1178965 - SourceSurfaceD2D1.cpp and Factory::GetD2D1Device...

Status: RESOLVED FIXED

(Core :: Graphics, defect)

Description

[Milan Sreckovic [:milan] (needinfo for best results)]

Attachment: Check for the failing GetD2D1Device.

+++ This bug was initially created as a clone of Bug #1155626 +++

Bug 1155626 fixed our assumption that Factory::GetD2D1Device always succeeds inside of DrawTargetD2D1.cpp, but not inside of SourceSurfaceD2D1.cpp.  Should we?

Comment 1

[Milan Sreckovic [:milan] (needinfo for best results)]

Comment on attachment 8627877 [details] [diff] [review]
Check for the failing GetD2D1Device.

Also, is mDevice used for anything?  Can we just get rid of it?

Comment 2

[Nicolas Silva [:nical]]

Comment on attachment 8627877 [details] [diff] [review]
Check for the failing GetD2D1Device.

Review of attachment 8627877 [details] [diff] [review]:
-----------------------------------------------------------------

::: gfx/2d/SourceSurfaceD2D1.cpp
@@ +78,5 @@
>    if (mRealizedBitmap) {
> +    return true;
> +  }
> +
> +  // Why aren't we using mDevice here or anywhere else?

Looking at the hg history, it looks like SourceSurfaceD2D1 used to always query the device from the factory until Bug 1078693, in which Bas added a mDevice member to check whether the device has changed since the creation of the surface (in which case the surface becomes invalid).

We could use mDevice in most places in this file.

Comment 3

[Milan Sreckovic [:milan] (needinfo for best results)]

Attachment: Check for the failing GetD2D1Device. Carry r=nical

Same code, just updated the comment in the patch with the bug number, etc.

Comment 4

[Pulsebot]

https://hg.mozilla.org/integration/mozilla-inbound/rev/de77ffb0f768

Comment 5

[Carsten Book [:Tomcat]]

https://hg.mozilla.org/mozilla-central/rev/de77ffb0f768