Philip Martin wrote:
>Brandon Ehle <behle@pipedreaminteractive.com> writes:
>
>
>
>>On my working copy of 21011 files in 3528 directories, roughly 81%
>>of the time is spent doing either IO or CPU intensive operations
>>reading XML data in XML_Parse() & XML_ParseBuffer(). Roughly 76% of
>>this time comes from svn_wc_entries_read.
>>
>>
>
>Here's what oprofile shows. It's for an 'svn update' that modified
>100 files in a directory containing 400 small files. It's Subversion
>HEAD and both the repository and working copy were on a RAM based
>filesystem.
>
>
>
Try cachegrind, I've found that oprofile doesn't show when a function is
slow because its waiting on IO or some other function in the kernel or
glibc, cachegrind gives you this information as well and is just as easy
to run as oprofile. You get the self spent times with both, but you can
see the parent+self spent times in cachegrind that shows the rest of the
picture.
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Received on Tue Nov 26 04:13:45 2002