summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* perf/tool: Make the event parser re-entrantZheng Yan2012-06-184-64/+98
| | | | | | | | | | | | | Make the event parser reentrant by creating separate scanner for each parsing. The scanner is passed to the bison as and argument to the lexer. Signed-off-by: Zheng Yan <zheng.z.yan@intel.com> [ Cleaned up the patch. ] Signed-off-by: Jiri Olsa <jolsa@redhat.com> Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Link: http://lkml.kernel.org/r/1339741902-8449-11-git-send-email-zheng.z.yan@intel.com Signed-off-by: Ingo Molnar <mingo@kernel.org>
* perf/tool: Use data struct for arg passing in event parse functionJiri Olsa2012-06-183-26/+50
| | | | | | | | | | | | | | | Moving all the bison arguments into the structure. In upcomming patches we are going to: - add more arguments - reuse the grammer for term parsing so it's more clear to pack/separate related arguments. Signed-off-by: Jiri Olsa <jolsa@redhat.com> Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Link: http://lkml.kernel.org/r/1339741902-8449-10-git-send-email-zheng.z.yan@intel.com Signed-off-by: Ingo Molnar <mingo@kernel.org>
* perf/x86: Add Intel Nehalem and Sandy Bridge-EP uncore supportYan, Zheng2012-06-183-0/+581
| | | | | | | | | | | | | | | The uncore subsystem in Sandy Bridge-EP consists of 8 components: Ubox, Cacheing Agent, Home Agent, Memory controller, Power Control, QPI Link Layer, R2PCIe, R3QPI. Signed-off-by: Zheng Yan <zheng.z.yan@intel.com> Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Andrew Morton <akpm@linux-foundation.org> Cc: Thomas Gleixner <tglx@linutronix.de> Link: http://lkml.kernel.org/r/1339741902-8449-9-git-send-email-zheng.z.yan@intel.com Signed-off-by: Ingo Molnar <mingo@kernel.org>
* perf: Add generic PCI uncore PMU device supportYan, Zheng2012-06-182-5/+236
| | | | | | | | | | | This patch adds generic support for uncore PMUs presented as PCI devices. (These come in addition to the CPU/MSR based uncores.) Signed-off-by: Zheng Yan <zheng.z.yan@intel.com> Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Link: http://lkml.kernel.org/r/1339741902-8449-8-git-send-email-zheng.z.yan@intel.com Signed-off-by: Ingo Molnar <mingo@kernel.org>
* perf/x86: Add Intel Nehalem and Sandy Bridge uncore PMU supportYan, Zheng2012-06-182-0/+245
| | | | | | | | | | Signed-off-by: Zheng Yan <zheng.z.yan@intel.com> Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Andrew Morton <akpm@linux-foundation.org> Cc: Thomas Gleixner <tglx@linutronix.de> Link: http://lkml.kernel.org/r/1339741902-8449-7-git-send-email-zheng.z.yan@intel.com Signed-off-by: Ingo Molnar <mingo@kernel.org>
* perf/x86: Add generic Intel uncore PMU supportYan, Zheng2012-06-183-1/+1085
| | | | | | | | | | | | | | | | | | | | | This patch adds the generic Intel uncore PMU support, including helper functions that add/delete uncore events, a hrtimer that periodically polls the counters to avoid overflow and code that places all events for a particular socket onto a single cpu. The code design is based on the structure of Sandy Bridge-EP's uncore subsystem, which consists of a variety of components, each component contains one or more "boxes". (Tooling support follows in the next patches.) Signed-off-by: Zheng Yan <zheng.z.yan@intel.com> Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Andrew Morton <akpm@linux-foundation.org> Cc: Thomas Gleixner <tglx@linutronix.de> Link: http://lkml.kernel.org/r/1339741902-8449-6-git-send-email-zheng.z.yan@intel.com Signed-off-by: Ingo Molnar <mingo@kernel.org>
* perf: Introduce perf_pmu_migrate_context()Yan, Zheng2012-06-182-0/+38
| | | | | | | | | | Originally from Peter Zijlstra. The helper migrates perf events from one cpu to another cpu. Signed-off-by: Zheng Yan <zheng.z.yan@intel.com> Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Link: http://lkml.kernel.org/r/1339741902-8449-5-git-send-email-zheng.z.yan@intel.com Signed-off-by: Ingo Molnar <mingo@kernel.org>
* perf: Allow the PMU driver to choose the CPU on which to install eventsYan, Zheng2012-06-181-4/+4
| | | | | | | | | | Allow the pmu->event_init callback to change event->cpu, so the PMU driver can choose the CPU on which to install events. Signed-off-by: Zheng Yan <zheng.z.yan@intel.com> Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Link: http://lkml.kernel.org/r/1339741902-8449-4-git-send-email-zheng.z.yan@intel.com Signed-off-by: Ingo Molnar <mingo@kernel.org>
* perf: Avoid race between cpu hotplug and installing eventYan, Zheng2012-06-181-0/+5
| | | | | | | | | | | perf_event_open() requires the cpu on which to install event is online, but the cpu can go offline after perf_event_open checks that. Add a get_online_cpus()/put_online_cpus() pair to avoid the race. Signed-off-by: Zheng Yan <zheng.z.yan@intel.com> Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Link: http://lkml.kernel.org/r/1339741902-8449-3-git-send-email-zheng.z.yan@intel.com Signed-off-by: Ingo Molnar <mingo@kernel.org>
* perf: Export perf_assign_events()Yan, Zheng2012-06-182-3/+5
| | | | | | | | | | Export perf_assign_events() so the uncore code can use it to schedule events. Signed-off-by: Zheng Yan <zheng.z.yan@intel.com> Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Link: http://lkml.kernel.org/r/1339741902-8449-2-git-send-email-zheng.z.yan@intel.com Signed-off-by: Ingo Molnar <mingo@kernel.org>
* Merge branch 'perf/urgent' into perf/coreIngo Molnar2012-06-189-14/+88
|\ | | | | | | | | | | Merge in all fixes before applying more changes. Signed-off-by: Ingo Molnar <mingo@kernel.org>
| * perf: Use css_tryget() to avoid propping up css refcountSalman Qazi2012-06-181-3/+7
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | An rmdir pushes css's ref count to zero. However, if the associated directory is open at the time, the dentry ref count is non-zero. If the fd for this directory is then passed into perf_event_open, it does a css_get(). This bounces the ref count back up from zero. This is a problem by itself. But what makes it turn into a crash is the fact that we end up doing an extra dput, since we perform a dput when css_put sees the ref count go down to zero. css_tryget() does not fall into that trap. So, we use that instead. Reproduction test-case for the bug: #include <unistd.h> #include <sys/types.h> #include <sys/stat.h> #include <fcntl.h> #include <linux/unistd.h> #include <linux/perf_event.h> #include <string.h> #include <errno.h> #include <stdio.h> #define PERF_FLAG_PID_CGROUP (1U << 2) int perf_event_open(struct perf_event_attr *hw_event_uptr, pid_t pid, int cpu, int group_fd, unsigned long flags) { return syscall(__NR_perf_event_open,hw_event_uptr, pid, cpu, group_fd, flags); } /* * Directly poke at the perf_event bug, since it's proving hard to repro * depending on where in the kernel tree. what moved? */ int main(int argc, char **argv) { int fd; struct perf_event_attr attr; memset(&attr, 0, sizeof(attr)); attr.exclude_kernel = 1; attr.size = sizeof(attr); mkdir("/dev/cgroup/perf_event/blah", 0777); fd = open("/dev/cgroup/perf_event/blah", O_RDONLY); perror("open"); rmdir("/dev/cgroup/perf_event/blah"); sleep(2); perf_event_open(&attr, fd, 0, -1, PERF_FLAG_PID_CGROUP); perror("perf_event_open"); close(fd); return 0; } Signed-off-by: Salman Qazi <sqazi@google.com> Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Acked-by: Tejun Heo <tj@kernel.org> Link: http://lkml.kernel.org/r/20120614223108.1025.2503.stgit@dungbeetle.mtv.corp.google.com Signed-off-by: Ingo Molnar <mingo@kernel.org>
| * Merge tag 'perf-urgent-for-mingo' of ↵Ingo Molnar2012-06-155-8/+61
| |\ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux into perf/urgent Pull perf fixes from Arnaldo Carvalho de Melo: * Set name of tracepoints when reading the perf.data headers, so that we don't end up using the local ones, from /sys. * Fix default output file for perf stat, from Stephane Eranian. * Fix endian handling of features bitmask in perf.data header, from David Ahern. Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> Signed-off-by: Ingo Molnar <mingo@kernel.org>
| | * perf tools: Fix synthesizing tracepoint names from the perf.data headersArnaldo Carvalho de Melo2012-06-121-0/+32
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | We need to use the per event info snapshoted at record time to synthesize the events name, so do it just after reading the perf.data headers, when we already processed the /sys events data, otherwise we'll end up using the local /sys that only by sheer luck will have the same tracepoint ID -> real event association. Example: # uname -a Linux felicio.ghostprotocols.net 3.4.0-rc5+ #1 SMP Sat May 19 15:27:11 BRT 2012 x86_64 x86_64 x86_64 GNU/Linux # perf record -e sched:sched_switch usleep 1 [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.015 MB perf.data (~648 samples) ] # cat /t/events/sched/sched_switch/id 279 # perf evlist -v sched:sched_switch: sample_freq=1, type: 2, config: 279, size: 80, sample_type: 1159, read_format: 7, disabled: 1, inherit: 1, mmap: 1, comm: 1, enable_on_exec: 1, sample_id_all: 1, exclude_guest: 1 # So on the above machine the sched:sched_switch has tracepoint id 279, but on the machine were we'll analyse it it has a different id: $ cat /t/events/sched/sched_switch/id 56 $ perf evlist -i /tmp/perf.data kmem:mm_balancedirty_writeout $ cat /t/events/kmem/mm_balancedirty_writeout/id 279 With this fix: $ perf evlist -i /tmp/perf.data sched:sched_switch Reported-by: Dmitry Antipov <dmitry.antipov@linaro.org> Cc: David Ahern <dsahern@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Mike Galbraith <efault@gmx.de> Cc: Namhyung Kim <namhyung@gmail.com> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Stephane Eranian <eranian@google.com> Link: http://lkml.kernel.org/n/tip-auwks8fpuhmrdpiefs55o5oz@git.kernel.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
| | * perf stat: Fix default output fileStephane Eranian2012-06-111-1/+7
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The following commit: commit 56f3bae70638b33477a6015fd362ccfe354fd3ee Author: Jim Cromie <jim.cromie@gmail.com> Date: Wed Sep 7 17:14:00 2011 -0600 perf stat: Add --log-fd <N> option to redirect stderr elsewhere introduced a bug in the way perf stat outputs the results by default, i.e., without the --log-fd or --output option. It would default to writing to file descriptor 0, i.e., stdin. Writing to stdin is allowed and is equivalent to writing to stdout. However, there is a major difference for any script that was already capturing the output of perf stat via redirection: perf stat >/tmp/log .... or perf stat 2>/tmp/log .... They would not capture anything anymore. They would have to do: perf stat 0>/tmp/log ... This breaks compatibility with existing scripts and does not look very natural. This patch fixes the problem by looking at output_fd only when it was modified by user (> 0). It also checks that the value if positive. Passing --log-fd 0 is ignored. I would also argue that defaulting to stderr for the results is not the right thing to do, though this patch does not address this specific issue. Signed-off-by: Stephane Eranian <eranian@google.com> Cc: David Ahern <dsahern@gmail.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Jim Cromie <jim.cromie@gmail.com> Link: http://lkml.kernel.org/r/20120515111111.GA9870@quad Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
| | * perf tools: Fix endianity swapping for adds_features bitmaskDavid Ahern2012-06-114-7/+22
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Based on Jiri's latest attempt: https://lkml.org/lkml/2012/5/16/61 Basically, adds_features should be byte swapped assuming unsigned longs are either 8-bytes (u64) or 4-bytes (u32). Fixes 32-bit ppc dumping 64-bit x86 feature data: ======== captured on: Sun May 20 19:23:23 2012 hostname : nxos-vdc-dev3 os release : 3.4.0-rc7+ perf version : 3.4.rc4.137.g978da3 arch : x86_64 nrcpus online : 16 nrcpus avail : 16 cpudesc : Intel(R) Xeon(R) CPU E5540 @ 2.53GHz cpuid : GenuineIntel,6,26,5 total memory : 24680324 kB ... Verified 64-bit x86 can still dump feature data for 32-bit ppc. Signed-off-by: David Ahern <dsahern@gmail.com> Reviewed-by: Jiri Olsa <jolsa@redhat.com> Cc: Corey Ashford <cjashfor@linux.vnet.ibm.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Link: http://lkml.kernel.org/r/4FBBB539.5010805@gmail.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
| * | watchdog: Quiet down the boot messagesDon Zickus2012-06-141-1/+18
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | A bunch of bugzillas have complained how noisy the nmi_watchdog is during boot-up especially with its expected failure cases (like virt and bios resource contention). This is my attempt to quiet them down and keep it less confusing for the end user. What I did is print the message for cpu0 and save it for future comparisons. If future cpus have an identical message as cpu0, then don't print the redundant info. However, if a future cpu has a different message, happily print that loudly. Before the change, you would see something like: ..TIMER: vector=0x30 apic1=0 pin1=2 apic2=-1 pin2=-1 CPU0: Intel(R) Core(TM)2 Quad CPU Q9550 @ 2.83GHz stepping 0a Performance Events: PEBS fmt0+, Core2 events, Intel PMU driver. ... version: 2 ... bit width: 40 ... generic registers: 2 ... value mask: 000000ffffffffff ... max period: 000000007fffffff ... fixed-purpose events: 3 ... event mask: 0000000700000003 NMI watchdog enabled, takes one hw-pmu counter. Booting Node 0, Processors #1 NMI watchdog enabled, takes one hw-pmu counter. #2 NMI watchdog enabled, takes one hw-pmu counter. #3 Ok. NMI watchdog enabled, takes one hw-pmu counter. Brought up 4 CPUs Total of 4 processors activated (22607.24 BogoMIPS). After the change, it is simplified to: ..TIMER: vector=0x30 apic1=0 pin1=2 apic2=-1 pin2=-1 CPU0: Intel(R) Core(TM)2 Quad CPU Q9550 @ 2.83GHz stepping 0a Performance Events: PEBS fmt0+, Core2 events, Intel PMU driver. ... version: 2 ... bit width: 40 ... generic registers: 2 ... value mask: 000000ffffffffff ... max period: 000000007fffffff ... fixed-purpose events: 3 ... event mask: 0000000700000003 NMI watchdog: enabled on all CPUs, permanently consumes one hw-PMU counter. Booting Node 0, Processors #1 #2 #3 Ok. Brought up 4 CPUs V2: little changes based on Joe Perches' feedback V3: printk cleanup based on Ingo's feedback; checkpatch fix V4: keep printk as one long line V5: Ingo fix ups Reported-and-tested-by: Nathan Zimmer <nzimmer@sgi.com> Signed-off-by: Don Zickus <dzickus@redhat.com> Cc: nzimmer@sgi.com Cc: joe@perches.com Link: http://lkml.kernel.org/r/1339594548-17227-1-git-send-email-dzickus@redhat.com Signed-off-by: Ingo Molnar <mingo@kernel.org>
| * | perf/x86: Fix broken LBR fixup codeStephane Eranian2012-06-131-1/+1
| |/ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | I noticed that the LBR fixups were not working anymore on programs where they used to. I tracked this down to a recent change to copy_from_user_nmi(): db0dc75d640 ("perf/x86: Check user address explicitly in copy_from_user_nmi()") This commit added a call to __range_not_ok() to the copy_from_user_nmi() routine. The problem is that the logic of the test must be reversed. __range_not_ok() returns 0 if the range is VALID. We want to return early from copy_from_user_nmi() if the range is NOT valid. Signed-off-by: Stephane Eranian <eranian@google.com> Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Acked-by: Arun Sharma <asharma@fb.com> Link: http://lkml.kernel.org/r/20120611134426.GA7542@quad Signed-off-by: Ingo Molnar <mingo@kernel.org>
| * Merge branch 'tip/perf/urgent' of ↵Ingo Molnar2012-06-081-1/+1
| |\ | | | | | | | | | | | | | | | | | | | | | git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace into perf/urgent Pull brown paper bag fix from Steve Rostedt. Signed-off-by: Ingo Molnar <mingo@kernel.org>
| | * tracing: Have tracing_off() actually turn tracing offSteven Rostedt2012-06-061-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | A recent update to have tracing_on/off() only affect the ftrace ring buffers instead of all ring buffers had a cut and paste error. The tracing_off() did the exact same thing as tracing_on() and would not actually turn off tracing. Unfortunately, tracing_off() is more important to be working than tracing_on() as this is a key development tool, as it lets the developer turn off tracing as soon as a problem is discovered. It is also used by panic and oops code. This bug also breaks the 'echo func:traceoff > set_ftrace_filter' Cc: <stable@vger.kernel.org> # 3.4 Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
* | | perf/x86/amd: Fix RDPMC index calculation for AMD family 15hRobert Richter2012-06-181-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The RDPMC index calculation is wrong for AMD family 15h (X86_FEATURE_ PERFCTR_CORE set). This leads to a #GP when accessing the counter: Pid: 2237, comm: syslog-ng Not tainted 3.5.0-rc1-perf-x86_64-standard-g130ff90 #135 AMD Pike/Pike RIP: 0010:[<ffffffff8100dc33>] [<ffffffff8100dc33>] x86_perf_event_update+0x27/0x66 While the msr address offset is (index << 1) we must use index to select the correct rdpmc. Signed-off-by: Robert Richter <robert.richter@amd.com> Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Vince Weaver <vweaver1@eecs.utk.edu> Signed-off-by: Ingo Molnar <mingo@kernel.org>
* | | Merge branch 'tip/perf/core' of ↵Ingo Molnar2012-06-183-7/+6
|\ \ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace into perf/core Pull ftrace robustization fixes from Steve Rostedt. Signed-off-by: Ingo Molnar <mingo@kernel.org>
| * | | tracing: Register the ftrace internal events during early bootSteven Rostedt2012-06-141-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | All trace events including ftrace internel events (like trace_printk and function tracing), register functions that describe how to print their output. The events may be recorded as soon as the ring buffer is allocated, but they are just raw binary in the buffer. The mapping of event ids to how to print them are held within a structure that is registered on system boot. If a crash happens in boot up before these functions are registered then their output (via ftrace_dump_on_oops) will be useless: Dumping ftrace buffer: --------------------------------- <...>-1 0.... 319705us : Unknown type 6 --------------------------------- This can be quite frustrating for a kernel developer trying to see what is going wrong. There's no reason to register them so late in the boot up process. They can be registered by early_initcall(). Reported-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
| * | | tracing: Add comments for the other bits of ftrace_event_call.flagsSteven Rostedt2012-06-141-0/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | TRACE_EVENT_FL_ENABLED_BIT, TRACE_EVENT_FL_FILTERED_BIT, TRACE_EVENT_FL_RECORDED_CMD_BIT, Have comments about what they are, but: TRACE_EVENT_FL_CAP_ANY_BIT, TRACE_EVENT_FL_NO_SET_FILTER_BIT, TRACE_EVENT_FL_IGNORE_ENABLE_BIT, do not, making them second class citizens. To prevent another class warfare, these bits have protested for their right to be commented. And By Golly! I'll give them what they want! Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
| * | | ftrace: Remove a superfluous checkBorislav Petkov2012-06-141-6/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | register_ftrace_function() checks ftrace_disabled and calls __register_ftrace_function which does it again. Drop the first check and add the unlikely hint to the second one. Also, drop the label as John correctly notices. No functional change. Link: http://lkml.kernel.org/r/20120329171140.GE6409@aftab Cc: Borislav Petkov <bp@amd64.org> Cc: John Kacur <jkacur@redhat.com> Signed-off-by: Borislav Petkov <borislav.petkov@amd.com> Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
* | | | uprobes: Remove the unnecessary initialization in add_utask()Oleg Nesterov2012-06-161-1/+0
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Trivial cleanup. No need to nullify ->active_uprobe after kzalloc(). Signed-off-by: Oleg Nesterov <oleg@redhat.com> Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com> Cc: Anton Arapov <anton@redhat.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Srikar Dronamraju <srikar@linux.vnet.ibm.com> Link: http://lkml.kernel.org/r/20120615154401.GA9633@redhat.com Signed-off-by: Ingo Molnar <mingo@kernel.org>
* | | | uprobes: __copy_insn() needs "loff_t offset"Oleg Nesterov2012-06-161-6/+6
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | 1. __copy_insn() needs "loff_t offset", not "unsigned long", to read the file. 2. use pgoff_t for "idx" and remove the unnecessary typecast. 3. fix the typo, "&=" is not what we want 4. can't resist, rename off1 to off. Signed-off-by: Oleg Nesterov <oleg@redhat.com> Acked-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com> Acked-by: Ananth N Mavinakayanahalli <ananth@in.ibm.com> Cc: Anton Arapov <anton@redhat.com> Cc: Peter Zijlstra <peterz@infradead.org> Link: http://lkml.kernel.org/r/20120615154359.GA9625@redhat.com Signed-off-by: Ingo Molnar <mingo@kernel.org>
* | | | uprobes: Don't use loff_t for the valid virtual addressOleg Nesterov2012-06-161-17/+9
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | loff_t looks confusing when it is used for the virtual address. Change map_info and install_breakpoint/remove_breakpoint paths to use "unsigned long". The patch doesn't change vma_address(), it can't return "long" because it is used to verify the mapping. But probably this needs some cleanups too. Signed-off-by: Oleg Nesterov <oleg@redhat.com> Signed-off-by: Anton Arapov <anton@redhat.com> Acked-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com> Acked-by: Ananth N Mavinakayanahalli <ananth@in.ibm.com> Cc: Peter Zijlstra <peterz@infradead.org> Link: http://lkml.kernel.org/r/20120615154355.GA9622@redhat.com Signed-off-by: Ingo Molnar <mingo@kernel.org>
* | | | uprobes: Simplify the usage of uprobe->pending_listOleg Nesterov2012-06-161-8/+4
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | uprobe->pending_list is only used to create the temporary list, it has no meaning after we drop uprobes_mmap_hash(inode). No need to initialize this node or remove it from tmp_list, and we can use list_for_each_entry(). Signed-off-by: Oleg Nesterov <oleg@redhat.com> Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com> Cc: Anton Arapov <anton@redhat.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Srikar Dronamraju <srikar@linux.vnet.ibm.com> Link: http://lkml.kernel.org/r/20120615154353.GA9614@redhat.com Signed-off-by: Ingo Molnar <mingo@kernel.org>
* | | | uprobes: Move BUG_ON(UPROBE_SWBP_INSN_SIZE) from write_opcode() to ↵Oleg Nesterov2012-06-161-6/+5
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | install_breakpoint() write_opcode() ensures that UPROBE_SWBP_INSN doesn't cross the page boundary. This looks a bit confusing, the check does not depend on vaddr and it is enough to do it only once right after install_breakpoint()->arch_uprobe_analyze_insn(). Signed-off-by: Oleg Nesterov <oleg@redhat.com> Acked-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com> Acked-by: Ananth N Mavinakayanahalli <ananth@in.ibm.com> Cc: Anton Arapov <anton@redhat.com> Cc: Peter Zijlstra <peterz@infradead.org> Link: http://lkml.kernel.org/r/20120615154350.GA9611@redhat.com Signed-off-by: Ingo Molnar <mingo@kernel.org>
* | | | uprobes: No need to re-check vma_address() in write_opcode()Oleg Nesterov2012-06-161-5/+0
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | write_opcode() is called by register_for_each_vma() and uprobe_mmap() paths. In both cases the caller has already verified this vaddr under mmap_sem, no need to re-check. Note also that this check is wrong anyway, we should not truncate loff_t returned by vma_address() if we do not trust this mapping. Signed-off-by: Oleg Nesterov <oleg@redhat.com> Acked-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com> Acked-by: Ananth N Mavinakayanahalli <ananth@in.ibm.com> Cc: Anton Arapov <anton@redhat.com> Cc: Peter Zijlstra <peterz@infradead.org> Link: http://lkml.kernel.org/r/20120615154347.GA9604@redhat.com Signed-off-by: Ingo Molnar <mingo@kernel.org>
* | | | uprobes: Copy_insn() should not return -ENOMEM if __copy_insn() failsOleg Nesterov2012-06-161-4/+4
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | copy_insn() returns -ENOMEM if the first __copy_insn() fails, it should return the correct error code. Signed-off-by: Oleg Nesterov <oleg@redhat.com> Acked-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com> Acked-by: Ananth N Mavinakayanahalli <ananth@in.ibm.com> Cc: Anton Arapov <anton@redhat.com> Cc: Peter Zijlstra <peterz@infradead.org> Link: http://lkml.kernel.org/r/20120615154344.GA9601@redhat.com Signed-off-by: Ingo Molnar <mingo@kernel.org>
* | | | uprobes: Copy_insn() shouldn't depend on mm/vma/vaddrOleg Nesterov2012-06-161-9/+6
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | 1. copy_insn() doesn't need "addr", it can use uprobe->offset. Remove this argument. 2. Change copy_insn/__copy_insn to accept "struct file*" instead of vma. copy_insn() is called only once and mm/vma/vaddr are random, it shouldn't depend on them. Signed-off-by: Oleg Nesterov <oleg@redhat.com> Acked-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com> Acked-by: Ananth N Mavinakayanahalli <ananth@in.ibm.com> Cc: Anton Arapov <anton@redhat.com> Cc: Peter Zijlstra <peterz@infradead.org> Link: http://lkml.kernel.org/r/20120615154342.GA9598@redhat.com Signed-off-by: Ingo Molnar <mingo@kernel.org>
* | | | uprobes: Document uprobe_register() vs uprobe_mmap() racePeter Zijlstra2012-06-161-3/+28
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Because the mind is treacherous and makes us forget we need to write stuff down. Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Signed-off-by: Oleg Nesterov <oleg@redhat.com> Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com> Cc: Anton Arapov <anton@redhat.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Srikar Dronamraju <srikar@linux.vnet.ibm.com> Link: http://lkml.kernel.org/r/20120615154339.GA9591@redhat.com Signed-off-by: Ingo Molnar <mingo@kernel.org>
* | | | uprobes: Change build_map_info() to try kmalloc(GFP_NOWAIT) firstOleg Nesterov2012-06-161-0/+10
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | build_map_info() doesn't allocate the memory under i_mmap_mutex to avoid the deadlock with page reclaim. But it can try GFP_NOWAIT first, it should work in the likely case and thus we almost never need the pre-alloc-and-retry path. Signed-off-by: Oleg Nesterov <oleg@redhat.com> Acked-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com> Acked-by: Peter Zijlstra <peterz@infradead.org> Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com> Cc: Anton Arapov <anton@redhat.com> Link: http://lkml.kernel.org/r/20120615154336.GA9588@redhat.com Signed-off-by: Ingo Molnar <mingo@kernel.org>
* | | | uprobes: Rework register_for_each_vma() to make it O(n)Oleg Nesterov2012-06-161-113/+86
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Currently register_for_each_vma() is O(n ** 2) + O(n ** 3), every time find_next_vma_info() "restarts" the vma_prio_tree_foreach() loop and each iteration rechecks the whole try_list. This also means that try_list can grow "indefinitely" if register/unregister races with munmap/mmap activity even if the number of mapping is bounded at any time. With this patch register_for_each_vma() builds the list of mm/vaddr structures only once and does install_breakpoint() for each entry. We do not care about the new mappings which can be created after build_map_info() drops mapping->i_mmap_mutex, uprobe_mmap() should do its work. Note that we do not allocate map_info under i_mmap_mutex, this can deadlock with page reclaim (but see the next patch). So we use 2 lists, "curr" which we are going to return, and "prev" which holds the already allocated memory. The main loop deques the entry from "prev" (initially it is empty), and if "prev" becomes empty again it counts the number of entries we need to pre-allocate outside of i_mmap_mutex. Signed-off-by: Oleg Nesterov <oleg@redhat.com> Acked-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com> Acked-by: Peter Zijlstra <peterz@infradead.org> Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com> Cc: Anton Arapov <anton@redhat.com> Link: http://lkml.kernel.org/r/20120615154333.GA9581@redhat.com Signed-off-by: Ingo Molnar <mingo@kernel.org>
* | | | uprobes: Install_breakpoint() should fail if is_swbp_insn() == TOleg Nesterov2012-06-161-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | install_breakpoint() returns -EEXIST if is_swbp_insn(orig_insn) == T, the caller treats this code as success. This is doubly wrong. The successful return should set UPROBE_COPY_INSN, but the real problem is that it shouldn't succeed. If the probed insn is int3 the application should get SIGTRAP, this won't happen with uprobe. Probably we can fix this, we can add the UPROBE_SHARED_BP flag and teach handle_swbp/set_orig_insn to handle this case correctly. But this needs some complications and we have other insns which can't be probed, lets make a simple fix for now. I think this needs a cleanup. UPROBE_COPY_INSN should die, copy_insn() should be called by alloc_uprobe(). arch_uprobe_analyze_insn() depends on ->mm (ia32_compat) but it is called only once. Signed-off-by: Oleg Nesterov <oleg@redhat.com> Acked-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com> Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com> Cc: Anton Arapov <anton@redhat.com> Cc: Peter Zijlstra <peterz@infradead.org> Link: http://lkml.kernel.org/r/20120615154331.GA9578@redhat.com Signed-off-by: Ingo Molnar <mingo@kernel.org>
* | | | uprobes: Write_opcode()->__replace_page() can race with try_to_unmap()Oleg Nesterov2012-06-161-28/+13
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | write_opcode() gets old_page via get_user_pages() and then calls __replace_page() which assumes that this old_page is still mapped after pte_offset_map_lock(). This is not true if this old_page was already try_to_unmap()'ed, and in this case everything __replace_page() does with old_page is wrong. Just for example, put_page() is not balanced. I think it is possible to teach __replace_page() to handle this unlikely case correctly, but this patch simply changes it to use page_check_address() and return -EAGAIN if it fails. The caller should notice this error code and retry. Note: write_opcode() asks for the cleanups, I'll try to do this in a separate patch. Signed-off-by: Oleg Nesterov <oleg@redhat.com> Acked-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com> Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com> Cc: Anton Arapov <anton@redhat.com> Cc: Peter Zijlstra <peterz@infradead.org> Link: http://lkml.kernel.org/r/20120615154328.GA9571@redhat.com Signed-off-by: Ingo Molnar <mingo@kernel.org>
* | | | uprobes: __copy_insn() should ensure a_ops->readpage != NULLOleg Nesterov2012-06-161-0/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | __copy_insn() blindly calls read_mapping_page(), this will crash the kernel if ->readpage == NULL, add the necessary check. For example, hugetlbfs_aops->readpage is NULL. Perhaps we should change read_mapping_page() instead. Signed-off-by: Oleg Nesterov <oleg@redhat.com> Acked-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com> Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com> Cc: Anton Arapov <anton@redhat.com> Cc: Peter Zijlstra <peterz@infradead.org> Link: http://lkml.kernel.org/r/20120615154325.GA9568@redhat.com Signed-off-by: Ingo Molnar <mingo@kernel.org>
* | | | uprobes: Valid_vma() should reject VM_HUGETLBOleg Nesterov2012-06-161-1/+2
|/ / / | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | __replace_page() obviously can't work with the hugetlbfs mappings, uprobe_register() will likely crash the kernel. Change valid_vma() to check VM_HUGETLB as well. As for PageTransHuge() no need to worry, vma->vm_file != NULL. Signed-off-by: Oleg Nesterov <oleg@redhat.com> Acked-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com> Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com> Cc: Anton Arapov <anton@redhat.com> Cc: Peter Zijlstra <peterz@infradead.org> Link: http://lkml.kernel.org/r/20120615154322.GA9561@redhat.com Signed-off-by: Ingo Molnar <mingo@kernel.org>
* | | perf/x86: Convert obsolete simple_strtoul() usage to kstrtoul()Shuah Khan2012-06-111-1/+6
| | | | | | | | | | | | | | | | | | | | | | | | Signed-off-by: Shuah Khan <shuahkhan@gmail.com> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Frederic Weisbecker <fweisbec@gmail.com> Link: http://lkml.kernel.org/r/1339384421.3025.8.camel@lorien2 Signed-off-by: Ingo Molnar <mingo@kernel.org>
* | | Merge tag 'v3.5-rc2' into perf/coreIngo Molnar2012-06-11111-864/+1632
|\ \ \ | | | | | | | | | | | | | | | | | | | | Merge in Linux 3.5-rc2 - to pick up fixes. Signed-off-by: Ingo Molnar <mingo@kernel.org>
| * | | Linux 3.5-rc2v3.5-rc2Linus Torvalds2012-06-081-1/+1
| | | |
| * | | mm, oom: fix badness score underflowDavid Rientjes2012-06-081-2/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | If the privileges given to root threads (3% of allowable memory) or a negative value of /proc/pid/oom_score_adj happen to exceed the amount of rss of a thread, its badness score overflows as a result of commit a7f638f999ff ("mm, oom: normalize oom scores to oom_score_adj scale only for userspace"). Fix this by making the type signed and return 1, meaning the thread is still eligible for kill, if the value is negative. Reported-by: Dave Jones <davej@redhat.com> Acked-by: Oleg Nesterov <oleg@redhat.com> Signed-off-by: David Rientjes <rientjes@google.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
| * | | Merge branch 'sched-urgent-for-linus' of ↵Linus Torvalds2012-06-086-39/+179
| |\ \ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip Pull scheduler fixes from Ingo Molnar. * 'sched-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: sched: Fix the relax_domain_level boot parameter sched: Validate assumptions in sched_init_numa() sched: Always initialize cpu-power sched: Fix domain iteration sched/rt: Fix lockdep annotation within find_lock_lowest_rq() sched/numa: Load balance between remote nodes sched/x86: Calculate booted cores after construction of sibling_mask
| | * | | sched: Fix the relax_domain_level boot parameterDimitri Sivanich2012-06-061-6/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | It does not get processed because sched_domain_level_max is 0 at the time that setup_relax_domain_level() is run. Simply accept the value as it is, as we don't know the value of sched_domain_level_max until sched domain construction is completed. Fix sched_relax_domain_level in cpuset. The build_sched_domain() routine calls the set_domain_attribute() routine prior to setting the sd->level, however, the set_domain_attribute() routine relies on the sd->level to decide whether idle load balancing will be off/on. Signed-off-by: Dimitri Sivanich <sivanich@sgi.com> Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Link: http://lkml.kernel.org/r/20120605184436.GA15668@sgi.com Signed-off-by: Ingo Molnar <mingo@kernel.org>
| | * | | sched: Validate assumptions in sched_init_numa()Peter Zijlstra2012-06-061-19/+80
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Add some code to validate assumptions we're making and output warnings if they are not. If this trigger we want to know about it. Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Alex Shi <lkml.alex@gmail.com> Link: http://lkml.kernel.org/n/tip-6uc3wk5s9udxtdl9cnku0vtt@git.kernel.org Signed-off-by: Ingo Molnar <mingo@kernel.org>
| | * | | sched: Always initialize cpu-powerPeter Zijlstra2012-06-062-2/+13
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Often when we run into mis-shapen topologies the balance iteration fails to update the cpu power properly and we'll end up in /0 traps. Always initialize the cpu-power to a semi-sane value so that we can at least boot the machine, even if the load-balancer might not function correctly. Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Link: http://lkml.kernel.org/n/tip-3lbhyj25sr169ha7z3qht5na@git.kernel.org Signed-off-by: Ingo Molnar <mingo@kernel.org>
| | * | | sched: Fix domain iterationPeter Zijlstra2012-06-064-10/+72
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Weird topologies can lead to asymmetric domain setups. This needs further consideration since these setups are typically non-minimal too. For now, make it work by adding an extra mask selecting which CPUs are allowed to iterate up. The topology that triggered it is the one from David Rientjes: 10 20 20 30 20 10 20 20 20 20 10 20 30 20 20 10 resulting in boxes that wouldn't even boot. Reported-by: David Rientjes <rientjes@google.com> Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Link: http://lkml.kernel.org/n/tip-3p86l9cuaqnxz7uxsojmz5rm@git.kernel.org Signed-off-by: Ingo Molnar <mingo@kernel.org>
| | * | | sched/rt: Fix lockdep annotation within find_lock_lowest_rq()Peter Zijlstra2012-06-061-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Roland Dreier reported spurious, hard to trigger lockdep warnings within the scheduler - without any real lockup. This bit gives us the right clue: > [89945.640512] [<ffffffff8103fa1a>] double_lock_balance+0x5a/0x90 > [89945.640568] [<ffffffff8104c546>] push_rt_task+0xc6/0x290 if you look at that code you'll find the double_lock_balance() in question is the one in find_lock_lowest_rq() [yay for inlining]. Now find_lock_lowest_rq() has a bug.. it fails to use double_unlock_balance() in one exit path, if this results in a retry in push_rt_task() we'll call double_lock_balance() again, at which point we'll run into said lockdep confusion. Reported-by: Roland Dreier <roland@kernel.org> Acked-by: Steven Rostedt <rostedt@goodmis.org> Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Link: http://lkml.kernel.org/r/1337282386.4281.77.camel@twins Signed-off-by: Ingo Molnar <mingo@kernel.org>
generated by cgit v1.2.3 (git 2.25.1) at 2025-01-06 12:02:59 +0000