1186002 - Startup crash in @0x0 | AllocateCB(void*, _D3DDDICB_ALLOCATE*)

Status: RESOLVED FIXED

(Core :: Graphics, defect)

Description

[Tracy Walker [:tracy]]

[Tracking Requested - why for this release]:

This bug was filed from the Socorro interface and is 
report bp-c3d91c58-3d12-4d63-888c-274172150719.
=============================================================

@0x0 | AllocateCB(void*, _D3DDDICB_ALLOCATE*)and @0x0 | NDXGI::CDevice::AllocateCB(void*, _D3DDDICB_ALLOCATE*) are start up crashers that have made their way into #5 and #6 respectively.

Both of these crashes were fixed last month in Bug 1170211.  However, beginning with builds of 20150713, both signatures have returned.

Comment 1

[(Away)]

> Both of these crashes were fixed last month in Bug 1170211.  However,
> beginning with builds of 20150713, both signatures have returned.

1 	6.1.7600 	575 	60.65 %
3 	6.1.7601 	371 	39.14 %

Jeff, is this more WARP fallout?

Comment 2

[(Away)]

> Jeff, is this more WARP fallout?

Actually, no, the dates don't quite line up for that.

The CheckIfRenderTargetViewNeedsRecreating points more to bug 1109718.

Comment 3

[(Away)]

Sounds a lot like the comment in DoesD3D11TextureSharingWorkInternal:
  // CreateTexture2D is known to crash on lower feature levels, see bugs
  // 1170211 and 1089413.

Comment 4

[Milan Sreckovic [:milan] (needinfo for best results)]

We also get a message introduced in bug 1141783, which means that we would have blocklisted this system before, but not now.  Bug 1141783 landed in May though, so the timing doesn't quite work, but it would be related.

Comment 5

[Milan Sreckovic [:milan] (needinfo for best results)]

Or, at least, if we can get one of the local machines into the "mismatched driver" state, it may reproduce.

Comment 6

[Milan Sreckovic [:milan] (needinfo for best results)]

Right, the message I've been talking about - "Potential driver version mismatch ignored due to missing DLLs"

Comment 7

[David Anderson [:dvander] - inactive, e-mail if emergency]

Interestingly, I don't see a "GraphicsStartupTest" annotation here, which probably means we got a successful startup once and decided the drivers were okay. Then a second startup crashed. That's not very comforting.

Comment 8

[Jeff Muizelaar [:jrmuizel]]

Attachment: This should presumeably fix the issue

Comment 9

[Pulsebot]

https://hg.mozilla.org/integration/mozilla-inbound/rev/beed0226c597

Comment 10

[Carsten Book [:Tomcat]]

https://hg.mozilla.org/mozilla-central/rev/beed0226c597

Comment 11

[Kate Glazko]

Tracking to continue watching this fix/these crash signatures.