1281126 - Crash in eme-adobe.dll@0x3ec3f4 mostly on AuthenticAMD family 15 model 107 stepping 1 | 2

Status: RESOLVED WONTFIX

(Core :: Audio/Video: Playback, defect, P3)

Description

[Chris Pearce [:cpearce (Not reading bugmail)]]

This bug was filed from the Socorro interface and is 
report bp-1e1465c6-24c1-4383-ad33-dd2f92160616.
=============================================================

Someone added an info bar to prompt users to submit unreported crash reports.

I expect this is the crash we experienced using the Adobe GMP for unencrypted decoding.

These are mostly on "AuthenticAMD family 15 model 107 stepping 1 | 2" and similar.

Comment 1

[Anthony Jones (:ajones, :kentuckyfriedtakahe, :k17e)]

P2 because unencrypted decoding is disabled for now.

Comment 2

[Chris Pearce [:cpearce (Not reading bugmail)]]

Adding Joe Steele, from Adobe.

Debugging the minidump for this crash report:
https://crash-stats.mozilla.com/report/index/654a3d58-633d-411a-b94f-1a5fa2160623#allthreads

If I follow the breadcrumbs through the minidump, I can see that the Adobe DLL has thrown a Microsoft C++, "bad_alloc" C++ exception.

Joe, do you get anything from the thread call stacks listed in the above link? There's probably a new[] that's getting an invalid or too large size passed to it that's failing.

Comment 3

[Joe Steele]

I will see if I can re-symbolize the call stack. Unless you have some reason to suspect low memory, I would guess this is a memory stomper of some kind leading to try and alloc an unrealistic memory size.

Comment 4

[Anthony Jones (:ajones, :kentuckyfriedtakahe, :k17e)]

Mass change P2 -> P3

Comment 5

[Chris Pearce [:cpearce (Not reading bugmail)]]

We're removing Adobe EME support, so closing this bug.