Join GitHub today
GitHub is home to over 50 million developers working together to host and review code, manage projects, and build software together.
Sign upMake ClipScrollGroup per ClipScrollInfo #1500
Conversation
|
looks solid to me! |
|
Couple of minor comments. |
| @@ -294,6 +295,7 @@ pub struct FrameBuilder { | |||
|
|
|||
| stacking_context_store: Vec<StackingContext>, | |||
| clip_scroll_group_store: Vec<ClipScrollGroup>, | |||
| clip_scroll_group_indices: HashMap<ClipAndScrollInfo, ClipScrollGroupIndex>, | |||
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
| @@ -1050,6 +1046,7 @@ impl FrameBuilder { | |||
| run: TextRun { | |||
| font_key, | |||
| logical_font_size: size, | |||
| run_offset, | |||
This comment has been minimized.
This comment has been minimized.
glennw
Jul 19, 2017
Member
There is an offset field in text runs with this #1498 - perhaps we can make use of that here, which applies the offset in the vertex shader? That could be done as a follow up though.
This comment has been minimized.
This comment has been minimized.
mrobinson
Jul 20, 2017
Author
Member
The newest version just reuses this offset. Thanks for the suggestion!
|
|
6c5fa65
to
3fd614c
|
@bors-servo r+ |
|
|
|
|
|
|
|
Needs a rebase. |
Instead of making each ClipScrollGroup per stacking context and ClipScrollInfo combination, make them per ClipScrollInfo. This should reduce the amount of work done per stacking context and is the first step toward accepting all coordinates relative to reference frames.
3fd614c
to
028f2bf
|
Rebased. Going to land the new version. Thank you both for the reviews! @bors-servo r=glennw |
|
|
…ps, r=glennw Make ClipScrollGroup per ClipScrollInfo Instead of making each ClipScrollGroup per stacking context and ClipScrollInfo combination, make them per ClipScrollInfo. This should reduce the amount of work done per stacking context and is the first step toward accepting all coordinates relative to reference frames. <!-- Reviewable:start --> --- This change is [<img src="https://reviewable.io/review_button.svg" height="34" align="absmiddle" alt="Reviewable"/>](https://reviewable.io/reviews/servo/webrender/1500) <!-- Reviewable:end -->
|
|
mrobinson commentedJul 19, 2017
•
edited by larsbergstrom
Instead of making each ClipScrollGroup per stacking context and
ClipScrollInfo combination, make them per ClipScrollInfo. This should
reduce the amount of work done per stacking context and is the first
step toward accepting all coordinates relative to reference frames.
This change is