier from an atomic notifier chain. + * + * Returns zero on success or %-ENOENT on failure. + */ +int atomic_notifier_chain_unregister(struct atomic_notifier_head *nh, + struct notifier_block *n) +{ + unsigned long flags; + int ret; + + spin_lock_irqsave(&nh->lock, flags); + ret = notifier_chain_unregister(&nh->head, n); + spin_unlock_irqrestore(&nh->lock, flags); + synchronize_rcu(); + return ret; +} + +EXPORT_SYMBOL_GPL(atomic_notifier_chain_unregister); + +/** + * atomic_notifier_call_chain - Call functions in an atomic notifier chain + * @nh: Pointer to head of the atomic notifier chain + * @val: Value passed unmodified to notifier function + * @v: Pointer passed unmodified to notifier function + * + * Calls each function in a notifier chain in turn. The functions + * run in an atomic context, so they must not block. + * This routine uses RCU to synchronize with changes to the chain. + * + * If the return value of the notifier can be and'ed + * with %NOTIFY_STOP_MASK then atomic_notifier_call_chain + * will return immediately, with the return value of + * the notifier function which halted execution. + * Otherwise the return value is the return value + * of the last notifier function called. + */ -int notifier_chain_register(struct notifier_block **list, struct notifier_block *n) +int atomic_notifier_call_chain(struct atomic_notifier_head *nh, + unsigned long val, void *v) { - write_lock(¬ifier_lock); - while(*list) - { - if(n->priority > (*list)->priority) - break; - list= &((*list)->next); - } - n->next = *list; - *list=n; - write_unlock(¬ifier_lock); - return 0; + int ret; + + rcu_read_lock(); + ret = notifier_call_chain(&nh->head, val, v); + rcu_read_unlock(); + return ret; } -EXPORT_SYMBOL(notifier_chain_register); +EXPORT_SYMBOL_GPL(atomic_notifier_call_chain); + +/* + * Blocking notifier chain routines. All access to the chain is + * synchronized by an rwsem. + */ /** - * notifier_chain_unregister - Remove notifier from a notifier chain - * @nl: Pointer to root list pointer + * blocking_notifier_chain_register - Add notifier to a blocking notifier chain + * @nh: Pointer to head of the blocking notifier chain * @n: New entry in notifier chain * - * Removes a notifier from a notifier chain. + * Adds a notifier to a blocking notifier chain. + * Must be called in process context. * - * Returns zero on success, or %-ENOENT on failure. + * Currently always returns zero. */ -int notifier_chain_unregister(struct notifier_block **nl, struct notifier_block *n) +int blocking_notifier_chain_register(struct blocking_notifier_head *nh, + struct notifier_block *n) { - write_lock(¬ifier_lock); - while((*nl)!=NULL) - { - if((*nl)==n) - { - *nl=n->next; - write_unlock(¬ifier_lock); - return 0; - } - nl=&((*nl)->next); - } - write_unlock(¬ifier_lock); - return -ENOENT; + int ret; + + /* + * This code gets used during boot-up, when task switching is + * not yet working and interrupts must remain disabled. At + * such times we must not call down_write(). + */ + if (unlikely(system_state == SYSTEM_BOOTING)) + return notifier_chain_register(&nh->head, n); + + down_write(&nh->rwsem); + ret = notifier_chain_register(&nh->head, n); + up_write(&nh->rwsem); + return ret; } -EXPORT_SYMBOL(notifier_chain_unregister); +EXPORT_SYMBOL_GPL(blocking_notifier_chain_register); /** - * notifier_call_chain - Call functions in a notifier chain - * @n: Pointer to root pointer of notifier chain + * blocking_notifier_chain_unregister - Remove notifier from a blocking notifier chain + * @nh: Pointer to head of the blocking notifier chain + * @n: Entry to remove from notifier chain + * + * Removes a notifier from a blocking notifier chain. + * Must be called from process context. + * + * Returns zero on success or %-ENOENT on failure. + */ +int blocking_notifier_chain_unregister(struct blocking_notifier_head *nh, + struct notifier_block *n) +{ + int ret; + + /* + * This code gets used during boot-up, when task switching is + * not yet working and interrupts must remain disabled. At + * such times we must not call down_write(). + */ + if (unlikely(system_state == SYSTEM_BOOTING)) + return notifier_chain_unregister(&nh->head, n); + + down_write(&nh->rwsem); + ret = notifier_chain_unregister(&nh->head, n); + up_write(&nh->rwsem); + return ret; +} + +EXPORT_SYMBOL_GPL(blocking_notifier_chain_unregister); + +/** + * blocking_notifier_call_chain - Call functions in a blocking notifier chain + * @nh: Pointer to head of the blocking notifier chain * @val: Value passed unmodified to notifier function * @v: Pointer passed unmodified to notifier function * - * Calls each function in a notifier chain in turn. + * Calls each function in a notifier chain in turn. The functions + * run in a process context, so they are allowed to block. * - * If the return value of the notifier can be and'd - * with %NOTIFY_STOP_MASK, then notifier_call_chain + * If the return value of the notifier can be and'ed + * with %NOTIFY_STOP_MASK then blocking_notifier_call_chain * will return immediately, with the return value of * the notifier function which halted execution. - * Otherwise, the return value is the return value + * Otherwise the return value is the return value * of the last notifier function called. */ -int __kprobes notifier_call_chain(struct notifier_block **n, unsigned long val, void *v) +int blocking_notifier_call_chain(struct blocking_notifier_head *nh, + unsigned long val, void *v) { - int ret=NOTIFY_DONE; - struct notifier_block *nb = *n; + int ret; - while(nb) - { - ret=nb->notifier_call(nb,val,v); - if(ret&NOTIFY_STOP_MASK) - { - return ret; - } - nb=nb->next; - } + down_read(&nh->rwsem); + ret = notifier_call_chain(&nh->head, val, v); + up_read(&nh->rwsem); return ret; } -EXPORT_SYMBOL(notifier_call_chain); +EXPORT_SYMBOL_GPL(blocking_notifier_call_chain); + +/* + * Raw notifier chain routines. There is no protection; + * the caller must provide it. Use at your own risk! + */ + +/** + * raw_notifier_chain_register - Add notifier to a raw notifier chain + * @nh: Pointer to head of the raw notifier chain + * @n: New entry in notifier chain + * + * Adds a notifier to a raw notifier chain. + * All locking must be provided by the caller. + * + * Currently always returns zero. + */ + +int raw_notifier_chain_register(struct raw_notifier_head *nh, + struct notifier_block *n) +{ + return notifier_chain_register(&nh->head, n); +} + +EXPORT_SYMBOL_GPL(raw_notifier_chain_register); + +/** + * raw_notifier_chain_unregister - Remove notifier from a raw notifier chain + * @nh: Pointer to head of the raw notifier chain + * @n: Entry to remove from notifier chain + * + * Removes a notifier from a raw notifier chain. + * All locking must be provided by the caller. + * + * Returns zero on success or %-ENOENT on failure. + */ +int raw_notifier_chain_unregister(struct raw_notifier_head *nh, + struct notifier_block *n) +{ + return notifier_chain_unregister(&nh->head, n); +} + +EXPORT_SYMBOL_GPL(raw_notifier_chain_unregister); + +/** + * raw_notifier_call_chain - Call functions in a raw notifier chain + * @nh: Pointer to head of the raw notifier chain + * @val: Value passed unmodified to notifier function + * @v: Pointer passed unmodified to notifier function + * + * Calls each function in a notifier chain in turn. The functions + * run in an undefined context. + * All locking must be provided by the caller. + * + * If the return value of the notifier can be and'ed + * with %NOTIFY_STOP_MASK then raw_notifier_call_chain + * will return immediately, with the return value of + * the notifier function which halted execution. + * Otherwise the return value is the return value + * of the last notifier function called. + */ + +int raw_notifier_call_chain(struct raw_notifier_head *nh, + unsigned long val, void *v) +{ + return notifier_call_chain(&nh->head, val, v); +} + +EXPORT_SYMBOL_GPL(raw_notifier_call_chain); /** * register_reboot_notifier - Register function to be called at reboot time @@ -196,13 +401,13 @@ EXPORT_SYMBOL(notifier_call_chain); * Registers a function with the list of functions * to be called at reboot time. * - * Currently always returns zero, as notifier_chain_register + * Currently always returns zero, as blocking_notifier_chain_register * always returns zero. */ int register_reboot_notifier(struct notifier_block * nb) { - return notifier_chain_register(&reboot_notifier_list, nb); + return blocking_notifier_chain_register(&reboot_notifier_list, nb); } EXPORT_SYMBOL(register_reboot_notifier); @@ -219,7 +424,7 @@ EXPORT_SYMBOL(register_reboot_notifier); int unregister_reboot_notifier(struct notifier_block * nb) { - return notifier_chain_unregister(&reboot_notifier_list, nb); + return blocking_notifier_chain_unregister(&reboot_notifier_list, nb); } EXPORT_SYMBOL(unregister_reboot_notifier); @@ -380,7 +585,7 @@ EXPORT_SYMBOL_GPL(emergency_restart); void kernel_restart_prepare(char *cmd) { - notifier_call_chain(&reboot_notifier_list, SYS_RESTART, cmd); + blocking_notifier_call_chain(&reboot_notifier_list, SYS_RESTART, cmd); system_state = SYSTEM_RESTART; device_shutdown(); } @@ -430,7 +635,7 @@ EXPORT_SYMBOL_GPL(kernel_kexec); void kernel_shutdown_prepare(enum system_states state) { - notifier_call_chain(&reboot_notifier_list, + blocking_notifier_call_chain(&reboot_notifier_list, (state == SYSTEM_HALT)?SYS_HALT:SYS_POWER_OFF, NULL); system_state = state; device_shutdown(); diff --git a/net/bluetooth/hci_core.c b/net/bluetooth/hci_core.c index 9106354c781..a49a6975092 100644 --- a/net/bluetooth/hci_core.c +++ b/net/bluetooth/hci_core.c @@ -73,23 +73,23 @@ DEFINE_RWLOCK(hci_cb_list_lock); struct hci_proto *hci_proto[HCI_MAX_PROTO]; /* HCI notifiers list */ -static struct notifier_block *hci_notifier; +static ATOMIC_NOTIFIER_HEAD(hci_notifier); /* ---- HCI notifications ---- */ int hci_register_notifier(struct notifier_block *nb) { - return notifier_chain_register(&hci_notifier, nb); + return atomic_notifier_chain_register(&hci_notifier, nb); } int hci_unregister_notifier(struct notifier_block *nb) { - return notifier_chain_unregister(&hci_notifier, nb); + return atomic_notifier_chain_unregister(&hci_notifier, nb); } static void hci_notify(struct hci_dev *hdev, int event) { - notifier_call_chain(&hci_notifier, event, hdev); + atomic_notifier_call_chain(&hci_notifier, event, hdev); } /* ---- HCI requests ---- */ diff --git a/net/core/dev.c b/net/core/dev.c index 8e1dc305122..a3ab11f3415 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -193,7 +193,7 @@ static inline struct hlist_head *dev_index_hash(int ifindex) * Our notifier list */ -static struct notifier_block *netdev_chain; +static BLOCKING_NOTIFIER_HEAD(netdev_chain); /* * Device drivers call our routines to queue packets here. We empty the @@ -736,7 +736,8 @@ int dev_change_name(struct net_device *dev, char *newname) if (!err) { hlist_del(&dev->name_hlist); hlist_add_head(&dev->name_hlist, dev_name_hash(dev->name)); - notifier_call_chain(&netdev_chain, NETDEV_CHANGENAME, dev); + blocking_notifier_call_chain(&netdev_chain, + NETDEV_CHANGENAME, dev); } return err; @@ -750,7 +751,7 @@ int dev_change_name(struct net_device *dev, char *newname) */ void netdev_features_change(struct net_device *dev) { - notifier_call_chain(&netdev_chain, NETDEV_FEAT_CHANGE, dev); + blocking_notifier_call_chain(&netdev_chain, NETDEV_FEAT_CHANGE, dev); } EXPORT_SYMBOL(netdev_features_change); @@ -765,7 +766,8 @@ EXPORT_SYMBOL(netdev_features_change); void netdev_state_change(struct net_device *dev) { if (dev->flags & IFF_UP) { - notifier_call_chain(&netdev_chain, NETDEV_CHANGE, dev); + blocking_notifier_call_chain(&netdev_chain, + NETDEV_CHANGE, dev); rtmsg_ifinfo(RTM_NEWLINK, dev, 0); } } @@ -862,7 +864,7 @@ int dev_open(struct net_device *dev) /* * ... and announce new interface. */ - notifier_call_chain(&netdev_chain, NETDEV_UP, dev); + blocking_notifier_call_chain(&netdev_chain, NETDEV_UP, dev); } return ret; } @@ -885,7 +887,7 @@ int dev_close(struct net_device *dev) * Tell people we are going down, so that they can * prepare to death, when device is still operating. */ - notifier_call_chain(&netdev_chain, NETDEV_GOING_DOWN, dev); + blocking_notifier_call_chain(&netdev_chain, NETDEV_GOING_DOWN, dev); dev_deactivate(dev); @@ -922,7 +924,7 @@ int dev_close(struct net_device *dev) /* * Tell people we are down */ - notifier_call_chain(&netdev_chain, NETDEV_DOWN, dev); + blocking_notifier_call_chain(&netdev_chain, NETDEV_DOWN, dev); return 0; } @@ -953,7 +955,7 @@ int register_netdevice_notifier(struct notifier_block *nb) int err; rtnl_lock(); - err = notifier_chain_register(&netdev_chain, nb); + err = blocking_notifier_chain_register(&netdev_chain, nb); if (!err) { for (dev = dev_base; dev; dev = dev->next) { nb->notifier_call(nb, NETDEV_REGISTER, dev); @@ -981,7 +983,7 @@ int unregister_netdevice_notifier(struct notifier_block *nb) int err; rtnl_lock(); - err = notifier_chain_unregister(&netdev_chain, nb); + err = blocking_notifier_chain_unregister(&netdev_chain, nb); rtnl_unlock(); return err; } @@ -992,12 +994,12 @@ int unregister_netdevice_notifier(struct notifier_block *nb) * @v: pointer passed unmodified to notifier function * * Call all network notifier blocks. Parameters and return value - * are as for notifier_call_chain(). + * are as for blocking_notifier_call_chain(). */ int call_netdevice_notifiers(unsigned long val, void *v) { - return notifier_call_chain(&netdev_chain, val, v); + return blocking_notifier_call_chain(&netdev_chain, val, v); } /* When > 0 there are consumers of rx skb time stamps */ @@ -2242,7 +2244,8 @@ int dev_change_flags(struct net_device *dev, unsigned flags) if (dev->flags & IFF_UP && ((old_flags ^ dev->flags) &~ (IFF_UP | IFF_PROMISC | IFF_ALLMULTI | IFF_VOLATILE))) - notifier_call_chain(&netdev_chain, NETDEV_CHANGE, dev); + blocking_notifier_call_chain(&netdev_chain, + NETDEV_CHANGE, dev); if ((flags ^ dev->gflags) & IFF_PROMISC) { int inc = (flags & IFF_PROMISC) ? +1 : -1; @@ -2286,8 +2289,8 @@ int dev_set_mtu(struct net_device *dev, int new_mtu) else dev->mtu = new_mtu; if (!err && dev->flags & IFF_UP) - notifier_call_chain(&netdev_chain, - NETDEV_CHANGEMTU, dev); + blocking_notifier_call_chain(&netdev_chain, + NETDEV_CHANGEMTU, dev); return err; } @@ -2303,7 +2306,8 @@ int dev_set_mac_address(struct net_device *dev, struct sockaddr *sa) return -ENODEV; err = dev->set_mac_address(dev, sa); if (!err) - notifier_call_chain(&netdev_chain, NETDEV_CHANGEADDR, dev); + blocking_notifier_call_chain(&netdev_chain, + NETDEV_CHANGEADDR, dev); return err; } @@ -2359,7 +2363,7 @@ static int dev_ifsioc(struct ifreq *ifr, unsigned int cmd) return -EINVAL; memcpy(dev->broadcast, ifr->ifr_hwaddr.sa_data, min(sizeof ifr->ifr_hwaddr.sa_data, (size_t) dev->addr_len)); - notifier_call_chain(&netdev_chain, + blocking_notifier_call_chain(&netdev_chain, NETDEV_CHANGEADDR, dev); return 0; @@ -2813,7 +2817,7 @@ int register_netdevice(struct net_device *dev) write_unlock_bh(&dev_base_lock); /* Notify protocols, that a new device appeared. */ - notifier_call_chain(&netdev_chain, NETDEV_REGISTER, dev); + blocking_notifier_call_chain(&netdev_chain, NETDEV_REGISTER, dev); /* Finish registration after unlock */ net_set_todo(dev); @@ -2892,7 +2896,7 @@ static void netdev_wait_allrefs(struct net_device *dev) rtnl_lock(); /* Rebroadcast unregister notification */ - notifier_call_chain(&netdev_chain, + blocking_notifier_call_chain(&netdev_chain, NETDEV_UNREGISTER, dev); if (test_bit(__LINK_STATE_LINKWATCH_PENDING, @@ -3148,7 +3152,7 @@ int unregister_netdevice(struct net_device *dev) /* Notify protocols, that we are about to destroy this device. They should clean all the things. */ - notifier_call_chain(&netdev_chain, NETDEV_UNREGISTER, dev); + blocking_notifier_call_chain(&netdev_chain, NETDEV_UNREGISTER, dev); /* * Flush the multicast chain diff --git a/net/decnet/dn_dev.c b/net/decnet/dn_dev.c index cc7b9d9255e..d2ae9893ca1 100644 --- a/net/decnet/dn_dev.c +++ b/net/decnet/dn_dev.c @@ -68,7 +68,7 @@ __le16 decnet_address = 0; static DEFINE_RWLOCK(dndev_lock); static struct net_device *decnet_default_device; -static struct notifier_block *dnaddr_chain; +static BLOCKING_NOTIFIER_HEAD(dnaddr_chain); static struct dn_dev *dn_dev_create(struct net_device *dev, int *err); static void dn_dev_delete(struct net_device *dev); @@ -446,7 +446,7 @@ static void dn_dev_del_ifa(struct dn_dev *dn_db, struct dn_ifaddr **ifap, int de } rtmsg_ifa(RTM_DELADDR, ifa1); - notifier_call_chain(&dnaddr_chain, NETDEV_DOWN, ifa1); + blocking_notifier_call_chain(&dnaddr_chain, NETDEV_DOWN, ifa1); if (destroy) { dn_dev_free_ifa(ifa1); @@ -481,7 +481,7 @@ static int dn_dev_insert_ifa(struct dn_dev *dn_db, struct dn_ifaddr *ifa) dn_db->ifa_list = ifa; rtmsg_ifa(RTM_NEWADDR, ifa); - notifier_call_chain(&dnaddr_chain, NETDEV_UP, ifa); + blocking_notifier_call_chain(&dnaddr_chain, NETDEV_UP, ifa); return 0; } @@ -1285,12 +1285,12 @@ void dn_dev_devices_on(void) int register_dnaddr_notifier(struct notifier_block *nb) { - return notifier_chain_register(&dnaddr_chain, nb); + return blocking_notifier_chain_register(&dnaddr_chain, nb); } int unregister_dnaddr_notifier(struct notifier_block *nb) { - return notifier_chain_unregister(&dnaddr_chain, nb); + return blocking_notifier_chain_unregister(&dnaddr_chain, nb); } #ifdef CONFIG_PROC_FS diff --git a/net/ipv4/devinet.c b/net/ipv4/devinet.c index 44fdf1413e2..81c2f788529 100644 --- a/net/ipv4/devinet.c +++ b/net/ipv4/devinet.c @@ -81,7 +81,7 @@ static struct ipv4_devconf ipv4_devconf_dflt = { static void rtmsg_ifa(int event, struct in_ifaddr *); -static struct notifier_block *inetaddr_chain; +static BLOCKING_NOTIFIER_HEAD(inetaddr_chain); static void inet_del_ifa(struct in_device *in_dev, struct in_ifaddr **ifap, int destroy); #ifdef CONFIG_SYSCTL @@ -267,7 +267,8 @@ static void inet_del_ifa(struct in_device *in_dev, struct in_ifaddr **ifap, *ifap1 = ifa->ifa_next; rtmsg_ifa(RTM_DELADDR, ifa); - notifier_call_chain(&inetaddr_chain, NETDEV_DOWN, ifa); + blocking_notifier_call_chain(&inetaddr_chain, + NETDEV_DOWN, ifa); inet_free_ifa(ifa); } else { promote = ifa; @@ -291,7 +292,7 @@ static void inet_del_ifa(struct in_device *in_dev, struct in_ifaddr **ifap, So that, this order is correct. */ rtmsg_ifa(RTM_DELADDR, ifa1); - notifier_call_chain(&inetaddr_chain, NETDEV_DOWN, ifa1); + blocking_notifier_call_chain(&inetaddr_chain, NETDEV_DOWN, ifa1); if (promote) { @@ -303,7 +304,8 @@ static void inet_del_ifa(struct in_device *in_dev, struct in_ifaddr **ifap, promote->ifa_flags &= ~IFA_F_SECONDARY; rtmsg_ifa(RTM_NEWADDR, promote); - notifier_call_chain(&inetaddr_chain, NETDEV_UP, promote); + blocking_notifier_call_chain(&inetaddr_chain, + NETDEV_UP, promote); for (ifa = promote->ifa_next; ifa; ifa = ifa->ifa_next) { if (ifa1->ifa_mask != ifa->ifa_mask || !inet_ifa_match(ifa1->ifa_address, ifa)) @@ -366,7 +368,7 @@ static int inet_insert_ifa(struct in_ifaddr *ifa) Notifier will trigger FIB update, so that listeners of netlink will know about new ifaddr */ rtmsg_ifa(RTM_NEWADDR, ifa); - notifier_call_chain(&inetaddr_chain, NETDEV_UP, ifa); + blocking_notifier_call_chain(&inetaddr_chain, NETDEV_UP, ifa); return 0; } @@ -938,12 +940,12 @@ u32 inet_confirm_addr(const struct net_device *dev, u32 dst, u32 local, int scop int register_inetaddr_notifier(struct notifier_block *nb) { - return notifier_chain_register(&inetaddr_chain, nb); + return blocking_notifier_chain_register(&inetaddr_chain, nb); } int unregister_inetaddr_notifier(struct notifier_block *nb) { - return notifier_chain_unregister(&inetaddr_chain, nb); + return blocking_notifier_chain_unregister(&inetaddr_chain, nb); } /* Rename ifa_labels for a device name change. Make some effort to preserve existing diff --git a/net/ipv4/netfilter/ip_conntrack_core.c b/net/ipv4/netfilter/ip_conntrack_core.c index 9e34034729a..ceaabc18202 100644 --- a/net/ipv4/netfilter/ip_conntrack_core.c +++ b/net/ipv4/netfilter/ip_conntrack_core.c @@ -80,8 +80,8 @@ static int ip_conntrack_vmalloc; static unsigned int ip_conntrack_next_id; static unsigned int ip_conntrack_expect_next_id; #ifdef CONFIG_IP_NF_CONNTRACK_EVENTS -struct notifier_block *ip_conntrack_chain; -struct notifier_block *ip_conntrack_expect_chain; +ATOMIC_NOTIFIER_HEAD(ip_conntrack_chain); +ATOMIC_NOTIFIER_HEAD(ip_conntrack_expect_chain); DEFINE_PER_CPU(struct ip_conntrack_ecache, ip_conntrack_ecache); @@ -92,7 +92,7 @@ __ip_ct_deliver_cached_events(struct ip_conntrack_ecache *ecache) { DEBUGP("ecache: delivering events for %p\n", ecache->ct); if (is_confirmed(ecache->ct) && !is_dying(ecache->ct) && ecache->events) - notifier_call_chain(&ip_conntrack_chain, ecache->events, + atomic_notifier_call_chain(&ip_conntrack_chain, ecache->events, ecache->ct); ecache->events = 0; ip_conntrack_put(ecache->ct); diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index 01c62a0d374..445006ee452 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -143,7 +143,7 @@ static void inet6_prefix_notify(int event, struct inet6_dev *idev, struct prefix_info *pinfo); static int ipv6_chk_same_addr(const struct in6_addr *addr, struct net_device *dev); -static struct notifier_block *inet6addr_chain; +static ATOMIC_NOTIFIER_HEAD(inet6addr_chain); struct ipv6_devconf ipv6_devconf = { .forwarding = 0, @@ -593,7 +593,7 @@ out2: read_unlock_bh(&addrconf_lock); if (likely(err == 0)) - notifier_call_chain(&inet6addr_chain, NETDEV_UP, ifa); + atomic_notifier_call_chain(&inet6addr_chain, NETDEV_UP, ifa); else { kfree(ifa); ifa = ERR_PTR(err); @@ -688,7 +688,7 @@ static void ipv6_del_addr(struct inet6_ifaddr *ifp) ipv6_ifa_notify(RTM_DELADDR, ifp); - notifier_call_chain(&inet6addr_chain,NETDEV_DOWN,ifp); + atomic_notifier_call_chain(&inet6addr_chain, NETDEV_DOWN, ifp); addrconf_del_timer(ifp); @@ -3767,12 +3767,12 @@ static void addrconf_sysctl_unregister(struct ipv6_devconf *p) int register_inet6addr_notifier(struct notifier_block *nb) { - return notifier_chain_register(&inet6addr_chain, nb); + return atomic_notifier_chain_register(&inet6addr_chain, nb); } int unregister_inet6addr_notifier(struct notifier_block *nb) { - return notifier_chain_unregister(&inet6addr_chain,nb); + return atomic_notifier_chain_unregister(&inet6addr_chain,nb); } /* diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c index 0ae281d9bfc..56389c83557 100644 --- a/net/netfilter/nf_conntrack_core.c +++ b/net/netfilter/nf_conntrack_core.c @@ -90,8 +90,8 @@ static int nf_conntrack_vmalloc; static unsigned int nf_conntrack_next_id; static unsigned int nf_conntrack_expect_next_id; #ifdef CONFIG_NF_CONNTRACK_EVENTS -struct notifier_block *nf_conntrack_chain; -struct notifier_block *nf_conntrack_expect_chain; +ATOMIC_NOTIFIER_HEAD(nf_conntrack_chain); +ATOMIC_NOTIFIER_HEAD(nf_conntrack_expect_chain); DEFINE_PER_CPU(struct nf_conntrack_ecache, nf_conntrack_ecache); @@ -103,7 +103,7 @@ __nf_ct_deliver_cached_events(struct nf_conntrack_ecache *ecache) DEBUGP("ecache: delivering events for %p\n", ecache->ct); if (nf_ct_is_confirmed(ecache->ct) && !nf_ct_is_dying(ecache->ct) && ecache->events) - notifier_call_chain(&nf_conntrack_chain, ecache->events, + atomic_notifier_call_chain(&nf_conntrack_chain, ecache->events, ecache->ct); ecache->events = 0; diff --git a/net/netlink/af_netlink.c b/net/netlink/af_netlink.c index d00a9034cb5..2a233ffcf61 100644 --- a/net/netlink/af_netlink.c +++ b/net/netlink/af_netlink.c @@ -123,7 +123,7 @@ static void netlink_destroy_callback(struct netlink_callback *cb); static DEFINE_RWLOCK(nl_table_lock); static atomic_t nl_table_users = ATOMIC_INIT(0); -static struct notifier_block *netlink_chain; +static ATOMIC_NOTIFIER_HEAD(netlink_chain); static u32 netlink_group_mask(u32 group) { @@ -469,7 +469,8 @@ static int netlink_release(struct socket *sock) .protocol = sk->sk_protocol, .pid = nlk->pid, }; - notifier_call_chain(&netlink_chain, NETLINK_URELEASE, &n); + atomic_notifier_call_chain(&netlink_chain, + NETLINK_URELEASE, &n); } if (nlk->module) @@ -1695,12 +1696,12 @@ static struct file_operations netlink_seq_fops = { int netlink_register_notifier(struct notifier_block *nb) { - return notifier_chain_register(&netlink_chain, nb); + return atomic_notifier_chain_register(&netlink_chain, nb); } int netlink_unregister_notifier(struct notifier_block *nb) { - return notifier_chain_unregister(&netlink_chain, nb); + return atomic_notifier_chain_unregister(&netlink_chain, nb); } static const struct proto_ops netlink_ops = { -- cgit v1.2.3-70-g09d2 From 02dea0875b0f9b331a65fd6097dfd6115ca4ef24 Mon Sep 17 00:00:00 2001 From: Jeff Dike Date: Fri, 31 Mar 2006 02:30:08 -0800 Subject: [PATCH] UML: Hotplug memory, take 2 Changes since first version added check for MADV_REMOVE support on the host fixed error return botch shrunk sprintf array by one character This adds hotplug memory support to UML. The mconsole syntax is config mem=[+-]n[KMG] In other words, add or subtract some number of kilobytes, megabytes, or gigabytes. Unplugged pages are allocated and then madvise(MADV_TRUNCATE), which is a currently experimental madvise extension. These pages are tracked so they can be plugged back in later if the admin decides to give them back. The first page to be unplugged is used to keep track of about 4M of other pages. A list_head is the first thing on this page. The rest is filled with addresses of other unplugged pages. This first page is not madvised, obviously. When this page is filled, the next page is used in a similar way and linked onto a list with the first page. Etc. This whole process reverses when pages are plugged back in. When a tracking page no longer tracks any unplugged pages, then it is next in line for plugging, which is done by freeing pages back to the kernel. Signed-off-by: Jeff Dike Cc: Paolo 'Blaisorblade' Giarrusso Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/um/drivers/mconsole_kern.c | 138 ++++++++++++++++++++++++++++++++++++++++ arch/um/include/mem_user.h | 1 - arch/um/include/os.h | 2 + arch/um/os-Linux/mem.c | 27 +------- arch/um/os-Linux/process.c | 44 +++++++++++++ arch/um/os-Linux/start_up.c | 20 ------ 6 files changed, 185 insertions(+), 47 deletions(-) (limited to 'arch/um/drivers') diff --git a/arch/um/drivers/mconsole_kern.c b/arch/um/drivers/mconsole_kern.c index 1488816588e..d060fce4940 100644 --- a/arch/um/drivers/mconsole_kern.c +++ b/arch/um/drivers/mconsole_kern.c @@ -20,6 +20,8 @@ #include "linux/namei.h" #include "linux/proc_fs.h" #include "linux/syscalls.h" +#include "linux/list.h" +#include "linux/mm.h" #include "linux/console.h" #include "asm/irq.h" #include "asm/uaccess.h" @@ -347,6 +349,142 @@ static struct mc_device *mconsole_find_dev(char *name) return(NULL); } +#define UNPLUGGED_PER_PAGE \ + ((PAGE_SIZE - sizeof(struct list_head)) / sizeof(unsigned long)) + +struct unplugged_pages { + struct list_head list; + void *pages[UNPLUGGED_PER_PAGE]; +}; + +static unsigned long long unplugged_pages_count = 0; +static struct list_head unplugged_pages = LIST_HEAD_INIT(unplugged_pages); +static int unplug_index = UNPLUGGED_PER_PAGE; + +static int mem_config(char *str) +{ + unsigned long long diff; + int err = -EINVAL, i, add; + char *ret; + + if(str[0] != '=') + goto out; + + str++; + if(str[0] == '-') + add = 0; + else if(str[0] == '+'){ + add = 1; + } + else goto out; + + str++; + diff = memparse(str, &ret); + if(*ret != '\0') + goto out; + + diff /= PAGE_SIZE; + + for(i = 0; i < diff; i++){ + struct unplugged_pages *unplugged; + void *addr; + + if(add){ + if(list_empty(&unplugged_pages)) + break; + + unplugged = list_entry(unplugged_pages.next, + struct unplugged_pages, list); + if(unplug_index > 0) + addr = unplugged->pages[--unplug_index]; + else { + list_del(&unplugged->list); + addr = unplugged; + unplug_index = UNPLUGGED_PER_PAGE; + } + + free_page((unsigned long) addr); + unplugged_pages_count--; + } + else { + struct page *page; + + page = alloc_page(GFP_ATOMIC); + if(page == NULL) + break; + + unplugged = page_address(page); + if(unplug_index == UNPLUGGED_PER_PAGE){ + INIT_LIST_HEAD(&unplugged->list); + list_add(&unplugged->list, &unplugged_pages); + unplug_index = 0; + } + else { + struct list_head *entry = unplugged_pages.next; + addr = unplugged; + + unplugged = list_entry(entry, + struct unplugged_pages, + list); + unplugged->pages[unplug_index++] = addr; + err = os_drop_memory(addr, PAGE_SIZE); + if(err) + printk("Failed to release memory - " + "errno = %d\n", err); + } + + unplugged_pages_count++; + } + } + + err = 0; +out: + return err; +} + +static int mem_get_config(char *name, char *str, int size, char **error_out) +{ + char buf[sizeof("18446744073709551615")]; + int len = 0; + + sprintf(buf, "%ld", uml_physmem); + CONFIG_CHUNK(str, size, len, buf, 1); + + return len; +} + +static int mem_id(char **str, int *start_out, int *end_out) +{ + *start_out = 0; + *end_out = 0; + + return 0; +} + +static int mem_remove(int n) +{ + return -EBUSY; +} + +static struct mc_device mem_mc = { + .name = "mem", + .config = mem_config, + .get_config = mem_get_config, + .id = mem_id, + .remove = mem_remove, +}; + +static int mem_mc_init(void) +{ + if(can_drop_memory()) + mconsole_register_dev(&mem_mc); + else printk("Can't release memory to the host - memory hotplug won't " + "be supported\n"); + return 0; +} + +__initcall(mem_mc_init); + #define CONFIG_BUF_SIZE 64 static void mconsole_get_config(int (*get_config)(char *, char *, int, diff --git a/arch/um/include/mem_user.h b/arch/um/include/mem_user.h index a1064c5823b..a54514d2cc3 100644 --- a/arch/um/include/mem_user.h +++ b/arch/um/include/mem_user.h @@ -49,7 +49,6 @@ extern int iomem_size; extern unsigned long host_task_size; extern unsigned long task_size; -extern void check_devanon(void); extern int init_mem_user(void); extern void setup_memory(void *entry); extern unsigned long find_iomem(char *driver, unsigned long *len_out); diff --git a/arch/um/include/os.h b/arch/um/include/os.h index d3d1bc6074e..5fb84e889b2 100644 --- a/arch/um/include/os.h +++ b/arch/um/include/os.h @@ -205,6 +205,8 @@ extern int os_map_memory(void *virt, int fd, unsigned long long off, extern int os_protect_memory(void *addr, unsigned long len, int r, int w, int x); extern int os_unmap_memory(void *addr, int len); +extern int os_drop_memory(void *addr, int length); +extern int can_drop_memory(void); extern void os_flush_stdout(void); /* tt.c diff --git a/arch/um/os-Linux/mem.c b/arch/um/os-Linux/mem.c index 9d7d69a523b..6ab372da965 100644 --- a/arch/um/os-Linux/mem.c +++ b/arch/um/os-Linux/mem.c @@ -121,36 +121,11 @@ int create_tmp_file(unsigned long long len) return(fd); } -static int create_anon_file(unsigned long long len) -{ - void *addr; - int fd; - - fd = open("/dev/anon", O_RDWR); - if(fd < 0) { - perror("opening /dev/anon"); - exit(1); - } - - addr = mmap(NULL, len, PROT_READ | PROT_WRITE, MAP_PRIVATE, fd, 0); - if(addr == MAP_FAILED){ - perror("mapping physmem file"); - exit(1); - } - munmap(addr, len); - - return(fd); -} - -extern int have_devanon; - int create_mem_file(unsigned long long len) { int err, fd; - if(have_devanon) - fd = create_anon_file(len); - else fd = create_tmp_file(len); + fd = create_tmp_file(len); err = os_set_exec_close(fd, 1); if(err < 0){ diff --git a/arch/um/os-Linux/process.c b/arch/um/os-Linux/process.c index d261888f39c..8176b0b5204 100644 --- a/arch/um/os-Linux/process.c +++ b/arch/um/os-Linux/process.c @@ -11,6 +11,7 @@ #include #include #include +#include #include "ptrace_user.h" #include "os.h" #include "user.h" @@ -20,6 +21,7 @@ #include "kern_util.h" #include "longjmp.h" #include "skas_ptrace.h" +#include "kern_constants.h" #define ARBITRARY_ADDR -1 #define FAILURE_PID -1 @@ -187,6 +189,48 @@ int os_unmap_memory(void *addr, int len) return(0); } +#ifndef MADV_REMOVE +#define MADV_REMOVE 0x5 /* remove these pages & resources */ +#endif + +int os_drop_memory(void *addr, int length) +{ + int err; + + err = madvise(addr, length, MADV_REMOVE); + if(err < 0) + err = -errno; + return err; +} + +int can_drop_memory(void) +{ + void *addr; + int fd; + + printk("Checking host MADV_REMOVE support..."); + fd = create_mem_file(UM_KERN_PAGE_SIZE); + if(fd < 0){ + printk("Creating test memory file failed, err = %d\n", -fd); + return 0; + } + + addr = mmap64(NULL, UM_KERN_PAGE_SIZE, PROT_READ | PROT_WRITE, + MAP_PRIVATE, fd, 0); + if(addr == MAP_FAILED){ + printk("Mapping test memory file failed, err = %d\n", -errno); + return 0; + } + + if(madvise(addr, UM_KERN_PAGE_SIZE, MADV_REMOVE) != 0){ + printk("MADV_REMOVE failed, err = %d\n", -errno); + return 0; + } + + printk("OK\n"); + return 1; +} + void init_new_thread_stack(void *sig_stack, void (*usr1_handler)(int)) { int flags = 0, pages; diff --git a/arch/um/os-Linux/start_up.c b/arch/um/os-Linux/start_up.c index 32753131f8d..387e26af301 100644 --- a/arch/um/os-Linux/start_up.c +++ b/arch/um/os-Linux/start_up.c @@ -470,25 +470,6 @@ int can_do_skas(void) } #endif -int have_devanon = 0; - -/* Runs on boot kernel stack - already safe to use printk. */ - -void check_devanon(void) -{ - int fd; - - printk("Checking for /dev/anon on the host..."); - fd = open("/dev/anon", O_RDWR); - if(fd < 0){ - printk("Not available (open failed with errno %d)\n", errno); - return; - } - - printk("OK\n"); - have_devanon = 1; -} - int __init parse_iomem(char *str, int *add) { struct iomem_region *new; @@ -664,6 +645,5 @@ void os_check_bugs(void) { check_ptrace(); check_sigio(); - check_devanon(); } -- cgit v1.2.3-70-g09d2 From f4c57a78e2c49f188babf675ba0a9264b5374c26 Mon Sep 17 00:00:00 2001 From: Jeff Dike Date: Fri, 31 Mar 2006 02:30:10 -0800 Subject: [PATCH] uml: fix initcall return values A number of UML initcalls were improperly returning 1. Also removed any nearby emacs formatting comments. Signed-off-by: Jeff Dike Cc: Paolo 'Blaisorblade' Giarrusso Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/um/drivers/daemon_kern.c | 13 +------------ arch/um/drivers/mcast_kern.c | 13 +------------ arch/um/drivers/pcap_kern.c | 13 +------------ arch/um/drivers/slip_kern.c | 13 +------------ arch/um/drivers/slirp_kern.c | 13 +------------ arch/um/drivers/ubd_kern.c | 2 +- arch/um/os-Linux/drivers/ethertap_kern.c | 13 +------------ arch/um/os-Linux/drivers/tuntap_kern.c | 13 +------------ 8 files changed, 8 insertions(+), 85 deletions(-) (limited to 'arch/um/drivers') diff --git a/arch/um/drivers/daemon_kern.c b/arch/um/drivers/daemon_kern.c index a61b7b46bc0..53d09ed78b4 100644 --- a/arch/um/drivers/daemon_kern.c +++ b/arch/um/drivers/daemon_kern.c @@ -95,18 +95,7 @@ static struct transport daemon_transport = { static int register_daemon(void) { register_transport(&daemon_transport); - return(1); + return 0; } __initcall(register_daemon); - -/* - * Overrides for Emacs so that we follow Linus's tabbing style. - * Emacs will notice this stuff at the end of the file and automatically - * adjust the settings for this buffer only. This must remain at the end - * of the file. - * --------------------------------------------------------------------------- - * Local variables: - * c-file-style: "linux" - * End: - */ diff --git a/arch/um/drivers/mcast_kern.c b/arch/um/drivers/mcast_kern.c index c9b078fba03..3a7af18cf94 100644 --- a/arch/um/drivers/mcast_kern.c +++ b/arch/um/drivers/mcast_kern.c @@ -124,18 +124,7 @@ static struct transport mcast_transport = { static int register_mcast(void) { register_transport(&mcast_transport); - return(1); + return 0; } __initcall(register_mcast); - -/* - * Overrides for Emacs so that we follow Linus's tabbing style. - * Emacs will notice this stuff at the end of the file and automatically - * adjust the settings for this buffer only. This must remain at the end - * of the file. - * --------------------------------------------------------------------------- - * Local variables: - * c-file-style: "linux" - * End: - */ diff --git a/arch/um/drivers/pcap_kern.c b/arch/um/drivers/pcap_kern.c index 07c80f2156e..466ff2c2f91 100644 --- a/arch/um/drivers/pcap_kern.c +++ b/arch/um/drivers/pcap_kern.c @@ -106,18 +106,7 @@ static struct transport pcap_transport = { static int register_pcap(void) { register_transport(&pcap_transport); - return(1); + return 0; } __initcall(register_pcap); - -/* - * Overrides for Emacs so that we follow Linus's tabbing style. - * Emacs will notice this stuff at the end of the file and automatically - * adjust the settings for this buffer only. This must remain at the end - * of the file. - * --------------------------------------------------------------------------- - * Local variables: - * c-file-style: "linux" - * End: - */ diff --git a/arch/um/drivers/slip_kern.c b/arch/um/drivers/slip_kern.c index a62f5ef445c..163ee0d5f75 100644 --- a/arch/um/drivers/slip_kern.c +++ b/arch/um/drivers/slip_kern.c @@ -93,18 +93,7 @@ static struct transport slip_transport = { static int register_slip(void) { register_transport(&slip_transport); - return(1); + return 0; } __initcall(register_slip); - -/* - * Overrides for Emacs so that we follow Linus's tabbing style. - * Emacs will notice this stuff at the end of the file and automatically - * adjust the settings for this buffer only. This must remain at the end - * of the file. - * --------------------------------------------------------------------------- - * Local variables: - * c-file-style: "linux" - * End: - */ diff --git a/arch/um/drivers/slirp_kern.c b/arch/um/drivers/slirp_kern.c index 33d7982be5d..109c541e78f 100644 --- a/arch/um/drivers/slirp_kern.c +++ b/arch/um/drivers/slirp_kern.c @@ -116,18 +116,7 @@ static struct transport slirp_transport = { static int register_slirp(void) { register_transport(&slirp_transport); - return(1); + return 0; } __initcall(register_slirp); - -/* - * Overrides for Emacs so that we follow Linus's tabbing style. - * Emacs will notice this stuff at the end of the file and automatically - * adjust the settings for this buffer only. This must remain at the end - * of the file. - * --------------------------------------------------------------------------- - * Local variables: - * c-file-style: "linux" - * End: - */ diff --git a/arch/um/drivers/ubd_kern.c b/arch/um/drivers/ubd_kern.c index 0336575d244..0897852b09a 100644 --- a/arch/um/drivers/ubd_kern.c +++ b/arch/um/drivers/ubd_kern.c @@ -891,7 +891,7 @@ int ubd_driver_init(void){ SA_INTERRUPT, "ubd", ubd_dev); if(err != 0) printk(KERN_ERR "um_request_irq failed - errno = %d\n", -err); - return(err); + return 0; } device_initcall(ubd_driver_init); diff --git a/arch/um/os-Linux/drivers/ethertap_kern.c b/arch/um/os-Linux/drivers/ethertap_kern.c index 6ae4b19d9f5..768606bec23 100644 --- a/arch/um/os-Linux/drivers/ethertap_kern.c +++ b/arch/um/os-Linux/drivers/ethertap_kern.c @@ -102,18 +102,7 @@ static struct transport ethertap_transport = { static int register_ethertap(void) { register_transport(ðertap_transport); - return(1); + return 0; } __initcall(register_ethertap); - -/* - * Overrides for Emacs so that we follow Linus's tabbing style. - * Emacs will notice this stuff at the end of the file and automatically - * adjust the settings for this buffer only. This must remain at the end - * of the file. - * --------------------------------------------------------------------------- - * Local variables: - * c-file-style: "linux" - * End: - */ diff --git a/arch/um/os-Linux/drivers/tuntap_kern.c b/arch/um/os-Linux/drivers/tuntap_kern.c index 4202b9ebad4..190009a6f89 100644 --- a/arch/um/os-Linux/drivers/tuntap_kern.c +++ b/arch/um/os-Linux/drivers/tuntap_kern.c @@ -87,18 +87,7 @@ static struct transport tuntap_transport = { static int register_tuntap(void) { register_transport(&tuntap_transport); - return(1); + return 0; } __initcall(register_tuntap); - -/* - * Overrides for Emacs so that we follow Linus's tabbing style. - * Emacs will notice this stuff at the end of the file and automatically - * adjust the settings for this buffer only. This must remain at the end - * of the file. - * --------------------------------------------------------------------------- - * Local variables: - * c-file-style: "linux" - * End: - */ -- cgit v1.2.3-70-g09d2 From 4d338e1accfc3473f7e453427dfd4f1ebf4dbbe6 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Fri, 31 Mar 2006 02:30:15 -0800 Subject: [PATCH] uml: sparse cleanups misc sparse annotations Signed-off-by: Al Viro Signed-off-by: Jeff Dike Cc: Paolo 'Blaisorblade' Giarrusso Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/um/drivers/harddog_kern.c | 8 +++--- arch/um/drivers/hostaudio_kern.c | 10 +++---- arch/um/drivers/slirp_kern.c | 2 +- arch/um/include/line.h | 18 +++++-------- arch/um/include/sysdep-i386/checksum.h | 5 ++-- arch/um/kernel/exec_kern.c | 4 +-- arch/um/kernel/process_kern.c | 2 +- arch/um/kernel/ptrace.c | 34 +++++++++++------------- arch/um/kernel/syscall_kern.c | 4 +-- arch/um/kernel/trap_kern.c | 8 +++--- arch/um/sys-i386/ptrace.c | 28 ++++++++++---------- arch/um/sys-i386/signal.c | 48 ++++++++++++++++++---------------- arch/um/sys-i386/syscalls.c | 2 +- include/asm-um/ptrace-generic.h | 2 +- include/asm-um/thread_info.h | 16 ++++++------ include/asm-um/uaccess.h | 2 +- 16 files changed, 93 insertions(+), 100 deletions(-) (limited to 'arch/um/drivers') diff --git a/arch/um/drivers/harddog_kern.c b/arch/um/drivers/harddog_kern.c index 49acb2badf3..d18a974735e 100644 --- a/arch/um/drivers/harddog_kern.c +++ b/arch/um/drivers/harddog_kern.c @@ -104,7 +104,7 @@ static int harddog_release(struct inode *inode, struct file *file) extern int ping_watchdog(int fd); -static ssize_t harddog_write(struct file *file, const char *data, size_t len, +static ssize_t harddog_write(struct file *file, const char __user *data, size_t len, loff_t *ppos) { /* @@ -118,6 +118,7 @@ static ssize_t harddog_write(struct file *file, const char *data, size_t len, static int harddog_ioctl(struct inode *inode, struct file *file, unsigned int cmd, unsigned long arg) { + void __user *argp= (void __user *)arg; static struct watchdog_info ident = { WDIOC_SETTIMEOUT, 0, @@ -127,13 +128,12 @@ static int harddog_ioctl(struct inode *inode, struct file *file, default: return -ENOTTY; case WDIOC_GETSUPPORT: - if(copy_to_user((struct harddog_info *)arg, &ident, - sizeof(ident))) + if(copy_to_user(argp, &ident, sizeof(ident))) return -EFAULT; return 0; case WDIOC_GETSTATUS: case WDIOC_GETBOOTSTATUS: - return put_user(0,(int *)arg); + return put_user(0,(int __user *)argp); case WDIOC_KEEPALIVE: return(ping_watchdog(harddog_out_fd)); } diff --git a/arch/um/drivers/hostaudio_kern.c b/arch/um/drivers/hostaudio_kern.c index 59602b81b24..37232f908cd 100644 --- a/arch/um/drivers/hostaudio_kern.c +++ b/arch/um/drivers/hostaudio_kern.c @@ -67,8 +67,8 @@ MODULE_PARM_DESC(mixer, MIXER_HELP); /* /dev/dsp file operations */ -static ssize_t hostaudio_read(struct file *file, char *buffer, size_t count, - loff_t *ppos) +static ssize_t hostaudio_read(struct file *file, char __user *buffer, + size_t count, loff_t *ppos) { struct hostaudio_state *state = file->private_data; void *kbuf; @@ -94,7 +94,7 @@ static ssize_t hostaudio_read(struct file *file, char *buffer, size_t count, return(err); } -static ssize_t hostaudio_write(struct file *file, const char *buffer, +static ssize_t hostaudio_write(struct file *file, const char __user *buffer, size_t count, loff_t *ppos) { struct hostaudio_state *state = file->private_data; @@ -152,7 +152,7 @@ static int hostaudio_ioctl(struct inode *inode, struct file *file, case SNDCTL_DSP_CHANNELS: case SNDCTL_DSP_SUBDIVIDE: case SNDCTL_DSP_SETFRAGMENT: - if(get_user(data, (int *) arg)) + if(get_user(data, (int __user *) arg)) return(-EFAULT); break; default: @@ -168,7 +168,7 @@ static int hostaudio_ioctl(struct inode *inode, struct file *file, case SNDCTL_DSP_CHANNELS: case SNDCTL_DSP_SUBDIVIDE: case SNDCTL_DSP_SETFRAGMENT: - if(put_user(data, (int *) arg)) + if(put_user(data, (int __user *) arg)) return(-EFAULT); break; default: diff --git a/arch/um/drivers/slirp_kern.c b/arch/um/drivers/slirp_kern.c index 109c541e78f..95e50c943e1 100644 --- a/arch/um/drivers/slirp_kern.c +++ b/arch/um/drivers/slirp_kern.c @@ -77,7 +77,7 @@ static int slirp_setup(char *str, char **mac_out, void *data) int i=0; *init = ((struct slirp_init) - { argw : { { "slirp", NULL } } }); + { .argw = { { "slirp", NULL } } }); str = split_if_spec(str, mac_out, NULL); diff --git a/arch/um/include/line.h b/arch/um/include/line.h index 6f4d680dc1d..6ac0f8252e2 100644 --- a/arch/um/include/line.h +++ b/arch/um/include/line.h @@ -58,23 +58,17 @@ struct line { }; #define LINE_INIT(str, d) \ - { init_str : str, \ - init_pri : INIT_STATIC, \ - valid : 1, \ - throttled : 0, \ - lock : SPIN_LOCK_UNLOCKED, \ - buffer : NULL, \ - head : NULL, \ - tail : NULL, \ - sigio : 0, \ - driver : d, \ - have_irq : 0 } + { .init_str = str, \ + .init_pri = INIT_STATIC, \ + .valid = 1, \ + .lock = SPIN_LOCK_UNLOCKED, \ + .driver = d } struct lines { int num; }; -#define LINES_INIT(n) { num : n } +#define LINES_INIT(n) { .num = n } extern void line_close(struct tty_struct *tty, struct file * filp); extern int line_open(struct line *lines, struct tty_struct *tty); diff --git a/arch/um/include/sysdep-i386/checksum.h b/arch/um/include/sysdep-i386/checksum.h index 7d3d202d7ff..052bb061a97 100644 --- a/arch/um/include/sysdep-i386/checksum.h +++ b/arch/um/include/sysdep-i386/checksum.h @@ -48,7 +48,8 @@ unsigned int csum_partial_copy_nocheck(const unsigned char *src, unsigned char * */ static __inline__ -unsigned int csum_partial_copy_from_user(const unsigned char *src, unsigned char *dst, +unsigned int csum_partial_copy_from_user(const unsigned char __user *src, + unsigned char *dst, int len, int sum, int *err_ptr) { if(copy_from_user(dst, src, len)){ @@ -192,7 +193,7 @@ static __inline__ unsigned short int csum_ipv6_magic(struct in6_addr *saddr, */ #define HAVE_CSUM_COPY_USER static __inline__ unsigned int csum_and_copy_to_user(const unsigned char *src, - unsigned char *dst, + unsigned char __user *dst, int len, int sum, int *err_ptr) { if (access_ok(VERIFY_WRITE, dst, len)){ diff --git a/arch/um/kernel/exec_kern.c b/arch/um/kernel/exec_kern.c index 1ca84319317..f9b346e05b0 100644 --- a/arch/um/kernel/exec_kern.c +++ b/arch/um/kernel/exec_kern.c @@ -58,14 +58,14 @@ long um_execve(char *file, char __user *__user *argv, char __user *__user *env) return(err); } -long sys_execve(char *file, char __user *__user *argv, +long sys_execve(char __user *file, char __user *__user *argv, char __user *__user *env) { long error; char *filename; lock_kernel(); - filename = getname((char __user *) file); + filename = getname(file); error = PTR_ERR(filename); if (IS_ERR(filename)) goto out; error = execve1(filename, argv, env); diff --git a/arch/um/kernel/process_kern.c b/arch/um/kernel/process_kern.c index 3113cab8675..6922bdfb5ad 100644 --- a/arch/um/kernel/process_kern.c +++ b/arch/um/kernel/process_kern.c @@ -407,7 +407,7 @@ static int proc_read_sysemu(char *buf, char **start, off_t offset, int size,int return strlen(buf); } -static int proc_write_sysemu(struct file *file,const char *buf, unsigned long count,void *data) +static int proc_write_sysemu(struct file *file,const char __user *buf, unsigned long count,void *data) { char tmp[2]; diff --git a/arch/um/kernel/ptrace.c b/arch/um/kernel/ptrace.c index 98e09395c09..394582202ce 100644 --- a/arch/um/kernel/ptrace.c +++ b/arch/um/kernel/ptrace.c @@ -46,6 +46,7 @@ extern int poke_user(struct task_struct * child, long addr, long data); long arch_ptrace(struct task_struct *child, long request, long addr, long data) { int i, ret; + unsigned long __user *p = (void __user *)(unsigned long)data; switch (request) { /* when I and D space are separate, these will need to be fixed. */ @@ -58,7 +59,7 @@ long arch_ptrace(struct task_struct *child, long request, long addr, long data) copied = access_process_vm(child, addr, &tmp, sizeof(tmp), 0); if (copied != sizeof(tmp)) break; - ret = put_user(tmp, (unsigned long __user *) data); + ret = put_user(tmp, p); break; } @@ -136,15 +137,13 @@ long arch_ptrace(struct task_struct *child, long request, long addr, long data) #ifdef PTRACE_GETREGS case PTRACE_GETREGS: { /* Get all gp regs from the child. */ - if (!access_ok(VERIFY_WRITE, (unsigned long *)data, - MAX_REG_OFFSET)) { + if (!access_ok(VERIFY_WRITE, p, MAX_REG_OFFSET)) { ret = -EIO; break; } for ( i = 0; i < MAX_REG_OFFSET; i += sizeof(long) ) { - __put_user(getreg(child, i), - (unsigned long __user *) data); - data += sizeof(long); + __put_user(getreg(child, i), p); + p++; } ret = 0; break; @@ -153,15 +152,14 @@ long arch_ptrace(struct task_struct *child, long request, long addr, long data) #ifdef PTRACE_SETREGS case PTRACE_SETREGS: { /* Set all gp regs in the child. */ unsigned long tmp = 0; - if (!access_ok(VERIFY_READ, (unsigned *)data, - MAX_REG_OFFSET)) { + if (!access_ok(VERIFY_READ, p, MAX_REG_OFFSET)) { ret = -EIO; break; } for ( i = 0; i < MAX_REG_OFFSET; i += sizeof(long) ) { - __get_user(tmp, (unsigned long __user *) data); + __get_user(tmp, p); putreg(child, i, tmp); - data += sizeof(long); + p++; } ret = 0; break; @@ -188,13 +186,12 @@ long arch_ptrace(struct task_struct *child, long request, long addr, long data) break; #endif case PTRACE_FAULTINFO: { - /* Take the info from thread->arch->faultinfo, - * but transfer max. sizeof(struct ptrace_faultinfo). - * On i386, ptrace_faultinfo is smaller! - */ - ret = copy_to_user((unsigned long __user *) data, - &child->thread.arch.faultinfo, - sizeof(struct ptrace_faultinfo)); + /* Take the info from thread->arch->faultinfo, + * but transfer max. sizeof(struct ptrace_faultinfo). + * On i386, ptrace_faultinfo is smaller! + */ + ret = copy_to_user(p, &child->thread.arch.faultinfo, + sizeof(struct ptrace_faultinfo)); if(ret) break; break; @@ -204,8 +201,7 @@ long arch_ptrace(struct task_struct *child, long request, long addr, long data) case PTRACE_LDT: { struct ptrace_ldt ldt; - if(copy_from_user(&ldt, (unsigned long __user *) data, - sizeof(ldt))){ + if(copy_from_user(&ldt, p, sizeof(ldt))){ ret = -EIO; break; } diff --git a/arch/um/kernel/syscall_kern.c b/arch/um/kernel/syscall_kern.c index 8e1a3501ff4..37d3978337d 100644 --- a/arch/um/kernel/syscall_kern.c +++ b/arch/um/kernel/syscall_kern.c @@ -104,7 +104,7 @@ long sys_pipe(unsigned long __user * fildes) } -long sys_uname(struct old_utsname * name) +long sys_uname(struct old_utsname __user * name) { long err; if (!name) @@ -115,7 +115,7 @@ long sys_uname(struct old_utsname * name) return err?-EFAULT:0; } -long sys_olduname(struct oldold_utsname * name) +long sys_olduname(struct oldold_utsname __user * name) { long error; diff --git a/arch/um/kernel/trap_kern.c b/arch/um/kernel/trap_kern.c index d56046c2aba..02f6d4d8dc3 100644 --- a/arch/um/kernel/trap_kern.c +++ b/arch/um/kernel/trap_kern.c @@ -198,7 +198,7 @@ unsigned long segv(struct faultinfo fi, unsigned long ip, int is_user, void *sc) si.si_signo = SIGBUS; si.si_errno = 0; si.si_code = BUS_ADRERR; - si.si_addr = (void *)address; + si.si_addr = (void __user *)address; current->thread.arch.faultinfo = fi; force_sig_info(SIGBUS, &si, current); } else if (err == -ENOMEM) { @@ -207,7 +207,7 @@ unsigned long segv(struct faultinfo fi, unsigned long ip, int is_user, void *sc) } else { BUG_ON(err != -EFAULT); si.si_signo = SIGSEGV; - si.si_addr = (void *) address; + si.si_addr = (void __user *) address; current->thread.arch.faultinfo = fi; force_sig_info(SIGSEGV, &si, current); } @@ -220,8 +220,8 @@ void bad_segv(struct faultinfo fi, unsigned long ip) si.si_signo = SIGSEGV; si.si_code = SEGV_ACCERR; - si.si_addr = (void *) FAULT_ADDRESS(fi); - current->thread.arch.faultinfo = fi; + si.si_addr = (void __user *) FAULT_ADDRESS(fi); + current->thread.arch.faultinfo = fi; force_sig_info(SIGSEGV, &si, current); } diff --git a/arch/um/sys-i386/ptrace.c b/arch/um/sys-i386/ptrace.c index 8032a105949..ff94eded93e 100644 --- a/arch/um/sys-i386/ptrace.c +++ b/arch/um/sys-i386/ptrace.c @@ -124,22 +124,22 @@ unsigned long getreg(struct task_struct *child, int regno) int peek_user(struct task_struct *child, long addr, long data) { /* read the word at location addr in the USER area. */ - unsigned long tmp; + unsigned long tmp; - if ((addr & 3) || addr < 0) - return -EIO; + if ((addr & 3) || addr < 0) + return -EIO; - tmp = 0; /* Default return condition */ - if(addr < MAX_REG_OFFSET){ - tmp = getreg(child, addr); - } - else if((addr >= offsetof(struct user, u_debugreg[0])) && - (addr <= offsetof(struct user, u_debugreg[7]))){ - addr -= offsetof(struct user, u_debugreg[0]); - addr = addr >> 2; - tmp = child->thread.arch.debugregs[addr]; - } - return put_user(tmp, (unsigned long *) data); + tmp = 0; /* Default return condition */ + if(addr < MAX_REG_OFFSET){ + tmp = getreg(child, addr); + } + else if((addr >= offsetof(struct user, u_debugreg[0])) && + (addr <= offsetof(struct user, u_debugreg[7]))){ + addr -= offsetof(struct user, u_debugreg[0]); + addr = addr >> 2; + tmp = child->thread.arch.debugregs[addr]; + } + return put_user(tmp, (unsigned long __user *) data); } struct i387_fxsave_struct { diff --git a/arch/um/sys-i386/signal.c b/arch/um/sys-i386/signal.c index 33a40f5ef0d..f5d0e1c37ea 100644 --- a/arch/um/sys-i386/signal.c +++ b/arch/um/sys-i386/signal.c @@ -19,7 +19,7 @@ #include "skas.h" static int copy_sc_from_user_skas(struct pt_regs *regs, - struct sigcontext *from) + struct sigcontext __user *from) { struct sigcontext sc; unsigned long fpregs[HOST_FP_SIZE]; @@ -57,7 +57,7 @@ static int copy_sc_from_user_skas(struct pt_regs *regs, return(0); } -int copy_sc_to_user_skas(struct sigcontext *to, struct _fpstate *to_fp, +int copy_sc_to_user_skas(struct sigcontext *to, struct _fpstate __user *to_fp, struct pt_regs *regs, unsigned long sp) { struct sigcontext sc; @@ -92,7 +92,7 @@ int copy_sc_to_user_skas(struct sigcontext *to, struct _fpstate *to_fp, "errno = %d\n", err); return(1); } - to_fp = (to_fp ? to_fp : (struct _fpstate *) (to + 1)); + to_fp = (to_fp ? to_fp : (struct _fpstate __user *) (to + 1)); sc.fpstate = to_fp; if(err) @@ -113,10 +113,11 @@ int copy_sc_to_user_skas(struct sigcontext *to, struct _fpstate *to_fp, * saved pointer is in the kernel, but the sigcontext is in userspace, so we * copy_to_user it. */ -int copy_sc_from_user_tt(struct sigcontext *to, struct sigcontext *from, +int copy_sc_from_user_tt(struct sigcontext *to, struct sigcontext __user *from, int fpsize) { - struct _fpstate *to_fp, *from_fp; + struct _fpstate *to_fp; + struct _fpstate __user *from_fp; unsigned long sigs; int err; @@ -131,13 +132,14 @@ int copy_sc_from_user_tt(struct sigcontext *to, struct sigcontext *from, return(err); } -int copy_sc_to_user_tt(struct sigcontext *to, struct _fpstate *fp, +int copy_sc_to_user_tt(struct sigcontext *to, struct _fpstate __user *fp, struct sigcontext *from, int fpsize, unsigned long sp) { - struct _fpstate *to_fp, *from_fp; + struct _fpstate __user *to_fp; + struct _fpstate *from_fp; int err; - to_fp = (fp ? fp : (struct _fpstate *) (to + 1)); + to_fp = (fp ? fp : (struct _fpstate __user *) (to + 1)); from_fp = from->fpstate; err = copy_to_user(to, from, sizeof(*to)); @@ -165,7 +167,7 @@ static int copy_sc_from_user(struct pt_regs *to, void __user *from) return(ret); } -static int copy_sc_to_user(struct sigcontext *to, struct _fpstate *fp, +static int copy_sc_to_user(struct sigcontext *to, struct _fpstate __user *fp, struct pt_regs *from, unsigned long sp) { return(CHOOSE_MODE(copy_sc_to_user_tt(to, fp, UPT_SC(&from->regs), @@ -173,7 +175,7 @@ static int copy_sc_to_user(struct sigcontext *to, struct _fpstate *fp, copy_sc_to_user_skas(to, fp, from, sp))); } -static int copy_ucontext_to_user(struct ucontext *uc, struct _fpstate *fp, +static int copy_ucontext_to_user(struct ucontext __user *uc, struct _fpstate __user *fp, sigset_t *set, unsigned long sp) { int err = 0; @@ -188,7 +190,7 @@ static int copy_ucontext_to_user(struct ucontext *uc, struct _fpstate *fp, struct sigframe { - char *pretcode; + char __user *pretcode; int sig; struct sigcontext sc; struct _fpstate fpstate; @@ -198,10 +200,10 @@ struct sigframe struct rt_sigframe { - char *pretcode; + char __user *pretcode; int sig; - struct siginfo *pinfo; - void *puc; + struct siginfo __user *pinfo; + void __user *puc; struct siginfo info; struct ucontext uc; struct _fpstate fpstate; @@ -213,16 +215,16 @@ int setup_signal_stack_sc(unsigned long stack_top, int sig, sigset_t *mask) { struct sigframe __user *frame; - void *restorer; + void __user *restorer; unsigned long save_sp = PT_REGS_SP(regs); int err = 0; stack_top &= -8UL; - frame = (struct sigframe *) stack_top - 1; + frame = (struct sigframe __user *) stack_top - 1; if (!access_ok(VERIFY_WRITE, frame, sizeof(*frame))) return 1; - restorer = (void *) frame->retcode; + restorer = frame->retcode; if(ka->sa.sa_flags & SA_RESTORER) restorer = ka->sa.sa_restorer; @@ -278,16 +280,16 @@ int setup_signal_stack_si(unsigned long stack_top, int sig, siginfo_t *info, sigset_t *mask) { struct rt_sigframe __user *frame; - void *restorer; + void __user *restorer; unsigned long save_sp = PT_REGS_SP(regs); int err = 0; stack_top &= -8UL; - frame = (struct rt_sigframe *) stack_top - 1; + frame = (struct rt_sigframe __user *) stack_top - 1; if (!access_ok(VERIFY_WRITE, frame, sizeof(*frame))) return 1; - restorer = (void *) frame->retcode; + restorer = frame->retcode; if(ka->sa.sa_flags & SA_RESTORER) restorer = ka->sa.sa_restorer; @@ -333,7 +335,7 @@ err: long sys_sigreturn(struct pt_regs regs) { unsigned long sp = PT_REGS_SP(¤t->thread.regs); - struct sigframe __user *frame = (struct sigframe *)(sp - 8); + struct sigframe __user *frame = (struct sigframe __user *)(sp - 8); sigset_t set; struct sigcontext __user *sc = &frame->sc; unsigned long __user *oldmask = &sc->oldmask; @@ -365,8 +367,8 @@ long sys_sigreturn(struct pt_regs regs) long sys_rt_sigreturn(struct pt_regs regs) { - unsigned long __user sp = PT_REGS_SP(¤t->thread.regs); - struct rt_sigframe __user *frame = (struct rt_sigframe *) (sp - 4); + unsigned long sp = PT_REGS_SP(¤t->thread.regs); + struct rt_sigframe __user *frame = (struct rt_sigframe __user *) (sp - 4); sigset_t set; struct ucontext __user *uc = &frame->uc; int sig_size = _NSIG_WORDS * sizeof(unsigned long); diff --git a/arch/um/sys-i386/syscalls.c b/arch/um/sys-i386/syscalls.c index 83e9be820a8..1845123ed12 100644 --- a/arch/um/sys-i386/syscalls.c +++ b/arch/um/sys-i386/syscalls.c @@ -104,7 +104,7 @@ long sys_ipc (uint call, int first, int second, union semun fourth; if (!ptr) return -EINVAL; - if (get_user(fourth.__pad, (void **) ptr)) + if (get_user(fourth.__pad, (void __user * __user *) ptr)) return -EFAULT; return sys_semctl (first, second, third, fourth); } diff --git a/include/asm-um/ptrace-generic.h b/include/asm-um/ptrace-generic.h index 46599ac4403..8c57e384cb8 100644 --- a/include/asm-um/ptrace-generic.h +++ b/include/asm-um/ptrace-generic.h @@ -28,7 +28,7 @@ struct pt_regs { union uml_pt_regs regs; }; -#define EMPTY_REGS { regs : EMPTY_UML_PT_REGS } +#define EMPTY_REGS { .regs = EMPTY_UML_PT_REGS } #define PT_REGS_IP(r) UPT_IP(&(r)->regs) #define PT_REGS_SP(r) UPT_SP(&(r)->regs) diff --git a/include/asm-um/thread_info.h b/include/asm-um/thread_info.h index 17b6b07c433..f166b9837c6 100644 --- a/include/asm-um/thread_info.h +++ b/include/asm-um/thread_info.h @@ -27,14 +27,14 @@ struct thread_info { #define INIT_THREAD_INFO(tsk) \ { \ - task: &tsk, \ - exec_domain: &default_exec_domain, \ - flags: 0, \ - cpu: 0, \ - preempt_count: 1, \ - addr_limit: KERNEL_DS, \ - restart_block: { \ - fn: do_no_restart_syscall, \ + .task = &tsk, \ + .exec_domain = &default_exec_domain, \ + .flags = 0, \ + .cpu = 0, \ + .preempt_count = 1, \ + .addr_limit = KERNEL_DS, \ + .restart_block = { \ + .fn = do_no_restart_syscall, \ }, \ } diff --git a/include/asm-um/uaccess.h b/include/asm-um/uaccess.h index 4e460d6f5ac..bea5a015f66 100644 --- a/include/asm-um/uaccess.h +++ b/include/asm-um/uaccess.h @@ -57,7 +57,7 @@ ({ \ const __typeof__((*(ptr))) __user *private_ptr = (ptr); \ (access_ok(VERIFY_READ, private_ptr, sizeof(*private_ptr)) ? \ - __get_user(x, private_ptr) : ((x) = 0, -EFAULT)); \ + __get_user(x, private_ptr) : ((x) = (__typeof__(*ptr))0, -EFAULT)); \ }) #define __put_user(x, ptr) \ -- cgit v1.2.3-70-g09d2 From e11c0cdf4c6f7976e6f4fe221369a7b420245389 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Fri, 31 Mar 2006 02:30:17 -0800 Subject: [PATCH] uml: fix min usage type-safe min() in arch/um/drivers/mconsole_kern.c Signed-off-by: Al Viro Signed-off-by: Jeff Dike Cc: Paolo 'Blaisorblade' Giarrusso Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/um/drivers/mconsole_kern.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/um/drivers') diff --git a/arch/um/drivers/mconsole_kern.c b/arch/um/drivers/mconsole_kern.c index d060fce4940..28e3760e8b9 100644 --- a/arch/um/drivers/mconsole_kern.c +++ b/arch/um/drivers/mconsole_kern.c @@ -616,7 +616,7 @@ static void console_write(struct console *console, const char *string, return; while(1){ - n = min(len, ARRAY_SIZE(console_buf) - console_index); + n = min((size_t)len, ARRAY_SIZE(console_buf) - console_index); strncpy(&console_buf[console_index], string, n); console_index += n; string += n; -- cgit v1.2.3-70-g09d2