linux-perf-users.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Ravi Bangoria <ravi.bangoria@linux.vnet.ibm.com>
To: Hendrik Brueckner <brueckner@linux.vnet.ibm.com>
Cc: Arnaldo Carvalho de Melo <acme@kernel.org>,
	Thomas-Mich Richter <tmricht@linux.vnet.ibm.com>,
	Michael Petlan <mpetlan@redhat.com>, Jiri Olsa <jolsa@redhat.com>,
	linux-perf-users@vger.kernel.org, linux-s390@vger.kernel.org,
	Ravi Bangoria <ravi.bangoria@linux.vnet.ibm.com>
Subject: Re: [PATCH v2 0/5] perf trace: Introduce arch-specific errno code/name mappings
Date: Thu, 25 Jan 2018 13:30:13 +0530	[thread overview]
Message-ID: <23f01b64-0584-bbaf-2418-e34c229bcce0@linux.vnet.ibm.com> (raw)
In-Reply-To: <1516352177-11106-1-git-send-email-brueckner@linux.vnet.ibm.com>

Hi Hendrik,

On 01/19/2018 02:26 PM, Hendrik Brueckner wrote:
> Introduce an interface to perf to resolve errno codes to their names.
> This is used by perf trace to display the errno name, for example, ENOENT,
> for an system call.  The previous implementation used the errno_to_name()
> function from the audit library.
>
> With this change, a table of errno codes and their names are created for
> each architecture on which perf is supported (others can be easily added
> if necessary).  The errno_to_name() function from the audit library is
> replaced with an arch-specific one, arch_errno_to_name().
>
> The benefit is that perf trace does not longer have a direct dependency
> for the audit library and the perf trace output can be displayed across
> architectures.  For example, collect trace output on s390 and display it
> on x86.

I haven't tested this but are you sure that we can do cross-arch perf-trace
after this patch set?

Why I'm curious is because, we embed arch specific syscall table in the perf
binary at compile time:

    $ cat tools/perf/util/syscalltbl.c

    #if defined(__x86_64__)
    #include <asm/syscalls_64.c>
    const int syscalltbl_native_max_id = SYSCALLTBL_x86_64_MAX_ID;
    static const char **syscalltbl_native = syscalltbl_x86_64;
    #elif defined(__s390x__)
    #include <asm/syscalls_64.c>
    const int syscalltbl_native_max_id = SYSCALLTBL_S390_64_MAX_ID;
    static const char **syscalltbl_native = syscalltbl_s390_64;
    #endif

If we don't have, let's say, s390 syscall table in perf compiled on x86 machine,
we won't be able to do perf-trace across architecture. Right ?

Please let me know if I'm missing anything.

Thanks,
Ravi

P.S.: Seems I can remove libaudit dependency for syscalls on PowerPC as well.
Will prepare patches for that very soon.

>
> The audit library dependency can be further reduced and completely removed
> for architectures that support system call tables in perf (see last commit).
>
> See also discussion on:
> - https://www.spinics.net/lists/linux-perf-users/msg04782.html
> - https://www.spinics.net/lists/linux-perf-users/msg05005.html
>
> Please note that this patch set is bases on the perf/core branch.
>
> Changes in v2:
> As the arch_errno_to_name() mapping is used by perf trace only,
> move all the scripts and generated files from tools/perf/util/
> to the tools/perf/trace/beauty/{generated/} directories.
> See also https://www.spinics.net/lists/linux-perf-users/msg05037.html
>
>
> Hendrik Brueckner (5):
>   tools include arch: Grab a copy of errno.h for arch's supported by
>     perf
>   tools include asm-generic: Grab errno.h and errno-base.h
>   perf util: Introduce architecture specific errno/name mapping
>   perf trace: Obtain errno values by using arch_syscalls__strerrno()
>   perf trace: Remove audit-libs dependency if syscall tables are present
>
>  tools/arch/alpha/include/uapi/asm/errno.h   | 128 +++++++++++++++++++++++++++
>  tools/arch/mips/include/asm/errno.h         |  17 ++++
>  tools/arch/mips/include/uapi/asm/errno.h    | 130 ++++++++++++++++++++++++++++
>  tools/arch/parisc/include/uapi/asm/errno.h  | 128 +++++++++++++++++++++++++++
>  tools/arch/powerpc/include/uapi/asm/errno.h |  10 +++
>  tools/arch/sparc/include/uapi/asm/errno.h   | 118 +++++++++++++++++++++++++
>  tools/arch/x86/include/uapi/asm/errno.h     |   1 +
>  tools/include/uapi/asm-generic/errno-base.h |  40 +++++++++
>  tools/include/uapi/asm-generic/errno.h      | 123 ++++++++++++++++++++++++++
>  tools/perf/Makefile.config                  |  10 ++-
>  tools/perf/Makefile.perf                    |  13 ++-
>  tools/perf/builtin-help.c                   |   2 +-
>  tools/perf/builtin-trace.c                  |  14 ++-
>  tools/perf/check-headers.sh                 |   9 ++
>  tools/perf/perf.c                           |   4 +-
>  tools/perf/trace/beauty/arch_errno_names.c  |   1 +
>  tools/perf/trace/beauty/arch_errno_names.sh | 100 +++++++++++++++++++++
>  tools/perf/trace/beauty/beauty.h            |   2 +
>  tools/perf/util/generate-cmdlist.sh         |   2 +-
>  19 files changed, 841 insertions(+), 11 deletions(-)
>  create mode 100644 tools/arch/alpha/include/uapi/asm/errno.h
>  create mode 100644 tools/arch/mips/include/asm/errno.h
>  create mode 100644 tools/arch/mips/include/uapi/asm/errno.h
>  create mode 100644 tools/arch/parisc/include/uapi/asm/errno.h
>  create mode 100644 tools/arch/powerpc/include/uapi/asm/errno.h
>  create mode 100644 tools/arch/sparc/include/uapi/asm/errno.h
>  create mode 100644 tools/arch/x86/include/uapi/asm/errno.h
>  create mode 100644 tools/include/uapi/asm-generic/errno-base.h
>  create mode 100644 tools/include/uapi/asm-generic/errno.h
>  create mode 100644 tools/perf/trace/beauty/arch_errno_names.c
>  create mode 100755 tools/perf/trace/beauty/arch_errno_names.sh
>

  parent reply	other threads:[~2018-01-25  7:58 UTC|newest]

Thread overview: 10+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-01-19  8:56 [PATCH v2 0/5] perf trace: Introduce arch-specific errno code/name mappings Hendrik Brueckner
2018-01-19  8:56 ` [PATCH v2 1/5] tools include arch: Grab a copy of errno.h for arch's supported by perf Hendrik Brueckner
2018-01-19  8:56 ` [PATCH v2 2/5] tools include asm-generic: Grab errno.h and errno-base.h Hendrik Brueckner
2018-01-19  8:56 ` [PATCH v2 3/5] perf util: Introduce architecture specific errno/name mapping Hendrik Brueckner
2018-01-19  8:56 ` [PATCH v2 4/5] perf trace: Obtain errno values by using arch_syscalls__strerrno() Hendrik Brueckner
2018-01-19  8:56 ` [PATCH v2 5/5] perf trace: Remove audit-libs dependency if syscall tables are present Hendrik Brueckner
2018-01-19 14:04   ` Arnaldo Carvalho de Melo
2018-01-21 12:27     ` Arnaldo Carvalho de Melo
2018-01-25  8:00 ` Ravi Bangoria [this message]
2018-01-25  9:28   ` [PATCH v2 0/5] perf trace: Introduce arch-specific errno code/name mappings Arnaldo Carvalho de Melo

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=23f01b64-0584-bbaf-2418-e34c229bcce0@linux.vnet.ibm.com \
    --to=ravi.bangoria@linux.vnet.ibm.com \
    --cc=acme@kernel.org \
    --cc=brueckner@linux.vnet.ibm.com \
    --cc=jolsa@redhat.com \
    --cc=linux-perf-users@vger.kernel.org \
    --cc=linux-s390@vger.kernel.org \
    --cc=mpetlan@redhat.com \
    --cc=tmricht@linux.vnet.ibm.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).