Redesigned clip and scroll API #3251

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Merged

bors-servo merged 1 commit into servo:master from kvark:clip-id

Jan 8, 2019

Member

kvark commented Oct 30, 2018 •

edited

Loading

Fixes https://bugzilla.mozilla.org/show_bug.cgi?id=1503447
Note: this description is edited to contain the relevant up-to-date information about the code.

General idea

This is something I wanted for a long time: the spatial hierarchy that isn't mixed up with the clip hierarchy. The change dragged a bunch of other API fixes that were accumulated in our technical debt, so it grew rather big/scary.

The motivation is - removing the complexity of the API that we inherited from the past. Both Servo and Gecko are in a better position to explicitly tell us what the clip/spatial properties of items are, and they don't need to mix them up in the same type. The old API had more complex logic, more state, and caused the Gecko integration to be painful and non-solid (there are edge cases not handled properly).

API changes:

separate SpatialId from ClipId 🎉
remove the clip/scroll stack from the API, all the SpatialId and ClipId are passed explicitly for display items in the form of SpaceAndClipInfo, which roughly corresponds to the old ClipAndScrollInfo.
~~reference frame are no longer pushed/popped and instead are "defined", similar to scroll/sticky frames.~~ Edit: this ended up harder than I expected, postponing until the main change is done.
stacking contexts no longer have an internal ClipId passed in (and the standard one being ignored), instead they use the standard one from the associated SpaceAndClipInfo.
define_clip_with_parent and define_scroll_frame_with_parent are removed.

TODO:

reviews:
- here
- Gecko part
- Servo part
integrate with Gecko and provide a good try push:
- https://treeherder.mozilla.org/#/jobs?repo=try&selectedJob=218192168&revision=500bb4429beb003927afb5e14201587519dce462
integrate with Servo:
- kvark/servo@f671511
- there are failures, but they are exactly the same as without this change

This change is

kvark requested a review from gw3583

October 30, 2018 19:25

kvark force-pushed the clip-id branch from 372c76a to dea9fa5 Compare

October 30, 2018 19:37

Contributor

gw3583 commented Oct 30, 2018

It would be good to also provide a patch for Servo as part of this PR, or some detailed instructions on what will be required to change in the next WR update in Servo.

Contributor

gw3583 commented Oct 30, 2018

(I haven't reviewed this yet, btw - just an initial comment above)

kvark force-pushed the clip-id branch from dea9fa5 to 105eab1 Compare

October 31, 2018 00:33

mrobinson requested changes

View reviewed changes

Member

mrobinson left a comment

It's great to finally see this change! I did a quick review of most of this patch and I'm happy to take a closer look later.

webrender_api/src/display_list.rs Outdated

@@ @@ -63,6 +63,12 @@ impl<T> ItemRange<T> { @@
                   }
               }
+              #[derive(Clone, Debug)]
+              pub enum ClipParent {
+                  Inherit,

Member

mrobinson Oct 31, 2018

Maybe this should be something like FromStack? I mention this because the Clip isn't really inherited from a parent in a real sense. It's just taken from the ClipStack itself, which is a convenience.

webrender/src/display_list_flattener.rs

-                          ClipId::Spatial(..) => {
-                              self.spatial_node_map[&id]
-                          }
-                          ClipId::ClipChain(_) => panic!("Tried to use ClipChain as scroll node."),

Member

mrobinson Oct 31, 2018

It's great to see this go!

webrender/src/display_list_flattener.rs Outdated

    
                      let root_scroll_node = ClipId::root_scroll_node(pipeline_id);

                      let reference_frame_info = self.simple_scroll_and_clip_chain(&root_scroll_node);

Member

mrobinson Oct 31, 2018

Why are you adding the background color to the scroll frame here? Won't that mean it will scroll with the contents?

Member Author

kvark Oct 31, 2018

nicely spotted! This is a mistake

webrender/src/display_list_flattener.rs Outdated

@@ @@ -332,7 +315,7 @@ impl<'a> DisplayListFlattener<'a> { @@
                       item: &DisplayItemRef,
                       info: &StickyFrameDisplayItem,
                       clip_and_scroll: &ScrollNodeAndClipChain,
-                      parent_id: &ClipId,
+                      //parent_id: &ClipId,

Member

mrobinson Oct 31, 2018

This should probably be deleted.

Contributor

staktrace Dec 27, 2018

Yup please drop this

webrender/src/display_list_flattener.rs Outdated

    
                      new_node_id: ClipId,

                      parent_id: ClipId,

                      new_node_id: SpatialId,

                      cs_info: &ClipAndScrollInfo,

Member

mrobinson Oct 31, 2018

Why not just pass parent_spatial_id here since you don't need the ClipId?

webrender/src/display_list_flattener.rs Outdated

@@ @@ -396,12 +376,12 @@ impl<'a> DisplayListFlattener<'a> { @@
                       pipeline_id: PipelineId,
                       item: &DisplayItemRef,
                       reference_frame: &ReferenceFrame,
-                      scroll_node_id: ClipId,
+                      cs_info: &ClipAndScrollInfo,

Member

mrobinson Oct 31, 2018

It looks like you can just pass the SpatialId here.

webrender/src/display_list_flattener.rs Outdated

@@ @@ -417,7 +397,7 @@ impl<'a> DisplayListFlattener<'a> { @@
                       pipeline_id: PipelineId,
                       item: &DisplayItemRef,
                       stacking_context: &StackingContext,
-                      scroll_node_id: ClipId,
+                      scroll_node_id: SpatialId,

Member

mrobinson Oct 31, 2018

Maybe a good time to rename this to spatial_node_id? since the scroll_node_id thing is no longer accurate?

webrender/src/display_list_flattener.rs Outdated

@@ @@ -1310,8 +1283,10 @@ impl<'a> DisplayListFlattener<'a> { @@
                           self.prim_store.chase_id = Some(prim_index);
                       }
+                      let root_space_id = SpatialId::root_reference_frame(pipeline_id);

Member

mrobinson Oct 31, 2018

Maybe this could be root_reference_frame_id or something like that, since root_space_id is a little ambiguous.

webrender/src/display_list_flattener.rs Outdated

-                      let spatial_node = self.id_to_index_mapper.get_spatial_node_index(parent_id);
+                      let mut parent_clip_chain_index = match cs_info.clip_node_id {
+                          Some(clip_id) => self.id_to_index_mapper.get_clip_chain_id(clip_id),
+                          None => ClipChainId::NONE,

Member

mrobinson Oct 31, 2018

This looks like it might cause a lot of code changes in Servo and Gecko? Currently, when this value is None, it effectively uses whatever the "implied" parent clip chain is. Now Servo and Gecko will need to track that value themselves and set it explicitly. Won't this cause a "no clip" situation instead?

Member Author

kvark Oct 31, 2018

This piece is not a part of the API, it's the DL processing. At this level, we just strictly respect the parent clip chain specified. It's up to the webrender_api/src/display_list.rs to ensure that the "implied" parent semantics is preserved.

I'm going to test Gecko shortly and see what the implications are. I'd expect wrench reftests to fail if this semantics was ported incorrectly.

webrender_api/src/display_item.rs Outdated

    
                  pub clip_node_id: Option<ClipId>,

              }

              impl ClipAndScrollInfo {

                  pub fn simple(node_id: ClipId) -> ClipAndScrollInfo {

                  pub fn simple(scroll_node_id: SpatialId) -> Self {

Member

mrobinson Oct 31, 2018

Now that these are separate ids, I'm not sure it is meaningful to have clip_node_id an Option<...>. Previously None here meant "use the implied ancestor of my spatial node. It doesn't look like that is possible any longer. Perhaps "no clip" should just be represented by the root clip.

Member Author

kvark Oct 31, 2018

I removed ClipAndScrollInfo altogether now. No need to group together loosely related things.

Contributor

gw3583 commented Oct 31, 2018

@mrobinson Thanks for jumping in to help with reviewing this! 👍

Member Author

kvark commented Oct 31, 2018 •

edited

Loading

@mrobinson it's a pleasant surprise to see you, thanks for taking a look!

kvark force-pushed the clip-id branch 2 times, most recently from 5c418ac to 4b670a5 Compare

November 2, 2018 16:10

Member Author

kvark commented Nov 2, 2018

Asking for a real review now, while I'm finishing testing Gecko side.

gw3583 requested changes

View reviewed changes

Contributor

gw3583 left a comment

Reviewed 4 of 14 files at r2, 11 of 11 files at r3.
Reviewable status: all files reviewed, 10 unresolved discussions (waiting on @mrobinson and @kvark)

wrench/reftests/performance/no-clip-mask.png, line 0 at r3 (raw file):
I'm somewhat surprised that this patch causes a difference here. Do we know why?

Contributor

gw3583 commented Nov 4, 2018

Just one question above, since it's not clear to me that this should have affected any reftest outputs (and if that has any implications for Gecko / Servo).

Otherwise, it seems like we need:

Try run
Finish up any remaining issues from @mrobinson 's comments.
Get sign off from @staktrace

Then this should be ready to merge.

Member Author

kvark commented Nov 6, 2018

@gw3583 I'm working on even more changes now that remove the clip-scroll stack completely from the API. More updates to come... and I'll double check the test. I assume there is a slight change in YAML reading semantics (e.g. parsing the "clip-and-scroll" field), since it diverges further from the actual WR API, and I'm trying to preserve compatibility with the existing reftests.

kvark changed the title ~~[WIP] Separate clipping and scrolling for good~~ [WIP] Redesigned clip and scroll API

kvark force-pushed the clip-id branch 2 times, most recently from b43652d to 9acbac0 Compare

November 7, 2018 16:26

Contributor

bors-servo commented Nov 12, 2018

☔ The latest upstream changes (presumably #3300) made this pull request unmergeable. Please resolve the merge conflicts.

kvark mentioned this pull request

[WIP] Remove explicit clip ID from stacking context #3311

Closed

Member Author

kvark commented Nov 15, 2018

I've got to document this somewhere: we have a few places in DL building where the clip is literally smuggled through the scroll ID, while the clip ID is silently dropped:

push_reference_frame
define_clip
possibly other

What happens is scroll_node_id gets through, and the DL flattener associates the produced clip (for define_clip) or scroll (for push_reference_frame) with the clip chain of the parent, hence the clipping survives even though the original clip_node_id is ignored.

But the worst thing is that there is code in Gecko that relies on that behavior...

kvark mentioned this pull request

Don't ignore the clip on the stack for reference frames and clips #3315

Merged

bors-servo pushed a commit that referenced this pull request


          Auto merge of #3315 - kvark:clip-respect, r=gw3583

Don't ignore the clip on the stack for reference frames and clips

Similar to #3311, this is a prelude to #3251, with the main purpose of getting rid of Gecko's hack in https://searchfox.org/mozilla-central/rev/d850d799a0009f851b5535580e0a8b4bb2c591d7/gfx/layers/wr/ClipManager.cpp#224

It also makes sense for WR in isolation (for the current clip/stack model, at least): in those spots addressed, we used to ignore the clip ID on the stack, only to re-interpret (smuggle!) the scroll ID as a clip. Now we are actually using the clip ID more consistently. See also - #3251 (comment)

Gecko try (that disables the hack in addition to having the WR change):
https://treeherder.mozilla.org/#/jobs?repo=try&selectedJob=212038686&revision=c10e5d839e024ba451a6622fa611c3525457f3ff
(Looks green! 🎉 )

<!-- 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/3315)
<!-- Reviewable:end -->

kvark force-pushed the clip-id branch from 9acbac0 to 4663a06 Compare

November 30, 2018 22:03

Contributor

bors-servo commented Dec 5, 2018

☔ The latest upstream changes (presumably #3384) made this pull request unmergeable. Please resolve the merge conflicts.

kvark force-pushed the clip-id branch 4 times, most recently from 11b4121 to 7b33fbd Compare

December 21, 2018 03:26

Member Author

kvark commented Dec 21, 2018

@gw3583 looks like the try is green: https://treeherder.mozilla.org/#/jobs?repo=try&selectedJob=218192168&revision=500bb4429beb003927afb5e14201587519dce462

kvark changed the title ~~[WIP] Redesigned clip and scroll API~~ Redesigned clip and scroll API

kvark force-pushed the clip-id branch from a9370e0 to bd1322e Compare

December 28, 2018 21:05

Member Author

kvark commented Dec 28, 2018

@staktrace alright, now everything is addressed :)

Contributor

bors-servo commented Jan 5, 2019

☔ The latest upstream changes (presumably #3472) made this pull request unmergeable. Please resolve the merge conflicts.


          WR API change with separate SpatialId from ClipId

fae42df

kvark force-pushed the clip-id branch from bd1322e to fae42df Compare

January 8, 2019 19:49

Member Author

kvark commented Jan 8, 2019

@gw3583 @jrmuizel Here is the latest up-to-date status:

Code is considered ready by me and the reviewers (thanks @staktrace !). Gecko side: https://phabricator.services.mozilla.com/D15230

Rebased try push shows a number of unexpected passes and 2 failures in Windows/Debug: https://treeherder.mozilla.org/#/jobs?repo=try&selectedJob=219135236&revision=a0bc8595ca76b1cae0e53220bfbf01550f185fba
I find this quite odd, since it's only in debug, and I didn't see those before the rebase. Launched a try from the same FF revision without my changes to compare: https://treeherder.mozilla.org/#/jobs?repo=try&selectedJob=219135236&revision=40d93ef7d636034ff37e7c9aa79762437b099055

@Darkspirit gave it a short run on Linux, didn't find any issues, was going to also run on Win10 today and I hope ping back.

These are the only issues to date. If all is good by the end of the day, we may proceed with landing.

Contributor

staktrace commented Jan 8, 2019

Rebased try push shows a number of unexpected passes and 2 failures in Windows/Debug: https://treeherder.mozilla.org/#/jobs?repo=try&selectedJob=219135236&revision=a0bc8595ca76b1cae0e53220bfbf01550f185fba I find this quite odd, since it's only in debug, and I didn't see those before the rebase.

I retriggered the job a few more times and it looks like that was just a random intermittent. The retriggers are all green so far.

Darkspirit commented Jan 8, 2019

LGTM

Contributor

staktrace commented Jan 8, 2019

Let's do this

Contributor

bors-servo commented Jan 8, 2019

📌 Commit fae42df has been approved by staktrace

Contributor

bors-servo commented Jan 8, 2019

⌛ Testing commit fae42df with merge e16ab88...

bors-servo pushed a commit that referenced this pull request


          Auto merge of #3251 - kvark:clip-id, r=staktrace

e16ab88

Redesigned clip and scroll API

Fixes https://bugzilla.mozilla.org/show_bug.cgi?id=1503447
Note: this description is edited to contain the relevant up-to-date information about the code.

## General idea
This is something I wanted for a long time: the spatial hierarchy that isn't mixed up with the clip hierarchy. The change dragged a bunch of other API fixes that were accumulated in our technical debt, so it grew rather big/scary.

The motivation is - removing the complexity of the API that we inherited from the past. Both Servo and Gecko are in a better position to explicitly tell us what the clip/spatial properties of items are, and they don't need to mix them up in the same type. The old API had more complex logic, more state, and caused the Gecko integration to be painful and non-solid (there are edge cases not handled properly).

## API changes:
 - separate `SpatialId` from `ClipId` 🎉
 - remove the clip/scroll stack from the API, all the `SpatialId` and `ClipId` are passed explicitly for display items in the form of `SpaceAndClipInfo`, which roughly corresponds to the old `ClipAndScrollInfo`.
  - ~~reference frame are no longer pushed/popped and instead are "defined", similar to scroll/sticky frames.~~ Edit: this ended up [harder than I expected](https://phabricator.services.mozilla.com/D13081), postponing until the main change is done.
  - stacking contexts no longer have an internal `ClipId` passed in (and the standard one being ignored), instead they use the standard one from the associated `SpaceAndClipInfo`.
  - `define_clip_with_parent` and `define_scroll_frame_with_parent` are removed.

## TODO:
- [ ] reviews:
  - [ ] here
  - [ ] Gecko part
  - [ ] Servo part
- [x] integrate with Gecko and provide a good try push:
  - https://treeherder.mozilla.org/#/jobs?repo=try&selectedJob=218192168&revision=500bb4429beb003927afb5e14201587519dce462
- [x] integrate with Servo:
  - kvark/servo@f671511
  - there are failures, but they are exactly the same as without this change

<!-- 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/3251)
<!-- Reviewable:end -->

Contributor

bors-servo commented Jan 8, 2019

☀️ Test successful - status-appveyor, status-taskcluster
Approved by: staktrace
Pushing e16ab88 to master...

bors-servo merged commit fae42df into servo:master

kvark deleted the clip-id branch

January 8, 2019 23:26

Member Author

kvark commented Jan 8, 2019

Wow, it's happening! 🚀

moz-v2v-gh pushed a commit to mozilla/gecko-dev that referenced this pull request


          Bug 1518605 - Update webrender to commit e16ab884672151fbdebc149479f9…

44eff8f

…bfdecd1da216 (WR PR #3251). r=kats

servo/webrender#3251

Differential Revision: https://phabricator.services.mozilla.com/D16005

--HG--
extra : moz-landing-system : lando

mykmelez pushed a commit to mykmelez/gecko that referenced this pull request


          Bug 1518605 - Update webrender to commit e16ab884672151fbdebc149479f9…

5ad9469

…bfdecd1da216 (WR PR #3251). r=kats

servo/webrender#3251

Differential Revision: https://phabricator.services.mozilla.com/D16005

staktrace pushed a commit to staktrace/webrender that referenced this pull request


          Bug 1518605 - Update webrender to commit e16ab88 (WR PR servo#3251). …

35b615e

…r=kats

servo#3251

Differential Revision: https://phabricator.services.mozilla.com/D16005

jdm mentioned this pull request

Update webrender servo/servo#22598

Merged

moz-v2v-gh pushed a commit to mozilla/gecko-dev that referenced this pull request


          Bug 1519441 - WR fix scroll frame rectangle r=gw

90036e2

This is a follow-up fix to servo/webrender#3251
Accidentally, the reference frame relative offset was applied twice.

Differential Revision: https://phabricator.services.mozilla.com/D16658

--HG--
extra : moz-landing-system : lando

moz-gfx pushed a commit to moz-gfx/webrender that referenced this pull request


          Bug 1519441 - WR fix scroll frame rectangle r=gw

0d498a4

This is a follow-up fix to servo#3251
Accidentally, the reference frame relative offset was applied twice.

Differential Revision: https://phabricator.services.mozilla.com/D16658

[wrupdater] From https://hg.mozilla.org/mozilla-central/rev/078e5984f999de05c3987fd04679b8703a042cb3

mykmelez pushed a commit to mykmelez/gecko that referenced this pull request


          Bug 1519441 - WR fix scroll frame rectangle r=gw

b4c35d2

This is a follow-up fix to servo/webrender#3251
Accidentally, the reference frame relative offset was applied twice.

Differential Revision: https://phabricator.services.mozilla.com/D16658

gecko-dev-updater pushed a commit to marco-c/gecko-dev-wordified that referenced this pull request


          Bug 1518605 - Update webrender to commit e16ab884672151fbdebc149479f9…

cff7ffc

…bfdecd1da216 (WR PR #3251). r=kats

servo/webrender#3251

Differential Revision: https://phabricator.services.mozilla.com/D16005

UltraBlame original commit: 9e0410ca1106ca09b30d29e48387d4447094eb79

gecko-dev-updater pushed a commit to marco-c/gecko-dev-comments-removed that referenced this pull request


          Bug 1518605 - Update webrender to commit e16ab884672151fbdebc149479f9…

e69064d

…bfdecd1da216 (WR PR #3251). r=kats

servo/webrender#3251

Differential Revision: https://phabricator.services.mozilla.com/D16005

UltraBlame original commit: 9e0410ca1106ca09b30d29e48387d4447094eb79

gecko-dev-updater pushed a commit to marco-c/gecko-dev-wordified that referenced this pull request


          Bug 1519441 - WR fix scroll frame rectangle r=gw

bf9b8e7

This is a follow-up fix to servo/webrender#3251
Accidentally, the reference frame relative offset was applied twice.

Differential Revision: https://phabricator.services.mozilla.com/D16658

UltraBlame original commit: 078e5984f999de05c3987fd04679b8703a042cb3

gecko-dev-updater pushed a commit to marco-c/gecko-dev-comments-removed that referenced this pull request


          Bug 1519441 - WR fix scroll frame rectangle r=gw

3519db0

This is a follow-up fix to servo/webrender#3251
Accidentally, the reference frame relative offset was applied twice.

Differential Revision: https://phabricator.services.mozilla.com/D16658

UltraBlame original commit: 078e5984f999de05c3987fd04679b8703a042cb3

gecko-dev-updater pushed a commit to marco-c/gecko-dev-wordified-and-comments-removed that referenced this pull request


          Bug 1518605 - Update webrender to commit e16ab884672151fbdebc149479f9…

e397f5a

…bfdecd1da216 (WR PR #3251). r=kats

servo/webrender#3251

Differential Revision: https://phabricator.services.mozilla.com/D16005

UltraBlame original commit: 9e0410ca1106ca09b30d29e48387d4447094eb79

gecko-dev-updater pushed a commit to marco-c/gecko-dev-wordified-and-comments-removed that referenced this pull request


          Bug 1519441 - WR fix scroll frame rectangle r=gw

d75d814

This is a follow-up fix to servo/webrender#3251
Accidentally, the reference frame relative offset was applied twice.

Differential Revision: https://phabricator.services.mozilla.com/D16658

UltraBlame original commit: 078e5984f999de05c3987fd04679b8703a042cb3

bhearsum pushed a commit to mozilla-releng/staging-firefox that referenced this pull request


          Bug 1518605 - Update webrender to commit e16ab884672151fbdebc149479f9…

2c90254

…bfdecd1da216 (WR PR #3251). r=kats

servo/webrender#3251

Differential Revision: https://phabricator.services.mozilla.com/D16005

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet