git.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: "R. Tyler Ballance" <tyler@slide.com>
To: Linus Torvalds <torvalds@linux-foundation.org>
Cc: "Shawn O. Pearce" <spearce@spearce.org>,
	"Junio C Hamano" <gitster@pobox.com>,
	"Nicolas Pitre" <nico@cam.org>, "Jan Krüger" <jk@jk.gs>,
	"Git ML" <git@vger.kernel.org>,
	kb@slide.com
Subject: Re: [PATCH] Wrap inflateInit to retry allocation after releasing pack memory
Date: Thu, 08 Jan 2009 10:15:52 -0800	[thread overview]
Message-ID: <1231438552.8870.645.camel@starfruit> (raw)
In-Reply-To: <alpine.LFD.2.00.0901071941210.3283@localhost.localdomain>

[-- Attachment #1: Type: text/plain, Size: 2597 bytes --]

On Wed, 2009-01-07 at 19:54 -0800, Linus Torvalds wrote:
> 
> On Wed, 7 Jan 2009, Shawn O. Pearce wrote:
> >
> > If we are running low on virtual memory we should release pack
> > windows if zlib's inflateInit fails due to an out of memory error.
> > It may be that we are running under a low ulimit and are getting
> > tight on address space.  Shedding unused windows may get us
> > sufficient working space to continue.
> 
> Let's do this (more complete) wrapping instead, ok?
> 
> This one _just_ wraps things, btw - it doesn't do the "retry on low memory 
> error" part, at least not yet. I think that's an independent issue from 
> the reporting.
> 
> Hmm? 
> 
> Tyler - does this make the corruption errors go away, and be replaced by 
> hard failures with "out of memory" reporting?

Yeah, looks like it:


        tyler@grapefruit:~/source/git/linux-2.6> export
        START=56d18e9932ebf4e8eca42d2ce509450e6c9c1666
        tyler@grapefruit:~/source/git/linux-2.6> git reset --hard
        HEAD is now at 9e42d0c Merge
        git://git.kernel.org/pub/scm/linux/kernel/git/davem/sparc-2.6
        tyler@grapefruit:~/source/git/linux-2.6> git reset --hard $START
        HEAD is now at 56d18e9 Merge branch 'upstream' of
        git://ftp.linux-mips.org/pub/scm/upstream-linus
        tyler@grapefruit:~/source/git/linux-2.6> ulimit -v `echo "350 *
        1024" | bc -l`
        tyler@grapefruit:~/source/git/linux-2.6> limit
        cputime         unlimited
        filesize        unlimited
        datasize        unlimited
        stacksize       8MB
        coredumpsize    0kB
        memoryuse       2561MB
        maxproc         24564
        descriptors     1024
        memorylocked    64kB
        addressspace    350MB
        maxfilelocks    unlimited
        sigpending      24564
        msgqueue        819200
        nice            0
        rt_priority     0
        tyler@grapefruit:~/source/git/linux-2.6> git pull
        Updating 56d18e9..9e42d0c
        fatal: Out of memory? inflateInit failed
        tyler@grapefruit:~/source/git/linux-2.6> which git
        /home/tyler/bin/git
        tyler@grapefruit:~/source/git/linux-2.6> 



> 
> This patch is potentially pretty noisy, on purpose. I didn't remove the 
> reporting from places that already do so - some of them have stricter 
> errors than this.

I'm assuming this patch is going to be reworked, if so, I'll back it out
of our internal 1.6.1 build and anxiously await The Real Deal(tm)


Cheers
-- 
-R. Tyler Ballance
Slide, Inc.

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 197 bytes --]

  parent reply	other threads:[~2009-01-08 18:17 UTC|newest]

Thread overview: 49+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2008-12-09  8:36 [PATCH/RFC] Allow writing loose objects that are corrupted in a pack file Jan Krüger
2008-12-09  9:02 ` R. Tyler Ballance
2008-12-09 16:24 ` Shawn O. Pearce
2009-01-06 22:52 ` R. Tyler Ballance
2009-01-07  1:25   ` Nicolas Pitre
2009-01-07  1:39     ` R. Tyler Ballance
2009-01-07  2:09       ` Nicolas Pitre
2009-01-07  2:47         ` R. Tyler Ballance
2009-01-07  3:21           ` Nicolas Pitre
2009-01-07  4:54       ` Linus Torvalds
2009-01-07  7:41         ` R. Tyler Ballance
2009-01-07  8:16           ` Junio C Hamano
2009-01-07  8:32             ` R. Tyler Ballance
2009-01-07  9:42               ` Junio C Hamano
2009-01-07  9:05           ` R. Tyler Ballance
2009-01-07 15:31           ` Nicolas Pitre
2009-01-07 16:07           ` Linus Torvalds
2009-01-07 16:08             ` Linus Torvalds
2009-01-07 22:55             ` R. Tyler Ballance
2009-01-07 23:29               ` Linus Torvalds
2009-01-08  0:28                 ` Public repro case! " R. Tyler Ballance
2009-01-08  0:48                   ` Linus Torvalds
2009-01-08  0:57                     ` R. Tyler Ballance
2009-01-08  1:08                       ` Linus Torvalds
2009-01-08  1:29                         ` Linus Torvalds
2009-01-08  1:46                           ` Shawn O. Pearce
2009-01-08  2:21                     ` James Pickens
2009-01-08  2:43                       ` Shawn O. Pearce
2009-01-08  5:40                         ` Junio C Hamano
2009-01-08  6:04                           ` Shawn O. Pearce
2009-01-08  2:52                       ` Boyd Stephen Smith Jr.
2009-01-08  2:52                   ` Linus Torvalds
2009-01-08  3:01                     ` Shawn O. Pearce
2009-01-08  3:06                       ` Linus Torvalds
2009-01-08  3:13                         ` Shawn O. Pearce
2009-01-08  3:16                           ` [PATCH] Wrap inflateInit to retry allocation after releasing pack memory Shawn O. Pearce
2009-01-08  3:54                             ` Linus Torvalds
2009-01-08  5:23                               ` Junio C Hamano
2009-01-08 15:35                                 ` Linus Torvalds
2009-01-08 15:34                               ` Shawn O. Pearce
2009-01-08 16:14                                 ` Linus Torvalds
2009-01-08 18:15                               ` R. Tyler Ballance [this message]
2009-01-08 20:22                                 ` Linus Torvalds
2009-01-08 20:37                                   ` R. Tyler Ballance
2009-01-09  1:43                                   ` Junio C Hamano
2009-01-08  0:37                 ` [PATCH/RFC] Allow writing loose objects that are corrupted in a pack file Linus Torvalds
2009-01-08  0:49                   ` R. Tyler Ballance
2009-01-08  1:01                     ` Linus Torvalds
2009-01-08  1:06                       ` R. Tyler Ballance

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=1231438552.8870.645.camel@starfruit \
    --to=tyler@slide.com \
    --cc=git@vger.kernel.org \
    --cc=gitster@pobox.com \
    --cc=jk@jk.gs \
    --cc=kb@slide.com \
    --cc=nico@cam.org \
    --cc=spearce@spearce.org \
    --cc=torvalds@linux-foundation.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 NNTP newsgroup(s).