public inbox for gcc-bugs@sourceware.org
help / color / mirror / Atom feed
From: "marxin at gcc dot gnu.org" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug gcov-profile/95348] GCC records zero functions and modules in the profiling data file, ICC does NOT
Date: Fri, 29 May 2020 09:11:12 +0000 [thread overview]
Message-ID: <bug-95348-4-LY8Z7JmIBT@http.gcc.gnu.org/bugzilla/> (raw)
In-Reply-To: <bug-95348-4@http.gcc.gnu.org/bugzilla/>
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=95348
Martin Liška <marxin at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|ASSIGNED |WAITING
--- Comment #7 from Martin Liška <marxin at gcc dot gnu.org> ---
Ok, I spent some time thinking about your workload and I would recommend the
following steps:
1) You should not generate profile data for each process to a different folder,
but rather merge it.
GCC PGO bootstrap contains ~500 .gcda files where the process is executed
~2000x.
Note that .gcda file merging happens per-file and the file is locked. It should
be a reasonable small window that can delay parallel process execution.
2) I would like to know how long does one process run and what portion is spent
in merging (and dumping) of a profile.
3) You may consider shrinking training run, 10.000 executions seems like a
massive training run to me.
4) GCDA file format is not ideal and can be simply and rapidly shrank by e.g.
gzip. For GCC PGO, it shrinks 10x.
Please provide as much information about the workload so that we can find a
feasible solution.
next prev parent reply other threads:[~2020-05-29 9:11 UTC|newest]
Thread overview: 45+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-05-26 20:37 [Bug gcov-profile/95348] New: " qinzhao at gcc dot gnu.org
2020-05-27 7:42 ` [Bug gcov-profile/95348] " rguenth at gcc dot gnu.org
2020-05-27 9:31 ` marxin at gcc dot gnu.org
2020-05-27 14:13 ` qinzhao at gcc dot gnu.org
2020-05-27 14:22 ` marxin at gcc dot gnu.org
2020-05-27 20:36 ` qinzhao at gcc dot gnu.org
2020-05-28 14:08 ` marxin at gcc dot gnu.org
2020-05-29 9:11 ` marxin at gcc dot gnu.org [this message]
2020-05-29 15:28 ` qinzhao at gcc dot gnu.org
2020-05-29 15:31 ` qinzhao at gcc dot gnu.org
2020-05-29 15:38 ` marxin at gcc dot gnu.org
2020-05-29 15:40 ` marxin at gcc dot gnu.org
2020-05-29 15:43 ` marxin at gcc dot gnu.org
2020-05-29 16:37 ` qing.zhao at oracle dot com
2020-05-29 16:40 ` qing.zhao at oracle dot com
2020-05-29 16:42 ` qinzhao at gcc dot gnu.org
2020-05-30 7:23 ` marxin at gcc dot gnu.org
2020-06-01 15:24 ` qinzhao at gcc dot gnu.org
2020-06-01 15:39 ` qinzhao at gcc dot gnu.org
2020-06-01 15:45 ` qing.zhao at oracle dot com
2020-06-01 15:45 ` qing.zhao at oracle dot com
2020-06-02 8:23 ` marxin at gcc dot gnu.org
2020-06-02 17:42 ` qinzhao at gcc dot gnu.org
2020-06-03 6:35 ` marxin at gcc dot gnu.org
2020-06-03 15:17 ` qing.zhao at oracle dot com
2020-06-10 8:16 ` marxin at gcc dot gnu.org
2020-06-10 8:21 ` marxin at gcc dot gnu.org
2020-06-10 14:49 ` qing.zhao at oracle dot com
2020-06-10 19:43 ` marxin at gcc dot gnu.org
2020-06-10 20:17 ` qing.zhao at oracle dot com
2020-06-10 20:26 ` qing.zhao at oracle dot com
2020-06-11 9:14 ` marxin at gcc dot gnu.org
2020-06-11 14:12 ` qing.zhao at oracle dot com
2020-06-11 18:08 ` marxin at gcc dot gnu.org
2020-06-15 19:54 ` qinzhao at gcc dot gnu.org
2020-06-15 19:56 ` qinzhao at gcc dot gnu.org
2020-06-16 12:53 ` marxin at gcc dot gnu.org
2020-06-16 15:58 ` qinzhao at gcc dot gnu.org
2020-06-17 6:35 ` marxin at gcc dot gnu.org
2020-06-24 9:06 ` marxin at gcc dot gnu.org
2020-07-02 8:16 ` cvs-commit at gcc dot gnu.org
2020-07-02 8:24 ` marxin at gcc dot gnu.org
2020-07-02 8:24 ` marxin at gcc dot gnu.org
2021-04-27 11:38 ` jakub at gcc dot gnu.org
2021-07-28 7:04 ` rguenth at gcc dot gnu.org
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=bug-95348-4-LY8Z7JmIBT@http.gcc.gnu.org/bugzilla/ \
--to=gcc-bugzilla@gcc.gnu.org \
--cc=gcc-bugs@gcc.gnu.org \
/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 read-only IMAP folder(s) and NNTP newsgroup(s).