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>
This patch adds some missing pci-related calls to the suspend and resume
routines of the 3c59x driver. It also makes the driver free/request IRQ on
suspend/resume, in accordance with the proposal at:
http://lists.osdl.org/pipermail/linux-pm/2005-May/000955.html
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
This updates documentation a bit (mostly removing obsolete stuff), and
marks swsusp as no longer experimental in config.
Signed-off-by: Pavel Machek <pavel@suse.cz>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
This allows a valid iommu placed immediately after memory to work, to be
recognized as after the last byte of memory and not overlapping it.
Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
Acked-by: Andi Kleen <ak@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Need to create sysfs only for cpus that are present. Without which we see
NR_CPUS entries created when we have CONFIG_HOTPLUG and CONFIG_HOTPLUG_CPU
enabled.
Signed-off-by: Ashok Raj <ashok.raj@intel.com>
Acked-by: Andi Kleen <ak@muc.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Need to ensure we dont get prempted when we clear ourself from mask when using
clustered mode genapic code.
Signed-off-by: Ashok Raj <ashok.raj@intel.com>
Acked-by: Andi Kleen <ak@muc.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
This is a multi-part message in MIME format. If the cpu lacks 3DNOW
feature, we can use a normal prefetcht0 instruction instead of NOP5.
"prefetchw (%rxx)" and "prefetcht0 (%rxx)" have the same length, ranging
from 3 to 5 bytes depending on the register. So this patch even helps
AMD64, shortening the length of the code.
Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Acked-by: Andi Kleen <ak@muc.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Up to date I've been using the GS value to determine the processor number
in dumps from show_regs, however this can be cumbersome to do if you don't
have the vmlinux to verify with the address of cpu_pda, how about the
following? I considered using hard_smp_processor_id for robustness but we
already dereference current so we're already relying on MSR_GS_BASE being
sane.
Signed-off-by: Zwane Mwaikambo <zwane@arm.linux.org.uk>
Acked-by: Andi Kleen <ak@muc.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
When handling writes to /proc/irq, current code is re-programming rte
entries directly. This is not recommended and could potentially cause
chipset's to lockup, or cause missing interrupts.
CONFIG_IRQ_BALANCE does this correctly, where it re-programs only when the
interrupt is pending. The same needs to be done for /proc/irq handling as well.
Otherwise user space irq balancers are really not doing the right thing.
- Changed pending_irq_balance_cpumask to pending_irq_migrate_cpumask for
lack of a generic name.
- added move_irq out of IRQ_BALANCE, and added this same to X86_64
- Added new proc handler for write, so we can do deferred write at irq
handling time.
- Display of /proc/irq/XX/smp_affinity used to display CPU_MASKALL, instead
it now shows only active cpu masks, or exactly what was set.
- Provided a common move_irq implementation, instead of duplicating
when using generic irq framework.
Tested on i386/x86_64 and ia64 with CONFIG_PCI_MSI turned on and off.
Tested UP builds as well.
MSI testing: tbd: I have cards, need to look for a x-over cable, although I
did test an earlier version of this patch. Will test in a couple days.
Signed-off-by: Ashok Raj <ashok.raj@intel.com>
Acked-by: Zwane Mwaikambo <zwane@holomorphy.com>
Grudgingly-acked-by: Andi Kleen <ak@muc.de>
Signed-off-by: Coywolf Qi Hunt <coywolf@lovecn.org>
Signed-off-by: Ashok Raj <ashok.raj@intel.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Patch from Ben Dooks
This has been broken for a while now, so fix the
problems with the code, test and bring up to date.
This also makes the code conditional on an
Kconfig option
Signed-off-by: Ben Dooks <ben-linux@fluff.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Patch from Adam Brooks
The current gettimeofday implementation for the IOP3xx processors reads the contents of the timer interrupt register and does math on the value to figure out exactly what time it is. To do this it multiplies the contents of the timer register with a large constant. The result is then divided by a large constant. Unfortunately the result of the first multiplication is often too large for the register to hold. The solution is to combine the two large constants to a single smaller constant at compile time. Then the timer value can be divided by single smaller constant without any overflow issues.
Signed-off-by: Adam Brooks <adam.j.brooks@intel.com>
Signed-off-by: Deepak Saxena <dsaxena@plexity.net>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Patch from Richard Purdie
Add machine detection routines for several new models of PXA based Zaurii
(SL-C3000 - Spitz, SL-C1000 - Akita, SL-C3100 - Borzoi, SL-C6000 - Tosa).
Sharp continue to use broken bootloaders in ROM.
Signed-off-by: Richard Purdie <rpurdie@rpsys.net>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Patch from Tony Lindgren
This patch updates H2 defconfig
Signed-off-by: Tony Lindgren <tony@atomide.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Patch from Tony Lindgren
This patch syncs the mainline kernel with linux-omap tree.
The highlights of the patch are:
- Start adding 24xx support by Paul Mundt
- Clean-up of cpu detection by Dirk Behme and Tony Lindgren
- Add DSP header by Toshihiro Kobayashi
- Add support for mtd-xip by Vladimir Barinov
- Add various new mux registers
- Move OMAP specific serial defines back to serial.h
Signed-off-by: Tony Lindgren <tony@atomide.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Patch from Tony Lindgren
This patch syncs the mainline kernel with linux-omap tree.
The highlights of the patch are:
- Clock updates by Tuukka Tikkanen, Juha Yrjola,
Daniel Petrini and Tony Lindgren
- DMA fixes by Imre Deak, Juha Yrjola and Daniel Petrini
- Add support to dual-mode hardware timers by Lauri Leukkunen
- GPIO support for 24xx by Paul Mundt
- GPIO wake-up support by Tony Lindgren
- Better GPIO interrupt handler to not lose interrupts by
Ralph Walden and Ladislav Michl
- Power Management updates by Tuukka Tikkanen
- Make Power Management code use new SRAM functions by
Tony Lindgren
Signed-off-by: Tony Lindgren <tony@atomide.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Patch from Ben Dooks
Add the Simtec Anubis to the list of supported
machines in the arch/arm/mach-s3c2410 directory.
This ensures the core peripherals are registered,
the timer source is configured and the correct
power-management is enabled.
Signed-off-by: Ben Dooks <ben-linux@fluff.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
This is my libata compatible low level driver for the Marvell SATA
family. Currently it successfully runs in PIO mode on a 6081 chip.
EDMA support is in the works and should be done shortly. Review,
testing (especially on other flavors of Marvell), comments welcome.
Signed-off-by: Brett Russ <russb@emc.com>
Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
ata_get_mode_mask() uses bits 3 and 4 in the pio_mask to represent PIO
modes 3 and 4. The value read from the drive, which reports support
for PIO3 and PIO4 in bits 0 and 1, is shifted left by 3 bits and OR'd
with 0x7 (which then corresponds to PIO 2-0 in libata). Thus, the
drivers below need adjustments to comply with the way pio_mask is
used. I changed the masks from the commented values to all support
PIO4-0, since the spec mandates that PIO0-2 are supported and there's
no reason not to support PIO3 IMO.
Signed-off-by: Brett Russ <russb@emc.com>
Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
No code changes, just Lindent + manual fixups.
This prepares us for updating to the latest Intel driver code, plus
gives the source code a nice facelift.
Need to use compat struct sizes and compat_sys_ioctl().
Reported by Adrian Bunk via kernel bugzilla #2683
Signed-off-by: David S. Miller <davem@davemloft.net>
From: Max Kellermann <max@duempel.org>
The sunrpc stats are collected in unsigned integers, but they are printed
with '%d'. That can result in negative numbers in /proc/net/rpc when the
highest bit of a counter is set. The following patch changes '%d' to '%u'
where appropriate.
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
proto_unregister holds a lock while calling kmem_cache_destroy, which
can sleep.
Noticed by Daniele Orlandi <daniele@orlandi.com>.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
result of comma operator is not an lvalue
Signed-off-by: Al Viro <viro@parcelfarce.linux.theplanet.co.uk>
Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
Dan Williams already included most parts of my WE-19 patch for
the airo driver in the kernel. There was just a few bits he could not
do because WE-19 itself was not in the kernel. Those are the missing
bits.
Tested with 2.6.13 (with real HW).
Signed-off-by: Jean Tourrilhes <jt@hpl.hp.com>
Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
My patch that adds WE-17 support to the Prism54 driver went
already in the kernel, except for a tiny bit that was dropped on the
way. This is the missing bit....
Tested with 2.6.13 (with real HW).
Signed-off-by: Jean Tourrilhes <jt@hpl.hp.com>
Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
wl3501_cs won't compile with WE-19. This patches fixes it.
Signed-off-by: Jean Tourrilhes <jt@hpl.hp.com>
Acked-by: Arnaldo Carvalho de Melo <acme@conectiva.com.br>
Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
This adds support for WE-17 to the atmel_cs driver. Not
tested, I don't have the HW.
Signed-off-by: Jean Tourrilhes <jt@hpl.hp.com>
Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
This adds support for WE-17 to the netwave_cs driver. Tested
with 2.6.13 (with real HW).
Signed-off-by: Jean Tourrilhes <jt@hpl.hp.com>
Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
This adds support for WE-17 to the ray_cs driver. Tested
with 2.6.13 (with real HW).
Signed-off-by: Jean Tourrilhes <jt@hpl.hp.com>
Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
Hi Jeff,
This is version 19 of the Wireless Extensions. It was supposed
to be the fallback of the WPA API changes, but people seem quite happy
about it (especially Jouni), so the patch is rather small.
The patch has been fully tested with 2.6.13 and various
wireless drivers, and is in its final version. Would you mind pushing
that into Linus's kernel so that the driver and the apps can take
advantage ot it ?
It includes :
o iwstat improvement (explicit dBm). This is the result of
long discussions with Dan Williams, the authors of
NetworkManager. Thanks to him for all the fruitful feedback.
o remove pointer from event stream. I was not totally sure if
this pointer was 32-64 bits clean, so I'd rather remove it and be at
peace with it.
o remove linux header from wireless.h. This has long been
requested by people writting user space apps, now it's done, and it
was not even painful.
o final deprecation of spy_offset. You did not like it, it's
now gone for good.
o Start deprecating dev->get_wireless_stats -> debloat netdev
o Add "check" version of event macros for ieee802.11
stack. Jiri Benc doesn't like the current macros, we aim to please ;-)
All those changes, except the last one, have been bit-roting on
my web pages for a while...
Patches for most kernel drivers will follow. Patches for the
Orinoco and the HostAP drivers have been sent to their respective
maintainers.
Have fun...
Jean
Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
Another small update for the spidernet driver to fix a bug encountered
during testing our latest hardware with dual-ethernet support.
Signed-off-by: Arnd Bergmann <arndb@de.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
- Prevent PCI posting problems by using synchronous register access
in critical places
- Check return value from firmware device tree functions
- fix device cleanup
Signed-off-by: Arnd Bergmann <arndb@de.ibm.com>
Cc: Jeff Garzik <jgarzik@pobox.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
This patch adds a driver for a new 1000 Mbit ethernet NIC. It is
integrated on the south bridge that is used for our Cell Blades.
The code gets the MAC address from the Open Firmware device tree, so it
won't compile on platforms other than ppc64.
This is the first public release, so I don't expect the first version to
get merged, but I'd aim for integration within the 2.6.13 time frame.
Cc: Utz Bacher <utz.bacher@de.ibm.com>
Signed-off-by: Arnd Bergmann <arndb@de.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
My overlords have asked me to update the copyright notice for iseries_veth.
Signed-off-by: Michael Ellerman <michael@ellerman.id.au>
Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
no need to mess with (wrong) casts for ->mem_start, when we have the
original iomem pointer used to set ->mem_start in the first place...
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
Hi,
This patch contains the following hardware related fixes and other
miscellaneous bug fixes.
1. Updated the definition of single and double-bit ECC errors
2. Earlier we were allocating Transmit descriptors equal to
MAX_SKB_FRAGS. This was causing a boundary condition failure.
Need to allocate MAX_SKB_FRAGS+1 descriptors.
3. On some platforms(like PPC), pci_alloc_consistent() can return
a zero DMA address. Since the NIC cannot handle zero-addresses,
a workaround has been provided. Basically, we don't use such
that page. We reallocate.
4. If list_info allocation failed during driver load, check for
it during driver exit and return instead of trying to dereference
NULL pointer.
5. Increase the debug level of few non-critical debug messages.
6. Reset the card on critical ECC double errors only in case of
XframeI since XframeII can recover from such errors.
7. Print copyright message on driver load.
8. Bumped up the driver version no. to 2.0.8.1
Signed-off-by: Ravinandan Arakali <ravinandan.arakali@neterion.com>
Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
This was found by inspection while looking for checksum problems
with the skge driver that sets CHECKSUM_HW. It did not fix the
problem, but it looks like it is needed.
If IP reassembly is trimming an overlapping fragment, it
should reset (or adjust) the hardware checksum flag on the skb.
Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Ax2asc was still using a static buffer for all invocations which isn't
exactly SMP-safe. Change ax2asc to take an additional result buffer as
the argument. Change all callers to provide such a buffer.
Signed-off-by: Ralf Baechle DL5RB <ralf@linux-mips.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
The new timestamp get/set routines should have const attribute
on parameters (helps to indicate direction).
Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Jakub Jelinek