linux.git/tools/perf, branch v5.5-rc2

Merge git://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf

2019-12-02T18:50:29+00:00

Daniel Borkmann says:

====================
pull-request: bpf 2019-12-02

The following pull-request contains BPF updates for your *net* tree.

We've added 10 non-merge commits during the last 6 day(s) which contain
a total of 10 files changed, 60 insertions(+), 51 deletions(-).

The main changes are:

1) Fix vmlinux BTF generation for binutils pre v2.25, from Stanislav Fomichev.

2) Fix libbpf global variable relocation to take symbol's st_value offset
   into account, from Andrii Nakryiko.

3) Fix libbpf build on powerpc where check_abi target fails due to different
   readelf output format, from Aurelien Jarno.

4) Don't set BPF insns RO for the case when they are JITed in order to avoid
   fragmenting the direct map, from Daniel Borkmann.

5) Fix static checker warning in btf_distill_func_proto() as well as a build
   error due to empty enum when BPF is compiled out, from Alexei Starovoitov.

6) Fix up generation of bpf_helper_defs.h for perf, from Arnaldo Carvalho de Melo.
====================

Signed-off-by: David S. Miller

perf script: Fix invalid LBR/binary mismatch error

2019-11-28T11:08:38+00:00

The 'len' returned by grab_bb() includes an extra MAXINSN bytes to allow
for the last instruction, so the the final 'offs' will not be 'len'.
Fix the error condition logic accordingly.

Before:

  $ perf record -e '{intel_pt//,cpu/mem_inst_retired.all_loads,aux-sample-size=8192/pp}:u' grep -rqs jhgjhg /boot
  [ perf record: Woken up 19 times to write data ]
  [ perf record: Captured and wrote 2.274 MB perf.data ]
  $ perf script -F +brstackinsn --xed --itrace=i1usl100 | head
            grep 13759 [002]  8091.310257:       1862                                        instructions:uH:      5641d58069eb bmexec+0x86b (/bin/grep)
        bmexec+2485:
        00005641d5806b35                        jnz 0x5641d5806bd0              # MISPRED
        00005641d5806bd0                        movzxb  (%r13,%rdx,1), %eax
        00005641d5806bd6                        add %rdi, %rax
        00005641d5806bd9                        movzxb  -0x1(%rax), %edx
        00005641d5806bdd                        cmp %rax, %r14
        00005641d5806be0                        jnb 0x5641d58069c0              # MISPRED
        mismatch of LBR data and executable
        00005641d58069c0                        movzxb  (%r13,%rdx,1), %edi

After:

  $ perf script -F +brstackinsn --xed --itrace=i1usl100 | head
            grep 13759 [002]  8091.310257:       1862                                        instructions:uH:      5641d58069eb bmexec+0x86b (/bin/grep)
        bmexec+2485:
        00005641d5806b35                        jnz 0x5641d5806bd0              # MISPRED
        00005641d5806bd0                        movzxb  (%r13,%rdx,1), %eax
        00005641d5806bd6                        add %rdi, %rax
        00005641d5806bd9                        movzxb  -0x1(%rax), %edx
        00005641d5806bdd                        cmp %rax, %r14
        00005641d5806be0                        jnb 0x5641d58069c0              # MISPRED
        00005641d58069c0                        movzxb  (%r13,%rdx,1), %edi
        00005641d58069c6                        add %rax, %rdi

Fixes: e98df280bc2a ("perf script brstackinsn: Fix recovery from LBR/binary mismatch")
Reported-by: Andi Kleen 
Signed-off-by: Adrian Hunter 
Cc: Jiri Olsa 
Link: http://lore.kernel.org/lkml/20191127095631.15663-1-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo

perf script: Fix brstackinsn for AUXTRACE

2019-11-28T11:08:38+00:00

brstackinsn must be allowed to be set by the user when AUX area data has
been captured because, in that case, the branch stack might be
synthesized on the fly. This fixes the following error:

Before:

  $ perf record -e '{intel_pt//,cpu/mem_inst_retired.all_loads,aux-sample-size=8192/pp}:u' grep -rqs jhgjhg /boot
  [ perf record: Woken up 19 times to write data ]
  [ perf record: Captured and wrote 2.274 MB perf.data ]
  $ perf script -F +brstackinsn --xed --itrace=i1usl100 | head
  Display of branch stack assembler requested, but non all-branch filter set
  Hint: run 'perf record -b ...'

After:

  $ perf record -e '{intel_pt//,cpu/mem_inst_retired.all_loads,aux-sample-size=8192/pp}:u' grep -rqs jhgjhg /boot
  [ perf record: Woken up 19 times to write data ]
  [ perf record: Captured and wrote 2.274 MB perf.data ]
  $ perf script -F +brstackinsn --xed --itrace=i1usl100 | head
            grep 13759 [002]  8091.310257:       1862                                        instructions:uH:      5641d58069eb bmexec+0x86b (/bin/grep)
        bmexec+2485:
        00005641d5806b35                        jnz 0x5641d5806bd0              # MISPRED
        00005641d5806bd0                        movzxb  (%r13,%rdx,1), %eax
        00005641d5806bd6                        add %rdi, %rax
        00005641d5806bd9                        movzxb  -0x1(%rax), %edx
        00005641d5806bdd                        cmp %rax, %r14
        00005641d5806be0                        jnb 0x5641d58069c0              # MISPRED
        mismatch of LBR data and executable
        00005641d58069c0                        movzxb  (%r13,%rdx,1), %edi

Fixes: 48d02a1d5c13 ("perf script: Add 'brstackinsn' for branch stacks")
Reported-by: Andi Kleen 
Signed-off-by: Adrian Hunter 
Cc: Jiri Olsa 
Link: http://lore.kernel.org/lkml/20191127095322.15417-1-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo

perf affinity: Add infrastructure to save/restore affinity

2019-11-28T11:08:38+00:00

The kernel perf subsystem has to IPI to the target CPU for many
operations. On systems with many CPUs and when managing many events the
overhead can be dominated by lots of IPIs.

An alternative is to set up CPU affinity in the perf tool, then set up
all the events for that CPU, and then move on to the next CPU.

Add some affinity management infrastructure to enable such a model.
Used in followon patches.

Committer notes:

Use zfree() in some places, add missing stdbool.h header, some minor
coding style changes.

Signed-off-by: Andi Kleen 
Acked-by: Jiri Olsa 
Link: http://lore.kernel.org/lkml/20191121001522.180827-3-andi@firstfloor.org
Signed-off-by: Arnaldo Carvalho de Melo

perf pmu: Use file system cache to optimize sysfs access

2019-11-28T11:08:38+00:00

pmu.c does a lot of redundant /sys accesses while parsing aliases
and probing for PMUs. On large systems with a lot of PMUs this
can get expensive (>2s):

  % time     seconds  usecs/call     calls    errors syscall
  ------ ----------- ----------- --------- --------- ----------------
   27.25    1.227847           8    160888     16976 openat
   26.42    1.190481           7    164224    164077 stat

Add a cache to remember if specific file names exist or don't
exist, which eliminates most of this overhead.

Also optimize some stat() calls to be slightly cheaper access()

Resulting in:

    0.18    0.004166           2      1851       305 open
    0.08    0.001970           2       829       622 access

Signed-off-by: Andi Kleen 
Acked-by: Jiri Olsa 
Link: http://lore.kernel.org/lkml/20191121001522.180827-2-andi@firstfloor.org
Signed-off-by: Arnaldo Carvalho de Melo

perf regs: Make perf_reg_name() return "unknown" instead of NULL

2019-11-28T11:08:38+00:00

To avoid breaking the build on arches where this is not wired up, at
least all the other features should be made available and when using
this specific routine, the "unknown" should point the user/developer to
the need to wire this up on this particular hardware architecture.

Detected in a container mipsel debian cross build environment, where it
shows up as:

  In file included from /usr/mipsel-linux-gnu/include/stdio.h:867,
                   from /git/linux/tools/perf/lib/include/perf/cpumap.h:6,
                   from util/session.c:13:
  In function 'printf',
      inlined from 'regs_dump__printf' at util/session.c:1103:3,
      inlined from 'regs__printf' at util/session.c:1131:2:
  /usr/mipsel-linux-gnu/include/bits/stdio2.h:107:10: error: '%-5s' directive argument is null [-Werror=format-overflow=]
    107 |   return __printf_chk (__USE_FORTIFY_LEVEL - 1, __fmt, __va_arg_pack ());
        |          ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

cross compiler details:

  mipsel-linux-gnu-gcc (Debian 9.2.1-8) 9.2.1 20190909

Also on mips64:

  In file included from /usr/mips64-linux-gnuabi64/include/stdio.h:867,
                   from /git/linux/tools/perf/lib/include/perf/cpumap.h:6,
                   from util/session.c:13:
  In function 'printf',
      inlined from 'regs_dump__printf' at util/session.c:1103:3,
      inlined from 'regs__printf' at util/session.c:1131:2,
      inlined from 'regs_user__printf' at util/session.c:1139:3,
      inlined from 'dump_sample' at util/session.c:1246:3,
      inlined from 'machines__deliver_event' at util/session.c:1421:3:
  /usr/mips64-linux-gnuabi64/include/bits/stdio2.h:107:10: error: '%-5s' directive argument is null [-Werror=format-overflow=]
    107 |   return __printf_chk (__USE_FORTIFY_LEVEL - 1, __fmt, __va_arg_pack ());
        |          ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
  In function 'printf',
      inlined from 'regs_dump__printf' at util/session.c:1103:3,
      inlined from 'regs__printf' at util/session.c:1131:2,
      inlined from 'regs_intr__printf' at util/session.c:1147:3,
      inlined from 'dump_sample' at util/session.c:1249:3,
      inlined from 'machines__deliver_event' at util/session.c:1421:3:
  /usr/mips64-linux-gnuabi64/include/bits/stdio2.h:107:10: error: '%-5s' directive argument is null [-Werror=format-overflow=]
    107 |   return __printf_chk (__USE_FORTIFY_LEVEL - 1, __fmt, __va_arg_pack ());
        |          ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

cross compiler details:

  mips64-linux-gnuabi64-gcc (Debian 9.2.1-8) 9.2.1 20190909

Fixes: 2bcd355b71da ("perf tools: Add interface to arch registers sets")
Cc: Adrian Hunter 
Cc: Jiri Olsa 
Cc: Namhyung Kim 
Link: https://lkml.kernel.org/n/tip-95wjyv4o65nuaeweq31t7l1s@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo

perf diff: Use llabs() with 64-bit values

2019-11-28T11:08:38+00:00

To fix this build error on a debian mipsel cross build environment:

  builtin-diff.c: In function 'compute_cycles_diff':
  builtin-diff.c:649:10: error: absolute value function 'labs' given an argument of type 's64' {aka 'long long int'} but has parameter of type 'long int' which may cause truncation of value [-Werror=absolute-value]
    649 |    val = labs(pair->block_info->cycles_spark[i] -
        |          ^~~~

Fixes: cebf7d51a6c3 ("perf diff: Report noisy for cycles diff")
Cc: Jin Yao 
Cc: Adrian Hunter 
Cc: Jiri Olsa 
Cc: Namhyung Kim 
Link: https://lkml.kernel.org/n/tip-pn7szy5uw384ntjgk6zckh6a@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo

perf diff: Use llabs() with 64-bit values

2019-11-28T11:08:37+00:00

To fix these build errors on a debian mipsel cross build environment:

  builtin-diff.c: In function 'block_cycles_diff_cmp':
  builtin-diff.c:550:6: error: absolute value function 'labs' given an argument of type 's64' {aka 'long long int'} but has parameter of type 'long int' which may cause truncation of value [-Werror=absolute-value]
    550 |  l = labs(left->diff.cycles);
        |      ^~~~
  builtin-diff.c:551:6: error: absolute value function 'labs' given an argument of type 's64' {aka 'long long int'} but has parameter of type 'long int' which may cause truncation of value [-Werror=absolute-value]
    551 |  r = labs(right->diff.cycles);
        |      ^~~~

Fixes: 99150a1faab2 ("perf diff: Use hists to manage basic blocks per symbol")
Cc: Jin Yao 
Cc: Adrian Hunter 
Cc: Jiri Olsa 
Cc: Namhyung Kim 
Link: https://lkml.kernel.org/n/tip-pn7szy5uw384ntjgk6zckh6a@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo

libbpf: Fix up generation of bpf_helper_defs.h

2019-11-28T00:52:31+00:00

  $ make -C tools/perf build-test

does, ends up with these two problems:

  make[3]: *** No rule to make target '/tmp/tmp.zq13cHILGB/perf-5.3.0/include/uapi/linux/bpf.h', needed by 'bpf_helper_defs.h'.  Stop.
  make[3]: *** Waiting for unfinished jobs....
  make[2]: *** [Makefile.perf:757: /tmp/tmp.zq13cHILGB/perf-5.3.0/tools/lib/bpf/libbpf.a] Error 2
  make[2]: *** Waiting for unfinished jobs....

Because $(srcdir) points to the /tmp/tmp.zq13cHILGB/perf-5.3.0 directory
and we need '/tools/ after that variable, and after fixing this then we
get to another problem:

  /bin/sh: /home/acme/git/perf/tools/scripts/bpf_helpers_doc.py: No such file or directory
  make[3]: *** [Makefile:184: bpf_helper_defs.h] Error 127
  make[3]: *** Deleting file 'bpf_helper_defs.h'
    LD       /tmp/build/perf/libapi-in.o
  make[2]: *** [Makefile.perf:778: /tmp/build/perf/libbpf.a] Error 2
  make[2]: *** Waiting for unfinished jobs....

Because this requires something outside the tools/ directories that gets
collected into perf's detached tarballs, to fix it just add it to
tools/perf/MANIFEST, which this patch does, now it works for that case
and also for all these other cases.

Fixes: e01a75c15969 ("libbpf: Move bpf_{helpers, helper_defs, endian, tracing}.h into libbpf")
Signed-off-by: Arnaldo Carvalho de Melo 
Signed-off-by: Alexei Starovoitov 
Cc: Adrian Hunter 
Cc: Alexei Starovoitov 
Cc: Andrii Nakryiko 
Cc: Daniel Borkmann 
Cc: Jiri Olsa 
Cc: Martin KaFai Lau 
Cc: Namhyung Kim 
Link: https://lkml.kernel.org/n/tip-4pnkg2vmdvq5u6eivc887wen@git.kernel.org
Link: https://lore.kernel.org/bpf/20191126151045.GB19483@kernel.org

perf tools: Allow to link with libbpf dynamicaly

2019-11-26T14:17:45+00:00

Currently we support only static linking with kernel's libbpf
(tools/lib/bpf). This patch adds libbpf package detection and support to
link perf with it dynamically.

The libbpf package status is displayed with:

  $ make VF=1
  Auto-detecting system features:
  ...
  ...                        libbpf: [ on  ]

It's not checked by default, because it's quite new.  Once it's on most
distros we can switch it on.

For the same reason it's not added to the test-all check.

Perf does not need advanced version of libbpf, so we can check just for
the base bpf_object__open function.

Adding new compile variable to detect libbpf package and link bpf
dynamically:

  $ make LIBBPF_DYNAMIC=1
    ...
    LINK     perf
  $ ldd perf | grep bpf
    libbpf.so.0 => /lib64/libbpf.so.0 (0x00007f46818bc000)

If libbpf is not installed, build stops with:

  Makefile.config:486: *** Error: No libbpf devel library found,\
  please install libbpf-devel.  Stop.

Committer testing:

  $ make LIBBPF_DYNAMIC=1 -C tools/perf O=/tmp/build/perf
  make: Entering directory '/home/acme/git/perf/tools/perf'
    BUILD:   Doing 'make -j8' parallel build
  Makefile.config:493: *** Error: No libbpf devel library found, please install libbpf-devel.  Stop.
  make[1]: *** [Makefile.perf:225: sub-make] Error 2
  make: *** [Makefile:70: all] Error 2
  make: Leaving directory '/home/acme/git/perf/tools/perf'
  $

Signed-off-by: Jiri Olsa 
Tested-by: Arnaldo Carvalho de Melo 
Cc: Toke Høiland-Jørgensen 
Cc: Alexander Shishkin 
Cc: Alexei Starovoitov 
Cc: Andrii Nakryiko 
Cc: Daniel Borkmann 
Cc: Jesper Dangaard Brouer 
Cc: Martin KaFai Lau 
Cc: Michael Petlan 
Cc: Namhyung Kim 
Cc: Peter Zijlstra 
Cc: Song Liu 
Cc: Yonghong Song 
Cc: bpf@vger.kernel.org
Cc: netdev@vger.kernel.org
Link: http://lore.kernel.org/lkml/20191126121253.28253-1-jolsa@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo