powerpc: Merge semaphore.h
Adopted the ppc64 version of semaphore.h. The 32-bit version used
smp_wmb(), but recent updates to atomic.h mean this is no longer required.
The 64-bit version made use of unlikely(), which has been retained in the
combined version.
This patch requires the recent atomic.h patch.
Signed-off-by: Becky Bruce <becky.bruce@freescale.com>
Signed-off-by: Kumar Gala <kumar.gala@freescale.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
Merge asm-ppc*/rwsem.h into include/asm-powerpc.
Removed smp_*mb() memory barriers from the ppc32 code
as they are now burried in the atomic_*() functions as
suggested by Paul, implemented by Arnd, and pushed out
by Becky. I am not the droid you are looking for.
This patch depends on Becky's atomic.h merge patch.
Signed-off-by: Jon Loeliger <jdl@freescale.com>
Signed-off-by: Kumar Gala <kumar.gala@freescale.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
powerpc: Merge atomic.h and memory.h into powerpc
Merged atomic.h into include/powerpc. Moved asm-style HMT_ defines from
memory.h into ppc_asm.h, where there were already HMT_defines; moved c-style
HMT_ defines to processor.h. Renamed memory.h to synch.h to better reflect
its contents.
Signed-off-by: Kumar Gala <kumar.gala@freescale.com>
Signed-off-by: Becky Bruce <becky.bruce@freescale.com>
Signed-off-by: Jon Loeliger <linuxppc@jdl.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
Merge asm-ppc*/seccomp.h. Drop TIF_32BIT check.
Signed-off-by: Jon Loeliger <jdl@freescale.com>
Signed-off-by: Kumar Gala <kumar.gala@freescale.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
Merged hw_irq.h between ppc32 & ppc64. Added support to use the Book-E
wrtee[i] instructions that allow modifying MSR[EE] atomically.
Additionally, added get_irq_desc() macros to ppc32 to allow mask_irq(),
unmask_irq(), and ack_irq() to be common between ppc32 & ppc64.
Note: because 64-bit Book-E implementations only have a 32-bit MSR the
macro's for Book-E need to come before the PPC64 macro's to ensure the
right thing happends for 64-bit Book-E processors.
Signed-off-by: Kumar Gala <kumar.gala@freescale.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
This merges the asm-ppc*/dma.h files.
Signed-off-by: Jon Loeliger <jdl@freescale.com>
Signed-off-by: Kumar Gala <kumar.gala@freescale.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
Merge asm-ppc*/vga.h
Signed-off-by: Jon Loeliger <jdl@freescale.com>
Signed-off-by: Kumar Gala <kumar.gala@freescale.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
ppc/ppc64: Merge elf.h into include/asm-powerpc
Merge elf.h into a single include file for 32 and 64-bit ppc platforms. This
patch has been tested on 32-bit and built on 64-bit platforms.
Signed-off-by: Kumar Gala <kumar.gala@freescale.com>
Signed-off-by: Becky Bruce <Becky.Bruce@freescale.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
ppc32/ppc64: Merge bug.h into include/asm-powerpc
This patch merges bug.h into include/asm-powerpc. Changed the data
structure for bug_entry such that line is always an int on both 32 and
64-bit platforms; removed casts to int from the 64-bit trap code to
reflect this.
Signed-off-by: Kumar Gala <kumar.gala@freescale.com>
Signed-off-by: Becky Bruce <Becky.Bruce@freescale.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
This is a revised patch to merge asm-ppc*/hardirq.h.
It removes some unnecessary #includes, but then requires
the addition of #include <asm/irq.h> in PPC32's hw_irq.h
much like ppc64 already does. Furthermore, several
unnecessary #includes were removed from some ppc32 boards
in order to break resulting bad #include cycles.
Builds pSeries_defconfig and all ppc32 platforms except
the already b0rken bseip.
Signed-off-by: Jon Loeliger <jdl@freescale.com>
Signed-off-by: Kumar Gala <kumar.gala@freescale.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
On ppc64 timer_interrupt() returned a value that was never used. Changed
the ppc64 version of timer_interrupt() to no longer return a value so
that the signatures between ppc32 & ppc64 match. This will simplify
future merging of arch/powerpc.
Signed-off-by: Kumar Gala <kumar.gala@freescale.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
Merge include/asm-ppc64/oprofile_ipml.h and arch/ppc/oprofile/op_impl.h
into include/asm-powerpc/oprofile_ipml.h
Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Paul Mackerras <paulus@samba.org>
Merge asm-ppc/posix_types.h and asm-ppc64/posix_types.h.
Signed-off-by: Jon Loeliger <jdl@freescale.com>
Signed-off-by: Kumar Gala <kumar.gala@freescale.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
This patch slightly change the TLB flush batch mecanism so that we
store the full vaddr (including vsid) when adding an entry to the
batch so that the flush part doesn't have to get to the context.
This cleans it a bit, and paves the way to future updates like
dynamic vsids.
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
Replace some of the hard-coded constants with PAGE_SIZE/SHIFT/ORDER where
appropriate.
Likewise, in a couple of places it doesn't make sense to base some
allocations on page size when all that's required is a constant 4K,
etc.
Signed-off-by: Olof Johansson <olof@lixom.net>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
There are potential cases in the future where the IOMMU might be
mapping smaller pages than the regular MMU is using. Keep the
allocator working on MMU pagesizes, but the low-level mapping
functions need to map more than one TCE entry per page to deal with
this.
Signed-off-by: Olof Johansson <olof@lixom.net>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
Split out the implementation-specific parts of include/asm-ppc64/iommu.h
to separate include files (tce.h and dart.h respectively).
The generic iommu code really doesn't care about the underlying
implementation, and the TCE and DART stuff is completely different.
Signed-off-by: Olof Johansson <olof@lixom.net>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
Here is a new patch that removes all notion of the pmac, prep,
chrp and openfirmware initialization sections, and then unifies
the sections.h files without those __pmac, etc, sections identifiers
cluttering things up.
Signed-off-by: Jon Loeliger <jdl@freescale.com>
Signed-off-by: Kumar Gala <kumar.gala@freescale.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
Merged ppc_asm.h between ppc32 & ppc64. The majority of the file is
common between the two architectures excluding how a single GPR is
saved/restored and which GPRs are non-volatile.
Additionally, moved the ASM_CONST macro used on ppc64 into ppc_asm.h.
Signed-off-by: Kumar Gala <kumar.gala@freescale.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
Here is a patch to merge the ppc and pp64 version of kmap_types.h
Signed-off-by: Jon Loeliger <jdl@freescale.com>
Signed-off-by: Kumar Gala <kumar.gala@freescale.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
ppc64_attention_msg and ppc64_dump_msg are not used so remove them.
Signed-off-by: Anton Blanchard <anton@samba.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
- Add PTRACE_GET_DEBUGREG/PTRACE_SET_DEBUGREG. The definition is
as follows:
/*
* Get or set a debug register. The first 16 are DABR registers and the
* second 16 are IABR registers.
*/
#define PTRACE_GET_DEBUGREG 25
#define PTRACE_SET_DEBUGREG 26
DABR == data breakpoint and IABR = instruction breakpoint in IBM
speak. We could split out the IABR into 2 more ptrace calls but I
figured there was no need and 16 DABR registers should be more
than enough (POWER4/POWER5 have one).
- Add 2 new SIGTRAP si_codes: TRAP_HWBKPT and TRAP_BRANCH. I couldnt
find any standards on either of these so I copied what ia64 is
doing. Again this might be better placed in
include/asm-generic/siginfo.h
Signed-off-by: Anton Blanchard <anton@samba.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
- Remove the PPC_REG* defines
- Wrap some more stuff with ifdef __KERNEL__
- Add missing PT_TRAP, PT_DAR, PT_DSISR defines
- Add PTRACE_GETEVRREGS/PTRACE_SETEVRREGS, even though we dont use it on
ppc64 we dont want to allocate them for something else.
Signed-off-by: Anton Blanchard <anton@samba.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
The ptrace get and set methods for VMX/Altivec registers present in the
ppc tree were missing for ppc64. This patch adds the 32-bit and
64-bit methods. Updated with the suggestions from Anton following the lines
of his code snippet.
Added:
- flush_altivec_to_thread calls as suggested by Anton
- piecewise copy of structure to preserve 32-bit vrsave data as per
Anton
(I consolidated the 32 and 64bit versions with 2 helper macros - Anton)
Signed-off-by: Robert C Jennings <rcjenn@austin.ibm.com>
Signed-off-by: Anton Blanchard <anton@samba.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
This adds code which gives us the option on ppc64 of instantiating the
PCI tree (the tree of pci_bus and pci_dev structs) from the Open
Firmware device tree rather than by probing PCI configuration space.
The OF device tree has a node for each PCI device and bridge in the
system, with properties that tell us what addresses the firmware has
configured for them and other details.
There are a couple of reasons why this is needed. First, on systems
with a hypervisor, there is a PCI-PCI bridge per slot under the PCI
host bridges. These PCI-PCI bridges have special isolation features
for virtualization. We can't write to their config space, and we are
not supposed to be reading their config space either. The firmware
tells us about the address ranges that they pass in the OF device
tree.
Secondly, on powermacs, the interrupt controller is in a PCI device
that may be behind a PCI-PCI bridge. If we happened to take an
interrupt just at the point when the device or a bridge on the path to
it was disabled for probing, we would crash when we try to access the
interrupt controller.
I have implemented a platform-specific function which is called for
each PCI bridge (host or PCI-PCI) to say whether the code should look
in the device tree or use normal PCI probing for the devices under
that bridge. On pSeries machines we use the device tree if we're
running under a hypervisor, otherwise we use normal probing. On
powermacs we use normal probing for the AGP bridge, since the device
for the AGP bridge itself isn't shown in the device tree (at least on
my G5), and the device tree for everything else.
This has been tested on a dual G5 powermac, a partition on a POWER5
machine (running under the hypervisor), and a legacy iSeries
partition.
Signed-off-by: Paul Mackerras <paulus@samba.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>
This patch pulls the PCI-related junk out of struct device_node and
puts it in a separate structure, struct pci_dn. The device_node now
just has a void * pointer in it, which points to a struct pci_dn for
nodes that represent PCI devices. It could potentially be used in
future for device-specific data for other sorts of devices, such as
virtual I/O devices.
Signed-off-by: Paul Mackerras <paulus@samba.org>
Remove asm-ppc64/segment.h now that all users are gone.
Signed-off-by: Kumar Gala <kumar.gala@freescale.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
Merge a few asm-ppc and asm-ppc64 header files.
Note: the merge of setup.h intentionally does not carry
forward the m68k cruft. That means this patch continues
to break the already broken amiga on the ppc32.
Signed-off-by: Jon Loeliger <jdl@freescale.com>
Signed-off-by: Kumar Gala <kumar.gala@freescale.com>
Signed-off-by: Paul Mackerras <paulus@samba.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 fixes a race condition where in system used to hang or sometime
crash within minutes when kprobes are inserted on ISR routine and a task
routine.
The fix has been stress tested on i386, ia64, pp64 and on x86_64. To
reproduce the problem insert kprobes on schedule() and do_IRQ() functions
and you should see hang or system crash.
Signed-off-by: Anil S Keshavamurthy <anil.s.keshavamurthy@intel.com>
Signed-off-by: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Acked-by: Prasanna S Panchamukhi <prasanna@in.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
This patch contains the ppc64 architecture specific changes to prevent the
possible race conditions.
Signed-off-by: Prasanna S Panchamukhi <prasanna@in.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
This makes sense now that we have asm-powerpc.
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>
These two files are basically identical, so make one just include the other
(protecting the 32-bit-only parts with __powerpc64__). Also remove some
completely unused defines.
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>
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>
Currently, we set the class bit in kernel SLB entries, and clear it on
user SLB entries. On POWER5, ERAT entries created in real mode have
the class bit clear. So to avoid flushing kernel ERAT entries on each
context switch, this patch inverts our usage of the class bit, setting
it on user SLB entries and clearing it on kernel SLB entries.
Booted on POWER5 and G5.
Signed-off-by: David Gibson <dwg@au1.ibm.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
