1273223 - Stack-overflow in NetEQ -- webrtc::Merge::Process (Windows-only)

Status: RESOLVED WONTFIX

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

Description

[Randell Jesup [:jesup] (needinfo me)]

This bug was filed from the Socorro interface and is 
report bp-0dd37617-53a3-4b72-8ab2-e2caf2160509.
=============================================================

Goes back to at least 41 (so not full-duplex).  Low frequency (~50 in the last month).

Most have no useful backtrace; this has more detail it appears:
https://crash-stats.mozilla.com/report/index/f19184b8-371d-4aa0-8b14-f94ba2160516

But, that stack looks kinda odd, so it may be partly bogus -- it appears it should be called from NetEqImpl::DoMerge(), and that from NetEqImpl::GetAudioInternal() - I do see the GetAudioInternal() sitting on DoMerge (likely inlined), but higher frames look odd (we're not getting called from CONVOLVE_FUNC in the sinc_resampler)

Comment 1

[Paul Adenot (:padenot)]

We use small stacks for audio thread, we can probably just make them bigger. This code in particular uses large stack buffer so it's not that surprising.

Comment 2

[Bulk Bug Changes for mreavy's org]

Mass change P2->P3 to align with new Mozilla triage process.

Comment 3

[Sylvestre Ledru [:Sylvestre]]

Closing because no crash reported since 12 weeks.