linux.git/kernel/events, branch v5.1.16

perf/ring-buffer: Always use {READ,WRITE}_ONCE() for rb->user_page data

2019-06-22T06:09:09+00:00

[ Upstream commit 4d839dd9e4356bbacf3eb0ab13a549b83b008c21 ]

We must use {READ,WRITE}_ONCE() on rb->user_page data such that
concurrent usage will see whole values. A few key sites were missing
this.

Suggested-by: Yabin Cui 
Signed-off-by: Peter Zijlstra (Intel) 
Cc: Alexander Shishkin 
Cc: Arnaldo Carvalho de Melo 
Cc: Jiri Olsa 
Cc: Linus Torvalds 
Cc: Peter Zijlstra 
Cc: Stephane Eranian 
Cc: Thomas Gleixner 
Cc: Vince Weaver 
Cc: acme@kernel.org
Cc: mark.rutland@arm.com
Cc: namhyung@kernel.org
Fixes: 7b732a750477 ("perf_counter: new output ABI - part 1")
Link: http://lkml.kernel.org/r/20190517115418.394192145@infradead.org
Signed-off-by: Ingo Molnar 
Signed-off-by: Sasha Levin

perf/ring_buffer: Add ordering to rb->nest increment

2019-06-22T06:09:09+00:00

[ Upstream commit 3f9fbe9bd86c534eba2faf5d840fd44c6049f50e ]

Similar to how decrementing rb->next too early can cause data_head to
(temporarily) be observed to go backward, so too can this happen when
we increment too late.

This barrier() ensures the rb->head load happens after the increment,
both the one in the 'goto again' path, as the one from
perf_output_get_handle() -- albeit very unlikely to matter for the
latter.

Suggested-by: Yabin Cui 
Signed-off-by: Peter Zijlstra (Intel) 
Cc: Alexander Shishkin 
Cc: Arnaldo Carvalho de Melo 
Cc: Jiri Olsa 
Cc: Linus Torvalds 
Cc: Peter Zijlstra 
Cc: Stephane Eranian 
Cc: Thomas Gleixner 
Cc: Vince Weaver 
Cc: acme@kernel.org
Cc: mark.rutland@arm.com
Cc: namhyung@kernel.org
Fixes: ef60777c9abd ("perf: Optimize the perf_output() path by removing IRQ-disables")
Link: http://lkml.kernel.org/r/20190517115418.309516009@infradead.org
Signed-off-by: Ingo Molnar 
Signed-off-by: Sasha Levin

perf/ring_buffer: Fix exposing a temporarily decreased data_head

2019-06-22T06:09:09+00:00

[ Upstream commit 1b038c6e05ff70a1e66e3e571c2e6106bdb75f53 ]

In perf_output_put_handle(), an IRQ/NMI can happen in below location and
write records to the same ring buffer:

	...
	local_dec_and_test(&rb->nest)
	...                          <-- an IRQ/NMI can happen here
	rb->user_page->data_head = head;
	...

In this case, a value A is written to data_head in the IRQ, then a value
B is written to data_head after the IRQ. And A > B. As a result,
data_head is temporarily decreased from A to B. And a reader may see
data_head < data_tail if it read the buffer frequently enough, which
creates unexpected behaviors.

This can be fixed by moving dec(&rb->nest) to after updating data_head,
which prevents the IRQ/NMI above from updating data_head.

[ Split up by peterz. ]

Signed-off-by: Yabin Cui 
Signed-off-by: Peter Zijlstra (Intel) 
Cc: Alexander Shishkin 
Cc: Arnaldo Carvalho de Melo 
Cc: Arnaldo Carvalho de Melo 
Cc: Jiri Olsa 
Cc: Linus Torvalds 
Cc: Namhyung Kim 
Cc: Peter Zijlstra 
Cc: Stephane Eranian 
Cc: Thomas Gleixner 
Cc: Vince Weaver 
Cc: mark.rutland@arm.com
Fixes: ef60777c9abd ("perf: Optimize the perf_output() path by removing IRQ-disables")
Link: http://lkml.kernel.org/r/20190517115418.224478157@infradead.org
Signed-off-by: Ingo Molnar 
Signed-off-by: Sasha Levin

perf/ring_buffer: Fix AUX software double buffering

2019-05-03T10:46:10+00:00

This recent commit:

  5768402fd9c6e87 ("perf/ring_buffer: Use high order allocations for AUX buffers optimistically")

overlooked the fact that the previous one page granularity of the AUX buffer
provided an implicit double buffering capability to the PMU driver, which
went away when the entire buffer became one high-order page.

Always make the full-trace mode AUX allocation at least two-part to preserve
the previous behavior and allow the implicit double buffering to continue.

Reported-by: Ammy Yi 
Signed-off-by: Alexander Shishkin 
Acked-by: Peter Zijlstra 
Cc: Arnaldo Carvalho de Melo 
Cc: Jiri Olsa 
Cc: Linus Torvalds 
Cc: Stephane Eranian 
Cc: Thomas Gleixner 
Cc: Vince Weaver 
Cc: adrian.hunter@intel.com
Fixes: 5768402fd9c6e87 ("perf/ring_buffer: Use high order allocations for AUX buffers optimistically")
Link: http://lkml.kernel.org/r/20190503085536.24119-2-alexander.shishkin@linux.intel.com
Signed-off-by: Ingo Molnar

perf/ring_buffer: Fix AUX record suppression

2019-04-16T10:13:57+00:00

The following commit:

  1627314fb54a33e ("perf: Suppress AUX/OVERWRITE records")

has an unintended side-effect of also suppressing all AUX records with no flags
and non-zero size, so all the regular records in the full trace mode.
This breaks some use cases for people.

Fix this by restoring "regular" AUX records.

Reported-by: Ben Gainey 
Tested-by: Ben Gainey 
Signed-off-by: Alexander Shishkin 
Signed-off-by: Peter Zijlstra (Intel) 
Cc: 
Cc: Arnaldo Carvalho de Melo 
Cc: Jiri Olsa 
Cc: Linus Torvalds 
Cc: Peter Zijlstra 
Cc: Stephane Eranian 
Cc: Thomas Gleixner 
Cc: Vince Weaver 
Fixes: 1627314fb54a33e ("perf: Suppress AUX/OVERWRITE records")
Link: https://lkml.kernel.org/r/20190329091338.29999-1-alexander.shishkin@linux.intel.com
Signed-off-by: Ingo Molnar

perf/core: Fix the address filtering fix

2019-04-16T10:13:57+00:00

The following recent commit:

  c60f83b813e5 ("perf, pt, coresight: Fix address filters for vmas with non-zero offset")

changes the address filtering logic to communicate filter ranges to the PMU driver
via a single address range object, instead of having the driver do the final bit of
math.

That change forgets to take into account kernel filters, which are not calculated
the same way as DSO based filters.

Fix that by passing the kernel filters the same way as file-based filters.
This doesn't require any additional changes in the drivers.

Reported-by: Adrian Hunter 
Signed-off-by: Alexander Shishkin 
Signed-off-by: Peter Zijlstra (Intel) 
Cc: Arnaldo Carvalho de Melo 
Cc: Jiri Olsa 
Cc: Linus Torvalds 
Cc: Peter Zijlstra 
Cc: Stephane Eranian 
Cc: Thomas Gleixner 
Cc: Vince Weaver 
Fixes: c60f83b813e5 ("perf, pt, coresight: Fix address filters for vmas with non-zero offset")
Link: https://lkml.kernel.org/r/20190329091212.29870-1-alexander.shishkin@linux.intel.com
Signed-off-by: Ingo Molnar

perf/core: Fix perf_event_disable_inatomic() race

2019-04-12T06:55:55+00:00

Thomas-Mich Richter reported he triggered a WARN()ing from event_function_local()
on his s390. The problem boils down to:

	CPU-A				CPU-B

	perf_event_overflow()
	  perf_event_disable_inatomic()
	    @pending_disable = 1
	    irq_work_queue();

	sched-out
	  event_sched_out()
	    @pending_disable = 0

					sched-in
					perf_event_overflow()
					  perf_event_disable_inatomic()
					    @pending_disable = 1;
					    irq_work_queue(); // FAILS

	irq_work_run()
	  perf_pending_event()
	    if (@pending_disable)
	      perf_event_disable_local(); // WHOOPS

The problem exists in generic, but s390 is particularly sensitive
because it doesn't implement arch_irq_work_raise(), nor does it call
irq_work_run() from it's PMU interrupt handler (nor would that be
sufficient in this case, because s390 also generates
perf_event_overflow() from pmu::stop). Add to that the fact that s390
is a virtual architecture and (virtual) CPU-A can stall long enough
for the above race to happen, even if it would self-IPI.

Adding a irq_work_sync() to event_sched_in() would work for all hardare
PMUs that properly use irq_work_run() but fails for software PMUs.

Instead encode the CPU number in @pending_disable, such that we can
tell which CPU requested the disable. This then allows us to detect
the above scenario and even redirect the IPI to make up for the failed
queue.

Reported-by: Thomas-Mich Richter 
Tested-by: Thomas Richter 
Signed-off-by: Peter Zijlstra (Intel) 
Acked-by: Mark Rutland 
Cc: Alexander Shishkin 
Cc: Arnaldo Carvalho de Melo 
Cc: Heiko Carstens 
Cc: Hendrik Brueckner 
Cc: Jiri Olsa 
Cc: Kees Cook 
Cc: Linus Torvalds 
Cc: Martin Schwidefsky 
Cc: Peter Zijlstra 
Cc: Thomas Gleixner 
Signed-off-by: Ingo Molnar

Merge tag 'perf-core-for-mingo-5.1-20190311' of git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux into perf/urgent

2019-03-22T21:50:41+00:00

Pull perf/core improvements and fixes from Arnaldo:

kernel:

  Stephane Eranian :

  - Restore mmap record type correctly when handling PERF_RECORD_MMAP2
    events, as the same template is used for all the threads interested
    in mmap events, some may want just PERF_RECORD_MMAP, while some
    may want the extra info in MMAP2 records.

perf probe:

  Adrian Hunter:

  - Fix getting the kernel map, because since changes related to x86 PTI
    entry trampolines handling, there are more than one kernel map.

perf script:

  Andi Kleen:

  - Support insn output for normal samples, i.e.:

    perf script -F ip,sym,insn --xed

    Will fetch the sample IP from the thread address space and feed it
    to Intel's XED disassembler, producing lines such as:

      ffffffffa4068804 native_write_msr            wrmsr
      ffffffffa415b95e __hrtimer_next_event_base   movq  0x18(%rax), %rdx

    That match 'perf annotate's output.

  - Make the --cpu filter apply to  PERF_RECORD_COMM/FORK/... events, in
    addition to PERF_RECORD_SAMPLE.

perf report:

  - Add a new --samples option to save a small random number of samples
    per hist entry, using a reservoir technique to select a representative
    number of samples.

    Then allow browsing the samples using 'perf script' as part of the hist
    entry context menu. This automatically adds the right filters, so only
    the thread or CPU of the sample is displayed. Then we use less' search
    functionality to directly jump to the time stamp of the selected sample.

    It uses different menus for assembler and source display.  Assembler
    needs xed installed and source needs debuginfo.

  - Fix the UI browser scripts pop up menu when there are many scripts
    available.

perf report:

  Andi Kleen:

  - Add 'time' sort option. E.g.:

    % perf report --sort time,overhead,symbol --time-quantum 1ms --stdio
    ...
         0.67%  277061.87300  [.] _dl_start
         0.50%  277061.87300  [.] f1
         0.50%  277061.87300  [.] f2
         0.33%  277061.87300  [.] main
         0.29%  277061.87300  [.] _dl_lookup_symbol_x
         0.29%  277061.87300  [.] dl_main
         0.29%  277061.87300  [.] do_lookup_x
         0.17%  277061.87300  [.] _dl_debug_initialize
         0.17%  277061.87300  [.] _dl_init_paths
         0.08%  277061.87300  [.] check_match
         0.04%  277061.87300  [.] _dl_count_modids
         1.33%  277061.87400  [.] f1
         1.33%  277061.87400  [.] f2
         1.33%  277061.87400  [.] main
         1.17%  277061.87500  [.] main
         1.08%  277061.87500  [.] f1
         1.08%  277061.87500  [.] f2
         1.00%  277061.87600  [.] main
         0.83%  277061.87600  [.] f1
         0.83%  277061.87600  [.] f2
         1.00%  277061.87700  [.] main

tools headers:

  Arnaldo Carvalho de Melo:

  - Update x86's syscall_64.tbl, no change in tools/perf behaviour.

  -  Sync copies asm-generic/unistd.h and linux/in with the kernel sources.

perf data:

  Jiri Olsa:

  - Prep work to support having perf.data stored as a directory, with one
    file per CPU, that ultimately will allow having one ring buffer reading
    thread per CPU.

Vendor events:

  Martin Liška:

  - perf PMU events for AMD Family 17h.

perf script python:

  Tony Jones:

  - Add python3 support for the remaining Intel PT related scripts, with
    these we should have a clean build of perf with python3 while still
    supporting the build with python2.

libbpf:

  Arnaldo Carvalho de Melo:

  - Fix the build on uCLibc, adding the missing stdarg.h since we use
    va_list in one typedef.

Signed-off-by: Arnaldo Carvalho de Melo

Merge tag 'trace-v5.1' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace

2019-03-12T00:01:32+00:00

Pull tracing updates from Steven Rostedt:
 "The biggest change for this release is in the histogram code:

   - Add "onchange(var)" histogram handler that executes a action when
     $var changes.

   - Add new "snapshot()" action for histogram handlers, that causes a
     snapshot of the ring buffer when triggered. ie.
     onchange(var).snapshot() will trigger a snapshot if var changes.

   - Add alternative for "trace()" action. Currently, to trigger a
     synthetic event, the name of that event is used as the handler
     name, which is inconsistent with the other actions.
     onchange(var).synthetic(param) where it can now be
     onchange(var).trace(synthetic, param). The older method will still
     be allowed, as long as the synthetic events do not overlap with
     other handler names.

   - The histogram documentation at testcases were updated for the new
     changes.

  Outside of the histogram code, we have:

   - Added a quicker way to enable set_ftrace_filter files, that will
     make it much quicker to bisect tracing a function that shouldn't be
     traced and crashes the kernel. (You can echo in numbers to
     set_ftrace_filter, and it will select the corresponding function
     that is in available_filter_functions).

   - Some better displaying of the tracing data (and more information
     was added).

  The rest are small fixes and more clean ups to the code"

* tag 'trace-v5.1' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace: (37 commits)
  tracing: Use strncpy instead of memcpy when copying comm in trace.c
  tracing: Use strncpy instead of memcpy when copying comm for hist triggers
  tracing: Use strncpy instead of memcpy for string keys in hist triggers
  tracing: Use str_has_prefix() in synth_event_create()
  x86/ftrace: Fix warning and considate ftrace_jmp_replace() and ftrace_call_replace()
  tracing/perf: Use strndup_user() instead of buggy open-coded version
  doc: trace: Fix documentation for uprobe_profile
  tracing: Fix spelling mistake: "analagous" -> "analogous"
  tracing: Comment why cond_snapshot is checked outside of max_lock protection
  tracing: Add hist trigger action 'expected fail' test case
  tracing: Add alternative synthetic event trace action test case
  tracing: Add hist trigger onchange() handler test case
  tracing: Add hist trigger snapshot() action test case
  tracing: Add SPDX license GPL-2.0 license identifier to inter-event testcases
  tracing: Add alternative synthetic event trace action syntax
  tracing: Add hist trigger onchange() handler Documentation
  tracing: Add hist trigger onchange() handler
  tracing: Add hist trigger snapshot() action Documentation
  tracing: Add hist trigger snapshot() action
  tracing: Add conditional snapshot
  ...

perf/core: Restore mmap record type correctly

2019-03-11T14:56:02+00:00

On mmap(), perf_events generates a RECORD_MMAP record and then checks
which events are interested in this record. There are currently 2
versions of mmap records: RECORD_MMAP and RECORD_MMAP2. MMAP2 is larger.
The event configuration controls which version the user level tool
accepts.

If the event->attr.mmap2=1 field then MMAP2 record is returned.  The
perf_event_mmap_output() takes care of this. It checks attr->mmap2 and
corrects the record fields before putting it in the sampling buffer of
the event.  At the end the function restores the modified MMAP record
fields.

The problem is that the function restores the size but not the type.
Thus, if a subsequent event only accepts MMAP type, then it would
instead receive an MMAP2 record with a size of MMAP record.

This patch fixes the problem by restoring the record type on exit.

Signed-off-by: Stephane Eranian 
Acked-by: Peter Zijlstra (Intel) 
Cc: Andi Kleen 
Cc: Jiri Olsa 
Cc: Kan Liang 
Fixes: 13d7a2410fa6 ("perf: Add attr->mmap2 attribute to an event")
Link: http://lkml.kernel.org/r/20190307185233.225521-1-eranian@google.com
Signed-off-by: Arnaldo Carvalho de Melo