summaryrefslogtreecommitdiffstats
AgeCommit message (Collapse)Author
2005-05-05[IA64] Merge audit fix for fsyscalls with syscall-optimizationsDavid Mosberger-Tang
Signed-off-by: David Mosberger-Tang <davidm@hpl.hp.com> Signed-off-by: Tony Luck <tony.luck@intel.com>
2005-04-27[IA64] need r29=psr *after* rsm psr.iDavid Mosberger-Tang
Yanmin Zhang pointed out a sequence problem when saving the psr. David Mosberger provided this patch (which gave up a cycle). Signed-off-by: Tony Luck <tony.luck@intel.com>
2005-04-27[IA64] use srlz.d instead of srlz.i in ia64_leave_kernel()David Mosberger-Tang
This patch switches the srlz.i in ia64_leave_kernel() to srlz.d. As per architecture manual, the former is needed only to ensure that the clearing of PSR.IC is seen by the VHPT for subsequent instruction fetches. However, since the remainder of the code (up to and including the RFI instruction) is mapped by a pinned TLB entry, there is no chance of an iTLB miss and we don't care whether or not the VHPT sees PSR.IC cleared. Since srlz.d is substantially cheaper than srlz.i, this should shave off a few cycles off the interrupt path (unverified though; I'm not setup to measure this at the moment). Signed-off-by: David Mosberger-Tang <davidm@hpl.hp.com> Signed-off-by: Tony Luck <tony.luck@intel.com>
2005-04-27[IA64] Annotate fsys_bubble_down() with McKinley dispatch info.David Mosberger-Tang
This patch changes comments & formatting only. There is no code change. Signed-off-by: David Mosberger-Tang <davidm@hpl.hp.com> Signed-off-by: Tony Luck <tony.luck@intel.com>
2005-04-27[IA64] Reschedule fsys_bubble_down().David Mosberger-Tang
Improvements come from eliminating srlz.i, not scheduling AR/CR-reads too early (while there are others still pending), scheduling the backing-store switch as well as possible, splitting the BBB bundle into a MIB/MBB pair. Why is it safe to eliminate the srlz.i? Observe that we used to clear bits ~PSR_PRESERVED_BITS in PSR.L. Since PSR_PRESERVED_BITS==PSR.{UP,MFL,MFH,PK,DT,PP,SP,RT,IC}, we ended up clearing PSR.{BE,AC,I,DFL,DFH,DI,DB,SI,TB}. However, PSR.BE : already is turned off in __kernel_syscall_via_epc() PSR.AC : don't care (kernel normally turns PSR.AC on) PSR.I : already turned off by the time fsys_bubble_down gets invoked PSR.DFL: always 0 (kernel never turns it on) PSR.DFH: don't care --- kernel never touches f32-f127 on its own initiative PSR.DI : always 0 (kernel never turns it on) PSR.SI : always 0 (kernel never turns it on) PSR.DB : don't care --- kernel never enables kernel-level breakpoints PSR.TB : must be 0 already; if it wasn't zero on entry to __kernel_syscall_via_epc, the branch to fsys_bubble_down will trigger a taken branch; the taken-trap-handler then converts the syscall into a break-based system-call. In other words: all the bits we're clearying are either 0 already or are don't cares! Thus, we don't have to write PSR.L at all and we don't have to do a srlz.i either. Good for another ~20 cycle improvement for EPC-based heavy-weight syscalls. Signed-off-by: David Mosberger-Tang <davidm@hpl.hp.com> Signed-off-by: Tony Luck <tony.luck@intel.com>
2005-04-27[IA64] Annotate __kernel_syscall_via_epc() with McKinley dispatch info.David Mosberger-Tang
Two other very minor changes: use "mov.i" instead of "mov" for reading ar.pfs (for clarity; doesn't affect the code at all). Also, predicate the load of r14 for consistency. Signed-off-by: David Mosberger-Tang <davidm@hpl.hp.com> Signed-off-by: Tony Luck <tony.luck@intel.com>
2005-04-27[IA64] Reschedule __kernel_syscall_via_epc().David Mosberger-Tang
Avoid some stalls, which is good for about 2 cycles when invoking a light-weight handler. When invoking a heavy-weight handler, this helps by about 7 cycles, with most of the improvement coming from the improved branch-prediction achieved by splitting the BBB bundle into two MIB bundles. Signed-off-by: David Mosberger-Tang <davidm@hpl.hp.com> Signed-off-by: Tony Luck <tony.luck@intel.com>
2005-04-27[IA64] Reschedule break_fault() for better performance.David Mosberger-Tang
This patch reorganizes break_fault() to optimistically assume that a system-call is being performed from user-space (which is almost always the case). If it turns out that (a) we're not being called due to a system call or (b) we're being called from within the kernel, we fixup the no-longer-valid assumptions in non_syscall() and .break_fixup(), respectively. With this approach, there are 3 major phases: - Phase 1: Read various control & application registers, in particular the current task pointer from AR.K6. - Phase 2: Do all memory loads (load system-call entry, load current_thread_info()->flags, prefetch kernel register-backing store) and switch to kernel register-stack. - Phase 3: Call ia64_syscall_setup() and invoke syscall-handler. Good for 26-30 cycles of improvement on break-based syscall-path. Signed-off-by: David Mosberger-Tang <davidm@hpl.hp.com> Signed-off-by: Tony Luck <tony.luck@intel.com>
2005-04-27[IA64] In ia64_leave_syscall(), fix comments and whitespace only.David Mosberger-Tang
Signed-off-by: David Mosberger-Tang <davidm@hpl.hp.com> Signed-off-by: Tony Luck <tony.luck@intel.com>
2005-04-27[IA64] Schedule ia64_leave_syscall() to read ar.bsp earlierDavid Mosberger-Tang
Reschedule code to read ar.bsp as early as possible. To enable this, don't bother clearing some of the registers when we're returning to kernel stacks. Also, instead of trying to support the pNonSys case (which makes no sense), do a bugcheck instead (with break 0). Finally, remove a clear of r14 which is a left-over from the previous patch. Signed-off-by: David Mosberger-Tang <davidm@hpl.hp.com> Signed-off-by: Tony Luck <tony.luck@intel.com>
2005-04-27[IA64] In syscall-entry, use st8 instead of stf8 to clear pt_regs.r8David Mosberger-Tang
Using stf8 seemed like a clever idea at the time, but stf8 forces the cache-line to be invalidated in the L1D (if it happens to be there already). This patch eliminates a guaranteed L1D cache-miss and, by itself, is good for a 1-2 cycle improvement for heavy-weight syscalls. Signed-off-by: David Mosberger-Tang <davidm@hpl.hp.com> Signed-off-by: Tony Luck <tony.luck@intel.com>
2005-04-27[IA64] On return from syscall, hint b7 with __kernel_syscall_via_epc().David Mosberger-Tang
Why is this a good idea? Clearing b7 to 0 is guaranteed to do us no good and writing it with __kernel_syscall_via_epc() yields a 6 cycle improvement _if_ the application performs another EPC-based system- call without overwriting b7, which is not all that uncommon. Well worth the minimal cost of 1 bundle of code. Signed-off-by: David Mosberger-Tang <davidm@hpl.hp.com> Signed-off-by: Tony Luck <tony.luck@intel.com>
2005-04-27[IA64] Schedule fp-clearing insns at least 6 cycles after reading ar.bsp.David Mosberger-Tang
Decreases syscall overhead by approximately 6 cycles. Signed-off-by: David Mosberger-Tang <davidm@hpl.hp.com> Signed-off-by: Tony Luck <tony.luck@intel.com>
2005-04-27[IA64] Use dynamic prediction for RSE-clearing branches.David Mosberger-Tang
This by itself is good for a 1-2 cycle speed up. Effect is bigger when combined with the later patches. Signed-off-by: David Mosberger-Tang <davidm@hpl.hp.com> Signed-off-by: Tony Luck <tony.luck@intel.com>
2005-04-27[IA64] __ia64_syscall() is no longer used anywhere in the kernel. Remove it.David Mosberger-Tang
Signed-off-by: David Mosberger-Tang <davidm@hpl.hp.com> Signed-off-by: Tony Luck <tony.luck@intel.com>
2005-04-27Automatic merge of ↵Linus Torvalds
rsync://rsync.kernel.org/pub/scm/linux/kernel/git/aegl/linux-2.6.git
2005-04-26[PATCH] PC300 pci_enable_device fixMarcelo Tosatti
Call pci_enable_device() before looking at IRQ and resources, and pci_disable_device() when shutting the interface down. The driver requires this fix or the "pci=routeirq" workaround on 2.6.10 and later kernels. Reported and tested by Artur Lipowski. From: Bjorn Helgaas <bjorn.helgaas@hp.com> Signed-off-by: Marcelo Tosatti <marcelo.tosatti@cyclades.com> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-04-26[PATCH] ppc64: trivial user annotationsAl Viro
Signed-off-by: Al Viro <viro@parcelfarce.linux.theplanet.co.uk> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-04-26[PATCH] size_t portability fixes - drivers/usbAl Viro
size_t is zu, ssize_t is zd... Signed-off-by: Al Viro <viro@parcelfarce.linux.theplanet.co.uk> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-04-26[IA64] re-enable preempt before page allocation for pgtable quicklistMartin Hicks
This is a fix to the pgtable_quicklist code. There is a GFP_KERNEL allocation in pgtable_quicklist_alloc(), which spews the usual warnings if the kernel is under heavy VM pressure and the reclaim code is invoked. re-enable preempt before we allocate the new page. This patch is against 2.6.12-rc2-mm2 Signed-off-by: Martin Hicks <mort@sgi.com> Signed-off-by: Tony Luck <tony.luckintel.com>
2005-04-26Automatic merge of kernel.org:/home/rmk/linux-2.6-serial.gitLinus Torvalds
2005-04-26Automatic merge of kernel.org:/home/rmk/linux-2.6-rmk.gitLinus Torvalds
2005-04-26[PATCH] qla trivial iomem annotationAl Viro
Signed-off-by: Al Viro <viro@parcelfarce.linux.theplanet.co.uk> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-04-26[PATCH] oprofile trivial user annotationsAl Viro
Signed-off-by: Al Viro <viro@parcelfarce.linux.theplanet.co.uk> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-04-26[PATCH] amd64 rt_sigframe user annotationAl Viro
->pretcode in struct rt_sigframe is a userland pointer (and already treated as such by code using that field). Signed-off-by: Al Viro <viro@parcelfarce.linux.theplanet.co.uk> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-04-26[PATCH] drivers/scsi NULL noise removalAl Viro
NULL noise in sbus scsi drivers got missed Signed-off-by: Al Viro <viro@parcelfarce.linux.theplanet.co.uk> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-04-26[PATCH] imsttfb missing iomem annotationsAl Viro
write_reg_le32() and read_reg_le32() expect iomem pointers... Signed-off-by: Al Viro <viro@parcelfarce.linux.theplanet.co.uk> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-04-26[PATCH] ppc-opc NULL noise removalAl Viro
Signed-off-by: Al Viro <viro@parcelfarce.linux.theplanet.co.uk> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-04-26[PATCH] savagefb iomem annotationsAl Viro
trivial iomem annotations + memset() replaced with memset_io() in a place that deals with ioremapped area. Signed-off-by: Al Viro <viro@parcelfarce.linux.theplanet.co.uk> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-04-26[PATCH] pegasus NULL noise removalAl Viro
Signed-off-by: Al Viro <viro@parcelfarce.linux.theplanet.co.uk> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-04-26[PATCH] tpm 64bit fixes (size_t)Al Viro
Signed-off-by: Al Viro <viro@parcelfarce.linux.theplanet.co.uk> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-04-26[PATCH] fix SND_POWERMAC dependenciesAl Viro
This beast is pmac-only; moreover, it won't build on other subarchitectures. Signed-off-by: Al Viro <viro@parcelfarce.linux.theplanet.co.uk> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-04-26[PATCH] Serial: Convert 8250_hp300 to use serial8250_{un,}register_portRussell King
Convert 8250_hp300 to use serial8250_register_port() and serial8250_unregister_port(). Tested by Kars de Jong, 4/4/2005. Signed-off-by: Russell King <rmk@arm.linux.org.uk>
2005-04-26[PATCH] Serial: Move error path processing inlineRussell King
With unlikely() there's no need for the error path to use gotos. Signed-off-by: Russell King <rmk@arm.linux.org.uk>
2005-04-26[PATCH] Serial: Ensure error paths are marked with unlikely()Russell King
Ensure ARM serial driver error paths are marked with the unlikely() compiler hint. Signed-off-by: Russell King <rmk@arm.linux.org.uk>
2005-04-26Automatic merge of ↵Linus Torvalds
rsync://rsync.kernel.org/pub/scm/linux/kernel/git/gregkh/usb-2.6.git/
2005-04-26[PATCH] ARM: make entry*.S includes more logicalRussell King
Move common includes to entry-header, and file specific includes to the relevant file. Signed-off-by: Russell King <rmk@arm.linux.org.uk>
2005-04-26[PATCH] ARM: Remove single-use user save/restore macrosRussell King
Assembly macros are pointless if they're only used once. Move them inline. Signed-off-by: Russell King <rmk@arm.linux.org.uk>
2005-04-26[PATCH] ARM: remove PT_TRACESYSRussell King
PT_TRACESYS is unused, remove it. Signed-off-by: Russell King <rmk@arm.linux.org.uk>
2005-04-26[PATCH] ARM: Remove SVC_MODE definitionRussell King
SVC_MODE reflects the MODE_SVC definition in asm/ptrace.h. Use the asm/ptrace.h definition instead, and remove SVC_MODE. Signed-off-by: Russell King <rmk@arm.linux.org.uk>
2005-04-26[PATCH] ARM: Use __NR_SYSCALL_BASE and __ARM_NR_BASE in asm codeRussell King
Don't define our own local constants, but use those already defined in asm/unistd.h instead. Signed-off-by: Russell King <rmk@arm.linux.org.uk>
2005-04-26Automatic merge of ↵Linus Torvalds
rsync://rsync.kernel.org/pub/scm/linux/kernel/git/davem/net-2.6.git
2005-04-26[PATCH] ARM: pt_regs offsetsRussell King
Generate pt_regs S_xx offsets from the structure itself instead of #defining them. Signed-off-by: Russell King <rmk@arm.linux.org.uk>
2005-04-26[PATCH] ARM: Remove argument for disable_irq/enable_irqRussell King
Since we do not require a register for these operations, we can remove this unnecessary argument. Signed-off-by: Russell King <rmk@arm.linux.org.uk>
2005-04-26[PATCH] ARM: remove some entry initialisation asm codeRussell King
Convert the trivial vector entry initialisation code to C code. Signed-off-by: Russell King <rmk@arm.linux.org.uk>
2005-04-25[NET]: kill gratitious includes of major.hAl Viro
A lot of places in there are including major.h for no reason whatsoever. Removed. And yes, it still builds. The history of that stuff is often amusing. E.g. for net/core/sock.c the story looks so, as far as I've been able to reconstruct it: we used to need major.h in net/socket.c circa 1.1.early. In 1.1.13 that need had disappeared, along with register_chrdev(SOCKET_MAJOR, "socket", &net_fops) in sock_init(). Include had not. When 1.2 -> 1.3 reorg of net/* had moved a lot of stuff from net/socket.c to net/core/sock.c, this crap had followed... Signed-off-by: Al Viro <viro@parcelfarce.linux.theplanet.co.uk> Signed-off-by: David S. Miller <davem@davemloft.net>
2005-04-25[TCP]: Trivial tcp_data_queue() cleanupJames Morris
This patch removes a superfluous intialization from tcp_data_queue(). Signed-off-by: James Morris <jmorris@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2005-04-25[PATCH] kill gratitious includes of major.h under net/*Al Viro
A lot of places in there are including major.h for no reason whatsoever. Removed. And yes, it still builds. The history of that stuff is often amusing. E.g. for net/core/sock.c the story looks so, as far as I've been able to reconstruct it: we used to need major.h in net/socket.c circa 1.1.early. In 1.1.13 that need had disappeared, along with register_chrdev(SOCKET_MAJOR, "socket", &net_fops) in sock_init(). Include had not. When 1.2 -> 1.3 reorg of net/* had moved a lot of stuff from net/socket.c to net/core/sock.c, this crap had followed... Signed-off-by: Al Viro <viro@parcelfarce.linux.theplanet.co.uk> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-04-25[PATCH] isofs includes sanitizedAl Viro
fs/isofs includes trimmed down to something resembling sanity. Kernel-only parts of linux/iso_fs.h and entire linux/iso_fs_{sb,i}.h moved to fs/isofs/isofs.h. A lot of useless #include in fs/isofs/*.c killed. Signed-off-by: Al Viro <viro@parcelfarce.linux.theplanet.co.uk> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-04-25[PATCH] ppc annotations: i2c-mpcAl Viro
Usual iomem annotations and NULL noise removal. Signed-off-by: Al Viro <viro@parcelfarce.linux.theplanet.co.uk> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
generated by cgit v1.2.3-70-g09d2 (git 2.46.0) at 2024-11-23 23:37:51 +0000