Bug 20873

Summary: [UXA tiling]X crash if set Virtual Desktop size >1024x1024 on 64bit system
Product: xorg Reporter: liuhaien <haien.liu>
Component: Driver/intelAssignee: Keith Packard <keithp>
Status: VERIFIED FIXED QA Contact: Xorg Project Team <xorg-team>
Severity: normal    
Priority: medium CC: jbarnes
Version: unspecified   
Hardware: Other   
OS: Linux (All)   
Whiteboard:
i915 platform: i915 features:
Attachments:
Description Flags
xorg.0.log
none
xorg conf file none

Description liuhaien 2009-03-26 01:22:32 UTC
Created attachment 24261 [details]
xorg.0.log

System Environment:
--------------------------
Arch:		x86_64
Platform:		G33
OSD:		Fedora release 8 (Werewolf)
Kernel_version:		2.6.29-rc8
Libdrm:		(master)c9cfeaa5542e08381a8144d0a5bd96036a19082f
Mesa:		(mesa_7_4_branch)14a794300e9d548e7426836b1dea1150e07199d0
Xserver:	(server-1.6-branch)60c161545af80eb78eb790a05bde79409dfdf16e
Xf86_video_intel:		(2.7)e2465249a90b9aefe6d7a96eb56a51fde54698a0
Kernel:       (for-airlied)a2e785c32b886dd7f0289d1cf15fc14e9c81bc01

Bug detailed description:
--------------------------
  startx with virtual desktop size >1024x1024(e.g 1025x1025)under UMS+UXA on  64bit system, X will crash with backtrace message.
  but it works on all 32bit system and below configurations on 64bit system:
   1. UMS+EXA 
   2. KMS+UXA
   3. UMS+UXA with tiling off 

following is the backtrace from gdb:
   #0  0x00007f6627a4a7d7 in i830_unbind_memory (pScrn=0x21da7f0, mem=0x222fe20)
    at i830_memory.c:1943
   #1  0x00007f6627a4a86d in i830_unbind_all_memory (pScrn=0x21da7f0)
     at i830_memory.c:2012
   #2  0x00007f6627a42d5b in I830LeaveVT (scrnIndex=<value optimized out>,
    flags=<value optimized out>) at i830_driver.c:3427
   #3  0x0000000000463e28 in AbortDDX () at xf86Init.c:1452
   #4  0x00000000004e936d in AbortServer () at log.c:407
   #5  0x00000000004e996f in FatalError (
    f=0x566978 "Caught signal %d.  Server aborting\n") at log.c:532
   #6  0x000000000046ee19 in xf86SigHandler (signo=11) at xf86Events.c:387
   #7  <signal handler called>
   #8  0x00007f6627a4a7d7 in i830_unbind_memory (pScrn=0x21da7f0, mem=0x2207900)
    at i830_memory.c:1943
   #9  0x00007f6627a4b00a in i830_free_memory (pScrn=0x21da7f0, mem=0xffffffff)
    at i830_memory.c:305
   #10 0x00007f6627a415fb in i830_xf86crtc_resize (scrn=0x21da7f0, width=1024,
    height=768) at i830_driver.c:1160
   #11 0x00000000004a72d5 in xf86RandR12ScreenSetSize (pScreen=0x2208a20,
    width=1024, height=768, mmWidth=270, mmHeight=203) at xf86RandR12.c:703
   #12 0x00000000004a79bd in xf86RandR12CreateScreenResources (pScreen=0x2208a20)
    at xf86RandR12.c:843
   #13 0x00000000004a052f in xf86CrtcCreateScreenResources (screen=0x2208a20)
    at xf86Crtc.c:701
   #14 0x000000000042dc5c in main (argc=2, argv=0x7fff31902fb8,
    envp=<value optimized out>) at main.c:325

Reproduce steps:
----------------
1. set virtual size >1024x1024
2. xinit&
Comment 1 liuhaien 2009-03-26 01:22:52 UTC
Created attachment 24262 [details]
xorg conf file
Comment 2 Keith Packard 2009-04-03 21:48:09 UTC
I cannot reproduce this.

chipset: GM965
kernel: drm-intel-2.6.29 0e56a4d653b66d4729f944b23935a00c4472f987
xserver: server-1.6-branch 667865b9eb98ed73461028b3d38be19619712533
driver: 2.7 10b5014c42dc055d9559ee112cc7a017e887d813
drm: master 51d6346f9f3c425f49e57d185530c6bcaeb94f5e

(fyi, with UXA and DRI2, you do not need a Virtual line any longer, but that doesn't excuse this bug)
Comment 3 Gordon Jin 2009-04-04 18:47:01 UTC
(In reply to comment #2)
> I cannot reproduce this.
> 
> chipset: GM965
> kernel: drm-intel-2.6.29 0e56a4d653b66d4729f944b23935a00c4472f987
> xserver: server-1.6-branch 667865b9eb98ed73461028b3d38be19619712533
> driver: 2.7 10b5014c42dc055d9559ee112cc7a017e887d813
> drm: master 51d6346f9f3c425f49e57d185530c6bcaeb94f5e

Are you using a 64-bit OS?

> (fyi, with UXA and DRI2, you do not need a Virtual line any longer, but that
> doesn't excuse this bug)

Thanks for this clarification. So I'm downgrading priority/severity.


Haien, please retest with the latest code. 

Comment 4 Keith Packard 2009-04-04 19:51:42 UTC
Yes, I installed a 64-bit user space to test against my 64-bit kernel.
Comment 5 liuhaien 2009-04-07 22:29:36 UTC
(In reply to comment #3)
> (In reply to comment #2)
> > I cannot reproduce this.
> > 
> > chipset: GM965
> > kernel: drm-intel-2.6.29 0e56a4d653b66d4729f944b23935a00c4472f987
> > xserver: server-1.6-branch 667865b9eb98ed73461028b3d38be19619712533
> > driver: 2.7 10b5014c42dc055d9559ee112cc7a017e887d813
> > drm: master 51d6346f9f3c425f49e57d185530c6bcaeb94f5e
> 
> Are you using a 64-bit OS?
> 
> > (fyi, with UXA and DRI2, you do not need a Virtual line any longer, but that
> > doesn't excuse this bug)
> 
> Thanks for this clarification. So I'm downgrading priority/severity.
> 
> 
> Haien, please retest with the latest code. 
> 

it works with the latest code:
Libdrm:		(master)1faab66cfd1a854925da6ff7109aa614292dea90
Mesa:		(mesa_7_4_branch)de197cf991416f0cd65ad2e2d2ca9aa599b52075
Xserver:	(server-1.6-branch)60c161545af80eb78eb790a05bde79409dfdf16e
Xf86_video_intel:		(2.7)3e5586cace98f73a9f8403a6446d380899ecbce9
Kernel:       (drm-intel-2.6.29)71d7aec4bbf923eaf43563c03459726a746cd332
Comment 6 Gordon Jin 2009-04-07 22:48:40 UTC
closing
Comment 7 liuhaien 2009-04-07 22:50:28 UTC
verified.

Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.