git.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Linus Torvalds <torvalds@linux-foundation.org>
To: "Carlos R. Mafra" <crmafra2@gmail.com>
Cc: Git Mailing List <git@vger.kernel.org>,
	Junio C Hamano <gitster@pobox.com>
Subject: Re: Performance issue of 'git branch'
Date: Wed, 22 Jul 2009 19:20:00 -0700 (PDT)	[thread overview]
Message-ID: <alpine.LFD.2.01.0907221850000.3352@localhost.localdomain> (raw)
In-Reply-To: <20090723012207.GA9368@Pilar.aei.mpg.de>



On Thu, 23 Jul 2009, Carlos R. Mafra wrote:
> > Let me guess: if you do a "ls -ld .git/refs/heads" you get a very big 
> > directory, despite it only having three entries in it. 
> 
> [mafra@Pilar:linux-2.6]$ ls -ld .git/refs/heads
> drwxr-xr-x 2 mafra mafra 4096 2009-07-22 23:01 .git/refs/heads/

Hmm. That's just a single block. 

Then I really don't see why the lstat takes so long.

> After 'echo 3 > /proc/sys/vm/drop_caches' it still takes too long,
> 
> 1248310449.693085 munmap(0x7f50bcd11000, 164) = 0
> 1248310449.693187 lstat(".git/refs/heads/sparse", 0x7fff618c0960) = -1 ENOENT (No such file or directory)
> 1248310449.719112 lstat(".git/refs/heads/stern", 0x7fff618c0960) = -1 ENOENT (No such file or directory)
> 1248310453.014041 fstat(1, {st_mode=S_IFCHR|0620, st_rdev=makedev(136, 3), ...}) = 0
> 1248310453.014183 mmap(NULL, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = 0x7f50bcd11000

Use 'strace -T', which shows how long the actual system calls take, rather 
than '-tt' which just shows when they started.

Maybe the four seconds is something else than the lstat - page faults on 
the pack-file in between the lstat and the fstat, for example.

> Perhaps I should delete the "stern" branch, but I would like to learn why
> it is slowing things, because it also happened before (in fact it is always
> like this, afaicr)

Absolutely. Don't delete it until we figure out what takes so long there.

> Do you have another theory? (now .git/refs/heads is empty)

Clearly it's IO, but if that 'lstat()' was just a red herring, then I 
suspect it's IO on the pack-file. If so, I'd further guess that your VAIO 
has some pitiful 4200rpm harddisk that is slow as hell and has horrible 
seek latencies, and the CPU is way overpowered compared to the cruddy 
disk.

It probably does the object lookup. You can see some debug output if you 
do

	GIT_DEBUG_LOOKUP=1 git branch

and that will show you the patterns. It won't be very pretty, especially 
if you have several pack-files, but maybe we can figure out what's up.

Hmm. I wonder.. I suspect 'git branch' looks up _all_ refs, and then 
afterwards it filters them. So even though it only prints out a few 
branches, maybe it will look at all the tags etc of the whole repository.

Ooh yes. That would do it. It's going to peel and look up every single ref 
it finds, so it's going to look up _hundreds_ of objects (all the tags, 
all the commits they point to, etc etc). Even if it then only shows a 
couple of branches.

Junio, any ideas?

		Linus

  reply	other threads:[~2009-07-23  2:20 UTC|newest]

Thread overview: 74+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2009-07-22 23:59 Performance issue of 'git branch' Carlos R. Mafra
2009-07-23  0:21 ` Linus Torvalds
2009-07-23  0:51   ` Linus Torvalds
2009-07-23  0:55     ` Linus Torvalds
2009-07-23  2:02       ` Carlos R. Mafra
2009-07-23  2:28         ` Linus Torvalds
2009-07-23 12:42           ` Jakub Narebski
2009-07-23 14:45             ` Carlos R. Mafra
2009-07-23 16:25             ` Linus Torvalds
2009-07-23  1:22   ` Carlos R. Mafra
2009-07-23  2:20     ` Linus Torvalds [this message]
2009-07-23  2:23       ` Linus Torvalds
2009-07-23  3:08         ` Linus Torvalds
2009-07-23  3:21           ` Linus Torvalds
2009-07-23 17:47             ` Tony Finch
2009-07-23 18:57               ` Linus Torvalds
2009-07-23 22:48                 ` Newton-Raphson, was " Tony Finch
2009-07-23 23:24                   ` Johannes Schindelin
2009-07-23 23:50                     ` Tony Finch
2009-07-24  0:43                       ` Johannes Schindelin
2009-07-23  3:18         ` Carlos R. Mafra
2009-07-23  3:27           ` Carlos R. Mafra
2009-07-23  3:40           ` Carlos R. Mafra
2009-07-23  3:47           ` Linus Torvalds
2009-07-23  4:10             ` Linus Torvalds
2009-07-23  5:13               ` Junio C Hamano
2009-07-23  5:17               ` Carlos R. Mafra
2009-07-23  4:40         ` Junio C Hamano
2009-07-23  5:36           ` Linus Torvalds
2009-07-23  5:52             ` Junio C Hamano
2009-07-23  6:04               ` Junio C Hamano
2009-07-23 17:19                 ` Linus Torvalds
2009-07-23 16:07           ` Carlos R. Mafra
2009-07-23 16:19             ` Linus Torvalds
2009-07-23 16:53               ` Carlos R. Mafra
2009-07-23 19:05                 ` Linus Torvalds
2009-07-23 19:13                   ` Linus Torvalds
2009-07-23 19:55                     ` Carlos R. Mafra
2009-07-24 20:36                       ` Linus Torvalds
2009-07-24 20:47                         ` Linus Torvalds
2009-07-24 21:21                           ` Linus Torvalds
2009-07-24 22:13                             ` Linus Torvalds
2009-07-24 22:18                               ` david
2009-07-24 22:42                                 ` Linus Torvalds
2009-07-24 22:46                                   ` david
2009-07-25  2:39                                     ` Linus Torvalds
2009-07-25  2:53                                       ` Daniel Barkalow
2009-08-07  4:21                               ` Jeff King
2009-07-24 22:54                             ` Theodore Tso
2009-07-24 22:59                               ` Shawn O. Pearce
2009-07-24 23:28                                 ` Junio C Hamano
2009-07-26 17:07                                 ` Avi Kivity
2009-07-26 17:16                                   ` Johannes Schindelin
2009-07-24 23:46                             ` Carlos R. Mafra
2009-07-25  0:41                               ` Carlos R. Mafra
2009-07-25 18:04                                 ` Linus Torvalds
2009-07-25 18:57                                   ` Timo Hirvonen
2009-07-25 19:06                                     ` Reece Dunn
2009-07-25 20:31                                     ` Mike Hommey
2009-07-25 21:02                                       ` Linus Torvalds
2009-07-25 21:13                                         ` Linus Torvalds
2009-07-25 23:23                                           ` Johannes Schindelin
2009-07-26  4:49                                             ` Linus Torvalds
2009-07-26 16:29                                               ` Theodore Tso
2009-07-26  7:54                                         ` Mike Hommey
2009-07-26 10:16                                           ` Johannes Schindelin
2009-07-26 10:23                                             ` demerphq
2009-07-26 10:27                                               ` demerphq
2009-07-25 21:04                                     ` Carlos R. Mafra
2009-07-23 16:48         ` Anders Kaseorg
2009-07-23 19:03           ` Carlos R. Mafra
2009-07-23  0:23 ` SZEDER Gábor
2009-07-23  2:25   ` Carlos R. Mafra
  -- strict thread matches above, loose matches on Subject: below --
2009-07-26 23:21 George Spelvin

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=alpine.LFD.2.01.0907221850000.3352@localhost.localdomain \
    --to=torvalds@linux-foundation.org \
    --cc=crmafra2@gmail.com \
    --cc=git@vger.kernel.org \
    --cc=gitster@pobox.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: 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).