Skip to content

cloud-kernel
cloud-kernel

openanolis / cloud-kernel
1 年多 前同步成功

通知 161
Star 36
Fork 7
cloud-kernel
cloud-kernel
切换分支/标签
  1. 02 6月, 2018 1 次提交
    • A
      perf trace beauty prctl: Default header_dir to cwd to work without parms · 0d690fc0
      由 Arnaldo Carvalho de Melo 提交于 
      Useful when checking the effects of header synchs for the files it uses
as a input to generate string tables, in retrospect this is how it
should've been done from day 1, not requiring the header_dir to be set
on the Makefile, will change everything later, so that the only parm,
common to all generators will be $(srctree) and $(beauty_outdir).

So, to see what it generates, just call it without any parameters:

  $ tools/perf/trace/beauty/prctl_option.sh
  static const char *prctl_options[] = {
	  [1] = "SET_PDEATHSIG",
	  [2] = "GET_PDEATHSIG",
	  [3] = "GET_DUMPABLE",
	  [4] = "SET_DUMPABLE",
	  [5] = "GET_UNALIGN",
	  [6] = "SET_UNALIGN",
	  [7] = "GET_KEEPCAPS",
	  [8] = "SET_KEEPCAPS",
	  [9] = "GET_FPEMU",
	  [10] = "SET_FPEMU",
	  [11] = "GET_FPEXC",
	  [12] = "SET_FPEXC",
	  [13] = "GET_TIMING",
	  [14] = "SET_TIMING",
	  [15] = "SET_NAME",
	  [16] = "GET_NAME",
	  [19] = "GET_ENDIAN",
	  [20] = "SET_ENDIAN",
	  [21] = "GET_SECCOMP",
	  [22] = "SET_SECCOMP",
	  [25] = "GET_TSC",
	  [26] = "SET_TSC",
	  [27] = "GET_SECUREBITS",
	  [28] = "SET_SECUREBITS",
	  [29] = "SET_TIMERSLACK",
	  [30] = "GET_TIMERSLACK",
	  [35] = "SET_MM",
	  [36] = "SET_CHILD_SUBREAPER",
	  [37] = "GET_CHILD_SUBREAPER",
	  [38] = "SET_NO_NEW_PRIVS",
	  [39] = "GET_NO_NEW_PRIVS",
	  [40] = "GET_TID_ADDRESS",
	  [41] = "SET_THP_DISABLE",
	  [42] = "GET_THP_DISABLE",
	  [45] = "SET_FP_MODE",
	  [46] = "GET_FP_MODE",
  };
  static const char *prctl_set_mm_options[] = {
	  [1] = "START_CODE",
	  [2] = "END_CODE",
	  [3] = "START_DATA",
	  [4] = "END_DATA",
	  [5] = "START_STACK",
	  [6] = "START_BRK",
	  [7] = "BRK",
	  [8] = "ARG_START",
	  [9] = "ARG_END",
	  [10] = "ENV_START",
	  [11] = "ENV_END",
	  [12] = "AUXV",
	  [13] = "EXE_FILE",
	  [14] = "MAP",
	  [15] = "MAP_SIZE",
  };
  $

Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Wang Nan <wangnan0@huawei.com>
Link: https://lkml.kernel.org/n/tip-qtotspuztydjttxi7k6mec6h@git.kernel.orgSigned-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
      0d690fc0
  3. 19 4月, 2018 1 次提交
    • A
      perf trace: Support MAP_FIXED_NOREPLACE · 4e5c01a7
      由 Arnaldo Carvalho de Melo 提交于 
      Introduced in a4ff8e86 ("mm: introduce MAP_FIXED_NOREPLACE"), and
now that we have that define in the just syncronized
tools/arch/*/include/uapi/asm/mman.h files, add support for it.

This should really transition to autogeneration of string tables as
done for various other things:

  $ ls /tmp/build/perf/trace/beauty/generated/*.c
  arch_errno_name_array.c kcmp_type_array.c madvise_behavior_array.c
  pkey_alloc_access_rights_array.c prctl_option_array.c
  $ head /tmp/build/perf/trace/beauty/generated/madvise_behavior_array.c
  static const char *madvise_advices[] = {
	[0] = "NORMAL",
	[1] = "RANDOM",
	[2] = "SEQUENTIAL",
	[3] = "WILLNEED",
	[4] = "DONTNEED",
	[8] = "FREE",
	[9] = "REMOVE",
	[10] = "DONTFORK",
	[11] = "DOFORK",
  $

Till then, add support for this the old way.

Also it has to be ifdef'ed, because arches like mips still don't define
it. The proper solution will be to have per-arch tables for these
values to support cross-analysis.

Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Michal Hocko <mhocko@suse.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Wang Nan <wangnan0@huawei.com>
Link: https://lkml.kernel.org/n/tip-td9t5vhjltqnlzaurkkgq8cn@git.kernel.orgSignef-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
      4e5c01a7
  5. 25 1月, 2018 2 次提交
  7. 23 1月, 2018 1 次提交
  9. 29 11月, 2017 1 次提交
  11. 02 11月, 2017 1 次提交
    • G
      License cleanup: add SPDX GPL-2.0 license identifier to files with no license · b2441318
      由 Greg Kroah-Hartman 提交于 
      Many source files in the tree are missing licensing information, which
makes it harder for compliance tools to determine the correct license.

By default all files without license information are under the default
license of the kernel, which is GPL version 2.

Update the files which contain no license information with the 'GPL-2.0'
SPDX license identifier.  The SPDX identifier is a legally binding
shorthand, which can be used instead of the full boiler plate text.

This patch is based on work done by Thomas Gleixner and Kate Stewart and
Philippe Ombredanne.

How this work was done:

Patches were generated and checked against linux-4.14-rc6 for a subset of
the use cases:
 - file had no licensing information it it.
 - file was a */uapi/* one with no licensing information in it,
 - file was a */uapi/* one with existing licensing information,

Further patches will be generated in subsequent months to fix up cases
where non-standard license headers were used, and references to license
had to be inferred by heuristics based on keywords.

The analysis to determine which SPDX License Identifier to be applied to
a file was done in a spreadsheet of side by side results from of the
output of two independent scanners (ScanCode & Windriver) producing SPDX
tag:value files created by Philippe Ombredanne.  Philippe prepared the
base worksheet, and did an initial spot review of a few 1000 files.

The 4.13 kernel was the starting point of the analysis with 60,537 files
assessed.  Kate Stewart did a file by file comparison of the scanner
results in the spreadsheet to determine which SPDX license identifier(s)
to be applied to the file. She confirmed any determination that was not
immediately clear with lawyers working with the Linux Foundation.

Criteria used to select files for SPDX license identifier tagging was:
 - Files considered eligible had to be source code files.
 - Make and config files were included as candidates if they contained >5
   lines of source
 - File already had some variant of a license header in it (even if <5
   lines).

All documentation files were explicitly excluded.

The following heuristics were used to determine which SPDX license
identifiers to apply.

 - when both scanners couldn't find any license traces, file was
   considered to have no license information in it, and the top level
   COPYING file license applied.

   For non */uapi/* files that summary was:

   SPDX license identifier                            # files
   ---------------------------------------------------|-------
   GPL-2.0                                              11139

   and resulted in the first patch in this series.

   If that file was a */uapi/* path one, it was "GPL-2.0 WITH
   Linux-syscall-note" otherwise it was "GPL-2.0".  Results of that was:

   SPDX license identifier                            # files
   ---------------------------------------------------|-------
   GPL-2.0 WITH Linux-syscall-note                        930

   and resulted in the second patch in this series.

 - if a file had some form of licensing information in it, and was one
   of the */uapi/* ones, it was denoted with the Linux-syscall-note if
   any GPL family license was found in the file or had no licensing in
   it (per prior point).  Results summary:

   SPDX license identifier                            # files
   ---------------------------------------------------|------
   GPL-2.0 WITH Linux-syscall-note                       270
   GPL-2.0+ WITH Linux-syscall-note                      169
   ((GPL-2.0 WITH Linux-syscall-note) OR BSD-2-Clause)    21
   ((GPL-2.0 WITH Linux-syscall-note) OR BSD-3-Clause)    17
   LGPL-2.1+ WITH Linux-syscall-note                      15
   GPL-1.0+ WITH Linux-syscall-note                       14
   ((GPL-2.0+ WITH Linux-syscall-note) OR BSD-3-Clause)    5
   LGPL-2.0+ WITH Linux-syscall-note                       4
   LGPL-2.1 WITH Linux-syscall-note                        3
   ((GPL-2.0 WITH Linux-syscall-note) OR MIT)              3
   ((GPL-2.0 WITH Linux-syscall-note) AND MIT)             1

   and that resulted in the third patch in this series.

 - when the two scanners agreed on the detected license(s), that became
   the concluded license(s).

 - when there was disagreement between the two scanners (one detected a
   license but the other didn't, or they both detected different
   licenses) a manual inspection of the file occurred.

 - In most cases a manual inspection of the information in the file
   resulted in a clear resolution of the license that should apply (and
   which scanner probably needed to revisit its heuristics).

 - When it was not immediately clear, the license identifier was
   confirmed with lawyers working with the Linux Foundation.

 - If there was any question as to the appropriate license identifier,
   the file was flagged for further research and to be revisited later
   in time.

In total, over 70 hours of logged manual review was done on the
spreadsheet to determine the SPDX license identifiers to apply to the
source files by Kate, Philippe, Thomas and, in some cases, confirmation
by lawyers working with the Linux Foundation.

Kate also obtained a third independent scan of the 4.13 code base from
FOSSology, and compared selected files where the other two scanners
disagreed against that SPDX file, to see if there was new insights.  The
Windriver scanner is based on an older version of FOSSology in part, so
they are related.

Thomas did random spot checks in about 500 files from the spreadsheets
for the uapi headers and agreed with SPDX license identifier in the
files he inspected. For the non-uapi files Thomas did random spot checks
in about 15000 files.

In initial set of patches against 4.14-rc6, 3 files were found to have
copy/paste license identifier errors, and have been fixed to reflect the
correct identifier.

Additionally Philippe spent 10 hours this week doing a detailed manual
inspection and review of the 12,461 patched files from the initial patch
version early this week with:
 - a full scancode scan run, collecting the matched texts, detected
   license ids and scores
 - reviewing anything where there was a license detected (about 500+
   files) to ensure that the applied SPDX license was correct
 - reviewing anything where there was no detection but the patch license
   was not GPL-2.0 WITH Linux-syscall-note to ensure that the applied
   SPDX license was correct

This produced a worksheet with 20 files needing minor correction.  This
worksheet was then exported into 3 different .csv files for the
different types of files to be modified.

These .csv files were then reviewed by Greg.  Thomas wrote a script to
parse the csv files and add the proper SPDX tag to the file, in the
format that the file expected.  This script was further refined by Greg
based on the output to detect more types of files automatically and to
distinguish between header and source .c files (which need different
comment types.)  Finally Greg ran the script using the .csv files to
generate the patches.
Reviewed-by: NKate Stewart <kstewart@linuxfoundation.org>
Reviewed-by: NPhilippe Ombredanne <pombredanne@nexb.com>
Reviewed-by: NThomas Gleixner <tglx@linutronix.de>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>
      b2441318
  13. 01 11月, 2017 2 次提交
    • A
      perf trace beauty kcmp: Beautify arguments · 1de3038d
      由 Arnaldo Carvalho de Melo 提交于 
      For some unknown reason there is no entry in tracefs's syscalls for
kcmp, i.e. no tracefs/events/syscalls/sys_{enter,exit}_kcmp, so we need
to provide a data dictionary for the fields.

To beautify the 'type' argument we automatically generate a strarray
from tools/include/uapi/kcmp.h, the idx1 and idx2 args, nowadays used
only if type == KCMP_FILE, are masked for all the other types and a
lookup is made for the thread and fd to show the path, if possible,
getting it from the probe:vfs_getname if in place or from procfs, races
allowing.

A system wide strace like tracing session, with callchains shows just
one user so far in this fedora 25 machine:

  # perf trace --max-stack 5 -e kcmp
  <SNIP>
  1502914.400 ( 0.001 ms): systemd/1 kcmp(pid1: 1 (systemd), pid2: 1 (systemd), type: FILE, idx1: 271<socket:[4723475]>, idx2: 25<socket:[4788686]>) = -1 ENOSYS Function not implemented
                                         syscall (/usr/lib64/libc-2.25.so)
                                         same_fd (/usr/lib/systemd/libsystemd-shared-233.so)
                                         service_add_fd_store (/usr/lib/systemd/systemd)
                                         service_notify_message.lto_priv.127 (/usr/lib/systemd/systemd)
  1502914.407 ( 0.001 ms): systemd/1 kcmp(pid1: 1 (systemd), pid2: 1 (systemd), type: FILE, idx1: 270<socket:[4726396]>, idx2: 25<socket:[4788686]>) = -1 ENOSYS Function not implemented
                                         syscall (/usr/lib64/libc-2.25.so)
                                         same_fd (/usr/lib/systemd/libsystemd-shared-233.so)
                                         service_add_fd_store (/usr/lib/systemd/systemd)
                                         service_notify_message.lto_priv.127 (/usr/lib/systemd/systemd)
  <SNIP>

The backtraces seem to agree this is really kcmp(), but this system
doesn't have the sys_kcmp(), bummer:

  # uname -a
  Linux jouet 4.14.0-rc3+ #1 SMP Fri Oct 13 12:21:12 -03 2017 x86_64 x86_64 x86_64 GNU/Linux
  # grep kcmp /proc/kallsyms
  ffffffffb60b8890 W sys_kcmp
  $ grep CONFIG_CHECKPOINT_RESTORE ../build/v4.14.0-rc3+/.config
  # CONFIG_CHECKPOINT_RESTORE is not set
  $

So systemd uses it, good fedora kernel config has it:

  $ grep CONFIG_CHECKPOINT_RESTORE /boot/config-4.13.4-200.fc26.x86_64
  CONFIG_CHECKPOINT_RESTORE=y
  [acme@jouet linux]$

/me goes to rebuild a kernel...

Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Andrey Vagin <avagin@openvz.org>
Cc: Cyrill Gorcunov <gorcunov@openvz.org>
Cc: David Ahern <dsahern@gmail.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Wang Nan <wangnan0@huawei.com>
Link: http://lkml.kernel.org/n/tip-gz5fca968viw8m7hryjqvrln@git.kernel.orgSigned-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
      1de3038d
    • A
      perf trace beauty: Implement pid_fd beautifier · 0a2f7540
      由 Arnaldo Carvalho de Melo 提交于 
      One that given a pid and a fd, will try to get the path for that fd.
Will be used in the upcoming kcmp's KCMP_FILE beautifier.

Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Andrey Vagin <avagin@openvz.org>
Cc: Cyrill Gorcunov <gorcunov@openvz.org>
Cc: David Ahern <dsahern@gmail.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Wang Nan <wangnan0@huawei.com>
Link: http://lkml.kernel.org/n/tip-7ketygp2dvs9h13wuakfncws@git.kernel.orgSigned-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
      0a2f7540
  15. 27 10月, 2017 1 次提交
    • A
      perf trace beauty prctl: Generate 'option' string table from kernel headers · d688d037
      由 Arnaldo Carvalho de Melo 提交于 
      This is one more case where the way that syscall parameter values are
defined in kernel headers are easy to parse using a shell script that
will then generate the string table that gets used by the prctl 'option'
argument beautifier.

This way as soon as the header syncronization mechanism in perf's build
system detects a change in a copy of a kernel ABI header and that file
is syncronized, we get 'perf trace' updated automagically.

Further work needed for the PR_SET_ values, as well for using eBPF to
copy the non-integer arguments to/from the kernel.

E.g.: System wide prctl tracing:

  # perf trace -e prctl
  1668.028 ( 0.025 ms): TaskSchedulerR/10649 prctl(option: SET_NAME, arg2: 0x2b61d5db15d0) = 0
  3365.663 ( 0.018 ms): chrome/10650 prctl(option: SET_SECCOMP, arg2: 2, arg4: 8         ) = -1 EFAULT Bad address
  3366.585 ( 0.010 ms): chrome/10650 prctl(option: SET_NO_NEW_PRIVS, arg2: 1             ) = 0
  3367.173 ( 0.009 ms): TaskSchedulerR/10652 prctl(option: SET_NAME, arg2: 0x2b61d2aaa300) = 0
  3367.222 ( 0.003 ms): TaskSchedulerR/10653 prctl(option: SET_NAME, arg2: 0x2b61d2aaa1e0) = 0
  3367.244 ( 0.002 ms): TaskSchedulerR/10654 prctl(option: SET_NAME, arg2: 0x2b61d2aaa0c0) = 0
  3367.265 ( 0.002 ms): TaskSchedulerR/10655 prctl(option: SET_NAME, arg2: 0x2b61d2ac7f90) = 0
  3367.281 ( 0.002 ms): Chrome_ChildIO/10656 prctl(option: SET_NAME, arg2: 0x7efbe406bb11) = 0
  3367.220 ( 0.004 ms): TaskSchedulerS/10651 prctl(option: SET_NAME, arg2: 0x2b61d2ac1be0) = 0
  3370.906 ( 0.010 ms): GpuMemoryThrea/10657 prctl(option: SET_NAME, arg2: 0x7efbe386ab11) = 0
  3370.983 ( 0.003 ms): File/10658 prctl(option: SET_NAME, arg2: 0x7efbe3069b11          ) = 0
  3384.272 ( 0.020 ms): Compositor/10659 prctl(option: SET_NAME, arg2: 0x7efbe2868b11    ) = 0
  3612.091 ( 0.012 ms): DOM Worker/11489 prctl(option: SET_NAME, arg2: 0x7f49ab97ebf2    ) = 0
<SNIP>
  4512.437 ( 0.004 ms): (sa1)/11490 prctl(option: SET_NAME, arg2: 0x7ffca15af844         ) = 0
  4512.468 ( 0.002 ms): (sa1)/11490 prctl(option: SET_MM, arg2: ARG_START, arg3: 0x7f5cb7c81000) = 0
  4512.472 ( 0.001 ms): (sa1)/11490 prctl(option: SET_MM, arg2: ARG_END, arg3: 0x7f5cb7c81006) = 0
  4514.667 ( 0.002 ms): (sa1)/11490 prctl(option: GET_SECUREBITS                         ) = 0

Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Andy Lutomirski <luto@kernel.org>
Cc: David Ahern <dsahern@gmail.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Wang Nan <wangnan0@huawei.com>
Link: http://lkml.kernel.org/n/tip-q0s2uw579o5ei6xlh2zjirgz@git.kernel.orgSigned-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
      d688d037
  17. 22 9月, 2017 1 次提交
    • A
      perf trace beauty madvise: Generate 'behavior' string table from kernel headers · 5a54c2f5
      由 Arnaldo Carvalho de Melo 提交于 
      This is one more case where the way that syscall parameter values are
defined in kernel headers are easy to parse using a shell script that
will then generate the string table that gets used by the madvise
'behaviour' argument beautifier.

This way as soon as the header syncronization mechanism in perf's build
system detects a change in a copy of a kernel ABI header and that file
is syncronized, we get 'perf trace' updated automagically.

So, when we syncronize this:

  Warning: Kernel ABI header at 'tools/include/uapi/asm-generic/mman-common.h' differs from latest version at 'include/uapi/asm-generic/mman-common.h'

We'll get these:

  #define MADV_WIPEONFORK 18              /* Zero memory on fork, child only */
  #define MADV_KEEPONFORK 19              /* Undo MADV_WIPEONFORK */

Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Rik van Riel <riel@redhat.com>
Cc: Wang Nan <wangnan0@huawei.com>
Link: http://lkml.kernel.org/n/tip-dolb0ghds4ui7wc1npgkchvc@git.kernel.orgSigned-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
      5a54c2f5
  19. 29 8月, 2017 1 次提交
    • A
      perf trace beauty: Beautify pkey_{alloc,free,mprotect} arguments · 83bc9c37
      由 Arnaldo Carvalho de Melo 提交于 
      Reuse 'mprotect' beautifiers for 'pkey_mprotect'.

System wide tracing pkey_alloc, pkey_free and pkey_mprotect calls, with
backtraces:

  # perf trace -e pkey_alloc,pkey_mprotect,pkey_free --max-stack=5
     0.000 ( 0.011 ms): pkey/7818 pkey_alloc(init_val: DISABLE_ACCESS|DISABLE_WRITE) = -1 EINVAL Invalid argument
                                       syscall (/usr/lib64/libc-2.25.so)
                                       pkey_alloc (/home/acme/c/pkey)
     0.022 ( 0.003 ms): pkey/7818 pkey_mprotect(start: 0x7f28c3890000, len: 4096, prot: READ|WRITE, pkey: -1) = 0
                                       syscall (/usr/lib64/libc-2.25.so)
                                       pkey_mprotect (/home/acme/c/pkey)
     0.030 ( 0.002 ms): pkey/7818 pkey_free(pkey: -1                               ) = -1 EINVAL Invalid argument
                                       syscall (/usr/lib64/libc-2.25.so)
                                       pkey_free (/home/acme/c/pkey)

The tools/include/uapi/asm-generic/mman-common.h file is used to find
the access rights defines for the pkey_alloc syscall second argument.

Since we have the detector of changes for the tools/include header files
versus its kernel origin (include/uapi/asm-generic/mman-common.h), we'll
get whatever new flag appears for that argument automatically.

This method should be used in other cases where it is easy to generate
those flags tables because the header has properly namespaced defines
like PKEY_DISABLE_ACCESS and PKEY_DISABLE_WRITE.

Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Arnd Bergmann <arnd@arndb.de>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Wang Nan <wangnan0@huawei.com>
Link: http://lkml.kernel.org/n/tip-3xq5312qlks7wtfzv2sk3nct@git.kernel.orgSigned-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
      83bc9c37
  21. 02 8月, 2017 6 次提交
    • A
      perf trace beautify ioctl: Beautify perf ioctl's 'cmd' arg · 81e3d8b2
      由 Arnaldo Carvalho de Melo 提交于 
      Also trying a new approach, using the copy of uapi/linux/perf_event.h we
auto generate the string tables, then include it in the ioctl cmd
beautifier.

This way either the perf developers will add the new commands to the
tools/ copy, like is happening with other areas of tools/include/ (bpf.h
comes to mind), or we'll be notified when building perf that our copy
drifted.

E.g., looking at some of the perf ioctls issued by the 'perf test' test cases:

  # (perf trace -e perf_event_open,ioctl perf test)  2>&1 | egrep "(cmd: PERF_|perf_event_open)"
  4: Read samples using the mmap interface      :
   348.811 ( 0.062 ms): perf/23351 perf_event_open(attr_uptr: 0x414a5e8, pid: 23351 (perf), group_fd: -1, flags: FD_CLOEXEC) = 3
   348.878 ( 0.039 ms): perf/23351 perf_event_open(attr_uptr: 0x414a5e8, pid: 23351 (perf), cpu: 1, group_fd: -1, flags: FD_CLOEXEC) = 4
   348.919 ( 0.036 ms): perf/23351 perf_event_open(attr_uptr: 0x414a5e8, pid: 23351 (perf), cpu: 2, group_fd: -1, flags: FD_CLOEXEC) = 5
   348.958 ( 0.036 ms): perf/23351 perf_event_open(attr_uptr: 0x414a5e8, pid: 23351 (perf), cpu: 3, group_fd: -1, flags: FD_CLOEXEC) = 6
   349.070 ( 0.046 ms): perf/23351 perf_event_open(attr_uptr: 0x414aa38, pid: 23351 (perf), group_fd: -1, flags: FD_CLOEXEC) = 7
   349.120 ( 0.037 ms): perf/23351 perf_event_open(attr_uptr: 0x414aa38, pid: 23351 (perf), cpu: 1, group_fd: -1, flags: FD_CLOEXEC) = 8
   349.161 ( 0.036 ms): perf/23351 perf_event_open(attr_uptr: 0x414aa38, pid: 23351 (perf), cpu: 2, group_fd: -1, flags: FD_CLOEXEC) = 9
   349.201 ( 0.035 ms): perf/23351 perf_event_open(attr_uptr: 0x414aa38, pid: 23351 (perf), cpu: 3, group_fd: -1, flags: FD_CLOEXEC) = 10
   349.306 ( 0.041 ms): perf/23351 perf_event_open(attr_uptr: 0x414b2d8, pid: 23351 (perf), group_fd: -1, flags: FD_CLOEXEC) = 11
   349.611 ( 0.005 ms): perf/23351 ioctl(fd: 3<anon_inode:[perf_event]>, cmd: PERF_ID, arg: 0x7fff025999b8) = 0
   349.619 ( 0.002 ms): perf/23351 ioctl(fd: 7<anon_inode:[perf_event]>, cmd: PERF_SET_OUTPUT, arg: 0x3  ) = 0
   349.623 ( 0.002 ms): perf/23351 ioctl(fd: 7<anon_inode:[perf_event]>, cmd: PERF_ID, arg: 0x7fff025999b8) = 0
   349.627 ( 0.002 ms): perf/23351 ioctl(fd: 11<anon_inode:[perf_event]>, cmd: PERF_SET_OUTPUT, arg: 0x3 ) = 0
   349.630 ( 0.001 ms): perf/23351 ioctl(fd: 11<anon_inode:[perf_event]>, cmd: PERF_ID, arg: 0x7fff025999b8) = 0
<SNIP>
  7: PERF_RECORD_* events & perf_sample fields  :
   647.150 ( 0.014 ms): perf/23354 perf_event_open(attr_uptr: 0x7fff02599920, pid: -1, cpu: 2, group_fd: -1, flags: FD_CLOEXEC) = 3
   647.197 ( 0.076 ms): perf/23354 perf_event_open(attr_uptr: 0x414b478, pid: -1, group_fd: -1, flags: FD_CLOEXEC) = 3
   647.289 ( 0.040 ms): perf/23354 perf_event_open(attr_uptr: 0x414b478, pid: -1, group_fd: -1, flags: FD_CLOEXEC) = 3
   647.368 ( 0.011 ms): perf/23354 perf_event_open(attr_uptr: 0x414a5e8, pid: 23355 (perf), group_fd: -1, flags: FD_CLOEXEC) = 3
   647.381 ( 0.005 ms): perf/23354 perf_event_open(attr_uptr: 0x414a5e8, pid: 23355 (perf), cpu: 1, group_fd: -1, flags: FD_CLOEXEC) = 4
   647.387 ( 0.005 ms): perf/23354 perf_event_open(attr_uptr: 0x414a5e8, pid: 23355 (perf), cpu: 2, group_fd: -1, flags: FD_CLOEXEC) = 5
   647.393 ( 0.004 ms): perf/23354 perf_event_open(attr_uptr: 0x414a5e8, pid: 23355 (perf), cpu: 3, group_fd: -1, flags: FD_CLOEXEC) = 7
   648.026 ( 0.011 ms): perf/23354 ioctl(fd: 3<anon_inode:[perf_event]>, cmd: PERF_ENABLE) = 0
   648.038 ( 0.002 ms): perf/23354 ioctl(fd: 4<anon_inode:[perf_event]>, cmd: PERF_ENABLE) = 0
   648.042 ( 0.002 ms): perf/23354 ioctl(fd: 5<anon_inode:[perf_event]>, cmd: PERF_ENABLE) = 0
   648.045 ( 0.002 ms): perf/23354 ioctl(fd: 7<anon_inode:[perf_event]>, cmd: PERF_ENABLE) = 0
<SNIP>
  18: Breakpoint overflow signal handler         :
  2772.721 ( 0.017 ms): perf/23375 perf_event_open(attr_uptr: 0x7fff02599d20, pid: -1, cpu: 3, group_fd: -1, flags: FD_CLOEXEC) = 3
  2772.748 ( 0.009 ms): perf/23375 perf_event_open(attr_uptr: 0x7fff02599e60, cpu: -1, group_fd: -1, flags: FD_CLOEXEC) = 3
  2772.768 ( 0.002 ms): perf/23375 ioctl(fd: 3, cmd: PERF_RESET) = 0
  2772.776 ( 0.008 ms): perf/23375 perf_event_open(attr_uptr: 0x7fff02599e60, cpu: -1, group_fd: -1, flags: FD_CLOEXEC) = 4
  2772.788 ( 0.002 ms): perf/23375 ioctl(fd: 4, cmd: PERF_RESET) = 0
  2772.791 ( 0.006 ms): perf/23375 perf_event_open(attr_uptr: 0x7fff02599e60, cpu: -1, group_fd: -1, flags: FD_CLOEXEC) = 5
  2772.800 ( 0.001 ms): perf/23375 ioctl(fd: 5, cmd: PERF_RESET) = 0
  2772.803 ( 0.005 ms): perf/23375 ioctl(fd: 3, cmd: PERF_ENABLE) = 0
  2772.810 ( 0.004 ms): perf/23375 ioctl(fd: 4, cmd: PERF_ENABLE) = 0
  2772.815 ( 0.004 ms): perf/23375 ioctl(fd: 5, cmd: PERF_ENABLE) = 0
<SNIP>

Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Wang Nan <wangnan0@huawei.com>
Link: http://lkml.kernel.org/n/tip-ahotwscqt080ae0ulu3zznh2@git.kernel.orgSigned-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
      81e3d8b2
    • A
      perf trace beautify ioctl: Beautify vhost virtio ioctl's 'cmd' arg · ec6dd85f
      由 Arnaldo Carvalho de Melo 提交于 
      Also trying a new approach, using a copy of uapi/linux/vhost.h we auto
generate the string tables, then include it in the ioctl cmd beautifier.

This way either the KVM developers will add the new commands to the
tools/ copy, like is happening with other areas of tools/include/ (bpf.h
comes to mind), or we'll be notified when building perf that our copy
drifted.

E.g., doing syswide tracing grepping for the newly beautified VHOST
ioctls:

  # perf trace -e ioctl 2>&1 | grep VHOST
  3873.064 ( 0.099 ms): qemu-system-x8/21238 ioctl(fd: 27</dev/vhost-net>, cmd: VHOST_NET_SET_BACKEND, arg: 0x7fff053dffe0) = 0
  3873.168 ( 0.019 ms): qemu-system-x8/21238 ioctl(fd: 27</dev/vhost-net>, cmd: VHOST_NET_SET_BACKEND, arg: 0x7fff053dffe0) = 0
  3873.226 ( 0.006 ms): qemu-system-x8/21238 ioctl(fd: 27</dev/vhost-net>, cmd: VHOST_GET_VRING_BASE, arg: 0x7fff053dff60) = 0
  3873.244 ( 0.002 ms): qemu-system-x8/21238 ioctl(fd: 27</dev/vhost-net>, cmd: VHOST_GET_VRING_BASE, arg: 0x7fff053dff60) = 0
  3873.817 ( 0.014 ms): qemu-system-x8/21238 ioctl(fd: 27</dev/vhost-net>, cmd: VHOST_SET_VRING_CALL, arg: 0x7fff053dff20) = 0
  3873.838 ( 0.004 ms): qemu-system-x8/21238 ioctl(fd: 27</dev/vhost-net>, cmd: VHOST_SET_VRING_CALL, arg: 0x7fff053dff20) = 0
  4701.372 ( 0.006 ms): qemu-system-x8/21238 ioctl(fd: 27</dev/vhost-net>, cmd: VHOST_SET_VRING_CALL, arg: 0x7fff053dfe20) = 0
  4701.417 ( 0.007 ms): qemu-system-x8/21238 ioctl(fd: 27</dev/vhost-net>, cmd: VHOST_SET_VRING_CALL, arg: 0x7fff053dfe20) = 0
  4701.563 ( 0.004 ms): qemu-system-x8/21238 ioctl(fd: 27</dev/vhost-net>, cmd: VHOST_SET_FEATURES, arg: 0x7fff053dfe88) = 0
  4701.571 ( 0.028 ms): qemu-system-x8/21238 ioctl(fd: 27</dev/vhost-net>, cmd: VHOST_SET_MEM_TABLE, arg: 0x563c7c906870) = 0
  4701.604 ( 0.003 ms): qemu-system-x8/21238 ioctl(fd: 27</dev/vhost-net>, cmd: VHOST_SET_VRING_NUM, arg: 0x7fff053dff00) = 0
  4701.609 ( 0.002 ms): qemu-system-x8/21238 ioctl(fd: 27</dev/vhost-net>, cmd: VHOST_SET_VRING_BASE, arg: 0x7fff053dff00) = 0
  4701.615 ( 0.002 ms): qemu-system-x8/21238 ioctl(fd: 27</dev/vhost-net>, cmd: VHOST_SET_VRING_ADDR, arg: 0x7fff053dfe70) = 0
  4701.619 ( 0.008 ms): qemu-system-x8/21238 ioctl(fd: 27</dev/vhost-net>, cmd: VHOST_SET_VRING_KICK, arg: 0x7fff053dfef0) = 0
  4701.634 ( 0.004 ms): qemu-system-x8/21238 ioctl(fd: 27</dev/vhost-net>, cmd: VHOST_SET_VRING_NUM, arg: 0x7fff053dff00) = 0
  4701.640 ( 0.002 ms): qemu-system-x8/21238 ioctl(fd: 27</dev/vhost-net>, cmd: VHOST_SET_VRING_BASE, arg: 0x7fff053dff00) = 0
  4701.644 ( 0.002 ms): qemu-system-x8/21238 ioctl(fd: 27</dev/vhost-net>, cmd: VHOST_SET_VRING_ADDR, arg: 0x7fff053dfe70) = 0
  4701.648 ( 0.009 ms): qemu-system-x8/21238 ioctl(fd: 27</dev/vhost-net>, cmd: VHOST_SET_VRING_KICK, arg: 0x7fff053dfef0) = 0
  4701.665 ( 0.005 ms): qemu-system-x8/21238 ioctl(fd: 27</dev/vhost-net>, cmd: VHOST_NET_SET_BACKEND, arg: 0x7fff053dff80) = 0
  4701.672 ( 0.004 ms): qemu-system-x8/21238 ioctl(fd: 27</dev/vhost-net>, cmd: VHOST_NET_SET_BACKEND, arg: 0x7fff053dff80) = 0
^C

 '-e ioctl' uses tracepoint filters, in time this will be replaces by
eBPF filters hooked at the syscall tracepoints and that "grep VHOST"
will also be done with eBPF, right at the kernel, to reduce overhead.

Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Jason Wang <jasowang@redhat.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: "Michael S. Tsirkin" <mst@redhat.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: Radim Krčmář <rkrcmar@redhat.com>
Cc: Wang Nan <wangnan0@huawei.com>
Link: http://lkml.kernel.org/n/tip-2gthnhpliunvakywjterrzz3@git.kernel.orgSigned-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
      ec6dd85f
    • A
      perf trace beauty ioctl: Pass _IOC_DIR to the per _IOC_TYPE scnprintf · 8ff69577
      由 Arnaldo Carvalho de Melo 提交于 
      Not all subsystems use the fact that we may have the same _IOC_NR for
different _IOC_DIR, as in the end it'll result in a different ioctl
number.

So, for instance, vhost virtio has:

  #define VHOST_GET_FEATURES      _IOR(VHOST_VIRTIO, 0x00, __u64)
  #define VHOST_SET_FEATURES      _IOW(VHOST_VIRTIO, 0x00, __u64)

So same _IOC_NR (0x00) but different _IOC_DIR (R versus W), but it also
have:

  #define VHOST_SET_VRING_ENDIAN _IOW(VHOST_VIRTIO, 0x13, struct vhost_vring_state)
  #define VHOST_GET_VRING_ENDIAN _IOW(VHOST_VIRTIO, 0x14, struct vhost_vring_state)

A "get" operation that uses a "W" _IOC_DIR, and its implementation, uses
copy_to_user, it should've probably been _IOR().

Then:

  /* Base value where queue looks for available descriptors */
  #define VHOST_SET_VRING_BASE _IOW(VHOST_VIRTIO, 0x12, struct vhost_vring_state)
  /* Get accessor: reads index, writes value in num */
  #define VHOST_GET_VRING_BASE _IOWR(VHOST_VIRTIO, 0x12, struct vhost_vring_state)

So we'll need to use _IOC_DIR() to disambiguate the VHOST_VIRTIO ioctl
bautifier.

Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: Radim Krčmář <rkrcmar@redhat.com>
Cc: Wang Nan <wangnan0@huawei.com>
Link: http://lkml.kernel.org/n/tip-rq6q717ql7j2z7kuccafgq84@git.kernel.orgSigned-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
      8ff69577
    • A
      perf trace beautify ioctl: Beautify KVM ioctl's 'cmd' arg · 45717b7f
      由 Arnaldo Carvalho de Melo 提交于 
      Also trying a new approach, using a copy of uapi/linux/kvm.h we auto
generate the string tables, then include it in the ioctl cmd beautifier.

This way either the KVM developers will add the new commands to the
tools/ copy, like is happening with other areas of tools/include/ (bpf.h
comes to mind), or we'll be notified when building perf that our copy
drifted.

E.g., a tracing a process and its threads, but would work for system wide as
well, just drop that '-p 21238', to see ioctls for DRM, tty, sound, etc:

  # perf trace -e ioctl -p 21238 2>&1 | grep -v KVM_RUN
    7801.536 ( 0.003 ms): CPU 0/KVM/21276 ioctl(fd: 13<anon_inode:kvm-vm>, cmd: KVM_IRQ_LINE_STATUS, arg: 0x7f484c6c73c0) = 0
  <SNIP lots of the last one>
    7801.715 ( 0.001 ms): CPU 0/KVM/21276 ioctl(fd: 13<anon_inode:kvm-vm>, cmd: KVM_IRQ_LINE_STATUS, arg: 0x7f484c6c73e0) = 0
   11001.051 ( 0.008 ms): qemu-system-x8/21238 ioctl(fd: 13<anon_inode:kvm-vm>, cmd: KVM_SIGNAL_MSI, arg: 0x563c83379d70) = 1
   11001.225 ( 0.005 ms): qemu-system-x8/21238 ioctl(fd: 13<anon_inode:kvm-vm>, cmd: KVM_SIGNAL_MSI, arg: 0x563c83379d70) = 1
   10750.377 (249.963 ms): CPU 0/KVM/21276  ... [continued]: ioctl()) = 0
   11011.780 ( 0.015 ms): qemu-system-x8/21238 ioctl(fd: 13<anon_inode:kvm-vm>, cmd: KVM_SIGNAL_MSI, arg: 0x563c83379d90) = 1
   11011.929 ( 0.005 ms): qemu-system-x8/21238 ioctl(fd: 13<anon_inode:kvm-vm>, cmd: KVM_SIGNAL_MSI, arg: 0x7fff053e1000) = 1
   11012.090 ( 0.004 ms): qemu-system-x8/21238 ioctl(fd: 13<anon_inode:kvm-vm>, cmd: KVM_SIGNAL_MSI, arg: 0x563c83379d70) = 1
   11023.127 ( 0.020 ms): qemu-system-x8/21238 ioctl(fd: 13<anon_inode:kvm-vm>, cmd: KVM_SIGNAL_MSI, arg: 0x563c83379d90) = 1
   11000.483 (249.807 ms): CPU 0/KVM/21276  ... [continued]: ioctl()) = 0
   25620.877 ( 0.042 ms): qemu-system-x8/21238 ioctl(fd: 13<anon_inode:kvm-vm>, cmd: KVM_IRQ_LINE_STATUS, arg: 0x7fff053e1080) = 0
  <SNIP several of the last one>
   25621.025 ( 0.002 ms): qemu-system-x8/21238 ioctl(fd: 13<anon_inode:kvm-vm>, cmd: KVM_IRQ_LINE_STATUS, arg: 0x7fff053e10a0) = 0
   25500.803 (120.186 ms): CPU 0/KVM/21276  ... [continued]: ioctl()) = 0
   25621.078 ( 0.005 ms): CPU 0/KVM/21276 ioctl(fd: 13<anon_inode:kvm-vm>, cmd: KVM_IRQ_LINE_STATUS, arg: 0x7f484c6c73c0) = 0
  <SNIP lots of the last one>
   25621.346 ( 0.001 ms): CPU 0/KVM/21276 ioctl(fd: 13<anon_inode:kvm-vm>, cmd: KVM_IRQ_LINE_STATUS, arg: 0x7f484c6c73e0) = 0
   40456.997 ( 0.100 ms): qemu-system-x8/21238 ioctl(fd: 27</dev/vhost-net>, cmd: (WRITE, 0xaf, 0x30, 0x8), arg: 0x7fff053dffe0) = 0
   40457.100 ( 0.019 ms): qemu-system-x8/21238 ioctl(fd: 27</dev/vhost-net>, cmd: (WRITE, 0xaf, 0x30, 0x8), arg: 0x7fff053dffe0) = 0
   40457.133 ( 0.002 ms): qemu-system-x8/21238 ioctl(fd: 27</dev/vhost-net>, cmd: (READ|WRITE, 0xaf, 0x12, 0x8), arg: 0x7fff053dff60) = 0
   40457.139 ( 0.001 ms): qemu-system-x8/21238 ioctl(fd: 27</dev/vhost-net>, cmd: (READ|WRITE, 0xaf, 0x12, 0x8), arg: 0x7fff053dff60) = 0
   40458.503 ( 0.027 ms): qemu-system-x8/21238 ioctl(fd: 13<anon_inode:kvm-vm>, cmd: KVM_IOEVENTFD, arg: 0x7fff053dfc80) = 0
   40458.601 ( 0.030 ms): qemu-system-x8/21238 ioctl(fd: 13<anon_inode:kvm-vm>, cmd: KVM_IOEVENTFD, arg: 0x7fff053dfc80) = 0
   40458.649 ( 0.003 ms): qemu-system-x8/21238 ioctl(fd: 27</dev/vhost-net>, cmd: (WRITE, 0xaf, 0x21, 0x8), arg: 0x7fff053dff20) = 0
   40458.654 ( 0.002 ms): qemu-system-x8/21238 ioctl(fd: 27</dev/vhost-net>, cmd: (WRITE, 0xaf, 0x21, 0x8), arg: 0x7fff053dff20) = 0
   40458.657 ( 0.018 ms): qemu-system-x8/21238 ioctl(fd: 13<anon_inode:kvm-vm>, cmd: KVM_IRQFD, arg: 0x7fff053dff00  ) = 0
   40459.077 ( 0.017 ms): qemu-system-x8/21238 ioctl(fd: 13<anon_inode:kvm-vm>, cmd: KVM_IRQFD, arg: 0x7fff053dff00  ) = 0
   40459.123 ( 0.017 ms): qemu-system-x8/21238 ioctl(fd: 13<anon_inode:kvm-vm>, cmd: KVM_IOEVENTFD, arg: 0x7fff053dfd20) = 0
  <SNIP lots of the last one>
   40463.477 ( 0.013 ms): qemu-system-x8/21238 ioctl(fd: 13<anon_inode:kvm-vm>, cmd: KVM_IOEVENTFD, arg: 0x7fff053dfd20) = 0
   40464.874 ( 0.010 ms): qemu-system-x8/21238 ioctl(fd: 13<anon_inode:kvm-vm>, cmd: KVM_GET_VCPU_EVENTS, arg: 0x7fff053e0000) = 0
   40464.892 ( 0.048 ms): qemu-system-x8/21238 ioctl(fd: 12</dev/kvm>, cmd: KVM_CHECK_EXTENSION, arg: 0x4c           ) = 1
   40464.991 ( 0.002 ms): qemu-system-x8/21238 ioctl(fd: 13<anon_inode:kvm-vm>, cmd: KVM_GET_CLOCK, arg: 0x7fff053e0040) = 0
   40464.962 ( 0.013 ms): CPU 0/KVM/21276 ioctl(fd: 20<anon_inode:kvm-vcpu>, cmd: KVM_GET_MSRS, arg: 0x7f484c6c7670) = 1
   44540.437 ( 0.103 ms): qemu-system-x8/21238 ioctl(fd: 13<anon_inode:kvm-vm>, cmd: KVM_SET_GSI_ROUTING, arg: 0x563c7c93c000) = 0
   44540.544 ( 0.008 ms): qemu-system-x8/21238 ioctl(fd: 13<anon_inode:kvm-vm>, cmd: KVM_IRQFD, arg: 0x7fff053dfea0  ) = 0
   44540.555 ( 0.029 ms): qemu-system-x8/21238 ioctl(fd: 13<anon_inode:kvm-vm>, cmd: KVM_SET_GSI_ROUTING, arg: 0x563c7c93c000) = 0
   44540.586 ( 0.003 ms): qemu-system-x8/21238 ioctl(fd: 13<anon_inode:kvm-vm>, cmd: KVM_IRQFD, arg: 0x7fff053dfea0  ) = 0
   44540.592 ( 0.027 ms): qemu-system-x8/21238 ioctl(fd: 13<anon_inode:kvm-vm>, cmd: KVM_SET_GSI_ROUTING, arg: 0x563c7c93c000) = 0
   44540.625 ( 0.005 ms): qemu-system-x8/21238 ioctl(fd: 27</dev/vhost-net>, cmd: (WRITE, 0xaf, 0x21, 0x8), arg: 0x7fff053dfe20) = 0
   44540.639 ( 0.018 ms): qemu-system-x8/21238 ioctl(fd: 13<anon_inode:kvm-vm>, cmd: KVM_SET_GSI_ROUTING, arg: 0x563c7c93c000) = 0
   44540.658 ( 0.003 ms): qemu-system-x8/21238 ioctl(fd: 27</dev/vhost-net>, cmd: (WRITE, 0xaf, 0x21, 0x8), arg: 0x7fff053dfe20) = 0
   44540.686 ( 0.015 ms): qemu-system-x8/21238 ioctl(fd: 13<anon_inode:kvm-vm>, cmd: KVM_IOEVENTFD, arg: 0x7fff053dfbe0) = 0
   44540.727 ( 0.014 ms): qemu-system-x8/21238 ioctl(fd: 13<anon_inode:kvm-vm>, cmd: KVM_IOEVENTFD, arg: 0x7fff053dfbe0) = 0
   44540.748 ( 0.005 ms): qemu-system-x8/21238 ioctl(fd: 27</dev/vhost-net>, cmd: (WRITE, 0xaf, 0, 0x8), arg: 0x7fff053dfe88) = 0
   44540.754 ( 0.026 ms): qemu-system-x8/21238 ioctl(fd: 27</dev/vhost-net>, cmd: (WRITE, 0xaf, 0x3, 0x8), arg: 0x563c7c906870) = 0
   44540.783 ( 0.002 ms): qemu-system-x8/21238 ioctl(fd: 27</dev/vhost-net>, cmd: (WRITE, 0xaf, 0x10, 0x8), arg: 0x7fff053dff00) = 0
   44540.787 ( 0.002 ms): qemu-system-x8/21238 ioctl(fd: 27</dev/vhost-net>, cmd: (WRITE, 0xaf, 0x12, 0x8), arg: 0x7fff053dff00) = 0
   44540.793 ( 0.002 ms): qemu-system-x8/21238 ioctl(fd: 27</dev/vhost-net>, cmd: (WRITE, 0xaf, 0x11, 0x28), arg: 0x7fff053dfe70) = 0
   44540.796 ( 0.010 ms): qemu-system-x8/21238 ioctl(fd: 27</dev/vhost-net>, cmd: (WRITE, 0xaf, 0x20, 0x8), arg: 0x7fff053dfef0) = 0
   44540.811 ( 0.002 ms): qemu-system-x8/21238 ioctl(fd: 27</dev/vhost-net>, cmd: (WRITE, 0xaf, 0x10, 0x8), arg: 0x7fff053dff00) = 0
   44540.814 ( 0.002 ms): qemu-system-x8/21238 ioctl(fd: 27</dev/vhost-net>, cmd: (WRITE, 0xaf, 0x12, 0x8), arg: 0x7fff053dff00) = 0
   44540.819 ( 0.002 ms): qemu-system-x8/21238 ioctl(fd: 27</dev/vhost-net>, cmd: (WRITE, 0xaf, 0x11, 0x28), arg: 0x7fff053dfe70) = 0
   44540.822 ( 0.005 ms): qemu-system-x8/21238 ioctl(fd: 27</dev/vhost-net>, cmd: (WRITE, 0xaf, 0x20, 0x8), arg: 0x7fff053dfef0) = 0
   44540.837 ( 0.006 ms): qemu-system-x8/21238 ioctl(fd: 27</dev/vhost-net>, cmd: (WRITE, 0xaf, 0x30, 0x8), arg: 0x7fff053dff80) = 0
   44540.862 ( 0.004 ms): qemu-system-x8/21238 ioctl(fd: 27</dev/vhost-net>, cmd: (WRITE, 0xaf, 0x30, 0x8), arg: 0x7fff053dff80) = 0
   44540.887 ( 0.014 ms): qemu-system-x8/21238 ioctl(fd: 13<anon_inode:kvm-vm>, cmd: KVM_IOEVENTFD, arg: 0x7fff053dfd00) = 0
  <SNIP lots of the last one>
   44542.756 ( 0.020 ms): qemu-system-x8/21238 ioctl(fd: 13<anon_inode:kvm-vm>, cmd: KVM_IOEVENTFD, arg: 0x7fff053dfd00) = 0
   44542.809 ( 0.007 ms): qemu-system-x8/21238 ioctl(fd: 13<anon_inode:kvm-vm>, cmd: KVM_SET_VCPU_EVENTS, arg: 0x7fff053dffb0) = 0
   44542.819 ( 0.003 ms): qemu-system-x8/21238 ioctl(fd: 12</dev/kvm>, cmd: KVM_CHECK_EXTENSION, arg: 0x4c           ) = 1
   44543.016 ( 0.004 ms): qemu-system-x8/21238 ioctl(fd: 13<anon_inode:kvm-vm>, cmd: KVM_SET_CLOCK, arg: 0x7fff053dfff0) = 0
   44543.022 ( 0.008 ms): qemu-system-x8/21238 ioctl(fd: 20<anon_inode:kvm-vcpu>, cmd: KVM_KVMCLOCK_CTRL             ) = 0
   46952.502 ( 0.010 ms): qemu-system-x8/21238 ioctl(fd: 13<anon_inode:kvm-vm>, cmd: KVM_SIGNAL_MSI, arg: 0x563c83379d70) = 1
   46829.292 (249.860 ms): CPU 0/KVM/21276  ... [continued]: ioctl()) = 0
  ^C
[root@jouet linux]#

Since there are clashes in _IOC_NR() for some cases, notably ioctls with
PPC_ and ARM_ in its name and some that depend on some internal state to
be valid, but use the same number as others, those were removed in the
shell script that builds the table, tools/perf/trace/beauty/kvm_ioctl.sh.

Since so far we're supporting only x86 in the 'cmd' ioctl arg beautifier
in perf trace, we can leave fully supporting these ioctls for later.

There are some more to handle here, notably the one for /dev/vhost-net, will
come later.

Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: Radim Krčmář <rkrcmar@redhat.com>
Cc: Wang Nan <wangnan0@huawei.com>
Link: http://lkml.kernel.org/n/tip-zxhebe579n338d7qrnjoctes@git.kernel.orgSigned-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
      45717b7f
    • A
      perf trace beautify ioctl: Beautify sound ioctl's 'cmd' arg · 2c3e9629
      由 Arnaldo Carvalho de Melo 提交于 
      This time we try a new approach, using a copy of uapi/sound/asound.h we
auto generate the string tables, then include it in the ioctl cmd
beautifier.

This way either the sound developers will add the new commands to the
tools/ copy, like is happening with other areas of tools/include/ (bpf.h
comes to mind), or we'll be notified when building perf that our copy
drifted.

E.g.:

  # perf trace -p 22084 -e ioctl 2>&1 | head -5
     0.000 ( 0.068 ms): alsa-sink-ALC3/22084 ioctl(fd: 49</dev/snd/pcmC1D0p>, cmd: SNDRV_PCM_HWSYNC, arg: 0x557f8d7fa0f0) = 0
     0.344 ( 0.041 ms): alsa-sink-ALC3/22084 ioctl(fd: 46</dev/snd/controlC1>, cmd: SNDRV_CTL_ELEM_READ, arg: 0x7fe764018ee0) = 0
     0.403 ( 0.011 ms): alsa-sink-ALC3/22084 ioctl(fd: 49</dev/snd/pcmC1D0p>, cmd: SNDRV_PCM_HWSYNC, arg: 0x557f8d7fa0f0) = 0
     0.427 ( 0.009 ms): alsa-sink-ALC3/22084 ioctl(fd: 49</dev/snd/pcmC1D0p>, cmd: SNDRV_PCM_STATUS_EXT, arg: 0x7fe76c2e0b30) = 0
     2.461 ( 0.042 ms): alsa-sink-ALC3/22084 ioctl(fd: 49</dev/snd/pcmC1D0p>, cmd: SNDRV_PCM_HWSYNC, arg: 0x557f8d7fa0f0) = 0
  #

Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Wang Nan <wangnan0@huawei.com>
Link: http://lkml.kernel.org/n/tip-8zuyf3e3u6jjcb2xzerw0kdi@git.kernel.orgSigned-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
      2c3e9629
    • A
      perf trace beauty ioctl: Beautify DRM ioctl cmds · ef9811f0
      由 Arnaldo Carvalho de Melo 提交于 
      This time we try a new approach, using uapi/drm/ copies of drm.h and
i915_drm.h we auto generate the string tables, then include it in the
ioctl cmd beautifier.

This way either the DRM developers will add the new commands to the
tools/ copy, like is happening with other areas of tools/include/ (bpf.h
comes to mind), or we'll be notified when building perf that our copy
drifted.

Either way the time from a new command being added to when 'perf trace'
gets to know it is greatly shortened, for instance:

  # strace -p 22401 -e ioctl
  ioctl(8, DRM_IOCTL_I915_GEM_BUSY, 0x7ffc934f7600) = 0
  ioctl(8, DRM_IOCTL_I915_GEM_SET_DOMAIN, 0x7ffc934f7550) = 0
  ioctl(8, DRM_IOCTL_I915_GEM_SW_FINISH, 0x7ffc934f76e0) = 0
  ioctl(8, DRM_IOCTL_I915_GEM_SW_FINISH, 0x7ffc934f7780) = 0
  ioctl(8, _IOC(_IOC_READ|_IOC_WRITE, 0x64, 0x69, 0x40), 0x7ffc934f7700) = 0
  ioctl(8, DRM_IOCTL_I915_GEM_SET_DOMAIN, 0x7ffc934f7780) = 0
  ioctl(8, DRM_IOCTL_I915_GEM_MADVISE, 0x7ffc934f76f0) = 0
  ioctl(8, DRM_IOCTL_I915_GEM_BUSY, 0x7ffc934f76c0) = 0
  ioctl(8, DRM_IOCTL_I915_GEM_MADVISE, 0x7ffc934f76b0) = 0
  ioctl(8, DRM_IOCTL_I915_GEM_SET_DOMAIN, 0x7ffc934f76d0) = 0
  ioctl(8, DRM_IOCTL_MODE_ADDFB, 0x7ffc934f7880) = 0
  ioctl(8, DRM_IOCTL_MODE_PAGE_FLIP, 0x7ffc934f77d0) = 0
  ^Cstrace: Process 22401 detached

versus:

  # perf trace -p 22401 -e ioctl
  1010.856 (0.006 ms): gnome-shell/22401 ioctl(fd: 8</dev/dri/card0>, cmd: DRM_I915_GEM_BUSY, arg: 0x7ffc934f7600) = 0
  1010.865 (0.003 ms): gnome-shell/22401 ioctl(fd: 8</dev/dri/card0>, cmd: DRM_I915_GEM_SET_DOMAIN, arg: 0x7ffc934f7550) = 0
  1010.872 (0.002 ms): gnome-shell/22401 ioctl(fd: 8</dev/dri/card0>, cmd: DRM_I915_GEM_SW_FINISH, arg: 0x7ffc934f76e0) = 0
  1010.939 (0.015 ms): gnome-shell/22401 ioctl(fd: 8</dev/dri/card0>, cmd: DRM_I915_GEM_SW_FINISH, arg: 0x7ffc934f7780) = 0
  1010.959 (0.085 ms): gnome-shell/22401 ioctl(fd: 8</dev/dri/card0>, cmd: DRM_I915_GEM_EXECBUFFER2, arg: 0x7ffc934f7700) = 0
  1011.048 (0.003 ms): gnome-shell/22401 ioctl(fd: 8</dev/dri/card0>, cmd: DRM_I915_GEM_SET_DOMAIN, arg: 0x7ffc934f7780) = 0
  1011.056 (0.002 ms): gnome-shell/22401 ioctl(fd: 8</dev/dri/card0>, cmd: DRM_I915_GEM_MADVISE, arg: 0x7ffc934f76f0) = 0
  1011.060 (0.002 ms): gnome-shell/22401 ioctl(fd: 8</dev/dri/card0>, cmd: DRM_I915_GEM_BUSY, arg: 0x7ffc934f76c0) = 0
  1011.064 (0.003 ms): gnome-shell/22401 ioctl(fd: 8</dev/dri/card0>, cmd: DRM_I915_GEM_MADVISE, arg: 0x7ffc934f76b0) = 0
  1011.068 (0.002 ms): gnome-shell/22401 ioctl(fd: 8</dev/dri/card0>, cmd: DRM_I915_GEM_SET_DOMAIN, arg: 0x7ffc934f76d0) = 0
  1011.074 (0.009 ms): gnome-shell/22401 ioctl(fd: 8</dev/dri/card0>, cmd: DRM_MODE_ADDFB, arg: 0x7ffc934f7880 ) = 0
  1011.096 (0.072 ms): gnome-shell/22401 ioctl(fd: 8</dev/dri/card0>, cmd: DRM_MODE_PAGE_FLIP, arg: 0x7ffc934f77d0) = 0
^C[root@jouet linux]#

Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Wang Nan <wangnan0@huawei.com>
Link: http://lkml.kernel.org/n/tip-mly2d7v9kf28rso81dijbixq@git.kernel.orgSigned-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
      ef9811f0
  23. 01 8月, 2017 1 次提交
    • A
      perf trace beauty ioctl: Improve 'cmd' beautifier · 1cc47f2d
      由 Arnaldo Carvalho de Melo 提交于 
      By using the _IOC_(DIR,NR,TYPE,SIZE) macros to lookup a 'type' keyed
table that then gets indexed by 'nr', falling back to a notation similar
to the one used by 'strace', only more compact, i.e.:

   474.356 ( 0.007 ms): gnome-shell/22401 ioctl(fd: 8</dev/dri/card0>, cmd: (READ|WRITE, 0x64, 0xae, 0x1c), arg: 0x7ffc934f7880) = 0
   474.369 ( 0.053 ms): gnome-shell/22401 ioctl(fd: 8</dev/dri/card0>, cmd: (READ|WRITE, 0x64, 0xb0, 0x18), arg: 0x7ffc934f77d0) = 0
   505.055 ( 0.014 ms): gnome-shell/22401 ioctl(fd: 8</dev/dri/card0>, cmd: (READ|WRITE, 0x64, 0xaf, 0x4), arg: 0x7ffc934f741c) = 0

This also moves it out of builtin-trace.c and into trace/beauty/ioctl.c
to better compartimentalize all these formatters.

Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Wang Nan <wangnan0@huawei.com>
Link: http://lkml.kernel.org/n/tip-s3enursdxsvnhdomh6qlte4g@git.kernel.orgSigned-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
      1cc47f2d
  25. 20 7月, 2017 3 次提交
    • A
      perf trace beauty clone: Suppress unused args according to 'flags' arg · 15bed274
      由 Arnaldo Carvalho de Melo 提交于 
      The 'parent_tidptr', 'child_tidptr' and 'tls' arguments to the 'clone'
syscall are only used when certain flags are set in 'flags', suppress
them when those aren't there.

E.g:

   9886.919 (0.236 ms): fetchmail/19298 clone(flags: CHILD_CLEARTID|CHILD_SETTID|0x11, child_stack: 0, child_tidptr: 0x7fe43f468590) = 19608 (fetchmail)
  12876.052 (0.249 ms): qemu-system-x8/21238 clone(flags: VM|FS|FILES|SIGHAND|THREAD|SYSVSEM|SETTLS|PARENT_SETTID|CHILD_CLEARTID, child_stack: 0x7f48117fc770, parent_tidptr: 0x7f48117ff9d0, child_tidptr: 0x7f48117ff9d0, tls: 0x7f48117ff700) = 19611 (qemu-system-x86)
  12876.555 (0.048 ms): worker/19611 clone(flags: VM|FS|FILES|SIGHAND|THREAD|SYSVSEM|SETTLS|PARENT_SETTID|CHILD_CLEARTID, child_stack: 0x7f480f7f8770, parent_tidptr: 0x7f480f7fb9d0, child_tidptr: 0x7f480f7fb9d0, tls: 0x7f480f7fb700) = 19612 (worker)
  16575.240 (0.469 ms): fetchmail/19298 clone(flags: CHILD_CLEARTID|CHILD_SETTID|0x11, child_stack: 0, child_tidptr: 0x7fe43f468590) = 19613 (fetchmail)
  20797.270 (0.335 ms): fetchmail/19298 clone(flags: CHILD_CLEARTID|CHILD_SETTID|0x11, child_stack: 0, child_tidptr: 0x7fe43f468590) = 19614 (fetchmail)
  21228.585 (0.501 ms): vim/19519 clone(flags: CHILD_CLEARTID|CHILD_SETTID|0x11, child_stack: 0, child_tidptr: 0x7fbad6ac27d0) = 19615 (vim)
  21232.193 (0.137 ms): bash/19615 clone(flags: CHILD_CLEARTID|CHILD_SETTID|0x11, child_stack: 0, child_tidptr: 0x7fad8bff49d0) = 19616 (bash)

Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Wang Nan <wangnan0@huawei.com>
Link: http://lkml.kernel.org/n/tip-0um93djul9knf239gwa5mpcb@git.kernel.orgSigned-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
      15bed274
    • A
      perf trace beauty clone: Beautify syscall arguments · 33396a3a
      由 Arnaldo Carvalho de Melo 提交于 
      Now, syswide tracing, selected entries:

  # trace -e clone
  24417.203 ( 0.158 ms): bash/11323 clone(flags: CHILD_CLEARTID|CHILD_SETTID|0x11, child_stack: 0, parent_tidptr: 0, child_tidptr: 0x7f0778e5c9d0, tls: 0x7f0778e5c700) = 11325 (bash)
          ? (     ?   ): bash/11325  ... [continued]: clone()) = 0
  24419.355 ( 0.093 ms): bash/10586 clone(flags: CHILD_CLEARTID|CHILD_SETTID|0x11, child_stack: 0, parent_tidptr: 0, child_tidptr: 0x7f0778e5c9d0, tls: 0x7f0778e5c700) = 11326 (bash)
          ? (     ?   ): bash/11326  ... [continued]: clone()) = 0
  24419.744 ( 0.102 ms): bash/11326 clone(flags: CHILD_CLEARTID|CHILD_SETTID|0x11, child_stack: 0, parent_tidptr: 0, child_tidptr: 0x7f0778e5c9d0, tls: 0x7f0778e5c700) = 11327 (bash)
          ? (     ?   ): bash/11327  ... [continued]: clone()) = 0
  24420.138 ( 0.105 ms): bash/11327 clone(flags: CHILD_CLEARTID|CHILD_SETTID|0x11, child_stack: 0, parent_tidptr: 0, child_tidptr: 0x7f0778e5c9d0, tls: 0x7f0778e5c700) = 11328 (bash)
          ? (     ?   ): bash/11328  ... [continued]: clone()) = 0
  35747.722 ( 0.044 ms): gpg-agent/18087 clone(flags: VM|FS|FILES|SIGHAND|THREAD|SYSVSEM|SETTLS|PARENT_SETTID|CHILD_CLEARTID, child_stack: 0x7ff0755f6ff0, parent_tidptr: 0x7ff0755f79d0, child_tidptr: 0x7ff0755f79d0, tls: 0x7ff0755f7700) = 11329 (gpg-agent)
          ? (     ?   ): gpg-agent/11329  ... [continued]: clone()) = 0
  35748.359 ( 0.022 ms): gpg-agent/18087 clone(flags: VM|FS|FILES|SIGHAND|THREAD|SYSVSEM|SETTLS|PARENT_SETTID|CHILD_CLEARTID, child_stack: 0x7ff075df7ff0, parent_tidptr: 0x7ff075df89d0, child_tidptr: 0x7ff075df89d0, tls: 0x7ff075df8700) = 11330 (gpg-agent)
          ? (     ?   ): gpg-agent/11330  ... [continued]: clone()) = 0
  35781.422 ( 0.452 ms): NetworkManager/1112 clone(flags: VM|FS|FILES|SIGHAND|THREAD|SYSVSEM|SETTLS|PARENT_SETTID|CHILD_CLEARTID, child_stack: 0x7f2f1fffedb0, parent_tidptr: 0x7f2f1ffff9d0, child_tidptr: 0x7f2f1ffff9d0, tls: 0x7f2f1ffff700) = 11331 (NetworkManager)
          ? (     ?   ): NetworkManager/11331  ... [continued]: clone()) = 0

Need to improve the formatting of the second return, to the child, this
cset only focused on the argument formatting.

If we trace just one pid:

  # trace -e clone -p 19863
     0.349 ( 0.025 ms): Chrome_IOThrea/19863 clone(flags: VM|FS|FILES|SIGHAND|THREAD|SYSVSEM|SETTLS|PARENT_SETTID|CHILD_CLEARTID, child_stack: 0x7ffb84eaac70, parent_tidptr: 0x7ffb84eab9d0, child_tidptr: 0x7ffb84eab9d0, tls: 0x7ffb84eab700) = 11637 (Chrome_IOThread)
     0.392 ( 0.013 ms): Chrome_IOThrea/19863 clone(flags: VM|FS|FILES|SIGHAND|THREAD|SYSVSEM|SETTLS|PARENT_SETTID|CHILD_CLEARTID, child_stack: 0x7ffb664b8c70, parent_tidptr: 0x7ffb664b99d0, child_tidptr: 0x7ffb664b99d0, tls: 0x7ffb664b9700) = 11638 (Chrome_IOThread)
     0.573 ( 0.015 ms): Chrome_IOThrea/19863 clone(flags: VM|FS|FILES|SIGHAND|THREAD|SYSVSEM|SETTLS|PARENT_SETTID|CHILD_CLEARTID, child_stack: 0x7ffb6046cc70, parent_tidptr: 0x7ffb6046d9d0, child_tidptr: 0x7ffb6046d9d0, tls: 0x7ffb6046d700) = 11639 (Chrome_IOThread)
     0.617 ( 0.014 ms): Chrome_IOThrea/19863 clone(flags: VM|FS|FILES|SIGHAND|THREAD|SYSVSEM|SETTLS|PARENT_SETTID|CHILD_CLEARTID, child_stack: 0x7ffb730dcc70, parent_tidptr: 0x7ffb730dd9d0, child_tidptr: 0x7ffb730dd9d0, tls: 0x7ffb730dd700) = 11640 (Chrome_IOThread)
     4.350 ( 0.065 ms): Chrome_IOThrea/19863 clone(flags: VM|FS|FILES|SIGHAND|THREAD|SYSVSEM|SETTLS|PARENT_SETTID|CHILD_CLEARTID, child_stack: 0x7ffb720d9c70, parent_tidptr: 0x7ffb720da9d0, child_tidptr: 0x7ffb720da9d0, tls: 0x7ffb720da700) = 11642 (Chrome_IOThread)
     5.642 ( 0.079 ms): Chrome_IOThrea/19863 clone(flags: VM|FS|FILES|SIGHAND|THREAD|SYSVSEM|SETTLS|PARENT_SETTID|CHILD_CLEARTID, child_stack: 0x7ffb718d8c70, parent_tidptr: 0x7ffb718d99d0, child_tidptr: 0x7ffb718d99d0, tls: 0x7ffb718d9700) = 11643 (Chrome_IOThread)
^C#

We'll also have to fix the argument ordering in different arches,
probably having multiple syscall_fmt entries with each possible order
and then use perf_evsel__env_arch() (if dealing with a perf.data file)
or the current system info, for live sessions.

Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Wang Nan <wangnan0@huawei.com>
Link: http://lkml.kernel.org/n/tip-am068uyubgj83snepolwhbfe@git.kernel.orgSigned-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
      33396a3a
    • A
      perf trace beauty mmap: Ignore 'fd' and 'offset' args for MAP_ANONYMOUS · d57da8c9
      由 Arnaldo Carvalho de Melo 提交于 
      Just suppress them, not used by the kernel.

Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Wang Nan <wangnan0@huawei.com>
Link: http://lkml.kernel.org/n/tip-atpt07y2x9a8ttlwja94ow3j@git.kernel.orgSigned-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
      d57da8c9
  27. 19 7月, 2017 18 次提交