| Age | Commit message (Collapse) | Author | Files | Lines |
|---|
|
For the memcg sock code, we'll need to register allocations
that are temporarily over limit. Let's make sure that margin
is 0 in this case.
I am keeping this as a separate patch, so that if any weirdness
interaction appears in the future, we can now exactly what caused
it.
Suggested by Johannes Weiner
Signed-off-by: Glauber Costa <[email protected]>
CC: KAMEZAWA Hiroyuki <[email protected]>
CC: Johannes Weiner <[email protected]>
CC: Michal Hocko <[email protected]>
CC: Tejun Heo <[email protected]>
CC: Li Zefan <[email protected]>
Acked-by: Tejun Heo <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
There is still a build bug with the sock memcg code, that triggers
with !CONFIG_NET, that survived my series of randconfig builds.
Signed-off-by: Glauber Costa <[email protected]>
Reported-by: Randy Dunlap <[email protected]>
CC: Hiroyouki Kamezawa <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
Fix new kernel-doc warning:
Warning(include/net/sock.h:372): No description found for parameter 'sk_cgrp_prioidx'
Signed-off-by: Randy Dunlap <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
Fix new kernel-doc warning:
Warning(drivers/net/phy/mdio_bus.c:49): No description found for parameter 'size'
Signed-off-by: Randy Dunlap <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
md5 key is added in socket through remote address.
remote address should be used in finding md5 key when
sending out reset packet.
Signed-off-by: shawnlu <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
Currently RSS rings are not created in a multi-channel config.
RSS rings can be created on one (out of four) interfaces per port in a
multi-channel config. Doing this insulates the driver from a FW bug wherin
multi-channel config is wrongly reported even when not enabled. This also
helps performance in a multi-channel config, as one interface per port gets
RSS rings.
Signed-off-by: Sathya Perla <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
Every call to num_args() immediately checks the return value for
less than zero, as it will return -EFAULT for a failed get_user()
call. So it makes no sense for the function to be declared as an
unsigned long.
Signed-off-by: Paul Gortmaker <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
Correctly implement a loss detection heuristic: New sequences (above
high_seq) sent during the fast recovery are deemed lost when higher
sequences are SACKed.
Current code does not catch these losses, because tcp_mark_head_lost()
does not check packets beyond high_seq. The fix is straight-forward by
checking packets until the highest sacked packet. In addition, all the
FLAG_DATA_LOST logic are in-effective and redundant and can be removed.
Update the loss heuristic comments. The algorithm above is documented
as heuristic B, but it is redundant too because heuristic A already
covers B.
Note that this change only marks some forward-retransmitted packets LOST.
It does NOT forbid TCP performing further CWR on new losses. A potential
follow-up patch under preparation is to perform another CWR on "new"
losses such as
1) sequence above high_seq is lost (by resetting high_seq to snd_nxt)
2) retransmission is lost.
Signed-off-by: Yuchung Cheng <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
Driver should check for mapping errors.
Machines with limited DMA maps may return an error when a PCI map is
requested (not an issue on standard x86).
Also use upper/lower 32 bits macros for clarity.
Signed-off-by: Stephen Hemminger <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
Skge device would assert carrier (link up) as soon as network device open
was called, rather than waiting until PHY has detected link.
Signed-off-by: Stephen Hemminger <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
With netem reordering, a gap of N is supposed to reorder every Nth packet with
given reorder probability. However, the code currently skips N packets and
reorders every (N+1)th packet.
Signed-off-by: Vijay Subramanian <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
Slave async events were mapped to single eq. This patch fixes this issue, so
the slaves can map the async events to any eq.
Signed-off-by: Marcel Apfelbaum <[email protected]>
Reviewed-by: Jack Morgenstein <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
Num mtts from profile is really the number of mtt segments.
Thus, in make profile, to get the proper number of MTT entries,
must multiply num_mtts by mtts per segment.
Signed-off-by: Marcel Apfelbaum <[email protected]>
Reviewed-by: Jack Morgenstein <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
The Virtual Functions should not be aware their function number.
Signed-off-by: Marcel Apfelbaum <[email protected]>
Reviewed-by: Jack Morgenstein <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
In native mode display all available staticstics.
In SRIOV mode on VF display only SW counters statistics,
in SRIOV mode on hypervisor display SW counters and errors (got from FW)
statistics.
Signed-off-by: Eugenia Emantayev <[email protected]>
Reviewed-by: Yevgeny Petrilin <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
In multifunction mode - DUMP_STATS command is not executed
for VFs.
Signed-off-by: Eugenia Emantayev <[email protected]>
Reviewed-by: Yevgeny Petrilin <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
Bug fix: Not all stats fields were cleared.
Signed-off-by: Eugenia Emantayev <[email protected]>
Reviewed-by: Yevgeny Petriln <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
We should be allowing a 5ms delay after the charge pump is started in
order to ensure it has finished ramping.
Signed-off-by: Mark Brown <[email protected]>
Cc: [email protected]
|
|
Passing the address of a variable as an operand to an asm statement
doesn't mark the value of this variable as used, so gcc may optimize its
initialisation away. Fix this by using the "m" constraint instead.
Signed-off-by: Andreas Schwab <[email protected]>
Signed-off-by: Geert Uytterhoeven <[email protected]>
Cc: [email protected]
|
|
Rename the "Mac ESP" irq as "ESP" to be consistent with all the other Mac
drivers and ESP drivers.
Signed-off-by: Finn Thain <[email protected]>
Signed-off-by: Geert Uytterhoeven <[email protected]>
|
|
Don't enable the SCSI irq when initialising the chip -- the irq has no
handler yet.
Signed-off-by: Finn Thain <[email protected]>
Signed-off-by: Geert Uytterhoeven <[email protected]>
|
|
macfb won't init in black & white modes since fb_alloc_cmap() no longer
works for zero cmap length. Fix this and also clean up a few printk's and
some stylistic inconsistencies.
Signed-off-by: Finn Thain <[email protected]>
Signed-off-by: Geert Uytterhoeven <[email protected]>
|
|
The m68k core irq code stopped honoring these flags during the irq
restructuring in 2006.
Signed-off-by: Geert Uytterhoeven <[email protected]>
|
|
gpiolib drivers should first set the output data before setting the
direction to avoid putting glitches on an output signal. As an
additional bonus, we tweak the code to avoid unnecessary register
writes to the output and direction registers if they have no need
to be updated.
Signed-off-by: Russell King <[email protected]>
|
|
We were not restoring the UCB1x00 gpio output data on resume, resulting
in incorrect GPIO output data after a resume. Add the missing register
write.
Signed-off-by: Russell King <[email protected]>
|
|
mcp_priv() does unexpected things when passed a void pointer. Make it
a typed inline function, which ensures that it works correctly in
these cases.
Signed-off-by: Russell King <[email protected]>
|
|
The perf_event_time() will call perf_cgroup_event_time()
if @event is a cgroup event. Just do it directly and avoid
the extra check..
Signed-off-by: Namhyung Kim <[email protected]>
Cc: Namhyung Kim <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Paul Mackerras <[email protected]>
Cc: Arnaldo Carvalho de Melo <[email protected]>
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Ingo Molnar <[email protected]>
|
|
When alloc_callchain_buffers() fails, it frees all of
entries before return. In addition, calling the
release_callchain_buffers() will cause a NULL pointer
dereference since callchain_cpu_entries is not set.
Signed-off-by: Namhyung Kim <[email protected]>
Acked-by: Frederic Weisbecker <[email protected]>
Cc: Namhyung Kim <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Paul Mackerras <[email protected]>
Cc: Arnaldo Carvalho de Melo <[email protected]>
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Ingo Molnar <[email protected]>
|
|
Sony Vaio VPCCW29FX does not resume correctly without
acpi_sleep=nonvs, so add it to the ACPI sleep blacklist.
https://bugzilla.kernel.org/show_bug.cgi?id=34722
Signed-off-by: Lan Tianyu <[email protected]>
Signed-off-by: Len Brown <[email protected]>
|
|
With the conversion of atomicio's routines in place (see commits
6f68c91c55e and 700130b41f4), atomicio.[ch] can be removed, replacing
the APEI specific pre-mapping capabilities with the more generalized
versions that drivers/acpi/osl.c provides.
Signed-off-by: Myron Stowe <[email protected]>
Signed-off-by: Len Brown <[email protected]>
|
|
This patch adds support for RAM to ACPI's mapping capabilities in order
to support APEI error injection (EINJ) actions.
This patch re-factors similar functionality introduced in commit
76da3fb3575, bringing it into osl.c in preparation for removing
./drivers/acpi/atomicio.[ch].
Signed-off-by: Myron Stowe <[email protected]>
Signed-off-by: Len Brown <[email protected]>
|
|
Base ACPI (CA) currently does not support atomic 64-bit reads and writes
(acpi_read() and acpi_write() split 64-bit loads/stores into two
32-bit transfers) yet APEI expects 64-bit transfer capability, even
when running on 32-bit systems.
This patch implements 64-bit read and write routines for APEI usage.
This patch re-factors similar functionality introduced in commit
04c25997c97, bringing it into the ACPI subsystem in preparation for
removing ./drivers/acpi/atomicio.[ch]. In the implementation I have
replicated acpi_os_read_memory() and acpi_os_write_memory(), creating
64-bit versions for APEI to utilize, as opposed to something more
elegant. My thinking is that we should attempt to see if we can get
ACPI's CA/OSL changed so that the existing acpi_read() and acpi_write()
interfaces are natively 64-bit capable and then subsequently remove the
replication.
Signed-off-by: Myron Stowe <[email protected]>
Signed-off-by: Len Brown <[email protected]>
|
|
This patch fixes CUBIC so that cwnd reductions made during RTOs can be
undone (just as they already can be undone when using the default/Reno
behavior).
When undoing cwnd reductions, BIC-derived congestion control modules
were restoring the cwnd from last_max_cwnd. There were two problems
with using last_max_cwnd to restore a cwnd during undo:
(a) last_max_cwnd was set to 0 on state transitions into TCP_CA_Loss
(by calling the module's reset() functions), so cwnd reductions from
RTOs could not be undone.
(b) when fast_covergence is enabled (which it is by default)
last_max_cwnd does not actually hold the value of snd_cwnd before the
loss; instead, it holds a scaled-down version of snd_cwnd.
This patch makes the following changes:
(1) upon undo, revert snd_cwnd to ca->loss_cwnd, which is already, as
the existing comment notes, the "congestion window at last loss"
(2) stop forgetting ca->loss_cwnd on TCP_CA_Loss events
(3) use ca->last_max_cwnd to check if we're in slow start
Signed-off-by: Neal Cardwell <[email protected]>
Acked-by: Stephen Hemminger <[email protected]>
Acked-by: Sangtae Ha <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
This patch fixes BIC so that cwnd reductions made during RTOs can be
undone (just as they already can be undone when using the default/Reno
behavior).
When undoing cwnd reductions, BIC-derived congestion control modules
were restoring the cwnd from last_max_cwnd. There were two problems
with using last_max_cwnd to restore a cwnd during undo:
(a) last_max_cwnd was set to 0 on state transitions into TCP_CA_Loss
(by calling the module's reset() functions), so cwnd reductions from
RTOs could not be undone.
(b) when fast_covergence is enabled (which it is by default)
last_max_cwnd does not actually hold the value of snd_cwnd before the
loss; instead, it holds a scaled-down version of snd_cwnd.
This patch makes the following changes:
(1) upon undo, revert snd_cwnd to ca->loss_cwnd, which is already, as
the existing comment notes, the "congestion window at last loss"
(2) stop forgetting ca->loss_cwnd on TCP_CA_Loss events
(3) use ca->last_max_cwnd to check if we're in slow start
Signed-off-by: Neal Cardwell <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
reverting back change that access enic->num_vfs outside
CONFIG_PCI_IOV
Reported-by: Randy Dunlap <[email protected]>
Signed-off-by: Roopa Prabhu <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
On a mx28evk board the following errors happens on mxs-sgtl5000 probe:
[ 0.660000] saif0_clk_set_rate: divider writing timeout
[ 0.670000] mxs-sgtl5000: probe of mxs-sgtl5000.0 failed with error -110
[ 0.670000] ALSA device list:
[ 0.680000] No soundcards found.
This timeout happens because clk_set_rate will result in writing to the DIV bits
of register HW_CLKCTRL_SAIF0 with the saif clock gated (CLKGATE bit set to one).
MX28 Reference states the following about CLKGATE:
"The DIV field can change ONLY when this clock gate bit field is low."
So call clk_prepare_enable prior to clk_set_rate to fix this problem.
After this change the mxs-saif driver can be correctly probed and audio is functional.
Signed-off-by: Fabio Estevam <[email protected]>
Signed-off-by: Mark Brown <[email protected]>
|
|
This reverts commit 42ff6572e5a4a7414330a4ca91f0335da67deca9.
New forcewake voodoo makes this no longer necessary.
Acked-by: Daniel Vetter <[email protected]>
Signed-off-by: Keith Packard <[email protected]>
|
|
The genirq layer complains if an interrupt handler returns with
interrupts enabled. The UCB1x00 handler does just this, because
ucb1x00_enable() calls mcp_enable(), which uses spin_lock_irq()
rather than spin_lock_irqsave(). Convert this, and the divisor
setting functions to use spin_lock_irqsave().
Signed-off-by: Russell King <[email protected]>
|
|
bus."
This reverts commit 5dd7bf59e0e8563265b3e5b33276099ef628fcc7.
Conflicts:
scripts/mod/file2alias.c
This change is wrong on many levels. First and foremost, it causes a
regression. On boot on Assabet, which this patch gives a codec id of
'ucb1x00', it gives:
ucb1x00 ID not found: 1005
0x1005 is a valid ID for the UCB1300 device.
Secondly, this patch is way over the top in terms of complexity. The
only device which has been seen to be connected with this MCP code is
the UCB1x00 (UCB1200, UCB1300 etc) devices, and they all use the same
driver. Adding a match table, requiring the codec string to match the
hardware ID read out of the ID register, etc is completely over the top
when we can just read the hardware ID register.
|
|
This reverts commit af9081ae64b941d32239b947882cd59ba855c5db.
This revert is necessary to revert 5dd7bf59e0e8563265b3e5b33276099ef628fcc7.
|
|
Properly clamp temperature limits set by the user. Without this fix,
attempts to write temperature limits above the maximum supported by
the chip (255 degrees Celsius) would arbitrarily and unexpectedly
result in the limit being set to 0 degree Celsius.
Signed-off-by: Jean Delvare <[email protected]>
Cc: [email protected]
Signed-off-by: Guenter Roeck <[email protected]>
|
|
Commit b6693005 (kmemleak: When the early log buffer is exceeded, report
the actual number) deferred the disabling of the early logging to
kmemleak_init(). However, when CONFIG_DEBUG_KMEMLEAK_DEFAULT_OFF=y, the
early logging was no longer disabled causing __init kmemleak functions
to be called even after the kernel freed the init memory. This patch
disables the early logging during kmemleak_init() if kmemleak is left
disabled.
Reported-by: Dirk Gouders <[email protected]>
Tested-by: Dirk Gouders <[email protected]>
Tested-by: Josh Boyer <[email protected]>
Signed-off-by: Catalin Marinas <[email protected]>
|
|
Commit 706e852 "pinctrl: correct a offset while enumerating pins"
modified the variable used by pinconf_pin_show()'s for loop, but didn't
update the for loop test expression.
Signed-off-by: Stephen Warren <[email protected]>
Signed-off-by: Linus Walleij <[email protected]>
|
|
Adding maintainer info for dma-buf buffer sharing framework;
some mailing lists interested in this work are also added.
Signed-off-by: Sumit Semwal <[email protected]>
Signed-off-by: Sumit Semwal <[email protected]>
Acked-by: Arnd Bergmann <[email protected]>
Acked-by: Daniel Vetter <[email protected]>
Acked-by: Dave Airlie <[email protected]>
|
|
Kmemleak should only track valid scan areas with a non-zero size.
Otherwise, such area may reside just at the end of an object and
kmemleak would report "Adding scan area to unknown object".
Signed-off-by: Tiejun Chen <[email protected]>
Signed-off-by: Catalin Marinas <[email protected]>
|
|
As on the other sam9 we need to cleanly shutdown the DDRAM before rebooting.
On those SoC the SDRAM/DDRAM controller is different. So, the assembly code
ends up being not cleanly combined with previous at91sam9_alt_restart function.
Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <[email protected]>
Acked-by: Nicolas Ferre <[email protected]>
|
|
Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <[email protected]>
Acked-by: Nicolas Ferre <[email protected]>
|
|
reset
Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <[email protected]>
Acked-by: Nicolas Ferre <[email protected]>
|
|
Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <[email protected]>
Acked-by: Nicolas Ferre <[email protected]>
|
|
fix AT91_DDRSDRC_MODE it's 3bit
add missing AT91_DDRSDRC_NR_14, AT91_DDRSDRC_DBW (16 and 32 bits support)
Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <[email protected]>
Acked-by: Nicolas Ferre <[email protected]>
|