public inbox for glibc-bugs@sourceware.org
help / color / mirror / Atom feed
From: "adhemerval.zanella at linaro dot org" <sourceware-bugzilla@sourceware.org>
To: glibc-bugs@sourceware.org
Subject: [Bug dynamic-link/15310] _dl_sort_fini is O(n^3) causing slow exit when many dsos
Date: Wed, 27 Oct 2021 14:58:58 +0000	[thread overview]
Message-ID: <bug-15310-131-TSAqQEOkuc@http.sourceware.org/bugzilla/> (raw)
In-Reply-To: <bug-15310-131@http.sourceware.org/bugzilla/>

https://sourceware.org/bugzilla/show_bug.cgi?id=15310

Adhemerval Zanella <adhemerval.zanella at linaro dot org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
                 CC|                            |adhemerval.zanella at linaro dot o
                   |                            |rg
         Resolution|---                         |FIXED

--- Comment #27 from Adhemerval Zanella <adhemerval.zanella at linaro dot org> ---
The testcase which creates a linear chain test to demostrate the O(n^3)
behavior shows an improvement with the new DSO sorting algorithm from
15a0c5730d1d5ae:

$ time ./testrun.sh /tmp/test/main1000 2>&1 >/dev/null 

real    0m1.641s
user    0m1.565s
sys     0m0.076s

$ time GLIBC_TUNABLES=glibc.rtld.dynamic_sort=2 ./testrun.sh /tmp/test/main1000
2>&1 >/dev/null 

real    0m0.316s
user    0m0.219s
sys     0m0.095s

And profile shows similar improvement:

$ perf record ./testrun.sh /tmp/test/main1000
[...]
$ perf report --stdio -q
[...]
    78.28%  ld-linux-x86-64  ld.so             [.] _dl_sort_maps
     6.77%  ld-linux-x86-64  ld.so             [.] do_lookup_x
     5.03%  ld-linux-x86-64  ld.so             [.] memmove
     2.64%  ld-linux-x86-64  ld.so             [.] strcmp

$ perf record env GLIBC_TUNABLES=glibc.rtld.dynamic_sort=2 ./testrun.sh
/tmp/test/main1000
[...]
$ perf report --stdio -q --dso=ld.so
[...]
    41.44%  ld-linux-x86-64  [.] do_lookup_x
    13.76%  ld-linux-x86-64  [.] strcmp
     6.01%  ld-linux-x86-64  [.] _dl_map_object
     5.17%  ld-linux-x86-64  [.] _dl_name_match_p
     1.34%  ld-linux-x86-64  [.] _dl_map_object_from_fd
     0.69%  ld-linux-x86-64  [.] _dl_add_to_namespace_list

-- 
You are receiving this mail because:
You are on the CC list for the bug.

  parent reply	other threads:[~2021-10-27 14:58 UTC|newest]

Thread overview: 31+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-03-27  7:47 [Bug dynamic-link/15310] New: " dhatch at ilm dot com
2013-03-27  8:12 ` [Bug dynamic-link/15310] " dhatch at ilm dot com
2013-03-27  8:45 ` dhatch at ilm dot com
2013-03-27 12:57 ` carlos at redhat dot com
2013-03-27 14:19 ` ppluzhnikov at google dot com
2013-03-27 20:33 ` dhatch at ilm dot com
2013-03-27 20:50 ` [Bug dynamic-link/15310] New: " Ondřej Bílka
2013-03-27 20:50 ` [Bug dynamic-link/15310] " neleai at seznam dot cz
2013-03-27 21:00 ` carlos at redhat dot com
2013-03-27 21:07 ` carlos at redhat dot com
2013-03-27 21:13 ` law at redhat dot com
2013-03-27 23:44 ` dhatch at ilm dot com
2013-03-28  0:31 ` dhatch at ilm dot com
2013-03-28  7:42   ` Ondřej Bílka
2013-03-28  7:42 ` neleai at seznam dot cz
2013-03-28 10:00 ` dhatch at ilm dot com
2013-03-28 10:19 ` dhatch at ilm dot com
2013-03-28 17:09 ` law at redhat dot com
2013-03-28 17:31 ` dhatch at ilm dot com
2013-04-02  9:54 ` dhatch at ilm dot com
2013-04-02 11:31   ` Ondřej Bílka
2013-04-02 11:31 ` neleai at seznam dot cz
2013-04-02 13:07 ` dhatch at ilm dot com
2013-04-02 23:37 ` dhatch at ilm dot com
2013-04-03  7:57   ` Ondřej Bílka
2013-04-03  7:57 ` neleai at seznam dot cz
2013-04-06 21:07 ` carlos at redhat dot com
2014-06-13 13:51 ` fweimer at redhat dot com
2014-06-13 18:37 ` fweimer at redhat dot com
2021-10-27 14:58 ` adhemerval.zanella at linaro dot org [this message]
2021-10-27 14:59 ` adhemerval.zanella at linaro dot 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-15310-131-TSAqQEOkuc@http.sourceware.org/bugzilla/ \
    --to=sourceware-bugzilla@sourceware.org \
    --cc=glibc-bugs@sourceware.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).