Thanks for filing the issue.
Could you please provide us a sample file/test case to triage the issue further.

I did provide an URL in my description ;-)
(NB: the bug at least appears with nVIDIA GPUs).

Thank you for providing more feedback. Adding requester "jmukthavaram@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

Unable to reproduce this issue on Ubuntu 14.04 using chrome latest stable #61.0.3163.100. 

fabrice.neyret@ Could you please confirm is this issue still observed on latest stable #61.0.3163.100? If yes, provide chrome://gpu details for further triage.

Thanks!

seems to have been solved. Same for return in switch.

For many other glslES bugs (either on linux or all OS), see acid tests here: https://www.shadertoy.com/results?query=bug&sort=newest&filter=
(should I really fill a form for each ? moreover it's unclear to me where to do it, since chippers + drivers + OS + browsers are concerned. Besides, "web-glsl" seems to differ to desktop-glsl bugs. )

Thank you for providing more feedback. Adding requester "brajkumar@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

kbr@ do you think there are specific actions to take here?

Can't reproduce on my Mac (AMD) and Linux (NVidia) in Chrome Stable (M61)

You will have to provide about:gpu from your Chrome on where you can reproduce this issue.

Either it's fixed already or it's a driver specific bug.

since it seems to be solved ( 6 month later), no, there is no action to do on this very bug.
Many more are acid tested here  https://www.shadertoy.com/results?query=bug&sort=newest , but they are most probably due to drivers than to the browser.

Thank you for providing more feedback. Adding requester "zmo@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

Thanks. I looked at a few of them, and I feel we should address these bugs, either work around them in browser (ANGLE's shader translator), or report them to driver vendors to get them fixed.

Do you think you are willing to file crbug.com bugs one for each these bugs? With shader source (reduced to minimum) in the bug report, and chrome's about:gpu content of the platform that can reproduce the bug. Especially when you discover new bugs.

That's the only way we can keep improving drivers and browsers conformance. On Chrome GPU team side, we don't have the bandwidth right now to keep track of client sites, so we can only react to bugs reported to us.

In the meantime, closing this as Fixed.

Ok, I just enter a bunch of them.

- it's never clear where to declare these bugs (so many layers involved ! ), so it's very cool if you can reroute where it should, depending on the very case.

- it was not easy to guess the right tags here. I opted for "Content". But "it's the browser" is ambiguous (are Angle and drivers considered part of it ?). And the question "is it ok on other browsers" stucks us when several yes and several no, typically when it's a windows vs linux+Mac(i.e. openGl) issue.

thanks !

Thanks Fabrice. Good tags to use are Internals>GPU, Blink>WebGL in components. I added those to your issues to make them more visible.

thanks for that (I didn't even saw where I could have set these :-) )