Bug 88136 – [chromium] Fix software rendering for device-scale-factor > 1

• Component: WebKit Misc.
• Status: RESOLVED FIXED

[Sadrul Habib Chowdhury]

The device-scale factor needs to be set correctly on the GraphicsContext.

[Sadrul Habib Chowdhury]

Created attachment 145377 [details]
Patch

[WebKit Review Bot]

Please wait for approval from [e-mail], [e-mail], [e-mail], [e-mail] or [e-mail] before submitting, as this patch contains changes to the Chromium public API. See also https://trac.webkit.org/wiki/ChromiumWebKitAPI.

[James Robinson]

Tests?

[Sadrul Habib Chowdhury]

Created attachment 145410 [details]
Patch

[Sadrul Habib Chowdhury]

(In reply to comment #3)
> Tests?

Added a reftest. Please take a look.

[Gustavo Noronha (kov)]

Comment on attachment 145410 [details]
Patch

Attachment 145410 [details] did not pass gtk-ews (gtk):
Output: http://queues.webkit.org/results/12863700

[Build Bot]

Comment on attachment 145410 [details]
Patch

Attachment 145410 [details] did not pass win-ews (win):
Output: http://queues.webkit.org/results/12870635

[Sadrul Habib Chowdhury]

Created attachment 145419 [details]
Patch

[Sadrul Habib Chowdhury]

Created attachment 145420 [details]
Patch

[WebKit Review Bot]

Comment on attachment 145420 [details]
Patch

Clearing flags on attachment: 145420

Committed r119423: <http://trac.webkit.org/changeset/119423>

[WebKit Review Bot]

All reviewed patches have been landed.  Closing bug.

[Alexey Proskuryakov]

The test added here crashes reproducibly on Mac WebKit2, <http://build.webkit.org/results/Lion%20Debug%20(WebKit2%20Tests)/r119496%20(7909)/fast/hidpi/device-scale-factor-paint-crash-log.txt>

Also, why are expected results for this test in /chromium? It's a retest.

Finally, [chromium] prefix is not appropriate here. Prefixes like this are meant to mark bugs that are of absolutely no interest to people working on other ports. This change is obviously touching non-chromium code.

[Sadrul Habib Chowdhury]

(In reply to comment #12)
> The test added here crashes reproducibly on Mac WebKit2, <http://build.webkit.org/results/Lion%20Debug%20(WebKit2%20Tests)/r119496%20(7909)/fast/hidpi/device-scale-factor-paint-crash-log.txt>
> 
> Also, why are expected results for this test in /chromium? It's a retest.

Indeed. I am unsure why the expected results were added (in http://trac.webkit.org/changeset/119442)

> 
> Finally, [chromium] prefix is not appropriate here. Prefixes like this are meant to mark bugs that are of absolutely no interest to people working on other ports. This change is obviously touching non-chromium code.

The 'main' change in this patch is in chromium, and the non-chromium changes are in InternalSettings (and related changes in WebCore.exp.in etc. files). So I had marked it as [chromium]. My bad.