public inbox for glibc-bugs@sourceware.org help / color / mirror / Atom feed
From: "rich at testardi dot com" <sourceware-bugzilla@sourceware.org> To: glibc-bugs@sources.redhat.com Subject: [Bug libc/11261] malloc uses excessive memory for multi-threaded applications Date: Wed, 10 Feb 2010 15:52:00 -0000 [thread overview] Message-ID: <20100210155235.539.qmail@sourceware.org> (raw) In-Reply-To: <20100208202339.11261.rich@testardi.com> ------- Additional Comments From rich at testardi dot com 2010-02-10 15:52 ------- Last mail... It turns out the arena_max and arena_test numbers are "fuzzy" (I am sure by design), since no lock is held here: static mstate internal_function arena_get2(mstate a_tsd, size_t size) { mstate a; #ifdef PER_THREAD if (__builtin_expect (use_per_thread, 0)) { if ((a = get_free_list ()) == NULL && (a = reused_arena ()) == NULL) /* Nothing immediately available, so generate a new arena. */ a = _int_new_arena(size); return a; } #endif Therefore, if narenas is less than the limit tested for in reused_arena(), and N threads get in to this code at once, narenas can then end up N-1 *above* the limit. The likelihood of this happening is proportional to the malloc arrival rate and the time spend in _int_new_arena(). This is exactly what I am seeing. So if you can live with 2 arenas, the critical thing to do is to make sure narenas is exactly 2 before going heavily multi-threaded, and then it won't be able to go above 2; otherwise, it can sneak up to 2+N-1, where N is the number of threads contending for allocations. If the ">=" in reused_arena() was changed to ">", then we could use this mechanism to limit narenas to exactly 1 right from the get-go. That would be ideal for our kind of applications (that can't live with 2 arenas). -- http://sourceware.org/bugzilla/show_bug.cgi?id=11261 ------- You are receiving this mail because: ------- You are on the CC list for the bug, or are watching someone who is.
next prev parent reply other threads:[~2010-02-10 15:52 UTC|newest] Thread overview: 21+ messages / expand[flat|nested] mbox.gz Atom feed top 2010-02-08 20:23 [Bug libc/11261] New: " rich at testardi dot com 2010-02-09 15:28 ` [Bug libc/11261] " drepper at redhat dot com 2010-02-09 16:02 ` rich at testardi dot com 2010-02-10 13:10 ` rich at testardi dot com 2010-02-10 13:21 ` drepper at redhat dot com 2010-02-10 13:42 ` rich at testardi dot com 2010-02-10 14:29 ` rich at testardi dot com 2010-02-10 15:52 ` rich at testardi dot com [this message] [not found] <bug-11261-131@http.sourceware.org/bugzilla/> 2011-08-27 21:45 ` heuler at infosim dot net 2011-08-27 22:02 ` rich at testardi dot com 2011-09-02 7:39 ` heuler at infosim dot net 2011-09-02 7:45 ` heuler at infosim dot net 2011-09-11 15:46 ` drepper.fsp at gmail dot com 2011-09-11 21:32 ` rich at testardi dot com 2012-07-29 10:10 ` zhannk at gmail dot com 2012-12-19 10:47 ` schwab@linux-m68k.org 2013-03-14 19:03 ` carlos at redhat dot com 2013-12-12 0:22 ` neleai at seznam dot cz 2013-12-12 3:32 ` siddhesh at redhat dot com 2013-12-12 8:41 ` neleai at seznam dot cz 2013-12-12 10:48 ` siddhesh at redhat dot com
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=20100210155235.539.qmail@sourceware.org \ --to=sourceware-bugzilla@sourceware.org \ --cc=glibc-bugs@sources.redhat.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: linkBe 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).