Since taking a spinlock disables preempt, and we need to spinlock tlb flush
on SMP for N class, we might as well just spinlock on uniprocessor machines
too.
Signed-off-by: Matthew Wilcox <willy@parisc-linux.org>
Signed-off-by: Kyle McMartin <kyle@parisc-linux.org>
We actually have two separate bad bugs
1. The read_lock implementation spins with disabled interrupts. This is
completely wrong
2. Our spin_lock_irqsave should check to see if interrupts were enabled
before the call and re-enable interrupts around the inner spin loop.
The problem is that if we spin with interrupts off, we can't receive
IPIs. This has resulted in a bug where SMP machines suddenly spit
smp_call_function timeout messages and hang.
The scenario I've caught is
CPU0 does a flush_tlb_all holding the vmlist_lock for write.
CPU1 tries a cat of /proc/meminfo which tries to acquire vmlist_lock for
read
CPU1 is now spinning with interrupts disabled
CPU0 tries to execute a smp_call_function to flush the local tlb caches
This is now a deadlock because CPU1 is spinning with interrupts disabled
and can never receive the IPI
Signed-off-by: James Bottomley <jejb@parisc-linux.org>
Signed-off-by: Kyle McMartin <kyle@parisc-linux.org>
This really only adds them for the machines I can check SMP on, which
is CPU interrupts and IOSAPIC (so not any of the GSC based machines).
With this patch, irqbalanced can be used to maintain irq balancing.
Unfortunately, irqbalanced is a bit x86 centric, so it doesn't do an
incredibly good job, but it does work.
Signed-off-by: James Bottomley <jejb@parisc-linux.org>
Signed-off-by: Kyle McMartin <kyle@parisc-linux.org>
Since irq.c uses smp_send_all_nop, we must define it for UP builds
as well. Make it a static inline so it gets optimized away. This forces
irq.c to include <asm/smp.h> though.
Signed-off-by: Kyle McMartin <kyle@parisc-linux.org>
Fix our interrupts not to use smp_call_function
On K and D class smp, the generic code calls this under an irq
spinlock, which causes the WARN_ON() message in smp_call_function()
(and is also illegal because it could deadlock).
The fix is to use a new scheme based on the IPI_NOP.
Signed-off-by: James Bottomley <jejb@parisc-linux.org>
Signed-off-by: Kyle McMartin <kyle@parisc-linux.org>
Introduce an atomic_inc_not_zero operation. Make this a special case of
atomic_add_unless because lockless pagecache actually wants
atomic_inc_not_negativeone due to its offset refcount.
Signed-off-by: Nick Piggin <npiggin@suse.de>
Cc: "Paul E. McKenney" <paulmck@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Fix more include file problems that surfaced since I submitted the previous
fix-missing-includes.patch. This should now allow not to include sched.h
from module.h, which is done by a followup patch.
Signed-off-by: Tim Schmielau <tim@physik3.uni-rostock.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
__MUTEX_INITIALIZER() has no users, and equates to the more commonly used
DECLARE_MUTEX(), thus making it pretty much redundant. Remove it for good.
Signed-off-by: Arthur Othieno <a.othieno@bluewin.ch>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Make the pid argument a long as on every other arcihtecture. Despite pid_t
beeing a 32bit type even on 64bit parisc this is not an ABI change due to
the parisc calling conventions. And even if it did it wouldn't matter too
much because 64bit userspace on parisc is in an embrionic stage.
Acked-by: Matthew Wilcox <matthew@wil.cx>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Removed some more references to check_region().
I checked these changes into the 'checkreg' branch of
rsync://rsync.kernel.org/pub/scm/linux/kernel/git/jgarzik/misc-2.6.git
The only valid references remaining are in:
drivers/scsi/advansys.c
drivers/scsi/BusLogic.c
drivers/cdrom/sbpcd.c
sound/oss/pss.c
Remove last vestiges of ide_check_region()
drivers/char/specialix: trim trailing whitespace
drivers/char/specialix: eliminate use of check_region()
Remove outdated and unused references to check_region()
[sound oss] remove check_region() usage from cs4232, wavfront
[netdrvr eepro] trim trailing whitespace
[netdrvr eepro] remove check_region() usage
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
The following series implements memory hot-add for ppc64 and i386. There are
x86_64 and ia64 implementations that will be submitted shortly as well,
through the normal maintainers.
This patch:
local_mapnr is unused, except for in an alpha header. Keep the alpha one,
kill the rest.
Signed-off-by: Dave Hansen <haveblue@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
There's a worrying function translation_exists in parisc cacheflush.h,
unaffected by split ptlock since flush_dcache_page is using it on some other
mm, without any relevant lock. Oh well, make it a slightly more robust by
factoring the pfn check within it. And it looked liable to confuse a
camouflaged swap or file entry with a good pte: fix that too.
Signed-off-by: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
There was one small but very significant change in the previous patch:
mprotect's flush_tlb_range fell outside the page_table_lock: as it is in 2.4,
but that doesn't prove it safe in 2.6.
On some architectures flush_tlb_range comes to the same as flush_tlb_mm, which
has always been called from outside page_table_lock in dup_mmap, and is so
proved safe. Others required a deeper audit: I could find no reliance on
page_table_lock in any; but in ia64 and parisc found some code which looks a
bit as if it might want preemption disabled. That won't do any actual harm,
so pending a decision from the maintainers, disable preemption there.
Remove comments on page_table_lock from flush_tlb_mm, flush_tlb_range and
flush_tlb_page entries in cachetlb.txt: they were rather misleading (what
generic code does is different from what usually happens), the rules are now
changing, and it's not yet clear where we'll end up (will the generic
tlb_flush_mmu happen always under lock? never under lock? or sometimes under
and sometimes not?).
Signed-off-by: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
fixup.S needs to specify .level and use correct LDREG macro.
New binutils has a bug where it doesn't "promote" from PA1.0 to PA1.1
correctly when using ",s" completer.
remove use of __LP64__ in assembly.h and add some white space.
Signed-off-by: Grant Grundler <grundler@parisc-linux.org>
Signed-off-by: Kyle McMartin <kyle@parisc-linux.org>
drivers/infiniband depends on definition of pgprot_noncached() macro.
Someone else will have to fix it's wrong.
Signed-off-by: Grant Grundler <grundler@parisc-linux.org
Signed-off-by: Kyle McMartin <kyle@parisc-linux.org>
add ECANCELED - SuSv3 wants one L. IB/SDP actually returns this error.
Signed-off-by: Grant Grundler <grundler@parisc-linux.org>
Signed-off-by: Kyle McMartin <kyle@parisc-linux.org>
Fix the alloc_slabmgmt panic
Hopefully this should also fix a lot of other intermittent kernel bugs.
The problem has been around since 2.6.9-rc2-pa6 when we allowed
floating point registers to be used in kernel code. The essence of
the problem is that gcc prefers to use floating point for integer
divides and multiples. Further, it can rely on the values in the no
clobber fp regs being correct across a function call. Unfortunately,
our task switch function only saves the integer no clobber registers,
not the fp ones, so if gcc makes a function call to any function in
the kernel which could sleep, the values it is relying on in any no
clobber floating point register may be lost. In the case of
alloc_slabmgmt, the value of the page offset is being stored in %fr12
across a call to kmem_getpages(), which sleeps if no pages are
available. Thus, the offset can be trashed and the slab code can end
up with a completely bogus address leading to corruption.
Kudos to Randolph who came up with the program to trip this problem at
will and thus allowed it to be tracked and fixed.
Signed-off-by: James Bottomley <jejb@parisc-linux.org>
Signed-off-by: Kyle McMartin <kyle@parisc-linux.org>
2.6.12-rc1-pa6 use work queue in LED/LCD driver instead of tasklet.
Main advantage is it allows use of msleep() in the led_LCD_driver to
"atomically" perform two MMIO writes (CMD, then DATA).
Lead to nice cleanup of the main led_work_func() and led_LCD_driver().
Kudos to David for being persistent.
From: David Pye <dmp@davidmpye.dyndns.org>
Signed-off-by: Grant Grundler <grundler@parisc-linux.org>
Signed-off-by: Kyle McMartin <kyle@parisc-linux.org>
Optimize ext2_find_next_zero_bit. Gives about 25% perf improvement with a
rsync test with ext3.
Signed-off-by: Randolph Chung <tausq@parisc-linux.org>
fix ext3 performance - ext2_find_next_zero() was culprit.
Kudos to jejb for pointing out the the possibility that ext2_test_bit
and ext2_find_next_zero() may in fact not be enumerating bits in
the bitmap because of endianess. Took sparc64 implementation and
adapted it to our tree. I suspect the real problem is ffz() wants
an unsigned long and was getting garbage in the top half of the
unsigned int. Not confirmed but that's what I suspect.
Signed-off-by: Grant Grundler <grundler@parisc-linux.org>
Fix find_next_bit for 32-bit
Make masking consistent for bitops
From: Joel Soete <soete.joel@tiscali.be>
Signed-off-by: Randolph Chung <tausq@parisc-linux.org>
Add back incorrectly removed ext2_find_first_zero_bit definition
Signed-off-by: James Bottomley <jejb@parisc-linux.org>
Fixup bitops.h to use volatile for *_bit() ops
Based on this email thread:
http://marc.theaimsgroup.com/?t=108826637900003
In a nutshell:
*_bit() want use of volatile.
__*_bit() are "relaxed" and don't use spinlock or volatile.
other minor changes:
o replaces hweight64() macro with alias to generic_hweight64() (Joel Soete)
o cleanup ext2* macros so (a) it's obvious what the XOR magic is about
and (b) one version that works for both 32/64-bit.
o replace 2 uses of CONFIG_64BIT with __LP64__. bitops.h used both.
I think header files that might go to user space should use
something userspace will know about (__LP64__).
Signed-off-by: Grant Grundler <grundler@parisc-linux.org>
Move SHIFT_PER_LONG to standard location for BITS_PER_LONG (asm/types.h)
and ditch the second definition of BITS_PER_LONG in bitops.h
Signed-off-by: Grant Grundler <grundler@parisc-linux.org>
Signed-off-by: Kyle McMartin <kyle@parisc-linux.org>
export profile_pc() symbol - oprofile needs it when built as a module.
Signed-off-by: Grant Grundler <grundler@parisc-linux.org>
Take into account nullified insn and lock functions for profiling
This is needed at the end of functions; it is typical that the return
branch nullifies the next insn, which is in the next function. This
causes profiling data to show up against the "wrong" function.
We also count lock times against the locker. This is consistent with
other architectures.
Signed-off-by: Randolph Chung <tausq@parisc-linux.org>
Signed-off-by: Kyle McMartin <kyle@parisc-linux.org>
Neaten up the CONFIG_PA20 ifdefs
More merge fixes, this time for SMP
Signed-off-by: Matthew Wilcox <willy@parisc-linux.org>
Prettify the CONFIG_DEBUG_SPINLOCK __SPIN_LOCK_UNLOCKED initializers.
Clean up some warnings with CONFIG_DEBUG_SPINLOCK enabled.
Fix build with spinlock debugging turned on. Patch is cleaner like this,
too.
Remove mandatory 16-byte alignment requirement on PA2.0 processors by
using the ldcw,CO completer. Provides a nice insn savings.
Signed-off-by: Kyle McMartin <kyle@parisc-linux.org>
move pa_tlb_lock and it's primary consumers to tlb_flush.h
Future step will be to move spinlock_t definition out of system.h.
Signed-off-by: Grant Grundler <grundler@parisc-linux.org>
Signed-off-by: Kyle McMartin <kyle@parisc-linux.org>
2.6.12-rc4-pa3 : first pass at making sure use of RFI conforms to
PA 2.0 arch pages F-4 and F-5, PA 1.1 Arch page 3-19 and 3-20.
The discussion revolves around all the rules for clearing PSW Q-bit.
The hard part is meeting all the rules for "relied upon translation".
.align directive is used to guarantee the critical sequence ends more than
8 instructions (32 bytes) from the end of page.
Signed-off-by: Grant Grundler <grundler@parisc-linux.org>
Signed-off-by: Kyle McMartin <kyle@parisc-linux.org>
Convert pa_dev->hpa from an unsigned long to a struct resource.
Signed-off-by: Matthew Wilcox <willy@parisc-linux.org>
Fix up users of ->hpa to use ->hpa.start instead.
Signed-off-by: Matthew Wilcox <willy@parisc-linux.org>
Signed-off-by: Kyle McMartin <kyle@parisc-linux.org>
Fix parse_tree_node. much more needs to be done to fix this file.
Signed-off-by: Matthew Wilcox <willy@parisc-linux.org>
Make drivers.c compile based on a patch from Pat Mochel.
From: Patrick Mochel <mochel@digitalimplant.org>
Signed-off-by: Kyle McMartin <kyle@parisc-linux.org>
Fix drivers.c to create new device tree nodes when no match is found.
Signed-off-by: Richard Hirst <rhirst@parisc-linux.org>
Do a proper depth-first search returning parents before children, using the
new klist infrastructure.
Signed-off-by: Richard Hirst <rhirst@parisc-linux.org>
Fixed parisc_device traversal so that pdc_stable works again
Fixed check_dev so it doesn't dereference a parisc_device until it
has verified the bus type
Signed-off-by: Randolph Chung <tausq@parisc-linux.org>
Convert pa_dev->hpa from an unsigned long to a struct resource.
Use insert_resource() instead of request_mem_region().
Request resources at bus walk time instead of driver probe time.
Don't release the resources as we don't have any hotplug parisc_device
support yet.
Add parisc_pathname() to conveniently get the textual representation
of the hwpath used in sysfs.
Inline the remnants of claim_device() into its caller.
Signed-off-by: Matthew Wilcox <willy@parisc-linux.org>
I noticed that some of the STI regions weren't showing up in iomem.
Reading the STI spec indicated that all STI devices occupy at least 32MB.
So check for STI HPAs and give them 32MB instead of 4kB.
Signed-off-by: Matthew Wilcox <willy@parisc-linux.org>
Signed-off-by: Kyle McMartin <kyle@parisc-linux.org>
As recently done by Russell King for ARM, commit
4732efbeb9 introduces a generic asm/futex.h copied
along most arches, which includes a "-ENOSYS support" to be changed if needed.
However, it includes an unused var (taken from the "real" version) which GCC
warns about.
Remove it from all arches having that file version (i.e. same GIT id).
$ git-diff-tree -r HEAD
and
$ git-ls-tree -r HEAD include/|grep 9feff4ce14
may be more interesting than looking at the patch itself, to make sure I've
just copied the arm header to all other archs having the original dummy version
of this file.
Cc: Jakub Jelinek <jakub@redhat.com>
Signed-off-by: Paolo 'Blaisorblade' Giarrusso <blaisorblade@yahoo.it>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
As written in Documentation/feature-removal-schedule.txt, remove the
io_remap_page_range() kernel API.
Signed-off-by: Randy Dunlap <rdunlap@xenotime.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
This patch (written by me and also containing many suggestions of Arjan van
de Ven) does a major cleanup of the spinlock code. It does the following
things:
- consolidates and enhances the spinlock/rwlock debugging code
- simplifies the asm/spinlock.h files
- encapsulates the raw spinlock type and moves generic spinlock
features (such as ->break_lock) into the generic code.
- cleans up the spinlock code hierarchy to get rid of the spaghetti.
Most notably there's now only a single variant of the debugging code,
located in lib/spinlock_debug.c. (previously we had one SMP debugging
variant per architecture, plus a separate generic one for UP builds)
Also, i've enhanced the rwlock debugging facility, it will now track
write-owners. There is new spinlock-owner/CPU-tracking on SMP builds too.
All locks have lockup detection now, which will work for both soft and hard
spin/rwlock lockups.
The arch-level include files now only contain the minimally necessary
subset of the spinlock code - all the rest that can be generalized now
lives in the generic headers:
include/asm-i386/spinlock_types.h | 16
include/asm-x86_64/spinlock_types.h | 16
I have also split up the various spinlock variants into separate files,
making it easier to see which does what. The new layout is:
SMP | UP
----------------------------|-----------------------------------
asm/spinlock_types_smp.h | linux/spinlock_types_up.h
linux/spinlock_types.h | linux/spinlock_types.h
asm/spinlock_smp.h | linux/spinlock_up.h
linux/spinlock_api_smp.h | linux/spinlock_api_up.h
linux/spinlock.h | linux/spinlock.h
/*
* here's the role of the various spinlock/rwlock related include files:
*
* on SMP builds:
*
* asm/spinlock_types.h: contains the raw_spinlock_t/raw_rwlock_t and the
* initializers
*
* linux/spinlock_types.h:
* defines the generic type and initializers
*
* asm/spinlock.h: contains the __raw_spin_*()/etc. lowlevel
* implementations, mostly inline assembly code
*
* (also included on UP-debug builds:)
*
* linux/spinlock_api_smp.h:
* contains the prototypes for the _spin_*() APIs.
*
* linux/spinlock.h: builds the final spin_*() APIs.
*
* on UP builds:
*
* linux/spinlock_type_up.h:
* contains the generic, simplified UP spinlock type.
* (which is an empty structure on non-debug builds)
*
* linux/spinlock_types.h:
* defines the generic type and initializers
*
* linux/spinlock_up.h:
* contains the __raw_spin_*()/etc. version of UP
* builds. (which are NOPs on non-debug, non-preempt
* builds)
*
* (included on UP-non-debug builds:)
*
* linux/spinlock_api_up.h:
* builds the _spin_*() APIs.
*
* linux/spinlock.h: builds the final spin_*() APIs.
*/
All SMP and UP architectures are converted by this patch.
arm, i386, ia64, ppc, ppc64, s390/s390x, x64 was build-tested via
crosscompilers. m32r, mips, sh, sparc, have not been tested yet, but should
be mostly fine.
From: Grant Grundler <grundler@parisc-linux.org>
Booted and lightly tested on a500-44 (64-bit, SMP kernel, dual CPU).
Builds 32-bit SMP kernel (not booted or tested). I did not try to build
non-SMP kernels. That should be trivial to fix up later if necessary.
I converted bit ops atomic_hash lock to raw_spinlock_t. Doing so avoids
some ugly nesting of linux/*.h and asm/*.h files. Those particular locks
are well tested and contained entirely inside arch specific code. I do NOT
expect any new issues to arise with them.
If someone does ever need to use debug/metrics with them, then they will
need to unravel this hairball between spinlocks, atomic ops, and bit ops
that exist only because parisc has exactly one atomic instruction: LDCW
(load and clear word).
From: "Luck, Tony" <tony.luck@intel.com>
ia64 fix
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Arjan van de Ven <arjanv@infradead.org>
Signed-off-by: Grant Grundler <grundler@parisc-linux.org>
Cc: Matthew Wilcox <willy@debian.org>
Signed-off-by: Hirokazu Takata <takata@linux-m32r.org>
Signed-off-by: Mikael Pettersson <mikpe@csd.uu.se>
Signed-off-by: Benoit Boissinot <benoit.boissinot@ens-lyon.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
There were three changes necessary in order to allow
sparc64 to use setup-res.c:
1) Sparc64 roots the PCI I/O and MEM address space using
parent resources contained in the PCI controller structure.
I'm actually surprised no other platforms do this, especially
ones like Alpha and PPC{,64}. These resources get linked into the
iomem/ioport tree when PCI controllers are probed.
So the hierarchy looks like this:
iomem --|
PCI controller 1 MEM space --|
device 1
device 2
etc.
PCI controller 2 MEM space --|
...
ioport --|
PCI controller 1 IO space --|
...
PCI controller 2 IO space --|
...
You get the idea. The drivers/pci/setup-res.c code allocates
using plain iomem_space and ioport_space as the root, so that
wouldn't work with the above setup.
So I added a pcibios_select_root() that is used to handle this.
It uses the PCI controller struct's io_space and mem_space on
sparc64, and io{port,mem}_resource on every other platform to
keep current behavior.
2) quirk_io_region() is buggy. It takes in raw BUS view addresses
and tries to use them as a PCI resource.
pci_claim_resource() expects the resource to be fully formed when
it gets called. The sparc64 implementation would do the translation
but that's absolutely wrong, because if the same resource gets
released then re-claimed we'll adjust things twice.
So I fixed up quirk_io_region() to do the proper pcibios_bus_to_resource()
conversion before passing it on to pci_claim_resource().
3) I was mistakedly __init'ing the function methods the PCI controller
drivers provide on sparc64 to implement some parts of these
routines. This was, of course, easy to fix.
So we end up with the following, and that nasty SPARC64 makefile
ifdef in drivers/pci/Makefile is finally zapped.
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
This patch gathers all the struct flock64 definitions (and the operations),
puts them under !CONFIG_64BIT and cleans up the arch files.
Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
This patch just gathers together all the struct flock definitions except
xtensa into asm-generic/fcntl.h.
Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
This patch puts the most popular of each fcntl operation/flag into
asm-generic/fcntl.h and cleans up the arch files.
Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
This patch puts the most popular of each open flag into asm-generic/fcntl.h
and cleans up the arch files.
Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
This set of patches creates asm-generic/fcntl.h and consolidates as much as
possible from the asm-*/fcntl.h files into it.
This patch just gathers all the identical bits of the asm-*/fcntl.h files into
asm-generic/fcntl.h.
Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Yoichi Yuasa <yuasa@hh.iij4u.or.jp>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
unused and useless..
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
IRQ_PER_CPU is not used by all architectures. This patch introduces the
macros ARCH_HAS_IRQ_PER_CPU and CHECK_IRQ_PER_CPU() to avoid the generation
of dead code in __do_IRQ().
ARCH_HAS_IRQ_PER_CPU is defined by architectures using IRQ_PER_CPU in their
include/asm_ARCH/irq.h file.
Through grepping the tree I found the following architectures currently use
IRQ_PER_CPU:
cris, ia64, ppc, ppc64 and parisc.
Signed-off-by: Karsten Wiese <annabellesgarden@yahoo.de>
Acked-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
The size of auxiliary vector is fixed at 42 in linux/sched.h. But it isn't
very obvious when looking at linux/elf.h. This patch adds AT_VECTOR_SIZE
so that we can change it if necessary when a new vector is added.
Because of include file ordering problems, doing this necessitated the
extraction of the AT_* symbols into a standalone header file.
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
When I first wrote the compat layer patches, I was somewhat cavalier about
the definition of compat_uid_t and compat_gid_t (or maybe I just
misunderstood :-)). This patch makes the compat types much more consistent
with the types we are being compatible with and hopefully will fix a few
bugs along the way.
compat type type in compat arch
__compat_[ug]id_t __kernel_[ug]id_t
__compat_[ug]id32_t __kernel_[ug]id32_t
compat_[ug]id_t [ug]id_t
The difference is that compat_uid_t is always 32 bits (for the archs we
care about) but __compat_uid_t may be 16 bits on some.
Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
ATM pthread_cond_signal is unnecessarily slow, because it wakes one waiter
(which at least on UP usually means an immediate context switch to one of
the waiter threads). This waiter wakes up and after a few instructions it
attempts to acquire the cv internal lock, but that lock is still held by
the thread calling pthread_cond_signal. So it goes to sleep and eventually
the signalling thread is scheduled in, unlocks the internal lock and wakes
the waiter again.
Now, before 2003-09-21 NPTL was using FUTEX_REQUEUE in pthread_cond_signal
to avoid this performance issue, but it was removed when locks were
redesigned to the 3 state scheme (unlocked, locked uncontended, locked
contended).
Following scenario shows why simply using FUTEX_REQUEUE in
pthread_cond_signal together with using lll_mutex_unlock_force in place of
lll_mutex_unlock is not enough and probably why it has been disabled at
that time:
The number is value in cv->__data.__lock.
thr1 thr2 thr3
0 pthread_cond_wait
1 lll_mutex_lock (cv->__data.__lock)
0 lll_mutex_unlock (cv->__data.__lock)
0 lll_futex_wait (&cv->__data.__futex, futexval)
0 pthread_cond_signal
1 lll_mutex_lock (cv->__data.__lock)
1 pthread_cond_signal
2 lll_mutex_lock (cv->__data.__lock)
2 lll_futex_wait (&cv->__data.__lock, 2)
2 lll_futex_requeue (&cv->__data.__futex, 0, 1, &cv->__data.__lock)
# FUTEX_REQUEUE, not FUTEX_CMP_REQUEUE
2 lll_mutex_unlock_force (cv->__data.__lock)
0 cv->__data.__lock = 0
0 lll_futex_wake (&cv->__data.__lock, 1)
1 lll_mutex_lock (cv->__data.__lock)
0 lll_mutex_unlock (cv->__data.__lock)
# Here, lll_mutex_unlock doesn't know there are threads waiting
# on the internal cv's lock
Now, I believe it is possible to use FUTEX_REQUEUE in pthread_cond_signal,
but it will cost us not one, but 2 extra syscalls and, what's worse, one of
these extra syscalls will be done for every single waiting loop in
pthread_cond_*wait.
We would need to use lll_mutex_unlock_force in pthread_cond_signal after
requeue and lll_mutex_cond_lock in pthread_cond_*wait after lll_futex_wait.
Another alternative is to do the unlocking pthread_cond_signal needs to do
(the lock can't be unlocked before lll_futex_wake, as that is racy) in the
kernel.
I have implemented both variants, futex-requeue-glibc.patch is the first
one and futex-wake_op{,-glibc}.patch is the unlocking inside of the kernel.
The kernel interface allows userland to specify how exactly an unlocking
operation should look like (some atomic arithmetic operation with optional
constant argument and comparison of the previous futex value with another
constant).
It has been implemented just for ppc*, x86_64 and i?86, for other
architectures I'm including just a stub header which can be used as a
starting point by maintainers to write support for their arches and ATM
will just return -ENOSYS for FUTEX_WAKE_OP. The requeue patch has been
(lightly) tested just on x86_64, the wake_op patch on ppc64 kernel running
32-bit and 64-bit NPTL and x86_64 kernel running 32-bit and 64-bit NPTL.
With the following benchmark on UP x86-64 I get:
for i in nptl-orig nptl-requeue nptl-wake_op; do echo time elf/ld.so --library-path .:$i /tmp/bench; \
for j in 1 2; do echo ( time elf/ld.so --library-path .:$i /tmp/bench ) 2>&1; done; done
time elf/ld.so --library-path .:nptl-orig /tmp/bench
real 0m0.655s user 0m0.253s sys 0m0.403s
real 0m0.657s user 0m0.269s sys 0m0.388s
time elf/ld.so --library-path .:nptl-requeue /tmp/bench
real 0m0.496s user 0m0.225s sys 0m0.271s
real 0m0.531s user 0m0.242s sys 0m0.288s
time elf/ld.so --library-path .:nptl-wake_op /tmp/bench
real 0m0.380s user 0m0.176s sys 0m0.204s
real 0m0.382s user 0m0.175s sys 0m0.207s
The benchmark is at:
http://sourceware.org/ml/libc-alpha/2005-03/txt00001.txt
Older futex-requeue-glibc.patch version is at:
http://sourceware.org/ml/libc-alpha/2005-03/txt00002.txt
Older futex-wake_op-glibc.patch version is at:
http://sourceware.org/ml/libc-alpha/2005-03/txt00003.txt
Will post a new version (just x86-64 fixes so that the patch
applies against pthread_cond_signal.S) to libc-hacker ml soon.
Attached is the kernel FUTEX_WAKE_OP patch as well as a simple-minded
testcase that will not test the atomicity of the operation, but at least
check if the threads that should have been woken up are woken up and
whether the arithmetic operation in the kernel gave the expected results.
Acked-by: Ingo Molnar <mingo@redhat.com>
Cc: Ulrich Drepper <drepper@redhat.com>
Cc: Jamie Lokier <jamie@shareable.org>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Yoichi Yuasa <yuasa@hh.iij4u.or.jp>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
