linux-stable.git/tools/perf, branch v5.4.8

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

2020-01-04T18:18:27+00:00

[ Upstream commit 5b596e0ff0e1852197d4c82d3314db5e43126bf7 ]

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 
Signed-off-by: Sasha Levin

perf script: Fix brstackinsn for AUXTRACE

2020-01-04T18:18:27+00:00

[ Upstream commit 0cd032d3b5fcebf5454315400ab310746a81ca53 ]

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 
Signed-off-by: Sasha Levin

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

2020-01-04T18:18:26+00:00

[ Upstream commit 98e93245113d0f5c279ef77f4a9e7d097323ad71 ]

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 
Signed-off-by: Sasha Levin

perf probe: Fix to show function entry line as probe-able

2019-12-31T15:45:41+00:00

commit 91e2f539eeda26ab00bd03fae8dc434c128c85ed upstream.

Fix die_walk_lines() to list the function entry line correctly.  Since
the dwarf_entrypc() does not return the entry pc if the DIE has only
range attribute, __die_walk_funclines() fails to list the declaration
line (entry line) in that case.

To solve this issue, this introduces die_entrypc() which correctly
returns the entry PC (the first address range) even if the DIE has only
range attribute. With this fix die_walk_lines() shows the function entry
line is able to probe correctly.

Fixes: 4cc9cec636e7 ("perf probe: Introduce lines walker interface")
Signed-off-by: Masami Hiramatsu 
Cc: Jiri Olsa 
Cc: Namhyung Kim 
Link: http://lore.kernel.org/lkml/157190837419.1859.4619125803596816752.stgit@devnote2
Signed-off-by: Arnaldo Carvalho de Melo 
Cc: Thomas Backlund 
Signed-off-by: Greg Kroah-Hartman

perf session: Fix decompression of PERF_RECORD_COMPRESSED records

2019-12-31T15:45:40+00:00

[ Upstream commit bb1835a3b86c73aa534ef6430ad40223728dfbc0 ]

Avoid termination of trace loading in case the last record in the
decompressed buffer partly resides in the following mmaped
PERF_RECORD_COMPRESSED record.

In this case NULL value returned by fetch_mmaped_event() means to
proceed to the next mmaped record then decompress it and load compressed
events.

The issue can be reproduced like this:

  $ perf record -z -- some_long_running_workload
  $ perf report --stdio -vv
  decomp (B): 44519 to 163000
  decomp (B): 48119 to 174800
  decomp (B): 65527 to 131072
  fetch_mmaped_event: head=0x1ffe0 event->header_size=0x28, mmap_size=0x20000: fuzzed perf.data?
  Error:
  failed to process sample
  ...

Testing:

  71: Zstd perf.data compression/decompression              : Ok

  $ tools/perf/perf report -vv --stdio
  decomp (B): 59593 to 262160
  decomp (B): 4438 to 16512
  decomp (B): 285 to 880
  Looking at the vmlinux_path (8 entries long)
  Using vmlinux for symbols
  decomp (B): 57474 to 261248
  prefetch_event: head=0x3fc78 event->header_size=0x28, mmap_size=0x3fc80: fuzzed or compressed perf.data?
  decomp (B): 25 to 32
  decomp (B): 52 to 120
  ...

Fixes: 57fc032ad643 ("perf session: Avoid infinite loop when seeing invalid header.size")
Link: https://marc.info/?l=linux-kernel&m=156580812427554&w=2
Co-developed-by: Jiri Olsa 
Acked-by: Jiri Olsa 
Signed-off-by: Alexey Budankov 
Cc: Alexander Shishkin 
Cc: Andi Kleen 
Cc: Namhyung Kim 
Cc: Peter Zijlstra 
Link: http://lore.kernel.org/lkml/cf782c34-f3f8-2f9f-d6ab-145cee0d5322@linux.intel.com
Signed-off-by: Arnaldo Carvalho de Melo 
Signed-off-by: Sasha Levin

perf probe: Filter out instances except for inlined subroutine and subprogram

2019-12-31T15:45:05+00:00

[ Upstream commit da6cb952a89efe24bb76c4971370d485737a2d85 ]

Filter out instances except for inlined_subroutine and subprogram DIE in
die_walk_instances() and die_is_func_instance().

This fixes an issue that perf probe sets some probes on calling address
instead of a target function itself.

When perf probe walks on instances of an abstruct origin (a kind of
function prototype of inlined function), die_walk_instances() can also
pass a GNU_call_site (a GNU extension for call site) to callback. Since
it is not an inlined instance of target function, we have to filter out
when searching a probe point.

Without this patch, perf probe sets probes on call site address too.This
can happen on some function which is marked "inlined", but has actual
symbol. (I'm not sure why GCC mark it "inlined"):

  # perf probe -D vfs_read
  p:probe/vfs_read _text+2500017
  p:probe/vfs_read_1 _text+2499468
  p:probe/vfs_read_2 _text+2499563
  p:probe/vfs_read_3 _text+2498876
  p:probe/vfs_read_4 _text+2498512
  p:probe/vfs_read_5 _text+2498627

With this patch:

Slightly different results, similar tho:

  # perf probe -D vfs_read
  p:probe/vfs_read _text+2498512

Committer testing:

  # uname -a
  Linux quaco 5.3.8-200.fc30.x86_64 #1 SMP Tue Oct 29 14:46:22 UTC 2019 x86_64 x86_64 x86_64 GNU/Linux

Before:

  # perf probe -D vfs_read
  p:probe/vfs_read _text+3131557
  p:probe/vfs_read_1 _text+3130975
  p:probe/vfs_read_2 _text+3131047
  p:probe/vfs_read_3 _text+3130380
  p:probe/vfs_read_4 _text+3130000
  # uname -a
  Linux quaco 5.3.8-200.fc30.x86_64 #1 SMP Tue Oct 29 14:46:22 UTC 2019 x86_64 x86_64 x86_64 GNU/Linux
  #

After:

  # perf probe -D vfs_read
  p:probe/vfs_read _text+3130000
  #

Fixes: db0d2c6420ee ("perf probe: Search concrete out-of-line instances")
Signed-off-by: Masami Hiramatsu 
Tested-by: Arnaldo Carvalho de Melo 
Cc: Jiri Olsa 
Cc: Namhyung Kim 
Link: http://lore.kernel.org/lkml/157241937063.32002.11024544873990816590.stgit@devnote2
Signed-off-by: Arnaldo Carvalho de Melo 
Signed-off-by: Sasha Levin

perf probe: Skip end-of-sequence and non statement lines

2019-12-31T15:45:04+00:00

[ Upstream commit f4d99bdfd124823a81878b44b5e8750b97f73902 ]

Skip end-of-sequence and non-statement lines while walking through lines
list.

The "end-of-sequence" line information means:

 "the current address is that of the first byte after the
  end of a sequence of target machine instructions."
 (DWARF version 4 spec 6.2.2)

This actually means out of scope and we can not probe on it.

On the other hand, the statement lines (is_stmt) means:

 "the current instruction is a recommended breakpoint location.
  A recommended breakpoint location is intended to “represent”
  a line, a statement and/or a semantically distinct subpart
  of a statement."

 (DWARF version 4 spec 6.2.2)

So, non-statement line info also should be skipped.

These can reduce unneeded probe points and also avoid an error.

E.g. without this patch:

  # perf probe -a "clear_tasks_mm_cpumask:1"
  Added new events:
    probe:clear_tasks_mm_cpumask (on clear_tasks_mm_cpumask:1)
    probe:clear_tasks_mm_cpumask_1 (on clear_tasks_mm_cpumask:1)
    probe:clear_tasks_mm_cpumask_2 (on clear_tasks_mm_cpumask:1)
    probe:clear_tasks_mm_cpumask_3 (on clear_tasks_mm_cpumask:1)
    probe:clear_tasks_mm_cpumask_4 (on clear_tasks_mm_cpumask:1)

  You can now use it in all perf tools, such as:

  	perf record -e probe:clear_tasks_mm_cpumask_4 -aR sleep 1

  #

This puts 5 probes on one line, but acutally it's not inlined function.
This is because there are many non statement instructions at the
function prologue.

With this patch:

  # perf probe -a "clear_tasks_mm_cpumask:1"
  Added new event:
    probe:clear_tasks_mm_cpumask (on clear_tasks_mm_cpumask:1)

  You can now use it in all perf tools, such as:

  	perf record -e probe:clear_tasks_mm_cpumask -aR sleep 1

  #

Now perf-probe skips unneeded addresses.

Committer testing:

Slightly different results, but similar:

Before:

  # uname -a
  Linux quaco 5.3.8-200.fc30.x86_64 #1 SMP Tue Oct 29 14:46:22 UTC 2019 x86_64 x86_64 x86_64 GNU/Linux
  #
  # perf probe -a "clear_tasks_mm_cpumask:1"
  Added new events:
    probe:clear_tasks_mm_cpumask (on clear_tasks_mm_cpumask:1)
    probe:clear_tasks_mm_cpumask_1 (on clear_tasks_mm_cpumask:1)
    probe:clear_tasks_mm_cpumask_2 (on clear_tasks_mm_cpumask:1)

  You can now use it in all perf tools, such as:

  	perf record -e probe:clear_tasks_mm_cpumask_2 -aR sleep 1

  #

After:

  # perf probe -a "clear_tasks_mm_cpumask:1"
  Added new event:
    probe:clear_tasks_mm_cpumask (on clear_tasks_mm_cpumask:1)

  You can now use it in all perf tools, such as:

  	perf record -e probe:clear_tasks_mm_cpumask -aR sleep 1

  # perf probe -l
    probe:clear_tasks_mm_cpumask (on clear_tasks_mm_cpumask@kernel/cpu.c)
  #

Fixes: 4cc9cec636e7 ("perf probe: Introduce lines walker interface")
Signed-off-by: Masami Hiramatsu 
Tested-by: Arnaldo Carvalho de Melo 
Cc: Jiri Olsa 
Cc: Namhyung Kim 
Link: http://lore.kernel.org/lkml/157241936090.32002.12156347518596111660.stgit@devnote2
Signed-off-by: Arnaldo Carvalho de Melo 
Signed-off-by: Sasha Levin

perf probe: Fix to show calling lines of inlined functions

2019-12-31T15:45:04+00:00

[ Upstream commit 86c0bf8539e7f46d91bd105e55eda96e0064caef ]

Fix to show calling lines of inlined functions (where an inline function
is called).

die_walk_lines() filtered out the lines inside inlined functions based
on the address. However this also filtered out the lines which call
those inlined functions from the target function.

To solve this issue, check the call_file and call_line attributes and do
not filter out if it matches to the line information.

Without this fix, perf probe -L doesn't show some lines correctly.
(don't see the lines after 17)

  # perf probe -L vfs_read
  
        0  ssize_t vfs_read(struct file *file, char __user *buf, size_t count, loff_t *pos)
        1  {
        2         ssize_t ret;

        4         if (!(file->f_mode & FMODE_READ))
                          return -EBADF;
        6         if (!(file->f_mode & FMODE_CAN_READ))
                          return -EINVAL;
        8         if (unlikely(!access_ok(buf, count)))
                          return -EFAULT;

       11         ret = rw_verify_area(READ, file, pos, count);
       12         if (!ret) {
       13                 if (count > MAX_RW_COUNT)
                                  count =  MAX_RW_COUNT;
       15                 ret = __vfs_read(file, buf, count, pos);
       16                 if (ret > 0) {
                                  fsnotify_access(file);
                                  add_rchar(current, ret);
                          }

With this fix:

  # perf probe -L vfs_read
  
        0  ssize_t vfs_read(struct file *file, char __user *buf, size_t count, loff_t *pos)
        1  {
        2         ssize_t ret;

        4         if (!(file->f_mode & FMODE_READ))
                          return -EBADF;
        6         if (!(file->f_mode & FMODE_CAN_READ))
                          return -EINVAL;
        8         if (unlikely(!access_ok(buf, count)))
                          return -EFAULT;

       11         ret = rw_verify_area(READ, file, pos, count);
       12         if (!ret) {
       13                 if (count > MAX_RW_COUNT)
                                  count =  MAX_RW_COUNT;
       15                 ret = __vfs_read(file, buf, count, pos);
       16                 if (ret > 0) {
       17                         fsnotify_access(file);
       18                         add_rchar(current, ret);
                          }
       20                 inc_syscr(current);
                  }

Fixes: 4cc9cec636e7 ("perf probe: Introduce lines walker interface")
Signed-off-by: Masami Hiramatsu 
Tested-by: Arnaldo Carvalho de Melo 
Cc: Jiri Olsa 
Cc: Namhyung Kim 
Link: http://lore.kernel.org/lkml/157241937995.32002.17899884017011512577.stgit@devnote2
Signed-off-by: Arnaldo Carvalho de Melo 
Signed-off-by: Sasha Levin

perf probe: Return a better scope DIE if there is no best scope

2019-12-31T15:45:03+00:00

[ Upstream commit c701636aeec4c173208697d68da6e4271125564b ]

Make find_best_scope() returns innermost DIE at given address if there
is no best matched scope DIE. Since Gcc sometimes generates intuitively
strange line info which is out of inlined function address range, we
need this fixup.

Without this, sometimes perf probe failed to probe on a line inside an
inlined function:

  # perf probe -D ksys_open:3
  Failed to find scope of probe point.
    Error: Failed to add events.

With this fix, 'perf probe' can probe it:

  # perf probe -D ksys_open:3
  p:probe/ksys_open _text+25707308
  p:probe/ksys_open_1 _text+25710596
  p:probe/ksys_open_2 _text+25711114
  p:probe/ksys_open_3 _text+25711343
  p:probe/ksys_open_4 _text+25714058
  p:probe/ksys_open_5 _text+2819653
  p:probe/ksys_open_6 _text+2819701

Signed-off-by: Masami Hiramatsu 
Tested-by: Arnaldo Carvalho de Melo 
Cc: Namhyung Kim 
Cc: Ravi Bangoria 
Cc: Steven Rostedt (VMware) 
Cc: Tom Zanussi 
Link: http://lore.kernel.org/lkml/157291300887.19771.14936015360963292236.stgit@devnote2
Signed-off-by: Arnaldo Carvalho de Melo 
Signed-off-by: Sasha Levin

perf probe: Skip overlapped location on searching variables

2019-12-31T15:45:02+00:00

[ Upstream commit dee36a2abb67c175265d49b9a8c7dfa564463d9a ]

Since debuginfo__find_probes() callback function can be called with  the
location which already passed, the callback function must filter out
such overlapped locations.

add_probe_trace_event() has already done it by commit 1a375ae7659a
("perf probe: Skip same probe address for a given line"), but
add_available_vars() doesn't. Thus perf probe -v shows same address
repeatedly as below:

  # perf probe -V vfs_read:18
  Available variables at vfs_read:18
          @
                  char*   buf
                  loff_t* pos
                  ssize_t ret
                  struct file*    file
          @
                  char*   buf
                  loff_t* pos
                  ssize_t ret
                  struct file*    file
          @
                  char*   buf
                  loff_t* pos
                  ssize_t ret
                  struct file*    file

With this fix, perf probe -V shows it correctly:

  # perf probe -V vfs_read:18
  Available variables at vfs_read:18
          @
                  char*   buf
                  loff_t* pos
                  ssize_t ret
                  struct file*    file
          @
                  char*   buf
                  loff_t* pos
                  ssize_t ret
                  struct file*    file

Fixes: cf6eb489e5c0 ("perf probe: Show accessible local variables")
Signed-off-by: Masami Hiramatsu 
Tested-by: Arnaldo Carvalho de Melo 
Cc: Jiri Olsa 
Cc: Namhyung Kim 
Link: http://lore.kernel.org/lkml/157241938927.32002.4026859017790562751.stgit@devnote2
Signed-off-by: Arnaldo Carvalho de Melo 
Signed-off-by: Sasha Levin