Issue 23350 - tiny pagein speedup
Summary: tiny pagein speedup
Status: CLOSED WONT_FIX
Alias: None
Product: General
Classification: Code
Component: code (show other issues)
Version: OOo 1.1
Hardware: All All
: P3 Trivial (vote)
Target Milestone: OOo 2.0
Assignee: mmeeks
QA Contact: issues@framework
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2003-12-09 16:02 UTC by mmeeks
Modified: 2004-02-16 09:29 UTC (History)
1 user (show)

See Also:
Issue Type: DEFECT
Latest Confirmation in: ---
Developer Difficulty: ---

Attachments
slight speedup (9.51 KB, patch)
2003-12-09 16:04 UTC, mmeeks 		no flags Details | Diff
Add an attachment (proposed patch, testcase, etc.)

Note You need to log in before you can comment on or make changes to this issue.
Description mmeeks 2003-12-09 16:02:43 UTC 
The attached patch knocks ~50ms off pagein time for me, hopefully more if you
have a highly slow disk; perhaps not worth loosing generality for that sort of
win but ...
Comment 1 mmeeks 2003-12-09 16:04:31 UTC 
Created attachment 11833 [details]
slight speedup
Comment 2 thorsten.martens 2003-12-12 14:10:44 UTC 
TM->MH: Please have a look, thanks !
Comment 3 Martin Hollmichel 2003-12-17 13:58:04 UTC 
reassigned.
Comment 4 matthias.huetsch 2004-02-13 17:25:04 UTC 
Hi Michael,

Well, weighting the 'loss of generality' versus this 'micro optimization', my
(purely personal) take on this is that I would wish you would join the 'framework'
team in their 'macro optimization' attempts for OOo 2.0, instead. What we really
need is ~5000ms of (cold) startup improvement, not ~50ms. I see that 100 such
patches get us there as well, but this entire 'pagein' thing is in itself an ugly
hack of last resort. I don't like to spend any more effort here, except adapting
to a possibly changing list of shared objects to preload.

Any contribution that (e.g.) helps to refactor 'libsvx...so' into a couple of
smaller shared objects would provide much higher overall benefit is desperately
needed and highly acknowedged.

Of course, this is all just my 2 cents. If you insist in applying this patch,
you a free to do so (you do have commit access, don't you?), but I won't.

Thanks anyway,
Matthias
Comment 5 matthias.huetsch 2004-02-13 17:25:47 UTC 
Reopen for reassignment...
Comment 6 matthias.huetsch 2004-02-13 17:26:34 UTC 
Reassigning...
Comment 7 matthias.huetsch 2004-02-13 17:27:56 UTC 
Resolving *now* (I will never learn this sequence...)
Comment 8 matthias.huetsch 2004-02-13 17:28:21 UTC 
Closing...
Comment 9 mmeeks 2004-02-16 09:29:54 UTC 
you're right - it's a pretty crackers patch.