The RT team has been searching for a nasty latency. This latency shows
up out of the blue and has been seen to be as big as 5ms!
Using ftrace I found the cause of the latency.
pcscd-2995 3dNh1 52360300us : irq_exit (smp_apic_timer_interrupt)
pcscd-2995 3dN.2 52360301us : idle_cpu (irq_exit)
pcscd-2995 3dN.2 52360301us : rcu_irq_exit (irq_exit)
pcscd-2995 3dN.1 52360771us : smp_apic_timer_interrupt (apic_timer_interrupt
)
pcscd-2995 3dN.1 52360771us : exit_idle (smp_apic_timer_interrupt)
Here's an example of a 400 us latency. pcscd took a timer interrupt and
returned with "need resched" enabled, but did not reschedule until after
the next interrupt came in at 52360771us 400us later!
At first I thought we somehow missed a preemption check in entry.S. But
I also noticed that this always seemed to happen during a __delay call.
pcscd-2995 3dN.2 52360836us : rcu_irq_exit (irq_exit)
pcscd-2995 3.N.. 52361265us : preempt_schedule (__delay)
Looking at the x86 delay, I found my problem.
In git commit 35d5d08a08, Andrew Morton
placed preempt_disable around the entire delay due to TSC's not working
nicely on SMP. Unfortunately for those that care about latencies this
is devastating! Especially when we have callers to mdelay(8).
Here I enable preemption during the loop and account for anytime the task
migrates to a new CPU. The delay asked for may be extended a bit by
the migration, but delay only guarantees that it will delay for that minimum
time. Delaying longer should not be an issue.
[
Thanks to Thomas Gleixner for spotting that cpu wasn't updated,
and to place the rep_nop between preempt_enabled/disable.
]
Signed-off-by: Steven Rostedt <srostedt@redhat.com>
Cc: akpm@osdl.org
Cc: Clark Williams <clark.williams@gmail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: "Luis Claudio R. Goncalves" <lclaudio@uudg.org>
Cc: Gregory Haskins <ghaskins@novell.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andi Kleen <andi-suse@firstfloor.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Fix this symbol export problem:
Building modules, stage 2.
MODPOST 193 modules
ERROR: "csum_partial" [fs/reiserfs/reiserfs.ko] undefined!
make[1]: *** [__modpost] Error 1
make: *** [modules] Error 2
This is due to a known weakness of symbol exports: if a symbol's
only in-core user is an EXPORT_SYMBOL from a lib-y section, the
symbol is not linked in.
The solution is to move the export to x8664_ksyms_64.c - but the real
solution would be to fix kbuild.
Signed-off-by: Ingo Molnar <mingo@elte.hu>
x86 has been switched to the generic versions of find_first_bit
and find_first_zero_bit, but the original versions were retained.
This patch just removes the now unused x86-specific versions.
also update UML.
Signed-off-by: Alexander van Heukelum <heukelum@fastmail.fm>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Switch x86_64 to generic find_first_bit. The x86_64-specific
implementation is not removed.
Signed-off-by: Alexander van Heukelum <heukelum@fastmail.fm>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
The versions with inline assembly are in fact slower on the machines I
tested them on (in userspace) (Athlon XP 2800+, p4-like Xeon 2.8GHz, AMD
Opteron 270). The i386-version needed a fix similar to 06024f21 to avoid
crashing the benchmark.
Benchmark using: gcc -fomit-frame-pointer -Os. For each bitmap size
1...512, for each possible bitmap with one bit set, for each possible
offset: find the position of the first bit starting at offset. If you
follow ;). Times include setup of the bitmap and checking of the
results.
Athlon Xeon Opteron 32/64bit
x86-specific: 0m3.692s 0m2.820s 0m3.196s / 0m2.480s
generic: 0m2.622s 0m1.662s 0m2.100s / 0m1.572s
If the bitmap size is not a multiple of BITS_PER_LONG, and no set
(cleared) bit is found, find_next_bit (find_next_zero_bit) returns a
value outside of the range [0, size]. The generic version always returns
exactly size. The generic version also uses unsigned long everywhere,
while the x86 versions use a mishmash of int, unsigned (int), long and
unsigned long.
Using the generic version does give a slightly bigger kernel, though.
defconfig: text data bss dec hex filename
x86-specific: 4738555 481232 626688 5846475 5935cb vmlinux (32 bit)
generic: 4738621 481232 626688 5846541 59360d vmlinux (32 bit)
x86-specific: 5392395 846568 724424 6963387 6a40bb vmlinux (64 bit)
generic: 5392458 846568 724424 6963450 6a40fa vmlinux (64 bit)
Signed-off-by: Alexander van Heukelum <heukelum@fastmail.fm>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
The patch kills 45 errors and a few warnings.
The file is now error/warning free:
total: 0 errors, 0 warnings, 237 lines checked
arch/x86/lib/string_32.c has no obvious style problems and is ready for submission.
no code changed:
arch/x86/lib/string_32.o:
text data bss dec hex filename
639 0 0 639 27f string_32.o.before
639 0 0 639 27f string_32.o.after
md5:
2db1c48187cf5113bb595153ee1fc73d string_32.o.before.asm
2db1c48187cf5113bb595153ee1fc73d string_32.o.after.asm
Signed-off-by: Paolo Ciarrocchi <paolo.ciarrocchi@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Semaphores are no longer performance-critical, so a generic C
implementation is better for maintainability, debuggability and
extensibility. Thanks to Peter Zijlstra for fixing the lockdep
warning. Thanks to Harvey Harrison for pointing out that the
unlikely() was unnecessary.
Signed-off-by: Matthew Wilcox <willy@linux.intel.com>
Acked-by: Ingo Molnar <mingo@elte.hu>
This simple patch makes the file error free (according to
checkpatch.pl)
no code changed:
arch/x86/lib/io_64.o:
text data bss dec hex filename
308 0 0 308 134 io_64.o.before
308 0 0 308 134 io_64.o.after
md5:
3c64f9ed83d091678e849b36ca27bee3 io_64.o.before.asm
3c64f9ed83d091678e849b36ca27bee3 io_64.o.after.asm
Signed-off-by: Paolo Ciarrocchi <paolo.ciarrocchi@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
- All implementations can be __devinit
- The function prototypes were in asm/timex.h but they all must be the same,
so create a single declaration in linux/timex.h.
- uninline the sparc64 version to match the other architectures
- Don't bother #defining ARCH_HAS_READ_CURRENT_TIMER to a particular value.
[ezk@cs.sunysb.edu: fix build]
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Haavard Skinnemoen <hskinnemoen@atmel.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Andi Kleen <ak@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This kills unused __clear_bit_string and find_next_zero_string (they
were used by only gart and calgary IOMMUs).
Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Cc: Jeff Garzik <jeff@garzik.org>
Cc: James Bottomley <James.Bottomley@steeleye.com>
Cc: Jens Axboe <jens.axboe@oracle.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Muli Ben-Yehuda <mulix@mulix.org>
Cc: Andi Kleen <ak@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Use the _ASM_EXTABLE macro from <asm/asm.h>, instead of open-coding
__ex_table entires in arch/x86/lib/usercopy_64.c.
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Use the _ASM_EXTABLE macro from <asm/asm.h>, instead of open-coding
__ex_table entires in arch/x86/lib/usercopy_32.c.
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Use the _ASM_EXTABLE macro from <asm/asm.h>, instead of open-coding
__ex_table entires in arch/x86/lib/mmx_32.c.
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Correct the obvious "copy and paste" errors explaining some of the
"find_next" routines.
Signed-off-by: Robert P. J. Day <rpjday@mindspring.com>
Acked-by: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: Adrian Bunk <bunk@kernel.org>
Without this patch the linker will generate a section
named .sched.text.1 which is unexpected.
This is because the gcc generated section has "ax" but the
assembler usage of .sched.text lacks the "ax" specifier.
It would be better to have a definition we could use from
assembler code but I did not find a suitable header
file for it.
Signed-off-by: Sam Ravnborg <sam@ravnborg.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
The ENDPROCs() were not used everywhere. Some code used just END() instead,
while other code used nothing. um/sys-i386/checksum.S didn't #include
<linux/linkage.h> . I also got confused because gcc puts the
.type near the ENTRY, while ENDPROC puts it on the opposite end.
Signed off by: John Reiser <jreiser@BitWagon.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Trivial unification of Makefiles for the
x86 specific library part.
Linking order is slightly modified but should be harmless.
Tested doing a defconfig build before and after and saw
no build changes.
It adds almost as many lines as it deletes - bacause
I broke a few lines up fo readability in the Makefile.
Signed-off-by: Sam Ravnborg <sam@ravnborg.org>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Marin Mitov points out that delay_tsc() can misbehave if it is preempted and
rescheduled on a different CPU which has a skewed TSC. Fix it by disabling
preemption.
(I assume that the worst-case behaviour here is a stall of 2^32 cycles)
Cc: Andi Kleen <ak@suse.de>
Cc: Marin Mitov <mitov@issp.bas.bg>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: <stable@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* ssh://master.kernel.org/pub/scm/linux/kernel/git/tglx/linux-2.6-x86: (33 commits)
x86: convert cpuinfo_x86 array to a per_cpu array
x86: introduce frame_pointer() and stack_pointer()
x86 & generic: change to __builtin_prefetch()
i386: do not BUG_ON() when MSR is unknown
x86: acpi use cpu_physical_id
x86: convert cpu_llc_id to be a per cpu variable
x86: convert cpu_to_apicid to be a per cpu variable
i386: introduce "used_vectors" bitmap which can be used to reserve vectors.
x86: use raw locks during oopses
x86: honor _PAGE_PSE bit on page walks
i386: do cpuid_device_create() in CPU_UP_PREPARE instead of CPU_ONLINE.
x86: implement missing x86_64 function smp_call_function_mask()
x86: use descriptor's functions instead of inline assembly
i386: consolidate show_regs and show_registers for i386
i386: make callgraph use dump_trace() on i386/x86_64
x86: enable iommu_merge by default
i386: i386 add AMD64 Barcelona PMU MSR definitions to msr.h
x86: Unify i386 and x86-64 early quirks
x86: enable HPET on ICH3 and ICH4
x86: force enable HPET on VT8235/8237 chipsets
...
Manually fix trivial conflict with task pid container helper changes in
arch/x86/kernel/process_32.c
is_init() is an ambiguous name for the pid==1 check. Split it into
is_global_init() and is_container_init().
A cgroup init has it's tsk->pid == 1.
A global init also has it's tsk->pid == 1 and it's active pid namespace
is the init_pid_ns. But rather than check the active pid namespace,
compare the task structure with 'init_pid_ns.child_reaper', which is
initialized during boot to the /sbin/init process and never changes.
Changelog:
2.6.22-rc4-mm2-pidns1:
- Use 'init_pid_ns.child_reaper' to determine if a given task is the
global init (/sbin/init) process. This would improve performance
and remove dependence on the task_pid().
2.6.21-mm2-pidns2:
- [Sukadev Bhattiprolu] Changed is_container_init() calls in {powerpc,
ppc,avr32}/traps.c for the _exception() call to is_global_init().
This way, we kill only the cgroup if the cgroup's init has a
bug rather than force a kernel panic.
[akpm@linux-foundation.org: fix comment]
[sukadev@us.ibm.com: Use is_global_init() in arch/m32r/mm/fault.c]
[bunk@stusta.de: kernel/pid.c: remove unused exports]
[sukadev@us.ibm.com: Fix capability.c to work with threaded init]
Signed-off-by: Serge E. Hallyn <serue@us.ibm.com>
Signed-off-by: Sukadev Bhattiprolu <sukadev@us.ibm.com>
Acked-by: Pavel Emelianov <xemul@openvz.org>
Cc: Eric W. Biederman <ebiederm@xmission.com>
Cc: Cedric Le Goater <clg@fr.ibm.com>
Cc: Dave Hansen <haveblue@us.ibm.com>
Cc: Herbert Poetzel <herbert@13thfloor.at>
Cc: Kirill Korotaev <dev@sw.ru>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
cpu_data is currently an array defined using NR_CPUS. This means that
we overallocate since we will rarely really use maximum configured cpus.
When NR_CPU count is raised to 4096 the size of cpu_data becomes
3,145,728 bytes.
These changes were adopted from the sparc64 (and ia64) code. An
additional field was added to cpuinfo_x86 to be a non-ambiguous cpu
index. This corresponds to the index into a cpumask_t as well as the
per_cpu index. It's used in various places like show_cpuinfo().
cpu_data is defined to be the boot_cpu_data structure for the NON-SMP
case.
Signed-off-by: Mike Travis <travis@sgi.com>
Acked-by: Christoph Lameter <clameter@sgi.com>
Cc: Andi Kleen <ak@suse.de>
Cc: James Bottomley <James.Bottomley@steeleye.com>
Cc: Dmitry Torokhov <dtor@mail.ru>
Cc: "Antonino A. Daplas" <adaplas@pol.net>
Cc: Mark M. Hoffman <mhoffman@lightlink.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
.i is an ending used for preprocessed stuff.
This patch therefore renames assembler include files to .h and guards
the contents with an #ifdef __ASSEMBLY__.
[ tglx: arch/x86 adaptation ]
Signed-off-by: Adrian Bunk <bunk@kernel.org>
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
The constraints in the inline assembler implementation of i386
strrchr() were incorrect and break the build with recent gcc 4.3.
Since there are only very few callers of strrchr() and none of them
are performance relevant just remove the assembler implementation
and use the C fallback instead.
[ tglx: arch/x86 adaptation ]
Cc: rguenther@suse.de
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
smp_call_function_single() now knows how to call the function on the
current cpu.
[ tglx: arch/x86 adaptation ]
Cc: H. Peter Anvin <hpa@zytor.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Fix an off-by-one error in find_next_zero_string which prevents
allocating the last bit.
[ tglx: arch/x86 adaptation ]
Signed-off-by: Andrew Hastings <abh@cray.com> on behalf of Cray Inc.
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Run the lockdep_sys_exit hook after all other C code on the syscall
return path.
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
movnt* instructions are not strongly ordered with respect to other stores,
so if we are to assume stores are strongly ordered in the rest of the 64
bit code, we must fence these off (see similar examples in 32 bit code).
[ The AMD memory ordering document seems to say that nontemporal stores can
also pass earlier regular stores, so maybe we need sfences _before_
movnt* everywhere too? ]
Signed-off-by: Nick Piggin <npiggin@suse.de>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
