882342 - Crash in libstagefright_omx on Motorola XT890 running Android 4.1

Status: RESOLVED WONTFIX

(Core :: Audio/Video, defect)

Description

[Aaron Train [:aaronmt]]

This bug was filed from the Socorro interface and is 
report bp-af9300f7-2717-4c98-b0c5-692492130612 .
============================================================= 

Device: Motorola XT890 (Razr I)
OS: Android 4.1
Build: Firefox Beta 22.0

I this this crash hitting play on http://vimeo.com/67913037

Comment 1

[Scoobidiver (away)]

Adding a close crash signature.

Comment 2

[Robert Kaiser]

libstagefright_omx.so@0x1303b has been the topcrash for x86 on 22 Beta all the time. I think we might need to blocklist something there.

Comment 3

[Robert Kaiser]

(In reply to Robert Kaiser (:[e-mail]) from comment #2)
> libstagefright_omx.so@0x1303b has been the topcrash for x86 on 22 Beta all
> the time.

This continues consistently with 23 Beta.

Comment 4

[cajbir (:cajbir)]

I'm thinking this should be blocklisted until we get a device we can test with.

Comment 5

[Edwin Flores [inactive from 2016-12-01] [:eflores] [:edwin]]

Is this the only device that needs to be blacklisted?

Comment 6

[Edwin Flores [inactive from 2016-12-01] [:eflores] [:edwin]]

Er, where "this" = Motorola XT890.

Comment 7

[Scoobidiver (away)]

Will bug 860599's patch fix it?

Comment 8

[cajbir (:cajbir)]

(In reply to Scoobidiver from comment #7)
> Will bug 860599's patch fix it?

It doesn't look likely based on the stack frame but it would be great if someone could try it on the device to confirm.

Comment 9

[Robert Kaiser]

From all I see, both the signatures in this bug are only hit with the Motorola XT890:

https://crash-stats.mozilla.com/report/list?product=FennecAndroid&signature=libstagefright_omx.so%400x1303b
https://crash-stats.mozilla.com/report/list?product=FennecAndroid&signature=libstagefright_omx.so%400x13055

Comment 10

[Edwin Flores [inactive from 2016-12-01] [:eflores] [:edwin]]

Attachment: 882342-blacklist-motorola-xt890-4.1.patch

Comment 11

[Edwin Flores [inactive from 2016-12-01] [:eflores] [:edwin]]

Kevin, you have an XT890 according to the wiki. Could you test that this blocklist patch works and if so, nominate for beta/aurora approval?

Comment 12

[Aaron Train [:aaronmt]]

Kevin? Ping.

Comment 13

[Kevin Brosnan [Ex-Mozilla]]

I am traveling and won't have access to the Razr I till the 13th of August. Aaron, Toronto has one, can you look at this?

Comment 14

[Aaron Train [:aaronmt]]

Confirming patch works for me on my XT890; used http://cd.pn/b

ro.product.model=XT890
ro.build.version.release=4.1.2
ro.build.version.sdk=16

Comment 15

[Aaron Train [:aaronmt]]

Comment on attachment 778741 [details] [diff] [review]
882342-blacklist-motorola-xt890-4.1.patch

[Approval Request Comment]
Bug caused by (feature/regressing bug #): 
User impact if declined: Crash on video playback
Testing completed (on m-c, etc.): Tested blocklist patch on an XT890
Risk to taking this patch (and alternatives if risky): No risk, patch adds one other device to check for. Ask Edwin.
String or IDL/UUID changes made by this patch:

Comment 16

[Aaron Train [:aaronmt]]

Is this needed on 25?

Comment 17

[Lukas Blakk [:lsblakk] use ?needinfo]

Comment on attachment 778741 [details] [diff] [review]
882342-blacklist-motorola-xt890-4.1.patch

This should uplift for FF24 on Aurora in order to be on Beta when it merges on Monday but since we're not shipping x86 builds for FF23, we don't need to respin our final beta/RC of FF23 for this.

Comment 18

[Robert Kaiser]

(In reply to Aaron Train [:aaronmt] from comment #16)
> Is this needed on 25?

I'm pretty sure 25 has the same crashes, but I would guess if we need it there depends on bug 860599 landing in this train or not. If it does, we probably should not blocklist there and see if this one is fixed as well by that mechanism. If that doesn't land, we'll need to block on the 25 train as well.

Comment 19

[Edwin Flores [inactive from 2016-12-01] [:eflores] [:edwin]]

(In reply to Robert Kaiser (:[e-mail]) from comment #18)
> (In reply to Aaron Train [:aaronmt] from comment #16)
> > Is this needed on 25?
> 
> I'm pretty sure 25 has the same crashes, but I would guess if we need it
> there depends on bug 860599 landing in this train or not. If it does, we
> probably should not blocklist there and see if this one is fixed as well by
> that mechanism. If that doesn't land, we'll need to block on the 25 train as
> well.

Just spun up an x86 build with the patch from bug 860599, and it still crashes with that; so will have to land this on 25.

Comment 20

[Edwin Flores [inactive from 2016-12-01] [:eflores] [:edwin]]

https://hg.mozilla.org/integration/mozilla-inbound/rev/6e7d9cb27760
http://hg.mozilla.org/releases/mozilla-aurora/rev/afefa40aebb4

Comment 21

[Ryan VanderMeulen [:RyanVM]]

https://hg.mozilla.org/mozilla-central/rev/6e7d9cb27760

Comment 22

[Scoobidiver (away)]

I've updated https://wiki.mozilla.org/Blocklisting/Blocked_Graphics_Drivers#On_Android_2 accordingly. It should have been done by the assignee.

Comment 23

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

This has been blacklisted. Leave it open for an actual fix.

Comment 24

[Tracy Walker [:tracy]]

no longer a topcrash

Comment 25

[Edwin Flores [inactive from 2016-12-01] [:eflores] [:edwin]]

Stagefright no longer used on 4.1.x after bug 1014614.