Shaohua Li [Mon, 15 Oct 2012 22:18:47 +0000 (09:18 +1100)]
raid5: create multiple threads to handle stripes
This is a new tempt to make raid5 handle stripes in multiple threads, as
suggested by Neil to have maxium flexibility and better numa binding. It
basically is a combination of my first and second generation patches. By
default, no multiple thread is enabled (all stripes are handled by raid5d).
An example to enable multiple threads:
#echo 3 > /sys/block/md0/md/auxthread_number
This will create 3 auxiliary threads to handle stripes. The threads can run
on any cpus and handle stripes produced by any cpus.
#echo 1-3 > /sys/block/md0/md/auxth0/cpulist
This will bind auxiliary thread 0 to cpu 1-3, and this thread will only handle
stripes produced by cpu 1-3. User tool can further change the thread's
affinity, but the thread can only handle stripes produced by cpu 1-3 till the
sysfs entry is changed again.
If stripes produced by a CPU aren't handled by any auxiliary thread, such
stripes will be handled by raid5d. Otherwise, raid5d doesn't handle any
stripes.
Signed-off-by: Shaohua Li <shli@fusionio.com> Signed-off-by: NeilBrown <neilb@suse.de>
David Rientjes [Mon, 15 Oct 2012 20:40:15 +0000 (13:40 -0700)]
thermal, cpufreq: Fix build when CPU_FREQ_TABLE isn't configured
Commit 023614183768 ("thermal: add generic cpufreq cooling
implementation") requires cpufreq_frequency_get_table(), but that
function is only defined for CONFIG_CPU_FREQ_TABLE resulting in the
following build error:
drivers/built-in.o: In function `cpufreq_get_max_state':
drivers/thermal/cpu_cooling.c:259: undefined reference to `cpufreq_frequency_get_table'
drivers/built-in.o: In function `get_cpu_frequency':
drivers/thermal/cpu_cooling.c:129: undefined reference to `cpufreq_frequency_get_table'
Fix it by selecting CONFIG_CPU_FREQ_TABLE for such a configuration.
It turns out CONFIG_EXYNOS_THERMAL also needs CONFIG_CPU_FREQ_TABLE, so
select it there as well.
Signed-off-by: David Rientjes <rientjes@google.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Alan Cox [Mon, 1 Oct 2012 15:24:17 +0000 (16:24 +0100)]
parport: dead code in pp_write
We always update bytes_written before we check signal_pending so it
follows that we can't get a signal return for 0 bytes so we don't
need to check in the singal path. The cases a signal causes an earlier
abort are handled before this and will not hit this path
Signed-off-by: Alan Cox <alan@linux.intel.com> Signed-off-by: Jiri Kosina <jkosina@suse.cz>
commit 2655a2c76f80d91da34faa8f4e114d1793435ed3 ("8250: use the 8250
register interface not the legacy one") forgot to fully switch one
instance of struct uart_port to struct uart_8250_port, causing the
following compile failure:
drivers/tty/serial/8250/8250_hp300.c: In function ‘hpdca_init_one’:
drivers/tty/serial/8250/8250_hp300.c:174: error: ‘uart’ undeclared (first use in this function)
drivers/tty/serial/8250/8250_hp300.c:174: error: (Each undeclared identifier is reported only once
drivers/tty/serial/8250/8250_hp300.c:174: error: for each function it appears in.)
This went unnoticed in -next, as CONFIG_HPDCA is not set to y by
allmodconfig.
Reported-by: Fengguang Wu <fengguang.wu@intel.com> Cc: Alan Cox <alan@linux.intel.com> Cc: Philip Blundell <philb@gnu.org> Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Commit fe04ddf7c291 ("kbuild: Do not package /boot and /lib in make
tar-pkg") accidentally reverted two previous kbuild commits. I don't
know what I was thinking.
This brings back changes made by commits 24cc7fb69a5b ("x86/kbuild:
archscripts depends on scripts_basic") and c1c1a59e37da ("firmware: fix
directory creation rule matching with make 3.80")
Reported-by: Jan Beulich <JBeulich@suse.com> Cc: <stable@vger.kernel.org> Signed-off-by: Michal Marek <mmarek@suse.cz> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Piotr Haber [Thu, 11 Oct 2012 12:05:15 +0000 (14:05 +0200)]
bcma: fix unregistration of cores
When cores are unregistered, entries
need to be removed from cores list in a safe manner.
Reported-by: Stanislaw Gruszka <sgruszka@redhat.com> Reviewed-by: Arend Van Spriel <arend@broadcom.com> Signed-off-by: Piotr Haber <phaber@broadcom.com> Cc: stable@vger.kernel.org Signed-off-by: John W. Linville <linville@tuxdriver.com>
Felix Fietkau [Wed, 10 Oct 2012 20:40:23 +0000 (22:40 +0200)]
mac80211: use ieee80211_free_txskb in a few more places
Free tx status skbs when draining power save buffers, pending frames, or
when tearing down a vif.
Fixes remaining conditions that can lead to hostapd/wpa_supplicant hangs when
running out of socket write memory.
Signed-off-by: Felix Fietkau <nbd@openwrt.org> Cc: stable@vger.kernel.org Signed-off-by: John W. Linville <linville@tuxdriver.com>
Dan Carpenter [Wed, 10 Oct 2012 18:13:12 +0000 (11:13 -0700)]
brcmfmac: fix end of loop check (signedness bug)
The problem here is that we loop until "remained_buf_len" is less than
zero, but since it is unsigned, it never is.
"remained_buf_len" has to be large enough to hold the value from
"mgmt_ie_buf_len". That variable is type u32, but it only holds small
values so I have changed to both variables to int.
Also I removed the bogus initialization from "mgmt_ie_buf_len" so that
GCC can detect if it is used unitialized. I moved the declaration of
"remained_buf_len" closer to where it is used so it's easier to read.
Reviewed-by: Pieter-Paul Giesberts <pieterpg@broadcom.com> Reviewed-by: Hante Meuleman <meuleman@broadcom.com> Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com> Signed-off-by: Franky Lin <frankyl@broadcom.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
Franky Lin [Wed, 10 Oct 2012 18:13:10 +0000 (11:13 -0700)]
brcmfmac: set dongle mode accordingly when interface up
The mode of WiFi dongle should be initialized in brcmf_cfg80211_up
which get called when network interface is brought up. Otherwise
brcmf_cfg80211_get_station would return error.
Signed-off-by: Franky Lin <frankyl@broadcom.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
Franky Lin [Wed, 10 Oct 2012 18:13:09 +0000 (11:13 -0700)]
brcmfmac: use control channel in roamed status reporting
Channel reported in scan results passed to cfg80211 is control
channel. But chanspec is reported while notifying cfg80211 about
roamed update. Cfg80211 complains because it could not find the
bss in the list. Report control channel while calling
cfg80211_roamed.
Signed-off-by: Franky Lin <frankyl@broadcom.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
Hante Meuleman [Wed, 10 Oct 2012 18:13:06 +0000 (11:13 -0700)]
brcmfmac: handle all exceptions as an error.
in brcmf_usb_probe_cb only return code ENOLINK was seen as an
error. This is wrong, all error codes should be returned to usb
subsystem.
Reviewed-by: Arend Van Spriel <arend@broadcom.com> Signed-off-by: Hante Meuleman <meuleman@broadcom.com> Signed-off-by: Franky Lin <frankyl@broadcom.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
mac80211: check if key has TKIP type before updating IV
This patch fix corruption which can manifest itself by following crash
when switching on rfkill switch with rt2x00 driver:
https://bugzilla.redhat.com/attachment.cgi?id=615362
Pointer key->u.ccmp.tfm of group key get corrupted in:
ieee80211_rx_h_michael_mic_verify():
/* update IV in key information to be able to detect replays */
rx->key->u.tkip.rx[rx->security_idx].iv32 = rx->tkip_iv32;
rx->key->u.tkip.rx[rx->security_idx].iv16 = rx->tkip_iv16;
because rt2x00 always set RX_FLAG_MMIC_STRIPPED, even if key is not TKIP.
We already check type of the key in different path in
ieee80211_rx_h_michael_mic_verify() function, so adding additional
check here is reasonable.
Cc: stable@vger.kernel.org # 3.0+ Signed-off-by: Stanislaw Gruszka <sgruszka@redhat.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
Ralf Baechle [Mon, 15 Oct 2012 11:44:56 +0000 (13:44 +0200)]
mm: huge_memory: Fix build error.
Certain configurations won't implicitly pull in <linux/pagemap.h> resulting
in the following build error:
mm/huge_memory.c: In function 'release_pte_page':
mm/huge_memory.c:1697:2: error: implicit declaration of function 'unlock_page' [-Werror=implicit-function-declaration]
mm/huge_memory.c: In function '__collapse_huge_page_isolate':
mm/huge_memory.c:1757:3: error: implicit declaration of function 'trylock_page' [-Werror=implicit-function-declaration]
cc1: some warnings being treated as errors
This creates a new thread, which falls through to the ret_from_fork
assembly, with r4 set NULL and r5 set to kernel_init. You can see
this in your oops dump register set - r5 is 0xc0344555, which is the
address of kernel_init plus 1 which marks the function as Thumb code.
Now, let's look at this code a little closer - this is what the
disassembly looks like:
This code was introduced in 9fff2fa0db911 (arm: switch to saner
kernel_execve() semantics). I have marked one instruction, and it's
the significant one - I'll come back to that later.
Eventually, having had a successful call to kernel_execve(), kernel_init()
returns zero.
In returning, it uses the value in 'lr' which was set by the instruction
I marked above. Unfortunately, this causes lr to contain 0xc000d18e -
an even address. This switches the ISA to ARM on return but with a non
word aligned PC value.
So, what do we end up executing? Well, not the instructions above - yes
the opcodes, but they don't mean the same thing in ARM mode. In ARM mode,
it looks like this instead:
I have included more above, because it's relevant. The PSR flags which
we can see in the oops dump are nZCv, so Z and C are set.
All the above ARM instructions are not executed, except for two. c000d1a0, which has no writeback, and writes below the current stack
pointer (and that data is lost when we take the next exception.) The
other instruction which is executed is c000d1ac, which takes us to...
0xc03493dc. However, remember that bit 1 of the PC got set. So that
makes the PC value 0xc03493de.
And that value is the value we find in the oops dump for PC. What is
the instruction here when interpreted in ARM mode?
and there we have our undefined instruction (remember that the 'never'
condition code, 0xf, has been deprecated and is now always executed as
it is now being used for additional instructions.)
This path also nicely explains the state of the stack we see in the oops
dump too.
The above is a consistent and sane story for how we got to the oops
dump, which all stems from the instruction at 0xc000d18a being wrong.
Reported-by: Daniel Mack <zonque@gmail.com> Tested-by: Daniel Mack <zonque@gmail.com> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Trond Myklebust [Thu, 11 Oct 2012 17:43:38 +0000 (13:43 -0400)]
NFSv4.1: Do not call pnfs_return_layout() from an rpciod context
Move the call to pnfs_return_layout() to the read and write rpc_release()
callbacks, so that it gets called from nfsiod, which is a more appropriate
context.
Trond Myklebust [Thu, 11 Oct 2012 16:26:04 +0000 (12:26 -0400)]
NFSv4.1: Kill nfs4_ds_disconnect()
There is nothing to prevent another thread from dereferencing ds->ds_clp
during or after the call to nfs4_ds_disconnect(), and Oopsing due to the
resulting NULL pointer.
Instead, we should just rely on filelayout_mark_devid_invalid() to keep
us out of trouble by avoiding that deviceid.
Salman Qazi [Fri, 5 Oct 2012 21:24:14 +0000 (14:24 -0700)]
crypto: vmac - Make VMAC work when blocks aren't aligned
VMAC implementation, as it is, does not work with blocks that
are not multiples of 128-bytes. Furthermore, this is a problem
when using the implementation on scatterlists, even
when the complete plain text is 128-byte multiple, as the pieces
that get passed to vmac_update can be pretty much any size.
I also added test cases for unaligned blocks.
Signed-off-by: Salman Qazi <sqazi@google.com> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Tim Chen [Thu, 27 Sep 2012 22:44:22 +0000 (15:44 -0700)]
crypto: crc32c - Optimize CRC32C calculation with PCLMULQDQ instruction
This patch adds the crc_pcl function that calculates CRC32C checksum using the
PCLMULQDQ instruction on processors that support this feature. This will
provide speedup over using CRC32 instruction only.
The usage of PCLMULQDQ necessitate the invocation of kernel_fpu_begin and
kernel_fpu_end and incur some overhead. So the new crc_pcl function is only
invoked for buffer size of 512 bytes or more. Larger sized
buffers will expect to see greater speedup. This feature is best used coupled
with eager_fpu which reduces the kernel_fpu_begin/end overhead. For
buffer size of 1K the speedup is around 1.6x and for buffer size greater than
4K, the speedup is around 3x compared to original implementation in crc32c-intel
module. Test was performed on Sandy Bridge based platform with constant frequency
set for cpu.
A white paper detailing the algorithm can be found here:
http://download.intel.com/design/intarch/papers/323405.pdf
Signed-off-by: Tim Chen <tim.c.chen@linux.intel.com> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
projects / karo-tx-linux.git / log