I've found out that the problem appears especially when I have nested structure with masks 

Could you provide an example? (That would eliminate the need for guessing.)

Based on the description I got the attached test. Does that reproduce the issue? If not, what is it missing?

Deleted: mask-image-and-clip-path.html 720 bytes

mask-image-and-clip-path.html 720 bytes View Download

I've prepared repo with exactly this issue. It worked well in the previous version on Chrome. Please look at the index.html. https://github.com/JustineOwl/svg-masks

Thank you for providing more feedback. Adding requester "durga.behera@chromium.org" to the cc list and removing "Needs-Feedback" label.

For more details visit https://www.chromium.org/issue-tracking/autotriage - Your friendly Sheriffbot

Thank you!

Setting P1 until we can find out what the breaking change was. Bisect please.

Able to reproduce the issue on Windows-7, Mac-10.12.3 and Linux Ubuntu-14.04 using chrome stable version 57.0.2987.133 and canary 59.0.3062.0.This is regression issue ,broken in M57.
Narrow Bisect::
===============
Good::57.0.2946.0 --   (build revision 437422)
Bad ::57.0.2947.0 --   (build revision 437705)
After executing the per-revision bisect script , got the following CL's between good and bad build versions
===========================================
https://chromium.googlesource.com/chromium/src/+log/9b853cfcf82d8c57487cb0dd8d1eeaa5ca40260f..3de2a45741756dfa0640ce1037a76ee9eedabfbb

Review-Url: https://codereview.chromium.org/2558633005

smcgruer- Could you please look into this issue, if it's related to your change, if not could you please help us to reassign this issue to the right owner.

Thanks.

FYI: To reproduce this with Chromium, you either need to build with proprietary codecs, or swap the MP4 with a webm.

This is my change, though I don't yet understand the full reason why (see below). We could revert, but it will then regress  issue 615870  . schenney, chrishtr I'm open to your thoughts there.



As for root-causing this, unfortunately I moved away from the clip-path stuff and so am not very familiar with the code. I compared the devtools-reported layer tree with/without my original change in https://crrev.com/3de2a45741756dfa0640ce1037a76ee9eedabfbb and found an interesting change in the reported position of some of the layers:

With my change:

document (810x834) (0,0)
  figure.figure (699x780) (48,16)
    .video-wrapper (699x780) (16777278,16777252)
      div (640x360) (16777215,16777215)
      div (640x318) (16777215,16777215)

Without my change:

document (810x834) (0,0)
  figure.figure (699x780) (48,16)
    .video-wrapper (699x780) (63,38)
      div (640x360) (0,0)
      div (640x318) (0,0)

Not P1 at this point. We make some compositing decisions based on whether there is a clip property on elements, so presumably that is causing the changes in position, but those are some really odd positions.

If this is video then it's being directly composited and the issue is probably in the compositor.

Just to Update,

Able to reproduce this issue on Windows 10 with chrome #59.0.3068.1 as per comment #6


Thank You...

Just to Update,
Still we are able to reproduce the issue on Windows-7 using chrome canary 60.0.3074.0 as per comment#6.

Thanks.

smcgruer, would you be so kind as to generate the test case with the webm and attach it here. Then you can mark it Available.

Reproduction attached. It could be made simpler I think, but this is a start.

Deleted: index.html 1.1 KB

index.html 1.1 KB View Download

Deleted: big-buck-bunny_trailer.webm 2.1 MB

	big-buck-bunny_trailer.webm 2.1 MB View Download

Deleted: screen3_ipad_large.jpg 230 KB

	screen3_ipad_large.jpg 230 KB View Download

Deleted: screen3_ipad_video_large_mask.svg 359 bytes

screen3_ipad_video_large_mask.svg 359 bytes Download

Deleted: screen3_ipad_video_large_clippath.svg 306 bytes

screen3_ipad_video_large_clippath.svg 306 bytes Download

Deleted: screen3_ipad_large_mask.svg 314 bytes

screen3_ipad_large_mask.svg 314 bytes Download

Deleted: screen3_ipad_large_clippath.svg 335 bytes

screen3_ipad_large_clippath.svg 335 bytes Download

This issue has been Available for over a year. If it's no longer important or seems unlikely to be fixed, please consider closing it out. If it is important, please re-triage the issue.

Sorry for the inconvenience if the bug really should have been left as Available.

For more details visit https://www.chromium.org/issue-tracking/autotriage - Your friendly Sheriffbot

This might just be a repeat of a bug filed a couple of days ago about our handling of layers when both clip and mask is present.