perf stat -a needs 10 open file descriptors per logical CPU
perf stat -a -dddd needs 20 open fds for each.
This implies that stat -a doesn't work on any system with the default
ulimit -n 1024 which has more than ~100 CPUs and stat -a -dddd doesn't
work on anything with more than 46 CPUs.
Longer term there needs to be probably some way to lower the file
descriptor requirements. This would need some changes in the kernel/user
interface.
But short term this patch just tries to increase the file descriptor
limit in perf itself, when it runs into a EMFILE.
It first sets it to the hard limit, and then tries to increase the hard
limit.
On Fedora systems the default seems to be soft limit 1024 and hard limit
4*1024. So even non root can support 409 or 186 CPUs respectively. root
can go far higher.
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Link: http://lkml.kernel.org/r/1375670486-15480-1-git-send-email-andi@firstfloor.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Freq events may not always be affine to a particular CPU. As such,
account_event_cpu() may crash if we account per cpu a freq event
that has event->cpu == -1.
To solve this, lets account freq events globally. In practice
this doesn't change much the picture because perf tools create
per-task perf events with one event per CPU by default. Profiling a
single CPU is usually a corner case so there is no much point in
optimizing things that way.
Reported-by: Jiri Olsa <jolsa@redhat.com>
Suggested-by: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Tested-by: Jiri Olsa <jolsa@redhat.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Stephane Eranian <eranian@google.com>
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/1375460996-16329-3-git-send-email-fweisbec@gmail.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
When we fail to allocate the callchain buffers, we roll back the refcount
we did and return from get_callchain_buffers().
However we take the refcount and allocate under the callchain lock
but the rollback is done outside the lock.
As a result, while we roll back, some concurrent callchain user may
call get_callchain_buffers(), see the non-zero refcount and give up
because the buffers are NULL without itself retrying the allocation.
The consequences aren't that bad but that behaviour looks weird enough and
it's better to give their chances to the following callchain users where
we failed.
Reported-by: Jiri Olsa <jolsa@redhat.com>
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Acked-by: Jiri Olsa <jolsa@redhat.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Stephane Eranian <eranian@google.com>
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/1375460996-16329-2-git-send-email-fweisbec@gmail.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
This patch adds support for the SNB-EP PCU uncore PMU extra_sel_bit
(bit 21) which is missing from the documentation in Table-2.75 of
Intel Xeon Processor E5-2600 Product Family Uncore Performance
Monitoring Guide. It is referred to later in Table-2.81. Without
this selection bit explicitly enabled by the kernel, some events
such as COREx_TRANSITION_CYCLES do not count correctly.
Signed-off-by: Yan, Zheng <zheng.z.yan@intel.com>
Reviewed-by: Stephane Eranian <eranian@google.com>
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/1376375382-21350-4-git-send-email-zheng.z.yan@intel.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
The QPI uncore boxes have two pairs of MATCH/MASK registers that
user to filter packet traffic serviced by QPI link layer. These
registers are in auxiliary PCI devices.
This patch changes the meaning of (struct pci_device_id)->driver_data.
The first 8 bits are device index of the same uncore type, the second
8 bytes are uncore type index. Auxiliary PCI device's type is defined
as UNCORE_EXTRA_PCI_DEV(0xff)
Signed-off-by: Yan, Zheng <zheng.z.yan@intel.com>
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/1375856245-10717-1-git-send-email-zheng.z.yan@intel.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Pull perf/core improvements and fixes from Arnaldo Carvalho de Melo:
* Allow specifying syscalls in 'perf trace', a la strace.
* Simplify symbol filtering by doing it at machine class level,
from Adrian Hunter.
* Add option to 'perf kvm' to print only events that exceed a specified time
duration, from David Ahern.
* 'perf sched' improvements, including removing some tracepoints that provide
the same information as the PERF_RECORD_{FORK,EXIT} events.
* Improve stack trace printing, from David Ahern.
* Update documentation with live command, from David Ahern
* Fix 'perf test' compile failure on do_sort_something, from David Ahern.
* Improve robustness of topology parsing code, from Stephane Eranian.
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Commit b55ae0a9 added code-reading.c which fails to compile on Fedora 16
with compiler version:
$ gcc --version
gcc (GCC) 4.6.3 20120306 (Red Hat 4.6.3-2)
Failure message is:
tests/code-reading.c: In function ‘do_sort_something’:
tests/code-reading.c:305:13: error: stack protector not protecting local variables: variable length buffer [-Werror=stack-protector]
cc1: all warnings being treated as errors
make: *** [/tmp/junk/tests/code-reading.o] Error 1
make: *** Waiting for unfinished jobs....
v2: as Adrian noticed changed sizeof to ARRAY_SIZE
Signed-off-by: David Ahern <dsahern@gmail.com>
Acked-by: Adrian Hunter <adrian.hunter@intel.com>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Link: http://lkml.kernel.org/r/1376454732-83728-1-git-send-email-dsahern@gmail.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
perf_trace_buf_prepare() + perf_trace_buf_submit(task => NULL)
make no sense if hlist_empty(head). Change perf_trace_##call()
to check ->perf_events beforehand and do nothing if it is empty.
This removes the overhead for tasks without events associated
with them. For example, "perf record -e sched:sched_switch -p1"
attaches the counter(s) to the single task, but every task in
system will do perf_trace_buf_prepare/submit() just to realize
that it was not attached to this event.
However, we can only do this if __task == NULL, so we also add
the __builtin_constant_p(__task) check.
With this patch "perf bench sched pipe" shows approximately 4%
improvement when "perf record -p1" runs in parallel, many thanks
to Steven for the testing.
Link: http://lkml.kernel.org/r/20130806160847.GA2746@redhat.com
Tested-by: David Ahern <dsahern@gmail.com>
Acked-by: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
The next patch tries to avoid the costly perf_trace_buf_* calls
when possible but there is a problem. We can only do this if
__task == NULL, perf_tp_event(task != NULL) has the additional
code for this case.
Unfortunately, TP_perf_assign/__perf_xxx which changes the default
values of __count/__task variables for perf_trace_buf_submit() is
called "too late", after we already did perf_trace_buf_prepare(),
and the optimization above can't work.
So this patch simply embeds __perf_xxx() into TP_ARGS(), this way
DECLARE_EVENT_CLASS() can use the result of assignments hidden in
"args" right after ftrace_get_offsets_##call() which is mostly
trivial. This allows us to have the fast-path "__task != NULL"
check at the start, see the next patch.
Link: http://lkml.kernel.org/r/20130806160844.GA2739@redhat.com
Tested-by: David Ahern <dsahern@gmail.com>
Acked-by: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
To simplify the review of the next patches:
1. We are going to reimplent __perf_task/counter and embedd them
into TP_ARGS(). expand TRACE_EVENT(sched_stat_runtime) into
DECLARE_EVENT_CLASS() + DEFINE_EVENT(), this way they can use
different TP_ARGS's.
2. Change perf_trace_##call() macro to do perf_fetch_caller_regs()
right before perf_trace_buf_prepare().
This way it evaluates TP_ARGS() asap, the next patch explores
this fact.
Note: after 87f44bbc perf_trace_buf_prepare() doesn't need
"struct pt_regs *regs", perhaps it makes sense to remove this
argument. And perhaps we can teach perf_trace_buf_submit()
to accept regs == NULL and do fetch_caller_regs(CALLER_ADDR1)
in this case.
3. Cosmetic, but the typecast from "void*" buys nothing. It just
adds the noise, remove it.
Link: http://lkml.kernel.org/r/20130806160841.GA2736@redhat.com
Acked-by: Peter Zijlstra <peterz@infradead.org>
Tested-by: David Ahern <dsahern@gmail.com>
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Pull perf/core improvements and fixes from Arnaldo Carvalho de Melo:
* Do annotation using /proc/kcore and /proc/kallsyms, removing the
need for a vmlinux file kernel assembly annotation. This also improves
this use case because vmlinux has just the initial kernel image, not
what is actually in use after various code patchings by things like
alternatives, etc. From Adrian Hunter.
* Add various improvements and fixes to the "vmlinux matches kallsyms"
'perf test' entry, related to the /proc/kcore annotation feature.
* Add --initial-delay option to 'perf stat' to skip measuring for
the startup phase, from Andi Kleen.
* Add perf kvm stat live mode that combines aspects of 'perf kvm stat' record
and report, from David Ahern.
* Add option to analyze specific VM in perf kvm stat report, from David Ahern.
* Do not require /lib/modules/* on a guest, fix from Jason Wessel.
* Group leader sampling, that allows just one event in a group to sample while
the other events have just its values read, from Jiri Olsa.
* Add support for a new modifier "D", which requests that the event, or group
of events, be pinned to the PMU, from Michael Ellerman.
* Fix segmentation fault on the gtk browser, from Namhyung Kim.
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Pull SCSI fixes from James Bottomley:
"This is three bug fixes: An fnic warning caused by sleeping under a
lock, a major regression with our updated WRITE SAME/UNMAP logic which
caused tons of USB devices (and one RAID card) to cease to function
and a megaraid_sas firmware initialisation problem which causes kdump
failures"
* tag 'scsi-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi:
[SCSI] Don't attempt to send extended INQUIRY command if skip_vpd_pages is set
[SCSI] fnic: BUG: sleeping function called from invalid context during probe
[SCSI] megaraid_sas: megaraid_sas driver init fails in kdump kernel
Pull powerpc fixes from Ben Herrenschmidt:
"This includes small series from Michael Neuling to fix a couple of
nasty remaining problems with the new Power8 support, also targeted at
stable 3.10, without which some new userspace accessible registers
aren't properly context switched, and in some case, can be clobbered
by the user of transactional memory.
Along with that, a few slightly more minor things, such as a missing
Kconfig option to enable handling of denorm exceptions when not
running under a hypervisor (or userspace will randomly crash when
hitting denorms with the vector unit), some nasty bugs in the new
pstore oops code, and other simple bug fixes worth having in now.
Note: I picked up the two powerpc KVM fixes as Alex Graf asked me to
handle KVM bits while he is on vacation. However I'll let him decide
whether they should go to -stable or not when he is back"
* 'merge' of git://git.kernel.org/pub/scm/linux/kernel/git/benh/powerpc:
powerpc/tm: Fix context switching TAR, PPR and DSCR SPRs
powerpc: Save the TAR register earlier
powerpc: Fix context switch DSCR on POWER8
powerpc: Rework setting up H/FSCR bit definitions
powerpc: Fix hypervisor facility unavaliable vector number
powerpc/kvm/book3s_pr: Return appropriate error when allocation fails
powerpc/kvm: Add signed type cast for comparation
powerpc/eeh: Add missing procfs entry for PowerNV
powerpc/pseries: Add backward compatibilty to read old kernel oops-log
powerpc/pseries: Fix buffer overflow when reading from pstore
powerpc: On POWERNV enable PPC_DENORMALISATION by default
Pull s390 kvm fixes from Paolo Bonzini:
"Two fixes for s390"
* tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm:
KVM: s390: fix pfmf non-quiescing control handling
KVM: s390: move kvm_guest_enter,exit closer to sie
Pull i2c fixes from Wolfram Sang:
"Some driver bugfixes for the I2C subsystem"
* 'i2c/for-current' of git://git.kernel.org/pub/scm/linux/kernel/git/wsa/linux:
i2c: mv64xxx: Document the newly introduced allwinner compatible
i2c: Fix Kontron PLD prescaler calculation
i2c: i2c-mxs: Use DMA mode even for small transfers
Pull btrfs fixes from Chris Mason:
"These are assorted fixes, mostly from Josef nailing down xfstests
runs. Zach also has a long standing fix for problems with readdir
wrapping f_pos (or ctx->pos)
These patches were spread out over different bases, so I rebased
things on top of rc4 and retested overnight"
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mason/linux-btrfs:
btrfs: don't loop on large offsets in readdir
Btrfs: check to see if root_list is empty before adding it to dead roots
Btrfs: release both paths before logging dir/changed extents
Btrfs: allow splitting of hole em's when dropping extent cache
Btrfs: make sure the backref walker catches all refs to our extent
Btrfs: fix backref walking when we hit a compressed extent
Btrfs: do not offset physical if we're compressed
Btrfs: fix extent buffer leak after backref walking
Btrfs: fix a bug of snapshot-aware defrag to make it work on partial extents
btrfs: fix file truncation if FALLOC_FL_KEEP_SIZE is specified
Pull NFS client bugfixes from Trond Myklebust:
- Stable patch for lockd to fix Oopses due to inappropriate calls to
utsname()->nodename
- Stable patches for sunrpc to fix Oopses on shutdown when using
AF_LOCAL sockets with rpcbind
- Fix memory leak and error checking issues in nfs4_proc_lookup_mountpoint
- Fix a regression with the sync mount option failing to work for nfs4
mounts
- Fix a writeback performance issue when doing cache invalidation
- Remove an incorrect call to nfs_setsecurity in nfs_fhget
* tag 'nfs-for-3.11-4' of git://git.linux-nfs.org/projects/trondmy/linux-nfs:
NFSv4: Fix up nfs4_proc_lookup_mountpoint
NFS: Remove unnecessary call to nfs_setsecurity in nfs_fhget()
NFSv4: Fix the sync mount option for nfs4 mounts
NFS: Fix writeback performance issue on cache invalidation
SUNRPC: If the rpcbind channel is disconnected, fail the call to unregister
SUNRPC: Don't auto-disconnect from the local rpcbind socket
LOCKD: Don't call utsname()->nodename from nlmclnt_setlockargs
Pull nfsd fixes from Bruce Fields:
"Some fixes for a 4.1 feature that in retrospect probably should have
waited for 3.12.... But it appears to be working now"
* 'for-3.11' of git://linux-nfs.org/~bfields/linux:
nfsd: Fix SP4_MACH_CRED negotiation in EXCHANGE_ID
nfsd4: Fix MACH_CRED NULL dereference
Pull sound fixes from Takashi Iwai:
"A couple of USB-audio fixes that should also go to stable kernels"
* tag 'sound-3.11' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound:
ALSA: usb-audio: do not trust too-big wMaxPacketSize values
ALSA: 6fire: fix DMA issues with URB transfer_buffer usage
Pull staging driver fixes from Greg KH:
"Here are 3 small fixes for staging/IIO drivers for 3.11-rc5. Nothing
huge, two IIO driver fixes, and a zcache fix. All of these have been
in linux-next for a while"
* tag 'staging-3.11-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/staging:
staging: zcache: fix "zcache=" kernel parameter
iio: ti_am335x_adc: Fix wrong samples received on 1st read
iio:trigger: Fix use_count race condition
Pull USB fixes from Greg KH:
"Here are 3 small USB fixes for 3.11-rc5.
One is a fix that the ChromeOS developers ran into on some Intel
hardware, one is a build fix, and the last is a MAINTAINERS update to
help people figure out where to send USB network driver patches.
All of these have been in linux-next for a while"
* tag 'usb-3.11-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb:
MAINTAINERS: Add separate section for USB NETWORKING DRIVERS
usb: xhci: add missing dma-mapping.h includes
usb: core: don't try to reset_device() a port that got just disconnected
When btrfs readdir() hits the last entry it sets the readdir offset to a
huge value to stop buggy apps from breaking when the same name is
returned by readdir() with concurrent rename()s.
But unconditionally setting the offset to INT_MAX causes readdir() to
loop returning any entries with offsets past INT_MAX. It only takes a
few hours of constant file creation and removal to create entries past
INT_MAX.
So let's set the huge offset to LLONG_MAX if the last entry has already
overflowed 32bit loff_t. Without large offsets behaviour is identical.
With large offsets 64bit apps will work and 32bit apps will be no more
broken than they currently are if they see large offsets.
Signed-off-by: Zach Brown <zab@redhat.com>
Signed-off-by: Josef Bacik <jbacik@fusionio.com>
Signed-off-by: Chris Mason <chris.mason@fusionio.com>
