Replace the use of the `S626_MULT_X1`, `S626_MULT_X2` and `S626_MULT_X4`
clock multiplier values with the equivalent `S626_CLKMULT_1X`,
`S626_CLKMULT_2X` and `S626_CLKMULT_4X` values to avoid duplication.
Replace the use of `S626_MULT_X0` with a new macro
`S626_CLKMULT_SPECIAL` (this is treated specially by the
'ClkMultA'/'ClkMultB' field of the 'CRA'/'CRB' register). Remove the
now unused `S626_MULT_X?` macros.
Signed-off-by: Ian Abbott <abbotti@mev.co.uk>
Reviewed-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The `S626_BF_*` bitfield position macros are no longer used and are just
a subset of the corresponding `S626_STDBIT_*` bitfield position macros.
Remove them.
Signed-off-by: Ian Abbott <abbotti@mev.co.uk>
Reviewed-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The 'IndxSrc' value for the standardized encoder setup is currently 1
bit wide and takes one of the following values:
S626_INDXSRC_HARD = 0 // index source from hardware encoder
S626_INDXSRC_SOFT = 1 // index source software controlled by IndxPol
However the hardware 'IndxSrcA' and 'IndxSrcB' values for the 'A' and
'B' counters are 2 bits wide. The above standardized values 0 and 1
correspond to the hardware values 0 and 2.
In order to simplify conversions between the standardized values and
hardware values, expand the range of standardized values to cover all
four possible values. The new values are as follows:
S626_INDXSRC_ENCODER = 0 // index source from hardware encoder
S626_INDXSRC_DIGIN = 1 // index source from digital inputs
S626_INDXSRC_SOFT = 2 // index source s/w controlled by IndxPol
S626_INDXSRC_DISABLED = 2 // index source disabled
(Note the change in value for `S626_INDXSRC_SOFT` and the replacement of
`S626_INDXSRC_HARD` with `S626_INDXSRC_ENCODER` for consistency with the
`CntSrc` values.)
Signed-off-by: Ian Abbott <abbotti@mev.co.uk>
Reviewed-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Use the new macros defined in "s626.h" for constructing and decomposing
'CRA', 'CRB' and standardized encoder setup values to make the
conversions between standardized encoder setup values, and CRA/CRB
register values easier to follow.
There is some messing about with the 'IndxSrc' values which are 1-bit
wide in the standardized encoder setup, and 2-bit wide in the 'CRA' and
'CRB' register values. This will be addressed by a later patch.
Signed-off-by: Ian Abbott <abbotti@mev.co.uk>
Reviewed-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
In case of error, the function platform_device_register_simple() returns
ERR_PTR() and never returns NULL. The NULL test in the return value check
should be replaced with IS_ERR().
Signed-off-by: Wei Yongjun <yongjun_wei@trendmicro.com.cn>
Acked-by: Sascha Hauer <s.hauer@pengutronix.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
As pointed out by Hartley Sweeten, one of my recent patches resulted in
the start of a multi-line comment ending up misaligned. Fix it.
Signed-off-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The latest version of NetworkManager does not recognize the device as wireless
without this change.
Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net>
Cc: Stable <stable@vger.kernel.org> # [3.12+]
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Emitting an OOM message isn't necessary after input_allocate_device
as there's a generic OOM and a dump_stack already done.
Signed-off-by: Joe Perches <joe@perches.com>
Acked-by: Jonathan Cameron <jic23@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Checking if MAC address is valid using is_valid_ether_addr() is already done in
of_get_mac_address().
Signed-off-by: Luka Perkov <luka@openwrt.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
imx27 pincontrol driver using the imx1 core driver. The DT bindings are
similar to other imx pincontrol drivers.
Signed-off-by: Markus Pargmann <mpa@pengutronix.de>
Acked-by: Sascha Hauer <s.hauer@pengutronix.de>
Acked-by: Shawn Guo <shawn.guo@linaro.org>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
This patch removes a space before semicolon as
specified by checkpatch.pl.
Signed-off-by: Kevin McKinney <klmckinney1@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The FUNCTION_TYPE parameter isn't associated with any NID, thus
showing the uninitialized nid in the error message is simply
nonsense.
Spotted by coverity CID 145068.
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Removed a developer debug statement per the TODO list. Additionally,
removed braces for the if-statement to match coding style.
Signed-off-by: Chuong Ngo <cngo.github@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Running checkpatch.pl on the file drivers/staging/bcm/Adapter.h gave
an error as it is a mistake to use typedef for structures
according to CodeingStyle as it reduces readability. The typedef was
removed and all occurrences of the typedef union were replaced with
union u_ip_address as types are all lowercase.
Signed-off-by: Himangi Saraogi <himangi774@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This patch fixes the Sparse Warnings "symbol was
not declared. Should it be static?" and "defined
but not used [-Wunused-variable]"
in reg.c
Signed-off-by: Ebru Akagunduz <ebru.akagunduz@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The size of the register cache array is actually 6 instead of 7,
as it caches up to AK4114_REG_INT1_MASK. This resulted in unexpected
access out of array range, although most of them aren't so serious
(just reading one more byte on the stack at snd_ak4114_create()).
Also, the check of cache size was wrongly done by checking with
sizeof() instead of ARRAY_SIZE(). Fixed this together.
(And yes, hardcoded numbers are bad, but I keep the coding style as is
for making it clear what this patch actually does.)
Spotted by coverity among several CIDs, e.g. 711621.
Signed-off-by: Takashi Iwai <tiwai@suse.de>
This patch fixes Sparse Warnings "symbol was not
declared. Should it be static?" and "defined
but not used [-Wunused-function]" in
phy_calibration.c
Signed-off-by: Ebru Akagunduz <ebru.akagunduz@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Use memdup_user rather than duplicating implementation. Fix following
coccinelle warnings:
drivers/staging/comedi/comedi_fops.c:1425:5-12: WARNING opportunity for memdup_user
drivers/staging/comedi/comedi_fops.c:1553:6-13: WARNING opportunity for memdup_user
Signed-off-by: Teodora Baluta <teobaluta@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Pull perf tooling fixes from Ingo Molnar:
"This contains five tooling fixes:
- fix a remaining mmap2 assumption which resulted in perf top output
breakage
- fix mmap ring-buffer processing bug that corrupts data
- fix for a severe python scripting memory leak
- fix broken (and user-visible) -g option handling
- fix stdio output
The diffstat size is larger than what we'd like to see this late :-/"
* 'perf-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
perf tools: Fixup mmap event consumption
perf top: Split -G and --call-graph
perf record: Split -g and --call-graph
perf hists: Add color overhead for stdio output buffer
perf tools: Fix up /proc/PID/maps parsing
perf script python: Fix mem leak due to missing Py_DECREFs on dict entries
Linux uses a return type of int for status codes. The file
ft1000_download.c uses a mixture of u16 and u32. This patch changes all
variables called status or Status to ints, whether they are returned
from the function or not. It also changes the return type of all
functions returning one of the variables to correspond. Also, the
declaration of scram_dnldr has been changed in ft1000_usb.h.
Signed-off-by: Kelley Nielsen <kelleynnn@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
function write_blk, in ft1000_download.c, contains many coding style
issues. It has indentations of 3 spaces, long lines, C99 comments, and
extra whitespace. It also has a return type of u32, and changing the
returned variable in the function triggers a checkpatch leading spaces
warning. Indentation should be fixed throughout the file for
consistency.
This patch fixes those issues, in preparation for correcting the status
return type throughout the file. The variable Status has been changed
from u32 to int and renamed status.
Signed-off-by: Kelley Nielsen <kelleynnn@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
function write_blk is long and overly complex, consisting of a triply
nested loop. It also has improper indentation and line lengths
throughout, and has return type of u32 rather than int. Some of the
lines, when converted to proper indentation, create checkpatch warnings
for too many leading tabs.
This patch extracts the innermost loop into its own function,
write_dpram32_and_check. This removes several levels of indentation from
the extracted lines and makes the original function simpler. Two local
variables from the original function, u16 resultbuffer[] and a loop
counter, have been made local variables of the new function. Two calls
to msleep() have been replaced with usleep_range() as per Documentation/
timers/timers-howto.txt (which was referred to in a checkpatch warning).
Several other style issues in the extracted code have been corrected as
well.
Signed-off-by: Kelley Nielsen <kelleynnn@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
function scram_dnldr, in ft1000_download.c, is very long and consists
mainly of nested switch statements inside a while loop. Some code in one
of the inner switch cases was almost identical to the code in the
previously extracted function request_code_segment. The duplicated code
was replaced with a call to request_code_segment, and
request_code_segment was slightly modified to work in both cases.
A new parameter was added to request_code_segment, a bool to distinguish
which case it was replacing. The name of an existing parameter (now
called endpoint) was changed to reflect the fact that it will be passed
in from more than one place. Several lines from the case containing the
duplicated code were moved to request_code_segment, and a test was added
to determine if these lines or a line from the original function should
be run.
Finally, an unused variable (tempword) was removed from scram_dnldr.
Signed-off-by: Kelley Nielsen <kelleynnn@gmail.com>
Reviewed-by: Peter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
function scram_dnldr in ft1000_download.c is very long and contains many
coding style errors and best practice violations. It consists of nested
switch statements inside a while loop. One of the inner switch cases has
been extracted as a helper function. Also, some style errors (such as
C99 comments) have been fixed, an assignment to an unread variable has
been removed, and break statements inside ifs have been converted to
returns.
Signed-off-by: Kelley Nielsen <kelleynnn@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Without the timer debugging, the delayed kobject release will just
result in undebuggable oopses if it triggers any latent bugs. That
doesn't actually help debugging at all.
So make DEBUG_KOBJECT_RELEASE depend on DEBUG_OBJECTS_TIMERS to avoid
having people enable one without the other.
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Just pass the error code returned from copy_from_user_toio() and
copy_to_user_fromio() helpers.
Spotted by coverity CID 114119.
Signed-off-by: Takashi Iwai <tiwai@suse.de>
The variable runtime is never used, and this might be even a source of
NULL-dereference. Nothing better than killing it.
Spotted by coverity CID 100862.
Signed-off-by: Takashi Iwai <tiwai@suse.de>
When no proper id string is given, the driver tries to fall back to
copy the proc_root name string via strcpy(), but this might overflow
the fixed string size. Let's use strlcpy().
Spotted by coverity CID 139008.
Signed-off-by: Takashi Iwai <tiwai@suse.de>
When nopcm=1 is set, some initializations based on hrtimer resolution
might be bogus because the driver checks the resolution only when
nopcm=0. Simply get the resolution always at first for fixing the
bug.
Spotted by coverity CID 139740.
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Core driver for register formats of imx1/imx21/imx27 processors.
The pins of those processors are grouped into ports. Each port has 32
pins. The pins mux configuration is controlled by registers with 1 or 2
bit per pin, depending on the specific control register.
Signed-off-by: Markus Pargmann <mpa@pengutronix.de>
Acked-by: Sascha Hauer <s.hauer@pengutronix.de>
Acked-by: Shawn Guo <shawn.guo@linaro.org>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Currently we only optimize the context switch between two
contexts that have the same parent; this forgoes the
optimization between parent and child context, even though these
contexts could be equivalent too.
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Shishkin, Alexander <alexander.shishkin@intel.com>
Link: http://lkml.kernel.org/r/20131007164257.GH3081@twins.programming.kicks-ass.net
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Originally I've thought that this is leftover hw state dirt from the
BIOS. But after way too much helpless flailing around on my part I've
noticed that the actual bug is when we change the state of an already
active pipe.
For example when we change the fdi lines from 2 to 3 without switching
off outputs in-between we'll never see the crucial on->off transition
in the ->modeset_global_resources hook the current logic relies on.
Patch version 2 got this right by instead also checking whether the
pipe is indeed active. But that in turn broke things when pipes have
been turned off through dpms since the bifurcate enabling is done in
the ->crtc_mode_set callback.
To address this issues discussed with Ville in the patch review move
the setting of the bifurcate bit into the ->crtc_enable hook. That way
we won't wreak havoc with this state when userspace puts all other
outputs into dpms off state. This also moves us forward with our
overall goal to unify the modeset and dpms on paths (which we need to
have to allow runtime pm in the dpms off state).
Unfortunately this requires us to move the bifurcate helpers around a
bit.
Also update the commit message, I've misanalyzed the bug rather badly.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=70507
Tested-by: Jan-Michael Brummer <jan.brummer@tabos.org>
Cc: stable@vger.kernel.org
Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
V3 of the NFQUEUE target ignores the --queue-bypass flag,
causing packets to be dropped when the userspace listener
isn't running.
Regression is in since 8746ddcf12 ("netfilter: xt_NFQUEUE:
introduce CPU fanout").
Reported-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Holger Eitzenberger <holger@eitzenberger.org>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Now that we can deal with nested NMI due to IRET re-enabling NMIs and
can deal with faults from NMI by making sure we preserve CR2 over NMIs
we can in fact simply access user-space memory from NMI context.
So rewrite copy_from_user_nmi() to use __copy_from_user_inatomic() and
rework the fault path to do the minimal required work before taking
the in_atomic() fault handler.
In particular avoid perf_sw_event() which would make perf recurse on
itself (it should be harmless as our recursion protections should be
able to deal with this -- but why tempt fate).
Also rename notify_page_fault() to kprobes_fault() as that is a much
better name; there is no notifier in it and its specific to kprobes.
Don measured that his worst case NMI path shrunk from ~300K cycles to
~150K cycles.
Cc: Stephane Eranian <eranian@google.com>
Cc: jmario@redhat.com
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: dave.hansen@linux.intel.com
Tested-by: Don Zickus <dzickus@redhat.com>
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/20131024105206.GM2490@laptop.programming.kicks-ass.net
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Oleg complained about the excessive 0-ing in perf_event_mmap_event(),
so try and be smarter about it while keeping it fairly fool proof and
avoid leaking random bits out to userspace.
Suggested-by: Oleg Nesterov <oleg@redhat.com>
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/n/tip-8jirlm99m6if2z13wd6rbyu6@git.kernel.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
perf_event_mmap_event() does kzalloc(PATH_MAX + sizeof(u64)) to
ensure we can align the size later. However this means that we
actually allocate PAGE_SIZE * 2 buffer, seems too much.
Change this code to allocate PATH_MAX==PAGE_SIZE bytes, but tell
d_path() to not use the last sizeof(u64) bytes.
Note: it is not clear why do we need __GFP_ZERO, see the next patch.
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/20131016201004.GC23214@redhat.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
1. perf_event_mmap(vma) is never called with a gate_vma-like arg,
remove the "if (!vma->vm_mm)" code.
2. arch_vma_name() can use the chached value of mmap_event->vma.
3. Change the code to not call arch_vma_name() twice.
4. Purely cosmetic, but since we use "goto got_name" all the time
remove "else" from "[stack]" branch just for symmetry.
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/20131016200945.GB23214@redhat.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
throttle_cfs_rq() doesn't check to make sure that period_timer is running,
and while update_curr/assign_cfs_runtime does, a concurrently running
period_timer on another cpu could cancel itself between this cpu's
update_curr and throttle_cfs_rq(). If there are no other cfs_rqs running
in the tg to restart the timer, this causes the cfs_rq to be stranded
forever.
Fix this by calling __start_cfs_bandwidth() in throttle if the timer is
inactive.
(Also add some sched_debug lines for cfs_bandwidth.)
Tested: make a run/sleep task in a cgroup, loop switching the cgroup
between 1ms/100ms quota and unlimited, checking for timer_active=0 and
throttled=1 as a failure. With the throttle_cfs_rq() change commented out
this fails, with the full patch it passes.
Signed-off-by: Ben Segall <bsegall@google.com>
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Cc: pjt@google.com
Link: http://lkml.kernel.org/r/20131016181632.22647.84174.stgit@sword-of-the-dawn.mtv.corp.google.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Currently, group entity load-weights are initialized to zero. This
admits some races with respect to the first time they are re-weighted in
earlty use. ( Let g[x] denote the se for "g" on cpu "x". )
Suppose that we have root->a and that a enters a throttled state,
immediately followed by a[0]->t1 (the only task running on cpu[0])
blocking:
put_prev_task(group_cfs_rq(a[0]), t1)
put_prev_entity(..., t1)
check_cfs_rq_runtime(group_cfs_rq(a[0]))
throttle_cfs_rq(group_cfs_rq(a[0]))
Then, before unthrottling occurs, let a[0]->b[0]->t2 wake for the first
time:
enqueue_task_fair(rq[0], t2)
enqueue_entity(group_cfs_rq(b[0]), t2)
enqueue_entity_load_avg(group_cfs_rq(b[0]), t2)
account_entity_enqueue(group_cfs_ra(b[0]), t2)
update_cfs_shares(group_cfs_rq(b[0]))
< skipped because b is part of a throttled hierarchy >
enqueue_entity(group_cfs_rq(a[0]), b[0])
...
We now have b[0] enqueued, yet group_cfs_rq(a[0])->load.weight == 0
which violates invariants in several code-paths. Eliminate the
possibility of this by initializing group entity weight.
Signed-off-by: Paul Turner <pjt@google.com>
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/20131016181627.22647.47543.stgit@sword-of-the-dawn.mtv.corp.google.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
__start_cfs_bandwidth calls hrtimer_cancel while holding rq->lock,
waiting for the hrtimer to finish. However, if sched_cfs_period_timer
runs for another loop iteration, the hrtimer can attempt to take
rq->lock, resulting in deadlock.
Fix this by ensuring that cfs_b->timer_active is cleared only if the
_latest_ call to do_sched_cfs_period_timer is returning as idle. Then
__start_cfs_bandwidth can just call hrtimer_try_to_cancel and wait for
that to succeed or timer_active == 1.
Signed-off-by: Ben Segall <bsegall@google.com>
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Cc: pjt@google.com
Link: http://lkml.kernel.org/r/20131016181622.22647.16643.stgit@sword-of-the-dawn.mtv.corp.google.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
hrtimer_expires_remaining does not take internal hrtimer locks and thus
must be guarded against concurrent __hrtimer_start_range_ns (but
returning HRTIMER_RESTART is safe). Use cfs_b->lock to make it safe.
Signed-off-by: Ben Segall <bsegall@google.com>
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Cc: pjt@google.com
Link: http://lkml.kernel.org/r/20131016181617.22647.73829.stgit@sword-of-the-dawn.mtv.corp.google.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
When we transition cfs_bandwidth_used to false, any currently
throttled groups will incorrectly return false from cfs_rq_throttled.
While tg_set_cfs_bandwidth will unthrottle them eventually, currently
running code (including at least dequeue_task_fair and
distribute_cfs_runtime) will cause errors.
Fix this by turning off cfs_bandwidth_used only after unthrottling all
cfs_rqs.
Tested: toggle bandwidth back and forth on a loaded cgroup. Caused
crashes in minutes without the patch, hasn't crashed with it.
Signed-off-by: Ben Segall <bsegall@google.com>
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Cc: pjt@google.com
Link: http://lkml.kernel.org/r/20131016181611.22647.80365.stgit@sword-of-the-dawn.mtv.corp.google.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
OK, so what I'm actually seeing on my WSM is that sched/clock.c is
'broken' for the purpose we're using it for.
What triggered it is that my WSM-EP is broken :-(
[ 0.001000] tsc: Fast TSC calibration using PIT
[ 0.002000] tsc: Detected 2533.715 MHz processor
[ 0.500180] TSC synchronization [CPU#0 -> CPU#6]:
[ 0.505197] Measured 3 cycles TSC warp between CPUs, turning off TSC clock.
[ 0.004000] tsc: Marking TSC unstable due to check_tsc_sync_source failed
For some reason it consistently detects TSC skew, even though NHM+
should have a single clock domain for 'reasonable' systems.
This marks sched_clock_stable=0, which means that we do fancy stuff to
try and get a 'sane' clock. Part of this fancy stuff relies on the tick,
clearly that's gone when NOHZ=y. So for idle cpus time gets stuck, until
it either wakes up or gets kicked by another cpu.
While this is perfectly fine for the scheduler -- it only cares about
actually running stuff, and when we're running stuff we're obviously not
idle. This does somewhat break down for perf which can trigger events
just fine on an otherwise idle cpu.
So I've got NMIs get get 'measured' as taking ~1ms, which actually
don't last nearly that long:
<idle>-0 [013] d.h. 886.311970: rcu_nmi_enter <-do_nmi
...
<idle>-0 [013] d.h. 886.311997: perf_sample_event_took: HERE!!! : 1040990
So ftrace (which uses sched_clock(), not the fancy bits) only sees
~27us, but we measure ~1ms !!
Now since all this measurement stuff lives in x86 code, we can actually
fix it.
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Cc: mingo@kernel.org
Cc: dave.hansen@linux.intel.com
Cc: eranian@google.com
Cc: Don Zickus <dzickus@redhat.com>
Cc: jmario@redhat.com
Cc: acme@infradead.org
Link: http://lkml.kernel.org/r/20131017133350.GG3364@laptop.programming.kicks-ass.net
Signed-off-by: Ingo Molnar <mingo@kernel.org>
