From: Shawn Pearce <spearce@spearce.org>
To: Marco Roeland <marco.roeland@xs4all.nl>
Cc: Linus Torvalds <torvalds@osdl.org>, Theodore Tso <tytso@mit.edu>,
Nicolas Pitre <nico@cam.org>,
Johannes Schindelin <Johannes.Schindelin@gmx.de>,
"Randal L. Schwartz" <merlyn@stonehenge.com>,
git@vger.kernel.org
Subject: Re: cloning the kernel - why long time in "Resolving 313037 deltas"
Date: Tue, 19 Dec 2006 03:49:08 -0500 [thread overview]
Message-ID: <20061219084908.GG2511@spearce.org> (raw)
In-Reply-To: <20061219084017.GB30448@fiberbit.xs4all.nl>
Marco Roeland <marco.roeland@xs4all.nl> wrote:
> On Tuesday December 19th 2006 at 03:32 Shawn Pearce wrote:
>
> > > Anyway, try the pread() version first, see if that fixes the OS X problem.
> >
> > It does. Without pread() (aka stock 'next') it takes me over an
> > hour to index a pack of linux-2.6. With pread() its 1m6s to run
> > index-pack on the same pack file. The indexes are (of course)
> > identically produced.
>
> I see the same here. From an (estimated) time of 37 minutes down to 52
> seconds with the pread() patch. Running the profiler (shark) on the old
> version showed that 85% of the time was spent in the Mac OS X mmap()
> system call.
More testing on Linux is probably needed, but if using pread()
on Linux is breakeven or slightly faster (as suggested by Johannes'
LilyPond test) this 60x performance improvement on initial clone
of largish projects on Mac OS X would be nice to have.
Not that I personally frequently clone large projects on Mac OS X.
But new users to Git might. :-)
--
next prev parent reply other threads:[~2006-12-19 8:49 UTC|newest]
Thread overview: 52+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <86y7p57y05.fsf@blue.stonehenge.com>
[not found] ` <Pine.LNX.4.63.0612182154170.19693@wbgn013.biozentrum.uni-wuerzburg.de>
[not found] ` <Pine.LNX.4.63.0612182213020.19693@wbgn013.biozentrum.uni-wuerzburg.de>
[not found] ` <Pine.LNX.4.64.0612181638220.18171@xanadu.home>
2006-12-18 21:55 ` [PATCH] fetch-pack: avoid fixing thin packs when unnecessary Johannes Schindelin
2006-12-18 22:17 ` Nicolas Pitre
[not found] ` <Pine.LNX.4.64.0612181251020.3479@woody.osdl.org>
[not found] ` <86r6uw9azn.fsf@blue.stonehenge.com>
[not found] ` <Pine.LNX.4.64.0612181625140.18171@xanadu.home>
2006-12-18 22:01 ` cloning the kernel - why long time in "Resolving 313037 deltas" Randal L. Schwartz
2006-12-18 22:09 ` Nicolas Pitre
2006-12-18 22:21 ` Randal L. Schwartz
2006-12-18 22:50 ` Nicolas Pitre
2006-12-18 22:22 ` Linus Torvalds
2006-12-18 22:26 ` Randal L. Schwartz
2006-12-18 23:02 ` Martin Langhoff
2006-12-22 1:44 ` Kyle Moffett
2006-12-22 1:56 ` Shawn Pearce
2006-12-22 8:04 ` Marco Roeland
2007-01-03 13:55 ` Andreas Ericsson
2006-12-18 23:28 ` Linus Torvalds
2006-12-19 0:13 ` Nicolas Pitre
2006-12-19 5:11 ` Theodore Tso
2006-12-19 6:39 ` Shawn Pearce
2006-12-19 6:51 ` Linus Torvalds
2006-12-19 7:26 ` Shawn Pearce
2006-12-19 7:52 ` Marco Roeland
2006-12-19 7:58 ` Shawn Pearce
2006-12-19 8:32 ` Shawn Pearce
2006-12-19 8:40 ` Marco Roeland
2006-12-19 8:49 ` Shawn Pearce [this message]
2006-12-19 9:13 ` Marco Roeland
2006-12-19 20:28 ` Alex Riesen
2006-12-21 20:35 ` Juergen Ruehle
2006-12-19 16:19 ` Theodore Tso
2006-12-19 16:57 ` Linus Torvalds
2006-12-20 1:54 ` Shawn Pearce
2006-12-20 1:58 ` Shawn Pearce
2006-12-19 6:47 ` Linus Torvalds
2006-12-19 8:32 ` Johannes Schindelin
2006-12-19 9:10 ` Junio C Hamano
2006-12-19 9:47 ` Jeff King
2006-12-19 10:24 ` Andy Whitcroft
2006-12-19 15:53 ` [PATCH] index-pack usage of mmap() is unacceptably slower on many OSes other than Linux Nicolas Pitre
2006-12-19 19:00 ` Junio C Hamano
2006-12-19 19:14 ` Nicolas Pitre
2006-12-19 19:55 ` Linus Torvalds
2006-12-19 19:57 ` Randal L. Schwartz
2006-12-19 20:03 ` Randal L. Schwartz
2006-12-19 20:02 ` Jeff Garzik
2006-12-20 0:30 ` Junio C Hamano
2006-12-20 0:40 ` Linus Torvalds
2006-12-20 0:50 ` Jeff Garzik
2006-12-20 1:12 ` Junio C Hamano
2006-12-20 20:17 ` Junio C Hamano
2006-12-20 20:53 ` Linus Torvalds
2006-12-20 21:52 ` Junio C Hamano
2006-12-20 22:13 ` Nikolai Weibull
2006-12-21 8:41 cloning the kernel - why long time in "Resolving 313037 deltas" linux
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=20061219084908.GG2511@spearce.org \
--to=spearce@spearce.org \
--cc=Johannes.Schindelin@gmx.de \
--cc=git@vger.kernel.org \
--cc=marco.roeland@xs4all.nl \
--cc=merlyn@stonehenge.com \
--cc=nico@cam.org \
--cc=torvalds@osdl.org \
--cc=tytso@mit.edu \
/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).