Comment 6 for bug 562429
Revision history for this message
| Andrew Bennetts (spiv) wrote Re: [Bug 562429] Re: performance regression for 'log' in trunk? | #6 |
© 2004
Canonical Ltd.
•
Terms of use
•
Data privacy
•
Contact Launchpad Support
•
Blog
•
Careers
•
System status
•
bbfa235
(Get the code!)
John A Meinel wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> Parth Malwankar wrote:
> > lsprof-file output attached for trunk (new_bzr.callgrind) and 2.1.1
> > (old_bzr.
> >
>
> I'll have to get a look at this tomorrow.
>
> In the short-term, I wonder if Andrew's 'move index to front' patch is
> to blame. I'm doing some profiling elsewhere, and I came across this
> taking 3-4s out of a total of 5.5s to do what I was profiling.
Looking at new_bzr.callgrind, I think you're right. _move_to_front is
apparently taking almost 25% of the total time.
Parth, if you try John's linked branch I expect you'll find it has fixed
the regression.