public inbox for glibc-bugs@sourceware.org help / color / mirror / Atom feed
From: "ppluzhnikov at google dot com" <sourceware-bugzilla@sourceware.org> To: glibc-bugs@sources.redhat.com Subject: [Bug libc/11787] Program with large TLS segment fails aio_write Date: Fri, 30 Mar 2012 21:28:00 -0000 [thread overview] Message-ID: <bug-11787-131-TbDxWAPBny@http.sourceware.org/bugzilla/> (raw) In-Reply-To: <bug-11787-131@http.sourceware.org/bugzilla/> http://sourceware.org/bugzilla/show_bug.cgi?id=11787 --- Comment #25 from Paul Pluzhnikov <ppluzhnikov at google dot com> 2012-03-30 16:44:43 UTC --- (In reply to comment #24) > I think the real question here isn't the change in the size of the stack but > the increase in the # of pages and potential wastage. > > If the TLS is small won't we end up burning an entire page for this small > amount of data? For apps that use small stacks and lots of threads, that extra > page could turn out to be significant. Yes, as I said in comment#12, an application that may notice this is one that uses lots of threads, and is *almost* running out of VM space. > I'm not sure I really buy the argument that if a thread asks for 16k that they > get exactly 16k usable. Let's also make "malloc(16K)" return 16K or less of usable memory ;-) > There's no perfect solution here. I wonder if we could come up with a good > heuristic based on the relative sizes of the TLS and thread requested stack. Great idea! > If the TLS is sufficiently small relative to the size of the requested stack, > then just fold the TLS into the requested stack like we've always done. > Otherwise, add the size of the TLS to the size of the requested stack (rounding > to a page of course). Harder to document and implement, but seems like it'd > strike a better balance. > > I don't know where the cutoff should be, 1%, 10%, 50%? Some experimentation may > guide us. Given the data Mike Frysinger and I collected (most binaries using <512 bytes of TLS), I say: if stack_size_requested >= 16 * __static_tls_sze use current calculation else increment size request by roundup(__static_tls_size, pagesize()) Most applications today request at least 64K stack (most actually default to *much* more than that), which would allow up to 4K of static TLS. But if that same application is instrumented for profiling and suddenly requires 128K of TLS, it would still work. -- Configure bugmail: http://sourceware.org/bugzilla/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.
next prev parent reply other threads:[~2012-03-30 16:46 UTC|newest] Thread overview: 64+ messages / expand[flat|nested] mbox.gz Atom feed top [not found] <bug-11787-131@http.sourceware.org/bugzilla/> 2012-02-24 4:35 ` ppluzhnikov at google dot com 2012-02-24 19:36 ` asharif.tools at gmail dot com 2012-03-22 21:48 ` asharif.tools at gmail dot com 2012-03-23 20:35 ` vapier at gentoo dot org 2012-03-23 20:37 ` carlos_odonell at mentor dot com 2012-03-23 20:39 ` carlos_odonell at mentor dot com 2012-03-23 21:33 ` carlos_odonell at mentor dot com 2012-03-23 22:07 ` ppluzhnikov at google dot com 2012-03-23 22:47 ` asharif.tools at gmail dot com 2012-03-23 22:51 ` carlos_odonell at mentor dot com 2012-03-23 23:02 ` carlos_odonell at mentor dot com 2012-03-23 23:03 ` carlos_odonell at mentor dot com 2012-03-24 2:22 ` carlos_odonell at mentor dot com 2012-03-24 10:30 ` ppluzhnikov at google dot com 2012-03-24 16:10 ` carlos_odonell at mentor dot com 2012-03-24 17:08 ` carlos_odonell at mentor dot com 2012-03-24 17:39 ` vapier at gentoo dot org 2012-03-24 18:30 ` ppluzhnikov at google dot com 2012-03-24 20:13 ` carlos_odonell at mentor dot com 2012-03-24 20:46 ` carlos_odonell at mentor dot com 2012-03-24 20:57 ` ppluzhnikov at google dot com 2012-03-25 2:01 ` carlos_odonell at mentor dot com 2012-03-25 23:16 ` vapier at gentoo dot org 2012-03-26 14:29 ` joseph at codesourcery dot com 2012-03-26 17:44 ` carlos_odonell at mentor dot com 2012-03-27 21:21 ` carlos_odonell at mentor dot com 2012-03-30 16:21 ` carlos_odonell at mentor dot com 2012-03-30 16:46 ` law at redhat dot com 2012-03-30 21:28 ` ppluzhnikov at google dot com [this message] 2012-04-01 19:47 ` carlos_odonell at mentor dot com 2012-04-24 22:03 ` asharif.tools at gmail dot com 2012-04-24 22:36 ` carlos_odonell at mentor dot com 2012-04-25 2:55 ` vapier at gentoo dot org 2012-04-25 18:25 ` asharif.tools at gmail dot com 2012-04-25 20:11 ` carlos_odonell at mentor dot com 2012-05-07 21:22 ` [Bug libc/11787] Program with large TLS segments fail carlos_odonell at mentor dot com 2012-05-17 14:11 ` carlos_odonell at mentor dot com 2012-05-22 7:05 ` siddhesh at redhat dot com 2012-05-23 16:41 ` carlos_odonell at mentor dot com 2012-05-23 17:09 ` siddhesh at redhat dot com 2012-05-23 17:09 ` siddhesh at redhat dot com 2012-06-05 15:15 ` ian at airs dot com 2012-11-19 6:18 ` carlos_odonell at mentor dot com 2012-12-03 23:52 ` carlos at systemhalted dot org 2012-12-03 23:53 ` carlos at systemhalted dot org 2013-04-26 18:06 ` ppluzhnikov at google dot com 2013-04-27 2:02 ` bugdal at aerifal dot cx 2013-05-22 18:40 ` carlos at redhat dot com 2013-09-13 21:45 ` bas at bmail dot ru 2013-11-23 9:05 ` david.abdurachmanov at gmail dot com 2013-11-23 9:20 ` david.abdurachmanov at gmail dot com 2013-11-24 15:58 ` carlos at redhat dot com 2014-02-16 19:42 ` jackie.rosen at hushmail dot com 2014-05-28 19:41 ` schwab at sourceware dot org 2014-06-30 17:33 ` fweimer at redhat dot com 2014-07-02 6:06 ` fweimer at redhat dot com 2015-03-23 2:50 ` andersk at mit dot edu 2015-08-22 21:36 ` [Bug dynamic-link/11787] " jsm28 at gcc dot gnu.org 2020-06-15 16:23 ` john.wellesz at hiventy dot com 2021-06-19 8:05 ` a3at.mail at gmail dot com 2022-01-26 6:44 ` sam at gentoo dot org 2022-06-07 8:09 ` i at maskray dot me 2022-11-06 20:12 ` i at maskray dot me 2010-07-02 23:02 [Bug libc/11787] New: Program with large TLS segment fails aio_write ppluzhnikov at google dot com 2010-07-02 23:03 ` [Bug libc/11787] " ppluzhnikov at google 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=bug-11787-131-TbDxWAPBny@http.sourceware.org/bugzilla/ \ --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).