Closed
Bug 46704
Opened 25 years ago
Closed 24 years ago
investigate imagelib cache performance
Categories
(Core :: Graphics: ImageLib, defect, P3)
Tracking
()
RESOLVED
FIXED
People
(Reporter: warrensomebody, Assigned: slogan)
Details
(Keywords: memory-footprint, perf, Whiteboard: [nsbeta3-])
Attachments
(2 files)
10.15 KB,
patch
|
Details | Diff | Splinter Review | |
4.08 KB,
patch
|
Details | Diff | Splinter Review |
From the footprint meeting 7/26/00:
3) Image Library
Description: We have anecdotal information that the image memory cache (not
the network memory cache) is not being hit
very often. We also don't know if the size of the cache is bounded and if it a
memory flush listener.
Module owner: pnunn@netscape.com
Task owner: syd@netscape.com (?)
Status: Syd to work with Pam to analyze image memory cache usage and size
limits.
Reporter | ||
Updated•25 years ago
|
Info I sent out responding to Vidur's email:
<quote>
In the interest of speeding up the process, here's some initial info:
1. The image cache is hit when the default channel load attribute for an image
is **not** set
to FORCE_VALIDATION, VALIDATE_ALWAYS, FORCE_RELOAD, and
INHIBIT_PERSISTANT_CACHING. When an image container is first created, it is
set
to not use the imgcache.
2. Size of cache is bounded to 2M. See IL_SetCacheSize()
in ImageManagerImpl::ImageManagerImpl() in mozilla/gfx/src/nsImageManager.cpp
3. I don't understand what you mean by a memory flush listener. Send me some
more
info about what you want to investigate and I'll dig out what I can.
-p
ps. I'm currently looking to see how internal (chrome) images are requested.
They should get pulled
out of the imgcache, but seem to get a force-validation from somewhere that
removes them from
the imgcache.
<unquote>
Reporter | ||
Comment 2•25 years ago
|
||
footprint meeting 8/9/00 status: dougt added a pref but its value is still set
to 2M. Someone needs to investigate what a good number will be.
Doug: Please update this bug with status.
Reporter | ||
Comment 3•25 years ago
|
||
Marking nsbeta3+ as per the 8/9/00 footprint meeting.
Whiteboard: [nsbeta3+]
This bug depends on the following bugs:
http://bugzilla.mozilla.org/show_bug.cgi?id=47662 duplicate images in
necko/imglib caches.
http://bugzilla.mozilla.org/show_bug.cgi?id=49097 only cache chrome, other types
of images.
http://bugzilla.mozilla.org/show_bug.cgi?id=49098 cache chrome animations
This bug depends on:
http://bugzilla.mozilla.org/show_bug.cgi?id=46704 -- keep chrome images in cache
Patch checked in. This patch only addresses giving
priority to chrome images in the image cache.
-p
Well....patch was backed out due to a crash.
Syd sent a new patch.
I've tested. I found one more place where chrome images
were being removed from the imgcache. I fixed that and
added an #ifdef we can change if we want to switch
"pinning chrome" on and off.
Tested on linux & nt.
Will check in asato (AsSoonAsTreeOpens)
-p
Comment 10•24 years ago
|
||
checked in with most recent changes.
-p
Comment 11•24 years ago
|
||
sorry cannto ahve blanket nsbeta3+ bugs at p3 or below. please
mark this fixed if your work is done.
Whiteboard: [nsbeta3+] → [nsbeta3-]
Updated•24 years ago
|
QA Contact: elig → tpreston
Comment 12•24 years ago
|
||
syd:
I'm closing this bug, ok. No need for it to be on our
radar. Imgcache issues are handled in other bugs.
-pn
Status: ASSIGNED → RESOLVED
Closed: 24 years ago
Resolution: --- → FIXED
You need to log in
before you can comment on or make changes to this bug.
Description
•