Comparing with kernel 2.6.24, tbench result has regression with
2.6.25-rc1.
1) On 2 quad-core processor stoakley: 4%.
2) On 4 quad-core processor tigerton: more than 30%.
bisect located below patch.
b4ce92775c is first bad commit
commit b4ce92775c
Author: Herbert Xu <herbert@gondor.apana.org.au>
Date: Tue Nov 13 21:33:32 2007 -0800
[IPV6]: Move nfheader_len into rt6_info
The dst member nfheader_len is only used by IPv6. It's also currently
creating a rather ugly alignment hole in struct dst. Therefore this patch
moves it from there into struct rt6_info.
Above patch changes the cache line alignment, especially member
__refcnt. I did a testing by adding 2 unsigned long pading before
lastuse, so the 3 members, lastuse/__refcnt/__use, are moved to next
cache line. The performance is recovered.
I created a patch to rearrange the members in struct dst_entry.
With Eric and Valdis Kletnieks's suggestion, I made finer arrangement.
1) Move tclassid under ops in case CONFIG_NET_CLS_ROUTE=y. So
sizeof(dst_entry)=200 no matter if CONFIG_NET_CLS_ROUTE=y/n. I
tested many patches on my 16-core tigerton by moving tclassid to
different place. It looks like tclassid could also have impact on
performance. If moving tclassid before metrics, or just don't move
tclassid, the performance isn't good. So I move it behind metrics.
2) Add comments before __refcnt.
On 16-core tigerton:
If CONFIG_NET_CLS_ROUTE=y, the result with below patch is about 18%
better than the one without the patch;
If CONFIG_NET_CLS_ROUTE=n, the result with below patch is about 30%
better than the one without the patch.
With 32bit 2.6.25-rc1 on 8-core stoakley, the new patch doesn't
introduce regression.
Thank Eric, Valdis, and David!
Signed-off-by: Zhang Yanmin <yanmin.zhang@intel.com>
Acked-by: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
There's a horrible slab abuse in net/netfilter/nf_conntrack_extend.c
that can be replaced with a call to ksize().
Cc: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Pekka Enberg <penberg@cs.helsinki.fi>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Use __KERNEL__ instead of __KERNEL to make sure the headers are not
usable by the kernel.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Current /proc/net is done with so called "shadows", but current
implementation is broken and has little chances to get fixed.
The problem is that dentries subtree of /proc/net directory has
fancy revalidation rules to make processes living in different
net namespaces see different entries in /proc/net subtree, but
currently, tasks see in the /proc/net subdir the contents of any
other namespace, depending on who opened the file first.
The proposed fix is to turn /proc/net into a symlink, which points
to /proc/self/net, which in turn shows what previously was in
/proc/net - the network-related info, from the net namespace the
appropriate task lives in.
# ls -l /proc/net
lrwxrwxrwx 1 root root 8 Mar 5 15:17 /proc/net -> self/net
In other words - this behaves like /proc/mounts, but unlike
"mounts", "net" is not a file, but a directory.
Changes from v2:
* Fixed discrepancy of /proc/net nlink count and selinux labeling
screwup pointed out by Stephen.
To get the correct nlink count the ->getattr callback for /proc/net
is overridden to read one from the net->proc_net entry.
To make selinux still work the net->proc_net entry is initialized
properly, i.e. with the "net" name and the proc_net parent.
Selinux fixes are
Acked-by: Stephen Smalley <sds@tycho.nsa.gov>
Changes from v1:
* Fixed a task_struct leak in get_proc_task_net, pointed out by Paul.
Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Acked-by: "Eric W. Biederman" <ebiederm@xmission.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
hci_sock_cleanup() always returns 0 and its return value isn't used
anywhere in the code.
Compile-tested with 'make allyesconfig && make net/bluetooth/bluetooth.ko'
Signed-off-by: Tobias Klauser <tklauser@distanz.ch>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Acked-by: Marcel Holtmann <marcel@holtmann.org>
Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
If all of the entropy is in the local and foreign addresses,
but xor'ing together would cancel out that entropy, the
current hash performs poorly.
Suggested by Cosmin Ratiu:
Basically, the situation is as follows: There is a client
machine and a server machine. Both create 15000 virtual
interfaces, open up a socket for each pair of interfaces and
do SIP traffic. By profiling I noticed that there is a lot of
time spent walking the established hash chains with this
particular setup.
The addresses were distributed like this: client interfaces
were 198.18.0.1/16 with increments of 1 and server interfaces
were 198.18.128.1/16 with increments of 1. As I said, there
were 15000 interfaces. Source and destination ports were 5060
for each connection. So in this case, ports don't matter for
hashing purposes, and the bits from the address pairs used
cancel each other, meaning there are no differences in the
whole lot of pairs, so they all end up in the same hash chain.
Signed-off-by: David S. Miller <davem@davemloft.net>
Based upon a report by Andrew Morton and code analysis done
by Jarek Poplawski.
This reverts 33f807ba0d ("[NETPOLL]:
Kill NETPOLL_RX_DROP, set but never tested.") and
c7b6ea24b4 ("[NETPOLL]: Don't need
rx_flags.").
The rx_flags did get tested for zero vs. non-zero and therefore we do
need those tests and that code which sets NETPOLL_RX_DROP et al.
Signed-off-by: David S. Miller <davem@davemloft.net>
The new SCTP socket api (draft 16) updates the AUTH API structures.
We never exported these since we knew they would change.
Update the rest to match the draft.
Signed-off-by: Vlad Yasevich <vladislav.yasevich@hp.com>
Properly add parens around the macro argument. This is not needed by
the kernel but the macro is exported to userspace, so it shouldn't
make any assumptions.
Also use NF_VERDICT_BITS instead of NF_VERDICT_QBTIS for the left-shift
since thats whats logically correct.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Don't require platform code to be #ifdeffed according to whether
I2C is enabled or not ... if it's not enabled, let GCC compile out
all I2C device declarations. (Issue noted on an NSLU2 build that
didn't configure I2C.)
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Jean Delvare <khali@linux-fr.org>
Commit 8b798c4d16 broke
alchemy build, fix it. Pointed out by Adrian Bunk.
Signed-off-by: Manuel Lauss <mano@roarinelk.homelinux.net>
Signed-off-by: Jean Delvare <khali@linux-fr.org>
The C99 specification states in section 6.11.5:
The placement of a storage-class specifier other than at the
beginning of the declaration specifiers in a declaration is an
obsolescent feature.
Signed-off-by: Tobias Klauser <tklauser@distanz.ch>
Signed-off-by: Jean Delvare <khali@linux-fr.org>
* 'upstream-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jgarzik/libata-dev:
libata-core: fix kernel-doc warning
sata_fsl: fix build with ATA_VERBOSE_DEBUG
[libata] ahci: AMD SB700/SB800 SATA support 64bit DMA
libata-pmp: clear hob for pmp register accesses
libata: automatically use DMADIR if drive/bridge requires it
power_state: get rid of write-only variable in SATA
pata_atiixp: Use 255 sector limit
Back in 2.6.17-rc2, a libata module parameter was added for atapi_dmadir.
That's nice, but most SATA devices which need it will tell us about it
in their IDENTIFY PACKET response, as bit-15 of word-62 of the
returned data (as per ATA7, ATA8 specifications).
So for those which specify it, we should automatically use the DMADIR bit.
Otherwise, disc writing will fail by default on many SATA-ATAPI drives.
This patch adds ATA_DFLAG_DMADIR and make ata_dev_configure() set it
if atapi_dmadir is set or identify data indicates DMADIR is necessary.
atapi_xlat() is converted to check ATA_DFLAG_DMADIR before setting
DMADIR.
Original patch is from Mark Lord.
Signed-off-by: Tejun Heo <htejun@gmail.com>
Cc: Mark Lord <mlord@pobox.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-2.6: (37 commits)
[NETFILTER]: fix ebtable targets return
[IP_TUNNEL]: Don't limit the number of tunnels with generic name explicitly.
[NET]: Restore sanity wrt. print_mac().
[NEIGH]: Fix race between neighbor lookup and table's hash_rnd update.
[RTNL]: Validate hardware and broadcast address attribute for RTM_NEWLINK
tg3: ethtool phys_id default
[BNX2]: Update version to 1.7.4.
[BNX2]: Disable parallel detect on an HP blade.
[BNX2]: More 5706S link down workaround.
ssb: Fix support for PCI devices behind a SSB->PCI bridge
zd1211rw: fix sparse warnings
rtl818x: fix sparse warnings
ssb: Fix pcicore cardbus mode
ssb: Make the GPIO API reentrancy safe
ssb: Fix the GPIO API
ssb: Fix watchdog access for devices without a chipcommon
ssb: Fix serial console on new bcm47xx devices
ath5k: Fix build warnings on some 64-bit platforms.
WDEV, ath5k, don't return int from bool function
WDEV: ath5k, fix lock imbalance
...
Use the added dev_alloc_name() call to create tunnel device name,
rather than iterate in a hand-made loop with an artificial limit.
Thanks Patrick for noticing this.
[ The way this works is, when the device is actually registered,
the generic code noticed the '%' in the name and invokes
dev_alloc_name() to fully resolve the name. -DaveM ]
Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
MAC_FMT had only one user and we tried to get rid of
that, but this created more problems than it solved.
As a result, this reverts three commits:
235365f3aa ("net/8021q/vlan_dev.c: Use
print_mac."), fea5fa875e ("[NET]: Remove
MAC_FMT"), and 8f789c4844 ("[NET]:
Elminate spurious print_mac() calls.")
Signed-off-by: David S. Miller <davem@davemloft.net>
- replace old name 'cont' with 'cgrp' (Paul Menage did this cleanup for
cgroup.c in commit bd89aabc67)
- remove a duplicate declaration of cgroup_path()
Signed-off-by: Li Zefan <lizf@cn.fujitsu.com>
Acked-by: Paul Menage <menage@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
fix:
- comments about need_forkexit_callback
- comments about release agent
- typo and comment style, etc.
Signed-off-by: Li Zefan <lizf@cn.fujitsu.com>
Acked-by: Paul Menage <menage@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Not all architectures implement futex_atomic_cmpxchg_inatomic(). The default
implementation returns -ENOSYS, which is currently not handled inside of the
futex guts.
Futex PI calls and robust list exits with a held futex result in an endless
loop in the futex code on architectures which have no support.
Fixing up every place where futex_atomic_cmpxchg_inatomic() is called would
add a fair amount of extra if/else constructs to the already complex code. It
is also not possible to disable the robust feature before user space tries to
register robust lists.
Compile time disabling is not a good idea either, as there are already
architectures with runtime detection of futex_atomic_cmpxchg_inatomic support.
Detect the functionality at runtime instead by calling
cmpxchg_futex_value_locked() with a NULL pointer from the futex initialization
code. This is guaranteed to fail, but the call of
futex_atomic_cmpxchg_inatomic() happens with pagefaults disabled.
On architectures, which use the asm-generic implementation or have a runtime
CPU feature detection, a -ENOSYS return value disables the PI/robust features.
On architectures with a working implementation the call returns -EFAULT and
the PI/robust features are enabled.
The relevant syscalls return -ENOSYS and the robust list exit code is blocked,
when the detection fails.
Fixes http://lkml.org/lkml/2008/2/11/149
Originally reported by: Lennart Buytenhek
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Acked-by: Ingo Molnar <mingo@elte.hu>
Cc: Lennert Buytenhek <buytenh@wantstofly.org>
Cc: Riku Voipio <riku.voipio@movial.fi>
Cc: <stable@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Merge include/linux/efs_fs{_i,_dir}.h into fs/efs/efs.h. efs_vh.h remains
there because this is the IRIX volume header and shouldn't really be
handled by efs but by the partitioning code. efs_sb.h remains there for
now because it's exported to userspace. Of course this wrong and aboot
should have a copy of it's own, but I'll leave that to a separate patch to
avoid any contention.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Fixes a sequencing bug in spi driver pxa2xx_spi.c in which the chip select
for a transfer may be asserted before the clock polarity is set on the
interface. As a result of this bug, the clock signal may have the wrong
polarity at transfer start, so it may need to make an extra half transition
before the intended clock/data signals begin. (This probably means all
transfers are one bit out of sequence.)
This only occurs on the first transfer following a change in clock polarity
in systems using more than one more than one such polarity. The fix
assures that the clock mode is properly set before asserting chip select.
This bug was introduced in a patch merged on 2006/12/10, kernel 2.6.20.
The patch defines an additional bit in: include/asm-arm/arch-pxa/regs-ssp.h
for 2.6.25 and newer kernels but this addition must be made in:
include/asm-arm/arch-pxa/pxa-regs.h for kernels between 2.6.20 and 2.6.24,
inclusive
Signed-off-by: Ned Forrester <nforrester@whoi.edu>
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Cc: Russell King <rmk@arm.linux.org.uk>
Cc: <stable@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Make is_vmalloc_addr() contingent on CONFIG_MMU=y, as it won't compile
in !MMU mode.
[ Bug introduced in commit 9e2779fa28:
"is_vmalloc_addr(): Check if an address is within the vmalloc
boundaries" ].
Signed-off-by: David Howells <dhowells@redhat.com>
Cc: Greg Ungerer <gerg@snapgear.com>
Cc: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Fix build failure on sparc:
In file included from include/linux/mm.h:39,
from include/linux/memcontrol.h:24,
from include/linux/swap.h:8,
from include/linux/suspend.h:7,
from init/do_mounts.c:6:
include/asm/pgtable.h:344: warning: parameter names (without
types) in function declaration
include/asm/pgtable.h:345: warning: parameter names (without
types) in function declaration
include/asm/pgtable.h:346: error: expected '=', ',', ';', 'asm' or
'__attribute__' before '___f___swp_entry'
viro sayeth:
I've run allmodconfig builds on a bunch of target, FWIW (essentially the
same patch). Note that these includes are recent addition caused by added
inline function that had since then become a define. So while I agree with
your comments in general, in _this_ case it's pretty safe.
The commit that had done it is 3062fc67da
("memcontrol: move mm_cgroup to header file") and the switch to #define
is in commit 60c12b1202 ("memcontrol: add
vm_match_cgroup()") (BTW, that probably warranted mentioning in the
changelog of the latter).
Cc: Adrian Bunk <bunk@kernel.org>
Cc: Robert Reif <reif@earthlink.net>
Signed-off-by: David Rientjes <rientjes@google.com>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Define HZ as a config option.
Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2.6.25-rc1 percpu changes broke CONFIG_DEBUG_PREEMPT's per_cpu checking
on several architectures. On s390, sparc64 and x86 it's been weakened to
not checking at all; whereas on powerpc64 it's become too strict, issuing
warnings from __raw_get_cpu_var in io_schedule and init_timer for example.
Fix this by weakening powerpc's __my_cpu_offset to use the non-checking
local_paca instead of get_paca (which itself contains such a check);
and strengthening the generic my_cpu_offset to go the old slow way via
smp_processor_id when CONFIG_DEBUG_PREEMPT (debug_smp_processor_id is
where all the knowledge of what's correct when lives).
Signed-off-by: Hugh Dickins <hugh@veritas.com>
Reviewed-by: Mike Travis <travis@sgi.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
During the last step of hibernation in the "platform" mode (with the
help of ACPI) we use the suspend code, including the devices'
->suspend() methods, to prepare the system for entering the ACPI S4
system sleep state.
But at least for some devices the operations performed by the
->suspend() callback in that case must be different from its operations
during regular suspend.
For this reason, introduce the new PM event type PM_EVENT_HIBERNATE and
pass it to the device drivers' ->suspend() methods during the last phase
of hibernation, so that they can distinguish this case and handle it as
appropriate. Modify the drivers that handle PM_EVENT_SUSPEND in a
special way and need to handle PM_EVENT_HIBERNATE in the same way.
These changes are necessary to fix a hibernation regression related
to the i915 driver (ref. http://lkml.org/lkml/2008/2/22/488).
Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Acked-by: Pavel Machek <pavel@ucw.cz>
Tested-by: Jeff Chua <jeff.chua.linux@gmail.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Add proper ifdef's to the patch loading code moved from the old instr
layer so that opl3 driver can be compiled without the sequencer support.
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Fix macro argument substitution in PageHead() and PageTail() - 'page' should
have brackets surrounding it (commit 6d7779538f).
Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/pci-2.6:
PCI: Fix wrong reference counter check for proc_dir_entry
PCI: fix up setup-bus.c #ifdef
PCI: don't load acpi_php when acpi is disabled
PCI: quirks: set 'En' bit of MSI Mapping for devices onHT-based nvidia platform
PCI: kernel-doc: fix pci-acpi warning
PCI: irq: patch for Intel ICH10 DeviceID's
PCI: pci_ids: patch for Intel ICH10 DeviceID's
PCI: AMD SATA IDE mode quirk
PCI: drivers/pcmcia/i82092.c: fix up after pci_bus_region changes
PCI: hotplug: acpiphp_ibm: Remove get device information
* 'for-linus' of master.kernel.org:/home/rmk/linux-2.6-arm:
[ARM] 4835/1: Fix stale comment in struct machine_desc description
[ARM] 4829/1: add .get method to pxa-cpufreq to silence a warning
[ARM] 4828/1: fix 3 warnings in drivers/video/pxafb.c
[ARM] 4827/1: fix two warnings in drivers/i2c/busses/i2c-pxa.c
[ARM] 4826/1: Orion: Register the RTC interrupt on the TS-209
[ARM] pxa: fix clock lookup to find specific device clocks
This patch adds the Intel ICH10 LPC and SMBus Controller DeviceID's.
Signed-off-by: Jason Gaston <jason.d.gaston@intel.com>
Cc: Jean Delvare <khali@linux-fr.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
This patch updates stale comment that pointed to nonexistent file.
Signed-off-by: Leonid Evdokimov <leon@darkk.net.ru>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Make hibernation work with CONFIG_DEBUG_PAGEALLOC set on x86, by
checking if the pages to be copied are marked as present in the
kernel mapping and temporarily marking them as present if that's not
the case. No functional modifications are introduced if
CONFIG_DEBUG_PAGEALLOC is unset.
Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Signed-off-by: Len Brown <len.brown@intel.com>
Change the FRV timerfd syscalls to be the same as i386 timerfd syscalls.
Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* 'upstream-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jgarzik/libata-dev:
libata: fix kernel-doc parameter warning
sata_mv: remove iounmap in mv_platform_remove and use devm_iomap
ata: fix sparse warning in libata.h
libata: implement libata.force module parameter
sata_mv: use hpriv->base instead of the host->iomap
sata_mv: Define module alias for platform device
ata: fix sparse warnings in pata_legacy.c
Noted by various people (Sam, Jeff, Roland..)
Commit 58b7983d15 intended to remove the
xfs "Makefile-linux-2.6" file, but it was mistakenly still left in the
tree as a empty file, and would cause git to correctly complain about a
tracked file being removed after a "make distclean" (which removes empty
files as garbage).
And the asm-x86/desc_64.h file was supposed to be removed by commit
c81c6ca45a, but instead stayed around
containing just a single newline.
Get rid of them both properly.
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This fixes the pcicore driver to not die a horrible
crash death when inserting a cardbus card.
Signed-off-by: Michael Buesch <mb@bu3sch.de>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
