SVG element with opacity does not display correct color
Reported by
mariko.k...@gmail.com,
Oct 9 2017
|
|||||||||||||||
Issue descriptionUserAgent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_12_6) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/61.0.3163.100 Safari/537.36 Example URL: https://codepen.io/kosamari/pen/XeqrYB Steps to reproduce the problem: 1. create 3+ SVG rect elements with fillColor 2. apply opacity to each rect What is the expected behavior? Each rect element displays less opaque color of its fillColor. What went wrong? 3rd rect element and after are displayed with same color as 2nd rect element. test case : https://codepen.io/kosamari/pen/XeqrYB screen shot also attached This bug was spotted on Mac OS and only happens when hardware acceleration is ON. Does it occur on multiple sites: N/A Is it a problem with a plugin? No Did this work before? N/A Does this work in other browsers? Yes Chrome version: 61.0.3163.100 Channel: stable OS Version: OS X 10.12.6 Flash Version: Shockwave Flash 27.0 r0
,
Oct 9 2017
,
Oct 10 2017
,
Oct 10 2017
Unable to reproduce the issue on reported chrome version 61.0.3163.100 as we have checked it on Mac 10.12.6 which has Intel graphics of 5000 by enabling hardware acceleration. @Reporter: Could you please confirm do we need to have Intel graphics 6000 in order to reproduce the issue? Thanks!
,
Oct 10 2017
,
Oct 11 2017
Experiencing the same issue, also only with hardware acceleration. Colours change as you scroll (see attached screenshots of the test case): Windows 10 - 1703 Build 15063.632 Version 61.0.3163.100 (Official Build) (64-bit) Intel HD 4600 - Intel Driver: 20.19.15.4624
,
Oct 11 2017
Rob, the initial report is from Mac, but I'm wondering if this is related to the clear bugs you've been looking at.
,
Oct 11 2017
From the symptoms it seems like it could be a duplicate of crbug.com/769760 . I haven't been able to repro it locally however.
,
Oct 11 2017
Apropos #8, could someone who can repro the issue see if it is fixed in the 63.0.3231.0 (or above) Canary?
,
Oct 11 2017
Hi there, sorry I forgot to mention on the original post. I'm seeing the same issue on Version 63.0.3236.0 (Official Build) canary (64-bit) as well.
,
Oct 11 2017
Thank you for providing more feedback. Adding requester "vamshi.kommuri@techmahindra.com" to the cc list and removing "Needs-Feedback" label. For more details visit https://www.chromium.org/issue-tracking/autotriage - Your friendly Sheriffbot
,
Oct 11 2017
robertphillips@, has the fix for the opacity issues made it into Canary yet? I'm having a hard time keeping track of it.
,
Oct 11 2017
That canary from #10 (63.0.3236.0) includes the fixes for both the clearing issues ( crbug.com/755871 & crbug.com/768134 ) and for one of the deferred proxy issues ( crbug.com/769760 ). It doesn't include the fix for the second deferred proxy issue ( crbug.com/769898 ). Note that, I don't think the fix for the Mac-specific clearing bug (https://skia-review.googlesource.com/c/skia/+/55800) would cover the HW under discussion here (Intel HD 4600 AFAICT).
,
Oct 11 2017
robertphillips@, assigning to you for now. Please assign back to me if this turns out to be unrelated to the existing Skia issues. Is it possible that this is due to the second deferred proxy issue you mention? If so, we should likely wait for the next canary and re-confirm that the issue is fixed/not-fixed.
,
Oct 11 2017
A bit more info: This also repros (w/ the 63.0.3236.0 canary) on a: MacBook Pro w/ a HD Graphics 4000 (Ivy Bridge) MacBook Pro w/ an Iris Graphics 6100 (Broadwell) It doesn't repro on a MacBook Pro w/ a HD Graphics 630 (Kaby Lake) (w/ 60.0.3112.101).
,
Oct 11 2017
,
Oct 11 2017
Replying to #14, it looks like this may be a duplicate of crbug.com/768134 . The Mac-specific work-around for that bug (https://skia-review.googlesource.com/55800) was restricted to Intel Iris Pro GPUs but it appears that the problem is more wide-spread.
,
Oct 12 2017
The following revision refers to this bug: https://skia.googlesource.com/skia/+/7f9b215695aae262797758b2e2a7b1d35fbe584f commit 7f9b215695aae262797758b2e2a7b1d35fbe584f Author: Robert Phillips <robertphillips@google.com> Date: Thu Oct 12 13:42:01 2017 Force all Intel GPUs to use draws instead of clears on Macs Bug: 773107 Change-Id: I931d9623a359e264dfee35a2ea6c591cdf83a052 Reviewed-on: https://skia-review.googlesource.com/58580 Reviewed-by: Brian Salomon <bsalomon@google.com> Commit-Queue: Robert Phillips <robertphillips@google.com> [modify] https://crrev.com/7f9b215695aae262797758b2e2a7b1d35fbe584f/src/gpu/gl/GrGLCaps.cpp
,
Oct 16 2017
The fix in #18 rolled into Chrome in https://chromium-review.googlesource.com/c/chromium/src/+/719419 at r508840 on 10/13.
,
Oct 16 2017
Expanding on #19, this fix has also made it into the 64.0.3241.0 Canary. I have verified that the issue is fixed in that Canary. Requesting cherry-pick back to M63.
,
Oct 17 2017
Your change meets the bar and is auto-approved for M63. Please go ahead and merge the CL to branch 3239 manually. Please contact milestone owner if you have questions. Owners: cmasso@(Android), cmasso@(iOS), gkihumba@(ChromeOS), govind@(Desktop) For more details visit https://www.chromium.org/issue-tracking/autotriage - Your friendly Sheriffbot
,
Oct 17 2017
Please merge your change to M63 branch 3239 before 4:00 PM PT so we can take it in for tomorrow's dev release. Thank you.
,
Oct 17 2017
The following revision refers to this bug: https://skia.googlesource.com/skia/+/5a13102e94e12ff900097e61d53f84029196a104 commit 5a13102e94e12ff900097e61d53f84029196a104 Author: Robert Phillips <robertphillips@google.com> Date: Tue Oct 17 17:46:46 2017 [M63 cherrypick] Force all Intel GPUs to use draws instead of clears on Macs No-Tree-Checks: true No-Try: true No-Presubmit: true Bug: 773107 Change-Id: I8ee0f9c08a8bec68ec76d8d7e42688a411b49f54 Reviewed-on: https://skia-review.googlesource.com/60528 Reviewed-by: Brian Osman <brianosman@google.com> [modify] https://crrev.com/5a13102e94e12ff900097e61d53f84029196a104/src/gpu/gl/GrGLCaps.cpp
,
Oct 17 2017
,
Nov 1 2017
The primary symptoms of this bug have been patched and we have several performance bugs to drive the improvement of the work around so I am closing this issue. |
|||||||||||||||
►
Sign in to add a comment |
|||||||||||||||
Comment 1 by pbomm...@chromium.org
, Oct 9 2017