Crash report: https://crash-stats.mozilla.org/report/index/3779fc4d-1e25-4a24-a132-b0bfa0230325

Reason: SIGSEGV / SEGV_MAPERR

Top 10 frames of crashing thread:

0  libGLES_mali.so  libGLES_mali.so@0x2bddb0  
1  libGLES_mali.so  libGLES_mali.so@0x446bda  
2  libGLES_mali.so  libGLES_mali.so@0x2b913e  
3  libGLES_mali.so  libGLES_mali.so@0x2b80da  
4  libGLES_mali.so  libGLES_mali.so@0x2b8a4a  
5  libGLES_mali.so  libGLES_mali.so@0x446bda  
6  libGLES_mali.so  libGLES_mali.so@0x4179c2  
7  libGLES_mali.so  libGLES_mali.so@0x408e02  
8  libGLES_mali.so  libGLES_mali.so@0x2badbe  
9  libGLES_mali.so  libGLES_mali.so@0x2b80da  

There are quite a few libGLES_mali.so signatures which have started crashing in 113. (The signatures I've added to this bug are not exhaustive.) All of them have brush_blend or brush_blend_ALPHA_PASS as the GraphicsCompileShader crash annotation. First build ID is March 24th. Bug 1823411 landed the previous day so is almost certainly the cause.

I have ordered a Galaxy J2 Prime, which is the most common device from the most common signature.

Set release status flags based on info from the regressing bug 1823411

The bug is linked to topcrash signatures, which match the following criterion:

• Top 10 AArch64 and ARM crashes on nightly

:gw, could you consider increasing the severity of this top-crash bug?

For more information, please visit auto_nag documentation.

The bug is marked as tracked for firefox113 (nightly). We have limited time to fix this, the soft freeze is in 3 days. However, the bug still isn't assigned.

:bhood, could you please find an assignee for this tracked bug? Given that it is a regression and we know the cause, we could also simply backout the regressor. If you disagree with the tracking decision, please talk with the release managers.

For more information, please visit auto_nag documentation.

Sorry I should have self assigned. I expect the fix to be simple, but am waiting for a device to arrive. It should arrive in the next day

For posterity, since we'll lose this data when the crash reports dissappear:

• This appears to affect basically the whole range of Mali-T GPUs, from T628 through to T820.
• Driver versions from r7p0 to r12p1
• Android SDK levels predominantely 23 and 24, but there are some 25 too.

I have the device, and this is the line which causes the crash. Reverting that the highp fixes it.

Since we're so late in the cycle I plan to simply land a fix ASAP. Then I will follow up by spending some more time trying to figure out what the actual underlying issue is, so that we can add a test and avoid running in to it again.

Since landing bug 1823411 we have been receiving crash reports on a
variety of Mali-T devices when attempting to compile the brush_blend
shader. This appears to be due to changing v_color_mat to mediump,
thougn the reason why that crashes is currently unknown. This patch
reverts it to highp to avoid the crash.

This is being landed as-is due to being so late in the cycle, in order
to prevent crashes making it to beta. Further work should be to
determine precisely what conditions cause the crash, and add a test to
ensure we do not encounter it again.

https://hg.mozilla.org/mozilla-central/rev/88039c2ec513

https://github.com/servo/webrender/wiki/Driver-issues#bug-1825518-crash-with-mediump-on-mali-t