I just tried again to block a single revision with 1.5.2, and
the performance is just as poor. Anyone have any thoughts, or am I out
of luck and need to figure out a work-around?
Thanks ...
Matt McHenry
412-690-2442 x150
Software Developer
Carnegie Learning, Inc.
Learning By Doing (r)
Helping over 500,000 students in 2,600 schools to succeed in math.
http://www.carnegielearning.com
> -----Original Message-----
> From: McHenry, Matt
> Sent: Friday, August 22, 2008 4:53 PM
> To: 'users_at_subversion.tigris.org'
> Subject: Merge: poor performance / slowness; out of memory error
>
>
> The only post I've found on this subject is
> http://svn.haxx.se/subusers/archive-2008-07/0224.shtml. I wonder if
> someone can tell me whether the performance I'm seeing is worse than
> expected, though.
>
> I've got a list of 20 branch revisions that I want to block from
> the trunk, in a file called 'ct-revs.txt' (comma-separated). The
trunk
> working copy is quite large (13 GB), and includes tens of thousands of
> small files as well as many larger files.
>
> $ svn merge --record-only -c `cat ct-revs.txt `
> svn://dev.carnegielearning.com/branches/${BRANCH}
> svn(80258) malloc: *** mmap(size=16777216) failed (error code=12)
> *** error: can't allocate region
> *** set a breakpoint in malloc_error_break to debug
> Out of memory - terminating application.
> Abort trap
>
> Well, yuck. As a work-around, I tried blocking only one
> revision. I tried this with the first two revisions listed. In both
> cases, it takes a *really* long time -- but watching in 'top', it
> doesn't seem to be eating memory, just using a lot of CPU. The first
> time, it was deleting svn:mergeinfo from a bunch of nested paths
> (eliding, as I understand it), so I hoped that was an isolated case.
> But the second time took about as long, and only modified the root
> svn:mergeinfo property:
>
> $ time svn merge --record-only -c 74407
> svn://dev.carnegielearning.com/branches/${BRANCH}
>
> real 16m19.251s
> user 12m0.161s
> sys 0m58.013s
>
> $ svn stat
> M .
>
> I haven't tried a normal merge yet because I haven't had time.
> Maybe that will be faster, but I'm not optimistic. :(
>
> This is running on a brand new 8-core Xserve w/ a 3-disk RAID
> stripe, BTW:
>
> $ uname -a
> Darwin build-machine.carnegielearning.com 9.4.0 Darwin Kernel Version
> 9.4.0: Mon Jun 9 19:30:53 PDT 2008; root:xnu-1228.5.20~1/RELEASE_I386
> i386
>
> $ svn --version
> svn, version 1.5.1 (r32289)
> compiled Aug 18 2008, 14:51:25
>
> Any help is appreciated -- thanks!
>
>
> Matt McHenry
> 412-690-2442 x150
> Software Developer
> Carnegie Learning, Inc.
> Learning By Doing (r)
> Helping over 500,000 students in 2,600 schools to succeed in math.
> http://www.carnegielearning.com
---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe_at_subversion.tigris.org
For additional commands, e-mail: users-help_at_subversion.tigris.org
Received on 2008-09-02 23:32:11 CEST