869011 - bmp-1bpp/bmp-size-1x1-1bpp.bmp reftest fails on b2g emulator after layers refactoring

Status: RESOLVED INVALID

(Core :: Graphics: Layers, defect)

Description

[Joe Drew (not getting mail)]

Attachment: use the correct program based on our format

+++ This bug was initially created as a clone of Bug #862787 +++

We were not using the correct program on RGBX gralloc textures. This patch fixes that.

Comment 1

[Joe Drew (not getting mail)]

FWIW, try results: https://tbpl.mozilla.org/?tree=Try&rev=5703575aa18d

Comment 2

[Benoit Jacob [:bjacob] (mostly away)]

Comment on attachment 745861 [details] [diff] [review]
use the correct program based on our format

Review of attachment 745861 [details] [diff] [review]:
-----------------------------------------------------------------

Sorry :) when I wrote this, I was still in "I'd be happy if this would even just boot" mode.

With the re-enabling of gralloc, a few reftests have been marked as failing in the reftest.list. That includes the one mentioned in the title of this bug, and also the WebGL one in bug 868892. I suppose that you will have to remove these fails-if annotations when landing this. Note that some of these reftests are hidden so you need to append &showall=1 to the TBPL URL.

Comment 3

[Joe Drew (not getting mail)]

Ah, OK. I've run it through Try again to see what's fixed by this patch: https://tbpl.mozilla.org/?tree=Try&rev=19821e7975a0

Comment 4

[Joe Drew (not getting mail)]

Sadly, this didn't fix bug 868892.

Comment 5

[Joe Drew (not getting mail)]

Ah, but Philor's changes hadn't made it to mozilla-central at that point. Retry: https://tbpl.mozilla.org/?tree=Try&rev=2b77742490b7

Comment 6

[Joe Drew (not getting mail)]

Aha! It did fix the alpha test.

Comment 7

[Joe Drew (not getting mail)]

https://hg.mozilla.org/integration/mozilla-inbound/rev/7e4d9be02b8f

Comment 8

[Ryan VanderMeulen [:RyanVM]]

https://hg.mozilla.org/mozilla-central/rev/7e4d9be02b8f

Comment 9

[[:fabrice] Fabrice Desré]

This patch makes the geeksphone keon crash early at startup. Since it's starting to be a quite popular device, this is pretty annoying.

Here's the gdb info:

Process /system/b2g/b2g created; pid = 29775
Listening on port 11000
prebuilt/linux-x86/toolchain/arm-linux-androideabi-4.4.x/bin/arm-linux-androideabi-gdb -x /tmp/b2g.gdbinit.fabrice /home/fabrice/dev/b2g/B2G/objdir-gecko/dist/bin/b2g
GNU gdb (GDB) 7.1-android-gg2
Copyright (C) 2010 Free Software Foundation, Inc.
License GPLv3+: GNU GPL version 3 or later <http://gnu.org/licenses/gpl.html>
This is free software: you are free to change and redistribute it.
There is NO WARRANTY, to the extent permitted by law.  Type "show copying"
and "show warranty" for details.
This GDB was configured as "--host=i686-linux-gnu --target=arm-elf-linux".
For bug reporting instructions, please see:
<http://www.gnu.org/software/gdb/bugs/>...
Reading symbols from /home/fabrice/dev/b2g/B2G/objdir-gecko/dist/bin/b2g...done.
Remote debugging from host 127.0.0.1
__dl__start () at bionic/linker/arch/arm/begin.S:35
35		mov	r0, sp
(gdb) c
Continuing.
[New Thread 29775.29804]
###################################### forms.js loaded
############################### browserElementPanning.js loaded
######################## BrowserElementChildPreload.js loaded
Crash reporter : Can't fetch app.reportCrashes. Exception: [Exception... "Component returned failure code: 0x8000ffff (NS_ERROR_UNEXPECTED) [nsIPrefBranch.getBoolPref]"  nsresult: "0x8000ffff (NS_ERROR_UNEXPECTED)"  location: "JS frame :: chrome://browser/content/shell.js :: shell_reportCrash :: line 130"  data: no]###################################### forms.js loaded
############################### browserElementPanning.js loaded
######################## BrowserElementChildPreload.js loaded
XXX FIXME : Got a mozContentEvent: system-message-listener-ready
[New Thread 29775.29796]

Program received signal SIGSEGV, Segmentation fault.
[Switching to Thread 29775.29796]
0x41dd0dfa in mozalloc_abort (msg=<value optimized out>) at /home/fabrice/dev/inbound/memory/mozalloc/mozalloc_abort.cpp:30
30	    MOZ_CRASH();
(gdb) bt
#0  0x41dd0dfa in mozalloc_abort (msg=<value optimized out>) at /home/fabrice/dev/inbound/memory/mozalloc/mozalloc_abort.cpp:30
#1  0x41dd0e10 in abort () at /home/fabrice/dev/inbound/memory/mozalloc/mozalloc_abort.cpp:39
#2  0x41dd0e10 in abort () at /home/fabrice/dev/inbound/memory/mozalloc/mozalloc_abort.cpp:39
Backtrace stopped: previous frame identical to this frame (corrupt stack?)

Comment 10

[Joe Drew (not getting mail)]

Fabrice, can you file a new bug for that? CC me and Benoit Jacob; I suspect it's related to gralloc.

(Did you bisect all the way down to this patch?)

Comment 11

[[:fabrice] Fabrice Desré]

(In reply to Joe Drew (:JOEDREW! \o/) from comment #10)
> Fabrice, can you file a new bug for that? CC me and Benoit Jacob; I suspect
> it's related to gralloc.

I filed bug 870375

> (Did you bisect all the way down to this patch?)

Yep. I'm running locally with just this patch backed out and that fixes the problem on the keon.

Comment 12

[Michael Wu [:mwu]]

Backed out to fix bug 870375

https://hg.mozilla.org/projects/birch/rev/7ef0ed6ea2db
https://hg.mozilla.org/projects/birch/rev/2795d46a9d46

Comment 13

[Benoit Jacob [:bjacob] (mostly away)]

Sotaro just ran into this with WebGL, we really need to land this!

Comment 14

[Sotaro Ikeda [:sotaro]]

(In reply to Benoit Jacob [:bjacob] from comment #13)
> Sotaro just ran into this with WebGL, we really need to land this!

The patch does not fix the problem. I am going to create a bug for it.