Mirror of the gdb-patches mailing list
 help / color / mirror / Atom feed
From: Ilya Leoshkevich via Gdb-patches <gdb-patches@sourceware.org>
To: Tom Tromey <tromey@adacore.com>, Andrew Burgess <aburgess@redhat.com>
Cc: Ulrich Weigand <ulrich.weigand@de.ibm.com>,
	Andreas Arnez <arnez@linux.ibm.com>,
	gdb-patches@sourceware.org, Pedro Alves <pedro@palves.net>,
	Ilya Leoshkevich <iii@linux.ibm.com>
Subject: [PATCH v2 0/7] gdb.perf/: Add JIT performance test
Date: Tue, 31 May 2022 00:11:40 +0200	[thread overview]
Message-ID: <20220530221147.1991835-1-iii@linux.ibm.com> (raw)

Hi,

This series implements a JIT performance test as discussed in [1].
Patch 1 is a small fix, patches 2-6 are preparations and patch 7 is
the test itself.

ChangeLog:
v1: https://sourceware.org/pipermail/gdb-patches/2022-May/189427.html
v1 -> v2:
* Make sure shared objects contain virtual addresses they are loaded
  at by reusing parts of jit-elf-helpers.exp.  By default, virtual
  addresses in all shared objects are relative to 0 - this creates a
  lots of overlaps and forces sort_cmp () to go through the slow "Sort
  on sequence number of the objfile in the chain" path, which is not a
  useful performance testing scenario.
* Fix DEFAULT_BINARY_LINK_WITH_SHLIBS style (Andrew).
* Rename SHLIB to TOTAL_NR_SHLIBS (Andrew).
* Add SOLIB_COUNT (Andrew).
* Drop "gdb.perf/: Compile the binary with -DSHLIB", it's superseded by
  "Allow adjusting GenPerfTest compile options".

The test can be used as follows:

$ make build-perf -j"$(nproc)" RUNTESTFLAGS=jit.exp
$ make check-perf RUNTESTFLAGS=jit-check.exp GDB_PERFTEST_MODE=run

The results can be converted to gnuplot-friendly format as follows:

$ perl -ne 'if (/:register_times:(\d+) (.+)/) { print "$1 $2\n"; }' \
      <gdb.perf/outputs/jit/pieces/perftest.sum >jit.txt

I've uploaded the results to [2].  They are similar to the ones from
Andrew and show that [1] is an improvement.  The third result shown on
the chart is from [3] - with it I can pass the 10k JITed objects mark
and debug CoreCLR JIT with GDB.  It's still WIP though - the testsuite
shows a couple of failures.

Best regards,
Ilya

[1] https://sourceware.org/pipermail/gdb-patches/2022-May/189341.html
[2] https://github.com/iii-i/binutils-gdb/blob/section-map-20220530-2/gdb/testsuite/jit.png
[3] https://github.com/iii-i/binutils-gdb/commits/section-map-20220530-2

Ilya Leoshkevich (7):
  gdb.perf/: Fix tcl_string_list_to_python_list {x}
  gdb.perf/: Add binary_link_with_shlibs setting to GenPerfTest
  gdb.perf/: Allow adjusting GenPerfTest compile options
  gdb.base/: Introduce jit-protocol-util.h
  gdb.base/: Introduce jit_compile_options
  gdb.base/: Introduce n_jit_so_address
  gdb.perf/: Add JIT performance test

 gdb/testsuite/gdb.base/jit-elf-fork-main.c | 48 ++----------
 gdb/testsuite/gdb.base/jit-elf-main.c      | 48 ++----------
 gdb/testsuite/gdb.base/jit-elf-util.h      | 17 +++++
 gdb/testsuite/gdb.base/jit-protocol-util.h | 74 ++++++++++++++++++
 gdb/testsuite/gdb.base/jit-reader-host.c   |  8 +-
 gdb/testsuite/gdb.perf/jit-check.exp       | 25 +++++++
 gdb/testsuite/gdb.perf/jit-check.py        | 60 +++++++++++++++
 gdb/testsuite/gdb.perf/jit-solib.c         | 27 +++++++
 gdb/testsuite/gdb.perf/jit.c               | 87 ++++++++++++++++++++++
 gdb/testsuite/gdb.perf/jit.exp             | 79 ++++++++++++++++++++
 gdb/testsuite/lib/gen-perf-test.exp        | 54 +++++++++++---
 gdb/testsuite/lib/jit-elf-helpers.exp      | 22 ++++--
 gdb/testsuite/lib/perftest.exp             |  2 +-
 13 files changed, 439 insertions(+), 112 deletions(-)
 create mode 100644 gdb/testsuite/gdb.base/jit-protocol-util.h
 create mode 100644 gdb/testsuite/gdb.perf/jit-check.exp
 create mode 100644 gdb/testsuite/gdb.perf/jit-check.py
 create mode 100644 gdb/testsuite/gdb.perf/jit-solib.c
 create mode 100644 gdb/testsuite/gdb.perf/jit.c
 create mode 100644 gdb/testsuite/gdb.perf/jit.exp

-- 
2.35.3


             reply	other threads:[~2022-05-30 22:12 UTC|newest]

Thread overview: 10+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-05-30 22:11 Ilya Leoshkevich via Gdb-patches [this message]
2022-05-30 22:11 ` [PATCH v2 1/7] gdb.perf/: Fix tcl_string_list_to_python_list {x} Ilya Leoshkevich via Gdb-patches
2022-05-30 22:11 ` [PATCH v2 2/7] gdb.perf/: Add binary_link_with_shlibs setting to GenPerfTest Ilya Leoshkevich via Gdb-patches
2022-05-30 22:11 ` [PATCH v2 3/7] gdb.perf/: Allow adjusting GenPerfTest compile options Ilya Leoshkevich via Gdb-patches
2022-05-30 22:11 ` [PATCH v2 4/7] gdb.base/: Introduce jit-protocol-util.h Ilya Leoshkevich via Gdb-patches
2022-05-30 22:11 ` [PATCH v2 5/7] gdb.base/: Introduce jit_compile_options Ilya Leoshkevich via Gdb-patches
2022-05-30 22:11 ` [PATCH v2 6/7] gdb.base/: Introduce n_jit_so_address Ilya Leoshkevich via Gdb-patches
2022-05-30 22:11 ` [PATCH v2 7/7] gdb.perf/: Add JIT performance test Ilya Leoshkevich via Gdb-patches
2022-06-22  8:48 ` [PING] [PATCH v2 0/7] " Ilya Leoshkevich via Gdb-patches
2022-08-03 10:52 ` [PING^2] " Ilya Leoshkevich via Gdb-patches

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=20220530221147.1991835-1-iii@linux.ibm.com \
    --to=gdb-patches@sourceware.org \
    --cc=aburgess@redhat.com \
    --cc=arnez@linux.ibm.com \
    --cc=iii@linux.ibm.com \
    --cc=pedro@palves.net \
    --cc=tromey@adacore.com \
    --cc=ulrich.weigand@de.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