Created attachment 63053 [details] Chromium Nightly build When background-size property is used, the background image moves, and the rendering is slow. Especially with move or dragdrop operation. I created a sample page, and I put in attachments screenshots of chromium with a recent webkit and midori with a old webkit. This bug has been verified with Chromium stable, Chromium nightly build, Safari stable, Safari nightly build, and Qt Webkit. According to tests, it would appear between version WebKit/531.2 (Midori) and AppleWebKit/532.4 version (Safari).
Created attachment 63054 [details] Midori (old version)
(In reply to comment #0) > Created an attachment (id=63053) [details] > Chromium Nightly build > > When background-size property is used, the background image moves, and the rendering is slow. Especially with move or dragdrop operation. > > I created a sample page, and I put in attachments screenshots of chromium with a recent webkit and midori with a old webkit. > > This bug has been verified with Chromium stable, Chromium nightly build, Safari stable, Safari nightly build, and Qt Webkit. > > According to tests, it would appear between version WebKit/531.2 (Midori) and AppleWebKit/532.4 version (Safari). According to tests, it seems that the bug was *introduced* between version WebKit/531.2 (Midori) and AppleWebKit/532.4 version (Safari).
What does the "defaults" mean in the title?
This looks like bug 26747, which <http://trac.webkit.org/changeset/46002> was supposed to fix.
Created attachment 63228 [details] Simper testcase
Created attachment 63230 [details] Just one file this time
Might this be the low-quality scaling heuristic kicking in and never bothering to clean up after itself?
(In reply to comment #7) > Might this be the low-quality scaling heuristic kicking in and never bothering to clean up after itself? This doesn’t appear to be the case.
I think this might be a side-effect of CG's scaling algorithm.
I just tried the example on Chrome 16.0.905.0 canary on Windows 7, and I can see the same artifacts. So It doesn't seem to be CG specific.
Safari, Chrome, and Firefox show the same rendering behavior for this test case. I do not believe any further compatibility issue remains.
<rdar://problem/96852062>
I confirm the bug has been killed at some point during the last 12 years.