David Howells [Tue, 13 Dec 2011 14:56:54 +0000 (14:56 +0000)]
x86_64, asm: Optimise fls(), ffs() and fls64()
fls(N), ffs(N) and fls64(N) can be optimised on x86_64. Currently they use a
CMOV instruction after the BSR/BSF to set the destination register to -1 if the
value to be scanned was 0 (in which case BSR/BSF set the Z flag).
Instead, according to the AMD64 specification, we can make use of the fact that
BSR/BSF doesn't modify its output register if its input is 0. By preloading
the output with -1 and incrementing the result, we achieve the desired result
without the need for a conditional check.
The Intel x86_64 specification, however, says that the result of BSR/BSF in
such a case is undefined. That said, when queried, one of the Intel CPU
architects said that the behaviour on all Intel CPUs is that:
(1) with BSRQ/BSFQ, the 64-bit destination register is written with its
original value if the source is 0, thus, in essence, giving the effect we
want. And,
(2) with BSRL/BSFL, the lower half of the 64-bit destination register is
written with its original value if the source is 0, and the upper half is
cleared, thus giving us the effect we want (we return a 4-byte int).
Further, it was indicated that they (Intel) are unlikely to get away with
changing the behaviour.
It might be possible to optimise the 32-bit versions of these functions, but
there's a lot more variation, and so the effective non-destructive property of
BSRL/BSRF cannot be relied on.
[ hpa: specifically, some 486 chips are known to NOT have this property. ]
I have benchmarked these functions on my Core2 Duo test machine using the
following program:
#include <stdlib.h>
#include <stdio.h>
#ifndef __x86_64__
#error
#endif
#define PAGE_SHIFT 12
typedef unsigned long long __u64, u64;
typedef unsigned int __u32, u32;
#define noinline __attribute__((noinline))
static __always_inline int fls64(__u64 x)
{
long bitpos = -1;
static inline unsigned long __fls(unsigned long word)
{
asm("bsr %1,%0"
: "=r" (word)
: "rm" (word));
return word;
}
static __always_inline int old_fls64(__u64 x)
{
if (x == 0)
return 0;
return __fls(x) + 1;
}
static noinline // __attribute__((const))
int old_get_order(unsigned long size)
{
int order;
size = (size - 1) >> (PAGE_SHIFT - 1);
order = -1;
do {
size >>= 1;
order++;
} while (size);
return order;
}
static inline __attribute__((const))
int get_order_old_fls64(unsigned long size)
{
int order;
size--;
size >>= PAGE_SHIFT;
order = old_fls64(size);
return order;
}
static inline __attribute__((const))
int get_order(unsigned long size)
{
int order;
size--;
size >>= PAGE_SHIFT;
order = fls64(size);
return order;
}
unsigned long prevent_optimise_out;
static noinline unsigned long test_old_get_order(void)
{
unsigned long n, total = 0;
long rep, loop;
for (rep = 1000000; rep > 0; rep--) {
for (loop = 0; loop <= 16384; loop += 4) {
n = 1UL << loop;
total += old_get_order(n);
}
}
return total;
}
static noinline unsigned long test_get_order_old_fls64(void)
{
unsigned long n, total = 0;
long rep, loop;
for (rep = 1000000; rep > 0; rep--) {
for (loop = 0; loop <= 16384; loop += 4) {
n = 1UL << loop;
total += get_order_old_fls64(n);
}
}
return total;
}
static noinline unsigned long test_get_order(void)
{
unsigned long n, total = 0;
long rep, loop;
for (rep = 1000000; rep > 0; rep--) {
for (loop = 0; loop <= 16384; loop += 4) {
n = 1UL << loop;
total += get_order(n);
}
}
return total;
}
int main(int argc, char **argv)
{
unsigned long total;
switch (argc) {
case 1: total = test_old_get_order(); break;
case 2: total = test_get_order_old_fls64(); break;
default: total = test_get_order(); break;
}
prevent_optimise_out = total;
return 0;
}
This allows me to test the use of the old fls64() implementation and the new
fls64() implementation and also to contrast these to the out-of-line loop-based
implementation of get_order(). The results were:
warthog>time ./get_order
real 1m37.191s
user 1m36.313s
sys 0m0.861s
warthog>time ./get_order x
real 0m16.892s
user 0m16.586s
sys 0m0.287s
warthog>time ./get_order x x
real 0m7.731s
user 0m7.727s
sys 0m0.002s
Using the current upstream fls64() as a basis for an inlined get_order() [the
second result above] is much faster than using the current out-of-line
loop-based get_order() [the first result above].
Using my optimised inline fls64()-based get_order() [the third result above]
is even faster still.
[ hpa: changed the selection of 32 vs 64 bits to use CONFIG_X86_64
instead of comparing BITS_PER_LONG, updated comments, rebased manually
on top of 83d99df7c4bf x86, bitops: Move fls64.h inside __KERNEL__ ]
H. Peter Anvin [Thu, 15 Dec 2011 22:55:53 +0000 (14:55 -0800)]
x86, bitops: Move fls64.h inside __KERNEL__
We would include <asm-generic/bitops/fls64.h> even without __KERNEL__,
but that doesn't make sense, as:
1. That file provides fls64(), but the corresponding function fls() is
not exported to user space.
2. The implementation of fls64.h uses kernel-only symbols.
3. fls64.h is not exported to user space.
This appears to have been a bug introduced in checkin:
d57594c203b1 bitops: use __fls for fls64 on 64-bit archs
Cc: Stephen Hemminger <shemminger@vyatta.com> Cc: Alexander van Heukelum <heukelum@mailshack.com> Cc: David Howells <dhowells@redhat.com> Signed-off-by: H. Peter Anvin <hpa@zytor.com> Link: http://lkml.kernel.org/r/4EEA77E1.6050009@zytor.com
Jan Beulich [Wed, 14 Dec 2011 08:33:25 +0000 (08:33 +0000)]
x86: Fix and improve percpu_cmpxchg{8,16}b_double()
They had several problems/shortcomings:
Only the first memory operand was mentioned in the 2x32bit asm()
operands, and 2x64-bit version had a memory clobber. The first
allowed the compiler to not recognize the need to re-load the
data in case it had it cached in some register, and the second
was overly destructive.
The memory operand in the 2x32-bit asm() was declared to only be
an output.
The types of the local copies of the old and new values were
incorrect (as in other per-CPU ops, the types of the per-CPU
variables accessed should be used here, to make sure the
respective types are compatible).
The __dummy variable was pointless (and needlessly initialized
in the 2x32-bit case), given that local copies of the inputs
already exist.
The 2x64-bit variant forced the address of the first object into
%rsi, even though this is needed only for the call to the
emulation function. The real cmpxchg16b can operate on an
memory.
At once also change the return value type to what it really is -
'bool'.
Signed-off-by: Jan Beulich <jbeulich@suse.com> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Andrew Morton <akpm@linux-foundation.org> Cc: David Howells <dhowells@redhat.com> Cc: Christoph Lameter <cl@linux.com> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Link: http://lkml.kernel.org/r/4EE86D6502000078000679FE@nat28.tlf.novell.com Signed-off-by: Ingo Molnar <mingo@elte.hu>
Alexey Dobriyan [Sun, 11 Dec 2011 18:13:19 +0000 (21:13 +0300)]
x86/i386: Use less assembly in strlen(), speed things up a bit
Current i386 strlen() hardcodes NOT/DEC sequence. DEC is
mentioned to be suboptimal on Core2. So, put only REPNE SCASB
sequence in assembly, compiler can do the rest.
The difference in generated code is like below (MCORE2=y):
The node_distance function is not x86 64-bit specific. Having
the #ifdef around the extern function declaration and the
#define causes the default node_distance macro to be used in
asm-generic/topology.h. This also causes a sparse warning in
arch/x86/mm/numa.c when CONFIG_X86_64 is not set:
warning: symbol '__node_distance' was not declared. Should it be
static?
The x86_64 kernel pushes the fake kernel stack in
arch/x86/kernel/entry_64.S:FAKE_STACK_FRAME, and
rflags register in it does not conform to the specification.
Although Intel's manual[1] says bit 1 of it shall be set to 1,
this bit is cleared to 0 on pushing the fake stack.
If it is not on purpose, it is better to be fixed, because
it can lead some tools misunderstanding the stack frame. For example,
"crash" utility[2] actually detects it and warns you like
below:
Since there is a possibility of !KPROBES int3 listeners
(such as kgdb) and since DIE_TRAP is currently not being
used by anybody, notify all listeners with DIE_INT3.
x86: Call do_notify_resume() with interrupts enabled
do_notify_resume() gets called with interrupts disabled on x86_32. This
is different from the x86_64 behavior, where interrupts are enabled at
the time.
Queries on lkml on this issue hasn't yielded any clear answer. Lets make
x86_32 behave the same as x86_64, unless there is a real reason to
maintain status quo.
Please refer https://lkml.org/lkml/2011/9/27/130 for more
details.
A similar change was suggested in ARM:
https://lkml.org/lkml/2011/8/25/231
My 32-bit machine works fine (tm) with this patch.
x86/div64: Add a micro-optimization shortcut if base is power of two
In the target code I have a do_div(x, PAGE_SIZE). The x86-64
version of it was doing a shift and a mask which is clever. The
32bit version of it had a div operation in it which made me
think. After digging I noticed that x86 has an optimized version
of it. This patch adds this shift and mask optimization if base
is constant so we don't have any runtime "checking" overhead
since most users use a power of ten.
Jan Beulich [Tue, 29 Nov 2011 11:24:10 +0000 (11:24 +0000)]
x86-64: Cleanup some assembly entry points
system_call_after_swapgs doesn't really benefit from forcing
alignment from it - quite the opposite, native code needlessly
so far got a big NOP instruction inserted in front of it. Xen
being the only user of the separate entry point can well live
with the branch going to three bytes into a cache line.
The compatibility mode ptregs entry points for one can make use
of the GLOBAL() macro, and should be suitably aligned. Their
shared continuation point (ia32_ptregs_common) otoh doesn't need
to be global at all, but should continue to be properly aligned.
Jan Beulich [Tue, 29 Nov 2011 11:17:45 +0000 (11:17 +0000)]
x86-64: Slightly shorten line system call entry and exit paths
GET_THREAD_INFO() involves a memory read immediately followed by
an "sub" on the value read, in turn (in several cases)
immediately followed by a use of the calculated value as the
base address of a memory access. This combination of
instructions has a non-negligible potential for stalls.
In the system call entry point code, however, the (fixed) offset
of the stack pointer from the end of the stack is generally
known, and hence we can instead avoid the memory load and
subtract, and instead do the memory reference using %rsp as the
base register. To do so in a legible fashion, introduce a
THREAD_INFO() macro which, provided a register (generally %rsp)
and the known offset from the end of the stack, produces a
suitable memory access operand.
The patch attempts to only touch the fast paths (no auditing and
alike), but manages to do so only in the 64-bit entry point
case; the compatibility mode entry points have so many
interdependencies between their various branch targets that it
was necessary to also adjust the slow paths to eliminate the
risk of having missed some register dependency during code
analysis.
Jan Beulich [Tue, 29 Nov 2011 11:03:46 +0000 (11:03 +0000)]
x86-64: Reduce amount of redundant code generated for invalidate_interruptNN
Previously these up to 32 entry points, consisting of all the
same code except for their very first instruction, consumed 0x70
bytes per instance. Just like for device interrupt entry points,
fold them together so that they all use a single instance of the
code after having pushed their vector indicator (resulting in
0x10 bytes per instance, to retain 16-byte alignment of the
individual entry points).
Jan Beulich [Tue, 29 Nov 2011 10:54:22 +0000 (10:54 +0000)]
x86-64: Slightly shorten int_ret_from_sys_call
Testing for a return to ring 0 was necessary here solely because
of the branch out of ret_from_fork. That branch, however, can be
directed to retint_restore_args, and thus the test-and-branch
can be eliminated here.
Maurice Ma [Tue, 11 Oct 2011 10:52:13 +0000 (11:52 +0100)]
x86, efi: Convert efi_phys_get_time() args to physical addresses
Because callers of efi_phys_get_time() pass virtual stack
addresses as arguments, we need to find their corresponding
physical addresses and when calling GetTime() in physical mode.
Without this patch the following line is printed on boot,
Andy Lutomirski [Tue, 8 Nov 2011 00:33:40 +0000 (16:33 -0800)]
x86-64: Set siginfo and context on vsyscall emulation faults
To make this work, we teach the page fault handler how to send
signals on failed uaccess. This only works for user addresses
(kernel addresses will never hit the page fault handler in the
first place), so we need to generate signals for those
separately.
This gets the tricky case right: if the user buffer spans
multiple pages and only the second page is invalid, we set
cr2 and si_addr correctly. UML relies on this behavior to
"fault in" pages as needed.
We steal a bit from thread_info.uaccess_err to enable this.
Before this change, uaccess_err was a 32-bit boolean value.
Linus Torvalds [Sun, 4 Dec 2011 19:57:09 +0000 (11:57 -0800)]
x86: Fix boot failures on older AMD CPU's
People with old AMD chips are getting hung boots, because commit bcb80e53877c ("x86, microcode, AMD: Add microcode revision to
/proc/cpuinfo") moved the microcode detection too early into
"early_init_amd()".
At that point we are *so* early in the booth that the exception tables
haven't even been set up yet, so the whole
doesn't actually work: if the rdmsr does a GP fault (due to non-existant
MSR register on older CPU's), we can't fix it up yet, and the boot fails.
Fix it by simply moving the code to a slightly later point in the boot
(init_amd() instead of early_init_amd()), since the kernel itself
doesn't even really care about the microcode patchlevel at this point
(or really ever: it's made available to user space in /proc/cpuinfo, and
updated if you do a microcode load).
Reported-tested-and-bisected-by: Larry Finger <Larry.Finger@lwfinger.net> Tested-by: Bob Tracy <rct@gherkin.frus.com> Acked-by: Borislav Petkov <borislav.petkov@amd.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Srivatsa S. Bhat <srivatsa.bhat@linux.vnet.ibm.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
xen/pm_idle: Make pm_idle be default_idle under Xen.
The idea behind commit d91ee5863b71 ("cpuidle: replace xen access to x86
pm_idle and default_idle") was to have one call - disable_cpuidle()
which would make pm_idle not be molested by other code. It disallows
cpuidle_idle_call to be set to pm_idle (which is excellent).
But in the select_idle_routine() and idle_setup(), the pm_idle can still
be set to either: amd_e400_idle, mwait_idle or default_idle. This
depends on some CPU flags (MWAIT) and in AMD case on the type of CPU.
In case of mwait_idle we can hit some instances where the hypervisor
(Amazon EC2 specifically) sets the MWAIT and we get:
In the case of amd_e400_idle we don't get so spectacular crashes, but we
do end up making an MSR which is trapped in the hypervisor, and then
follow it up with a yield hypercall. Meaning we end up going to
hypervisor twice instead of just once.
The previous behavior before v3.0 was that pm_idle was set to
default_idle regardless of select_idle_routine/idle_setup.
We want to do that, but only for one specific case: Xen. This patch
does that.
Fixes RH BZ #739499 and Ubuntu #881076 Reported-by: Stefan Bader <stefan.bader@canonical.com> Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Linus Torvalds [Fri, 2 Dec 2011 21:30:58 +0000 (13:30 -0800)]
Merge branch 'usb-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb
* 'usb-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb: (21 commits)
usb: ftdi_sio: add PID for Propox ISPcable III
Revert "xHCI: reset-on-resume quirk for NEC uPD720200"
xHCI: fix bug in xhci_clear_command_ring()
usb: gadget: fsl_udc: fix dequeuing a request in progress
usb: fsl_mxc_udc.c: Remove compile-time dependency of MX35 SoC type
usb: fsl_mxc_udc.c: Fix build issue by including missing header file
USB: fsl_udc_core: use usb_endpoint_xfer_isoc to judge ISO XFER
usb: udc: Fix gadget driver's speed check in various UDC drivers
usb: gadget: fix g_serial regression
usb: renesas_usbhs: fixup driver speed
usb: renesas_usbhs: fixup gadget.dev.driver when udc_stop.
usb: renesas_usbhs: fixup signal the driver that cable was disconnected
usb: renesas_usbhs: fixup device_register timing
usb: musb: PM: fix context save/restore in suspend/resume path
USB: linux-cdc-acm.inf: add support for the acm_ms gadget
EHCI : Fix a regression in the ISO scheduler
xHCI: reset-on-resume quirk for NEC uPD720200
USB: whci-hcd: fix endian conversion in qset_clear()
USB: usb-storage: unusual_devs entry for Kingston DT 101 G2
usb: option: add SIMCom SIM5218
...
Linus Torvalds [Fri, 2 Dec 2011 21:30:25 +0000 (13:30 -0800)]
Merge branch 'staging-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/staging
* 'staging-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/staging:
Staging: comedi: fix integer overflow in do_insnlist_ioctl()
Revert "Staging: comedi: integer overflow in do_insnlist_ioctl()"
Staging: comedi: integer overflow in do_insnlist_ioctl()
Staging: comedi: fix signal handling in read and write
Staging: comedi: fix mmap_count
staging: comedi: fix oops for USB DAQ devices.
staging: comedi: usbduxsigma: Fixed wrong range for the analogue channel.
staging:rts_pstor:Complete scanning_done variable
staging: usbip: bugfix for deadlock
Linus Torvalds [Fri, 2 Dec 2011 18:38:20 +0000 (10:38 -0800)]
Merge branch 'for-linus' of git://oss.sgi.com/xfs/xfs
* 'for-linus' of git://oss.sgi.com/xfs/xfs:
xfs: fix attr2 vs large data fork assert
xfs: force buffer writeback before blocking on the ilock in inode reclaim
xfs: validate acl count
Linus Torvalds [Fri, 2 Dec 2011 16:25:04 +0000 (08:25 -0800)]
Merge branch 'drm-fixes' of git://people.freedesktop.org/~airlied/linux
* 'drm-fixes' of git://people.freedesktop.org/~airlied/linux:
vmwgfx: integer overflow in vmw_kms_update_layout_ioctl()
drm/radeon/kms: fix 2D tiling CS support on EG/CM
drm/radeon/kms: fix scanout of 2D tiled buffers on EG/CM
drm: Fix lack of CRTC disable for drm_crtc_helper_set_config(.fb=NULL)
drm/radeon/kms: add some new pci ids
drm/radeon/kms: Skip ACPI call to ATIF when possible
drm/radeon/kms: Hide debugging message
drm/radeon/kms: add some loop timeouts in pageflip code
drm/nv50/disp: silence compiler warning
drm/nouveau: fix oopses caused by clear being called on unpopulated ttms
drm/nouveau: Keep RAMIN heap within the channel.
drm/nvd0/disp: fix sor dpms typo, preventing dpms on in some situations
drm/nvc0/gr: fix TP init for transform feedback offset queries
drm/nouveau: add dumb ioctl support
Linus Torvalds [Fri, 2 Dec 2011 16:10:51 +0000 (08:10 -0800)]
Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound:
ALSA: hda - Fix S3/S4 problem on machines with VREF-pin mute-LED
ALSA: hda_intel - revert a quirk that affect VIA chipsets
ALSA: hda - Avoid touching mute-VREF pin for IDT codecs
firmware: Sigma: Fix endianess issues
firmware: Sigma: Skip header during CRC generation
firmware: Sigma: Prevent out of bounds memory access
ALSA: usb-audio - Support for Roland GAIA SH-01 Synthesizer
ASoC: Supply dcs_codes for newer WM1811 revisions
ASoC: Error out if we can't generate a LRCLK at all for WM8994
ASoC: Correct name of Speyside Main Speaker widget
ASoC: skip resume of soc-audio devices without codecs
ASoC: cs42l51: Fix off-by-one for reg_cache_size
ASoC: drop support for PlayPaq with WM8510
ASoC: mpc8610: tell the CS4270 codec that it's the master
ASoC: cs4720: use snd_soc_cache_sync()
ASoC: SAMSUNG: Fix build error
ASoC: max9877: Update register if either val or val2 is changed
ASoC: Fix wrong define for AD1836_ADC_WORD_OFFSET
Xi Wang [Mon, 28 Nov 2011 11:25:43 +0000 (12:25 +0100)]
vmwgfx: integer overflow in vmw_kms_update_layout_ioctl()
There are two issues in vmw_kms_update_layout_ioctl(). First, the
for loop forgets to index rects and only checks the first element.
Second, there is a potential integer overflow if userspace passes
in a large arg->num_outputs. The call to kzalloc() would allocate
a small buffer, leading to out-of-bounds read.
Reported-by: Haogang Chen <haogangchen@gmail.com> Signed-off-by: Xi Wang <xi.wang@gmail.com> Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> Signed-off-by: Dave Airlie <airlied@redhat.com>
Chris Wilson [Mon, 28 Nov 2011 21:10:05 +0000 (21:10 +0000)]
drm: Fix lack of CRTC disable for drm_crtc_helper_set_config(.fb=NULL)
Disabling the CRTC by setting its framebuffer to NULL, as used by
drm_framebuffer_cleanup(), was failing to pass the current framebuffer
to the crtc_func->disable callback. This is because of the dance within
drm_crtc_helper_set_config to pass the new_fb (NULL in this case) to the
drm_crtc_helper_set_mode with the currently attached fb as a parameter.
drm_crtc_helper_set_mode treats this as a no-op and the encoder is still
enabled. And so the current fb is forgotten before the call to
drm_helper_disable_unused_functions.
This patch treats disabling the CRTC as a simple special case rather
than adding further complexity into the configuration logic.
This fixes a pin-leak of the fb bo on Xserver close.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Signed-off-by: Dave Airlie <airlied@redhat.com>
* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net: (73 commits)
netfilter: Remove ADVANCED dependency from NF_CONNTRACK_NETBIOS_NS
ipv4: flush route cache after change accept_local
sch_red: fix red_change
Revert "udp: remove redundant variable"
bridge: master device stuck in no-carrier state forever when in user-stp mode
ipv4: Perform peer validation on cached route lookup.
net/core: fix rollback handler in register_netdevice_notifier
sch_red: fix red_calc_qavg_from_idle_time
bonding: only use primary address for ARP
ipv4: fix lockdep splat in rt_cache_seq_show
sch_teql: fix lockdep splat
net: fec: Select the FEC driver by default for i.MX SoCs
isdn: avoid copying too long drvid
isdn: make sure strings are null terminated
netlabel: Fix build problems when IPv6 is not enabled
sctp: better integer overflow check in sctp_auth_create_key()
sctp: integer overflow in sctp_auth_create_key()
ipv6: Set mcast_hops to IPV6_DEFAULT_MCASTHOPS when -1 was given.
net: Fix corruption in /proc/*/net/dev_mcast
mac80211: fix race between the AGG SM and the Tx data path
...
Peter Pan(潘卫平) [Thu, 1 Dec 2011 15:47:06 +0000 (15:47 +0000)]
ipv4: flush route cache after change accept_local
After reset ipv4_devconf->data[IPV4_DEVCONF_ACCEPT_LOCAL] to 0,
we should flush route cache, or it will continue receive packets with local
source address, which should be dropped.
Signed-off-by: Weiping Pan <panweiping3@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Eric Dumazet [Thu, 1 Dec 2011 11:06:34 +0000 (11:06 +0000)]
sch_red: fix red_change
Le mercredi 30 novembre 2011 à 14:36 -0800, Stephen Hemminger a écrit :
> (Almost) nobody uses RED because they can't figure it out.
> According to Wikipedia, VJ says that:
> "there are not one, but two bugs in classic RED."
RED is useful for high throughput routers, I doubt many linux machines
act as such devices.
I was considering adding Adaptative RED (Sally Floyd, Ramakrishna
Gummadi, Scott Shender), August 2001
In this version, maxp is dynamic (from 1% to 50%), and user only have to
setup min_th (target average queue size)
(max_th and wq (burst in linux RED) are automatically setup)
By the way it seems we have a small bug in red_change()
if (skb_queue_empty(&sch->q))
red_end_of_idle_period(&q->parms);
First, if queue is empty, we should call
red_start_of_idle_period(&q->parms);
Second, since we dont use anymore sch->q, but q->qdisc, the test is
meaningless.
Oh well...
[PATCH] sch_red: fix red_change()
Now RED is classful, we must check q->qdisc->q.qlen, and if queue is empty,
we start an idle period, not end it.
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Linus Torvalds [Thu, 1 Dec 2011 22:55:34 +0000 (14:55 -0800)]
Merge branch 'upstream-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jlbec/ocfs2
* 'upstream-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jlbec/ocfs2: (31 commits)
ocfs2: avoid unaligned access to dqc_bitmap
ocfs2: Use filemap_write_and_wait() instead of write_inode_now()
ocfs2: honor O_(D)SYNC flag in fallocate
ocfs2: Add a missing journal credit in ocfs2_link_credits() -v2
ocfs2: send correct UUID to cleancache initialization
ocfs2: Commit transactions in error cases -v2
ocfs2: make direntry invalid when deleting it
fs/ocfs2/dlm/dlmlock.c: free kmem_cache_zalloc'd data using kmem_cache_free
ocfs2: Avoid livelock in ocfs2_readpage()
ocfs2: serialize unaligned aio
ocfs2: Implement llseek()
ocfs2: Fix ocfs2_page_mkwrite()
ocfs2: Add comment about orphan scanning
ocfs2: Clean up messages in the fs
ocfs2/cluster: Cluster up now includes network connections too
ocfs2/cluster: Add new function o2net_fill_node_map()
ocfs2/cluster: Fix output in file elapsed_time_in_ms
ocfs2/dlm: dlmlock_remote() needs to account for remastery
ocfs2/dlm: Take inflight reference count for remotely mastered resources too
ocfs2/dlm: Cleanup dlm_wait_for_node_death() and dlm_wait_for_node_recovery()
...
Akinobu Mita [Tue, 15 Nov 2011 22:56:34 +0000 (14:56 -0800)]
ocfs2: avoid unaligned access to dqc_bitmap
The dqc_bitmap field of struct ocfs2_local_disk_chunk is 32-bit aligned,
but not 64-bit aligned. The dqc_bitmap is accessed by ocfs2_set_bit(),
ocfs2_clear_bit(), ocfs2_test_bit(), or ocfs2_find_next_zero_bit(). These
are wrapper macros for ext2_*_bit() which need to take an unsigned long
aligned address (though some architectures are able to handle unaligned
address correctly)
So some 64bit architectures may not be able to access the dqc_bitmap
correctly.
This avoids such unaligned access by using another wrapper functions for
ext2_*_bit(). The code is taken from fs/ext4/mballoc.c which also need to
handle unaligned bitmap access.
Signed-off-by: Akinobu Mita <akinobu.mita@gmail.com> Acked-by: Joel Becker <jlbec@evilplan.org> Cc: Mark Fasheh <mfasheh@suse.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Joel Becker <jlbec@evilplan.org>
Linus Torvalds [Thu, 1 Dec 2011 19:53:54 +0000 (11:53 -0800)]
Merge branch 'fixes' of http://ftp.arm.linux.org.uk/pub/linux/arm/kernel/git-cur/linux-2.6-arm
* 'fixes' of http://ftp.arm.linux.org.uk/pub/linux/arm/kernel/git-cur/linux-2.6-arm:
ARM: 7182/1: ARM cpu topology: fix warning
ARM: 7181/1: Restrict kprobes probing SWP instructions to ARMv5 and below
ARM: 7180/1: Change kprobes testcase with unpredictable STRD instruction
ARM: 7177/1: GIC: avoid skipping non-existent PPIs in irq_start calculation
ARM: 7176/1: cpu_pm: register GIC PM notifier only once
ARM: 7175/1: add subname parameter to mfp_set_groupg callers
ARM: 7174/1: Fix build error in kprobes test code on Thumb2 kernels
ARM: 7172/1: dma: Drop GFP_COMP for DMA memory allocations
ARM: 7171/1: unwind: add unwind directives to bitops assembly macros
ARM: 7170/2: fix compilation breakage in entry-armv.S
ARM: 7168/1: use cache type functions for arch_get_unmapped_area
ARM: perf: check that we have a platform device when reserving PMU
ARM: 7166/1: Use PMD_SHIFT instead of PGDIR_SHIFT in dma-consistent.c
ARM: 7165/2: PL330: Fix typo in _prepare_ccr()
ARM: 7163/2: PL330: Only register usable channels
ARM: 7162/1: errata: tidy up Kconfig options for PL310 errata workarounds
ARM: 7161/1: errata: no automatic store buffer drain
ARM: perf: initialise used_mask for fake PMU during validation
ARM: PMU: remove pmu_init declaration
ARM: PMU: re-export release_pmu symbol to modules
If we take the "try_again" goto, due to a checksum error,
the 'len' has already been truncated. So we won't compute
the same values as the original code did.
Reported-by: paul bilke <fsmail@conspiracy.net> Signed-off-by: David S. Miller <davem@davemloft.net>
Merge branch 'for-usb-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/sarah/xhci into usb-linus
* 'for-usb-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/sarah/xhci:
Revert "xHCI: reset-on-resume quirk for NEC uPD720200"
xHCI: fix bug in xhci_clear_command_ring()
bridge: master device stuck in no-carrier state forever when in user-stp mode
When in user-stp mode, bridge master do not follow state of its slaves, so
after the following sequence of events it can stuck forever in no-carrier
state:
1) turn stp off
2) put all slaves down - master device will follow their state and also go in
no-carrier state
3) turn stp on with bridge-stp script returning 0 (go to the user-stp mode)
Now bridge master won't follow slaves' state and will never reach running
state.
This patch solves the problem by making user-stp and kernel-stp behavior
similar regarding master following slaves' states.
Signed-off-by: Vitalii Demianets <vitas@nppfactor.kiev.ua> Acked-by: Stephen Hemminger <shemminger@vyatta.com> Signed-off-by: David S. Miller <davem@davemloft.net>
The commit added a reset-on-resume quirk because the NEC chipset stopped
responding to commands about 30 minutes after a system resume from
suspend. We thought it was a chipset issue, but it turns out that the
xHCI driver was zeroing out the link TRB after a successful context
restore during resume. The host controller would fall off the command
ring sometime later, causing it to not respond to new commands.
The link TRB issue has been fixed with commit 158886cd2cf4599e04f9b7e10cb767f5f39b14f1 "xHCI: fix bug in
xhci_clear_command_ring()", so revert the reset-on-resume quirk, as it's
not necessary.
Commit df711fc9962b9491af2b92bd0d21ecbfefe4e5fa was marked for stable
trees back to 2.6.37, but according to my mail, it has not made it into
Linus' tree or the stable trees yet.
Andiry Xu [Wed, 30 Nov 2011 08:37:41 +0000 (16:37 +0800)]
xHCI: fix bug in xhci_clear_command_ring()
When system enters suspend, xHCI driver clears command ring by writing zero
to all the TRBs. However, this also writes zero to the Link TRB, and the ring
is mangled. This may cause driver accesses wrong memory address and the
result is unpredicted.
When clear the command ring, keep the last Link TRB intact, only clear its
cycle bit. This should fix the "command ring full" issue reported by Oliver
Neukum.
This should be backported to stable kernels as old as 2.6.37, since the
commit 89821320 "xhci: Fix command ring replay after resume" is merged.
Signed-off-by: Andiry Xu <andiry.xu@amd.com> Signed-off-by: Sarah Sharp <sarah.a.sharp@linux.intel.com> Reported-by: Oliver Neukum <oneukum@suse.de>
Linus Torvalds [Thu, 1 Dec 2011 16:28:53 +0000 (08:28 -0800)]
Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mason/linux-btrfs
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mason/linux-btrfs:
Btrfs: fix meta data raid-repair merge problem
Btrfs: skip allocation attempt from empty cluster
Btrfs: skip block groups without enough space for a cluster
Btrfs: start search for new cluster at the beginning
Btrfs: reset cluster's max_size when creating bitmap
Btrfs: initialize new bitmaps' list
Btrfs: fix oops when calling statfs on readonly device
Btrfs: Don't error on resizing FS to same size
Btrfs: fix deadlock on metadata reservation when evicting a inode
Fix URL of btrfs-progs git repository in docs
btrfs scrub: handle -ENOMEM from init_ipath()
Jan Schmidt [Thu, 1 Dec 2011 14:30:36 +0000 (09:30 -0500)]
Btrfs: fix meta data raid-repair merge problem
Commit 4a54c8c16 introduced raid-repair, killing the individual
readpage_io_failed_hook entries from inode.c and disk-io.c. Commit 4bb31e92 introduced new readahead code, adding a readpage_io_failed_hook to
disk-io.c.
The raid-repair commit had logic to disable raid-repair, if
readpage_io_failed_hook is set. Thus, the readahead commit effectively
disabled raid-repair for meta data.
This commit changes the logic to always attempt raid-repair when needed and
call the readpage_io_failed_hook in case raid-repair fails. This is much
more straight forward and should have been like that from the beginning.
Signed-off-by: Jan Schmidt <list.btrfs@jan-o-sch.net> Reported-by: Stefan Behrens <sbehrens@giantdisaster.de> Signed-off-by: Chris Mason <chris.mason@oracle.com>
Charles Chin [Thu, 1 Dec 2011 10:21:00 +0000 (11:21 +0100)]
ALSA: hda - Fix S3/S4 problem on machines with VREF-pin mute-LED
The verb command in stac92xx_post_suspend caused the audio to stop
working after resuming from S3 mode on HP laptops with the VREF-pin
mute-LED control. Removing relevant post_suspend registering.
Although removing D3 on AFG is no optimal solution, the impact should
be small in comparison with the broken S3/S4.
Signed-off-by: Charles Chin <Charles.Chin@idt.com> Cc: <stable@kernel.org> Signed-off-by: Takashi Iwai <tiwai@suse.de>
Jean Delvare [Wed, 30 Nov 2011 16:36:39 +0000 (17:36 +0100)]
drm/radeon/kms: Skip ACPI call to ATIF when possible
I am under the impression that it only makes sense to call the ATIF
method if the graphics device has an ACPI handle attached. So we could
skip the call altogether if there is no such handle.
Signed-off-by: Jean Delvare <jdelvare@suse.de> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Dave Airlie <airlied@redhat.com>
Jean Delvare [Wed, 30 Nov 2011 16:26:36 +0000 (17:26 +0100)]
drm/radeon/kms: Hide debugging message
Use the proper macro to issue the debugging message in
radeon_atif_call(). Otherwise we spam the log of many systems with a
message which looks like an error message of unknown origin, and could
thus confuse the user. Commit dc77de12dde95c8da39e4c417eb70c7d445cf84b
was a first step in this direction, but was not sufficient IMHO.
Signed-off-by: Jean Delvare <jdelvare@suse.de> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Dave Airlie <airlied@redhat.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com> Cc: stable@kernel.org Reviewed-by: Mario Kleiner <mario.kleiner@tuebingen.mpg.de> Tested-by: Simon Farnsworth <simon.farnsworth@onelan.co.uk> Signed-off-by: Dave Airlie <airlied@redhat.com>
Dave Airlie [Thu, 1 Dec 2011 09:01:55 +0000 (09:01 +0000)]
Merge branch 'drm-nouveau-fixes' of git://git.freedesktop.org/git/nouveau/linux-2.6 into drm-fixes
* 'drm-nouveau-fixes' of git://git.freedesktop.org/git/nouveau/linux-2.6:
drm/nv50/disp: silence compiler warning
drm/nouveau: fix oopses caused by clear being called on unpopulated ttms
drm/nouveau: Keep RAMIN heap within the channel.
drm/nvd0/disp: fix sor dpms typo, preventing dpms on in some situations
drm/nvc0/gr: fix TP init for transform feedback offset queries
drm/nouveau: add dumb ioctl support
RongQing.Li [Thu, 1 Dec 2011 04:43:07 +0000 (23:43 -0500)]
net/core: fix rollback handler in register_netdevice_notifier
Within nested statements, the break statement terminates only the
do, for, switch, or while statement that immediately encloses it,
So replace the break with goto.
Signed-off-by: RongQing.Li <roy.qing.li@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Eric Dumazet [Wed, 30 Nov 2011 12:10:53 +0000 (12:10 +0000)]
sch_red: fix red_calc_qavg_from_idle_time
Since commit a4a710c4a7490587 (pkt_sched: Change PSCHED_SHIFT from 10 to
6) it seems RED/GRED are broken.
red_calc_qavg_from_idle_time() computes a delay in us units, but this
delay is now 16 times bigger than real delay, so the final qavg result
smaller than expected.
Use standard kernel time services since there is no need to obfuscate
them.
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Only use the primary address of the bond device
for master_ip. This will prevent changing the ARP source
address in Active-Backup mode whenever a secondry address
is added to the bond device.
Signed-off-by: Henrik Saavedra Persson <henrik.e.persson@ericsson.com> Signed-off-by: Andy Gospodarek <andy@greyhouse.net> Signed-off-by: David S. Miller <davem@drr.davemloft.net>
Linus Torvalds [Thu, 1 Dec 2011 00:24:24 +0000 (16:24 -0800)]
Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/broonie/regulator
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/broonie/regulator:
regulator: twl: fix twl4030 support for smps regulators
regulator: fix use after free bug
regulator: aat2870: Fix the logic of checking if no id is matched in aat2870_get_regulator
ARM: 7181/1: Restrict kprobes probing SWP instructions to ARMv5 and below
The SWP instruction is deprecated on ARMv6 and with ARMv7 it will be
UNDEFINED when CONFIG_SWP_EMULATE is selected. In this case, probing a
SWP instruction will cause an oops when the kprobes emulation code
executes an undefined instruction.
As the SWP instruction should be rare or non-existent in kernels for
ARMv6 and later, we can simply avoid these problems by not allowing
probing of these.
Reported-by: Leif Lindholm <leif.lindholm@arm.com> Tested-by: Leif Lindholm <leif.lindholm@arm.com> Acked-by: Nicolas Pitre <nicolas.pitre@linaro.org> Signed-off-by: Jon Medhurst <tixy@yxit.co.uk> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
ARM: 7180/1: Change kprobes testcase with unpredictable STRD instruction
There is a kprobes testcase for the instruction "strd r2, [r3], r4".
This has unpredictable behaviour as it uses r3 for register writeback
addressing and also stores it to memory.
On a cortex A9, this testcase would fail because the instruction writes
the updated value of r3 to memory, whereas the kprobes emulation code
writes the original value.
Fix this by changing testcase to used r5 instead of r3.
Reported-by: Leif Lindholm <leif.lindholm@arm.com> Tested-by: Leif Lindholm <leif.lindholm@arm.com> Acked-by: Nicolas Pitre <nicolas.pitre@linaro.org> Signed-off-by: Jon Medhurst <tixy@yxit.co.uk> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Eric Dumazet [Tue, 29 Nov 2011 20:05:55 +0000 (20:05 +0000)]
ipv4: fix lockdep splat in rt_cache_seq_show
After commit f2c31e32b378 (fix NULL dereferences in check_peer_redir()),
dst_get_neighbour() should be guarded by rcu_read_lock() /
rcu_read_unlock() section.
Reported-by: Miles Lane <miles.lane@gmail.com> Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Alexandre Oliva [Wed, 30 Nov 2011 18:43:00 +0000 (13:43 -0500)]
Btrfs: skip block groups without enough space for a cluster
We test whether a block group has enough free space to hold the
requested block, but when we're doing clustered allocation, we can
save some cycles by testing whether it has enough room for the cluster
upfront, otherwise we end up attempting to set up a cluster and
failing. Only in the NO_EMPTY_SIZE loop do we attempt an unclustered
allocation, and by then we'll have zeroed the cluster size, so this
patch won't stop us from using the block group as a last resort.
Signed-off-by: Alexandre Oliva <oliva@lsd.ic.unicamp.br> Signed-off-by: Chris Mason <chris.mason@oracle.com>
Alexandre Oliva [Wed, 30 Nov 2011 18:43:00 +0000 (13:43 -0500)]
Btrfs: start search for new cluster at the beginning
Instead of starting at zero (offset is always zero), request a cluster
starting at search_start, that denotes the beginning of the current
block group.
Signed-off-by: Alexandre Oliva <oliva@lsd.ic.unicamp.br> Signed-off-by: Chris Mason <chris.mason@oracle.com>
Alexandre Oliva [Wed, 30 Nov 2011 18:43:00 +0000 (13:43 -0500)]
Btrfs: reset cluster's max_size when creating bitmap
The field that indicates the size of the largest contiguous chunk of
free space in the cluster is not initialized when setting up bitmaps,
it's only increased when we find a larger contiguous chunk. We end up
retaining a larger value than appropriate for highly-fragmented
clusters, which may cause pointless searches for large contiguous
groups, and even cause clusters that do not meet the density
requirements to be set up.
Signed-off-by: Alexandre Oliva <oliva@lsd.ic.unicamp.br> Signed-off-by: Chris Mason <chris.mason@oracle.com>
Alexandre Oliva [Mon, 28 Nov 2011 14:04:43 +0000 (12:04 -0200)]
Btrfs: initialize new bitmaps' list
We're failing to create clusters with bitmaps because
setup_cluster_no_bitmap checks that the list is empty before inserting
the bitmap entry in the list for setup_cluster_bitmap, but the list
field is only initialized when it is restored from the on-disk free
space cache, or when it is written out to disk.
Besides a potential race condition due to the multiple use of the list
field, filesystem performance severely degrades over time: as we use
up all non-bitmap free extents, the try-to-set-up-cluster dance is
done at every metadata block allocation. For every block group, we
fail to set up a cluster, and after failing on them all up to twice,
we fall back to the much slower unclustered allocation.
To make matters worse, before the unclustered allocation, we try to
create new block groups until we reach the 1% threshold, which
introduces additional bitmaps and thus block groups that we'll iterate
over at each metadata block request.
Mike Fleetwood [Fri, 18 Nov 2011 18:55:01 +0000 (18:55 +0000)]
Btrfs: Don't error on resizing FS to same size
It seems overly harsh to fail a resize of a btrfs file system to the
same size when a shrink or grow would succeed. User app GParted trips
over this error. Allow it by bypassing the shrink or grow operation.
Signed-off-by: Mike Fleetwood <mike.fleetwood@googlemail.com>
Miao Xie [Fri, 18 Nov 2011 09:43:00 +0000 (17:43 +0800)]
Btrfs: fix deadlock on metadata reservation when evicting a inode
When I ran the xfstests, I found the test tasks was blocked on meta-data
reservation.
By debugging, I found the reason of this bug:
start transaction
|
v
reserve meta-data space
|
v
flush delay allocation -> iput inode -> evict inode
^ |
| v
wait for delay allocation flush <- reserve meta-data space
And besides that, the flush on evicting inode will block the thread, which
is reclaiming the memory, and make oom happen easily.
Fix this bug by skipping the flush step when evicting inode.
Younes Manton [Tue, 22 Nov 2011 19:58:31 +0000 (14:58 -0500)]
drm/nouveau: Keep RAMIN heap within the channel.
The entire RAMIN is allocated to be 'size', but the heap is
specified as 'base' + 'size' inside RAMIN, so it will overflow
past RAMIN by 'base' bytes on NV50+ and clobber other allocatons
unless it's size is adjusted.
Signed-off-by: Younes Manton <younes.m@gmail.com> Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
Linus Torvalds [Tue, 29 Nov 2011 22:43:22 +0000 (14:43 -0800)]
Merge branch 'pm-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm
* 'pm-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm:
PM: Update comments describing device power management callbacks
PM / Sleep: Update documentation related to system wakeup
PM / Runtime: Make documentation follow the new behavior of irq_safe
PM / Sleep: Correct inaccurate information in devices.txt
PM / Domains: Document how PM domains are used by the PM core
PM / Hibernate: Do not leak memory in error/test code paths
Paul Moore [Tue, 29 Nov 2011 10:10:54 +0000 (10:10 +0000)]
netlabel: Fix build problems when IPv6 is not enabled
A recent fix to the the NetLabel code caused build problem with
configurations that did not have IPv6 enabled; see below:
netlabel_kapi.c: In function 'netlbl_cfg_unlbl_map_add':
netlabel_kapi.c:165:4:
error: implicit declaration of function 'netlbl_af6list_add'
This patch fixes this problem by making the IPv6 specific code conditional
on the IPv6 configuration flags as we done in the rest of NetLabel and the
network stack as a whole. We have to move some variable declarations
around as a result so things may not be quite as pretty, but at least it
builds cleanly now.
Some additional IPv6 conditionals were added to the NetLabel code as well
for the sake of consistency.
Reported-by: Randy Dunlap <rdunlap@xenotime.net> Signed-off-by: Paul Moore <pmoore@redhat.com> Acked-by: Randy Dunlap <rdunlap@xenotime.net> Signed-off-by: David S. Miller <davem@davemloft.net>
Eric Dumazet [Tue, 29 Nov 2011 21:31:23 +0000 (22:31 +0100)]
IB: Fix RCU lockdep splats
Commit f2c31e32b37 ("net: fix NULL dereferences in check_peer_redir()")
forgot to take care of infiniband uses of dst neighbours.
Many thanks to Marc Aurele who provided a nice bug report and feedback.
Reported-by: Marc Aurele La France <tsi@ualberta.ca> Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com> Cc: David Miller <davem@davemloft.net> Cc: <stable@kernel.org> Signed-off-by: Roland Dreier <roland@purestorage.com>
The while loop at the bottom of ipoib_mcast_join_finish() will attempt
to send queued multicast packets in mcast->pkt_queue and eventually
end up in ipoib_mcast_send():
if (!mcast->ah) {
if (skb_queue_len(&mcast->pkt_queue) < IPOIB_MAX_MCAST_QUEUE)
skb_queue_tail(&mcast->pkt_queue, skb);
else {
++dev->stats.tx_dropped;
dev_kfree_skb_any(skb);
}
My read is that the code will requeue the packet and return to the
ipoib_mcast_join_finish() while loop and the stage is set for the
"hung" task diagnostic as the while loop never sees a non-NULL ah, and
will do nothing to resolve.
There are GFP_ATOMIC allocates in the provider routines, so this is
possible and should be dealt with.
The test that induced the failure is associated with a host SM on the
same server during a shutdown.
This patch causes ipoib_mcast_join_finish() to exit with an error
which will flush the queued mcast packets. Nothing is done to unwind
the QP attached state so that subsequent sends from above will retry
the join.
Reviewed-by: Ram Vepa <ram.vepa@qlogic.com> Reviewed-by: Gary Leshner <gary.leshner@qlogic.com> Signed-off-by: Mike Marciniszyn <mike.marciniszyn@qlogic.com> Signed-off-by: Roland Dreier <roland@purestorage.com>
Xi Wang [Tue, 29 Nov 2011 09:26:30 +0000 (09:26 +0000)]
sctp: better integer overflow check in sctp_auth_create_key()
The check from commit 30c2235c is incomplete and cannot prevent
cases like key_len = 0x80000000 (INT_MAX + 1). In that case, the
left-hand side of the check (INT_MAX - key_len), which is unsigned,
becomes 0xffffffff (UINT_MAX) and bypasses the check.
However this shouldn't be a security issue. The function is called
from the following two code paths:
1) setsockopt()
2) sctp_auth_asoc_set_secret()
In case (1), sca_keylength is never going to exceed 65535 since it's
bounded by a u16 from the user API. As such, the key length will
never overflow.
In case (2), sca_keylength is computed based on the user key (1 short)
and 2 * key_vector (3 shorts) for a total of 7 * USHRT_MAX, which still
will not overflow.
In other words, this overflow check is not really necessary. Just
make it more correct.
Signed-off-by: Xi Wang <xi.wang@gmail.com> Cc: Vlad Yasevich <vladislav.yasevich@hp.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Linus Torvalds [Tue, 29 Nov 2011 19:13:22 +0000 (11:13 -0800)]
Merge branch 'slab/urgent' of git://git.kernel.org/pub/scm/linux/kernel/git/penberg/linux
* 'slab/urgent' of git://git.kernel.org/pub/scm/linux/kernel/git/penberg/linux:
slub: avoid potential NULL dereference or corruption
slub: use irqsafe_cpu_cmpxchg for put_cpu_partial
slub: move discard_slab out of node lock
slub: use correct parameter to add a page to partial list tail
With Dmitry fsstress updates I've seen very reproducible crashes in
xfs_attr_shortform_remove because xfs_attr_shortform_bytesfit claims that
the attributes would not fit inline into the inode after removing an
attribute. It turns out that we were operating on an inode with lots
of delalloc extents, and thus an if_bytes values for the data fork that
is larger than biggest possible on-disk storage for it which utterly
confuses the code near the end of xfs_attr_shortform_bytesfit.
Fix this by always allowing the current attribute fork, like we already
do for the attr1 format, given that delalloc conversion will take care
for moving either the data or attribute area out of line if it doesn't
fit at that point - or making the point moot by merging extents at this
point.
Also document the function better, and clean up some loose bits.
Reviewed-by: Dave Chinner <dchinner@redhat.com> Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Ben Myers <bpm@sgi.com>
xfs: force buffer writeback before blocking on the ilock in inode reclaim
If we are doing synchronous inode reclaim we block the VM from making
progress in memory reclaim. So if we encouter a flush locked inode
promote it in the delwri list and wake up xfsbufd to write it out now.
Without this we can get hangs of up to 30 seconds during workloads hitting
synchronous inode reclaim.
The scheme is copied from what we do for dquot reclaims.
Reported-by: Simon Kirby <sim@hostway.ca> Signed-off-by: Christoph Hellwig <hch@lst.de> Tested-by: Simon Kirby <sim@hostway.ca> Signed-off-by: Ben Myers <bpm@sgi.com>
Rob Herring [Mon, 28 Nov 2011 02:16:33 +0000 (20:16 -0600)]
of/irq: of_irq_init: add check for parent equal to child node
With the revert of "of/irq: of_irq_find_parent: check for parent equal to
child" (dc9372808412edb), we need another way to handle parent node equal
to the child node. This can simply be handled in of_irq_init by checking
for this condition.
Signed-off-by: Rob Herring <rob.herring@calxeda.com> Tested-by: Pawel Moll <pawel.moll@arm.com> Tested-by: Stephen Warren <swarren@nvidia.com>
Li Yang [Wed, 23 Nov 2011 12:12:56 +0000 (20:12 +0800)]
usb: gadget: fsl_udc: fix dequeuing a request in progress
The original implementation of dequeuing a request in progress
is not correct. Change to use a correct process and also clean
up the related functions a little bit.
Signed-off-by: Li Yang <leoli@freescale.com> Acked-by: Peter Chen <peter.chen@freescale.com> Signed-off-by: Felipe Balbi <balbi@ti.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
projects / mv-sheeva.git / log