Bug 483170

Summary: Radeon XPRESS 200M (RC410) - enabling and disabling compiz repeatedly eventually crashes xserver
Product: [Fedora] Fedora Reporter: Alex Villacís Lasso <alexvillacislasso>
Component: xorg-x11-drv-atiAssignee: Dave Airlie <airlied>
Status: CLOSED RAWHIDE QA Contact: Fedora Extras Quality Assurance <extras-qa>
Severity: medium Docs Contact:
Priority: low    
Version: 10CC: jglisse, samuel-rhbugs, xgl-maint, yulrottmann
Target Milestone: ---   
Target Release: ---   
Hardware: i386   
OS: Linux   
Whiteboard: card_RC410
Fixed In Version: Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2009-10-23 07:38:05 UTC Type: ---
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Embargoed:
Attachments:
Description Flags
dmesg output of session where xorg crashed
none
xserver log of crashed session
none
Output of lspci -v
none
xorg.conf used with xserver
none
My Xorg.0.log from the crash. none

Description Alex Villacís Lasso 2009-01-29 23:00:56 UTC
Created attachment 330414 [details]
dmesg output of session where xorg crashed

Description of problem:

On this machine, I installed Fedora 10 and updated it to the latest packages. Then I installed compiz-fusion in order to enable desktop effects. I kept switching between compiz and metacity in GNOME. Eventually, after two or three switches back and forth, the X server crashes on me and I am returned to the graphical login. Using ati driver from updates-testing does not help.

Version-Release number of selected component (if applicable):

xorg-x11-server-Xorg-1.5.3-6.fc10.i386
xorg-x11-drv-ati-6.10.0-1.fc10.i386
mesa-libGL-7.2-0.15.fc10.i386
mesa-dri-drivers-7.2-0.15.fc10.i386
kernel-2.6.27.12-170.2.5.fc10.i686

How reproducible:
Always.

Steps to Reproduce:
1. Get a Radeon RC410 chipset to test
2. In GNOME, install compiz-fusion and switch back and forth between metacity and compiz.
3. Repeat until crash
  
Actual results:
xserver crashes

Expected results:
no xserver crash

Additional info:

Comment 1 Alex Villacís Lasso 2009-01-29 23:02:18 UTC
Created attachment 330415 [details]
xserver log of crashed session

Comment 2 Alex Villacís Lasso 2009-01-29 23:03:15 UTC
Created attachment 330416 [details]
Output of lspci -v

Comment 3 Alex Villacís Lasso 2009-01-29 23:05:06 UTC
Created attachment 330417 [details]
xorg.conf used with xserver

Comment 4 Samuel Sieb 2009-02-24 21:02:45 UTC
I get this on a laptop with an X300:
kernel-2.6.27.15-170.2.24.fc10.i686
xorg-x11-drv-ati-6.10.0-2.fc10.i386

No xorg.conf.

Enable desktop effects, disable desktop effects, enable desktop effect, crash.

Program received signal SIGSEGV, Segmentation fault.
0x0055dd7f in dri_bo_reference (bo=0x9abeef8) at radeon_dri_bufmgr.c:58
58	   bo->bufmgr->bo_reference(bo);
(gdb) where
#0  0x0055dd7f in dri_bo_reference (bo=0x9abeef8) at radeon_dri_bufmgr.c:58
#1  0x0053b59b in bufmgr_classic_emit_reloc (batch_buf=0x99d3808, 
    flags=16777216, delta=0, offset=732, target=0x9abeef8) at r300_mem.c:792
#2  0x0055de43 in dri_emit_reloc (reloc_buf=0x99d3808, flags=16777216, 
    delta=0, offset=732, target_buf=0x9abeef8) at radeon_dri_bufmgr.c:137
#3  0x005425e1 in emit_cb_offset (r300=0x9eb1410) at r300_cmdbuf.c:356
#4  0x0054291c in r300EmitAtoms (r300=0x9eb1410, dirty=1 '\001')
    at r300_cmdbuf.c:247
#5  0x00542078 in r300EmitState (r300=0x9eb1410) at r300_cmdbuf.c:297
#6  0x0054214e in r300BeginBatch (r300=0x9eb1410, n=4, autostate=1 '\001', 
    function=0x56a774 "r300EmitCacheFlush", line=487) at r300_cmdbuf.c:192
#7  0x00556de4 in r300EmitCacheFlush (rmesa=0x9eb1410) at r300_emit.c:487
#8  0x00548286 in r300RunRender (ctx=0x9eda0e8, stage=<value optimized out>)
    at r300_render.c:316
#9  0x010bd059 in _tnl_run_pipeline (ctx=0x9eda0e8) at tnl/t_pipeline.c:158
#10 0x010bdaa4 in _tnl_draw_prims (ctx=0x9eda0e8, arrays=0x9ca2b20, 
    prim=0xbfe587b0, nr_prims=1, ib=0x0, min_index=0, max_index=15)
    at tnl/t_draw.c:402
#11 0x010b57ce in vbo_exec_DrawArrays (mode=7, start=0, count=16)
    at vbo/vbo_exec_array.c:267
#12 0x010a678b in neutral_DrawArrays (mode=7, start=0, count=16)
    at main/vtxfmt_tmp.h:328
#13 0x0017434d in __glXDisp_DrawArrays (pc=0x9aa85c4 "") at render2.c:248
#14 0x0016ddaf in __glXDisp_Render (cl=0x9a6e934, pc=0x9aa8594 "h\001�")
    at glxcmds.c:1789
#15 0x0017263a in __glXDispatch (client=0x9a6e748) at glxext.c:512
#16 0x08085e9f in Dispatch () at dispatch.c:454
#17 0x0806b71d in main (argc=8, argv=0xbfe58a14, envp=Cannot access memory at address 0x291
) at main.c:441

Comment 5 Samuel Sieb 2009-02-24 21:03:34 UTC
Created attachment 333088 [details]
My Xorg.0.log from the crash.

Comment 6 Yul Rottmann 2009-05-02 12:49:18 UTC
I get this on my X700Pro PCIe. Even if I log in and then the desktop effects are enabled it sometimes crashes and the screen goes black and then again to the gdm log in. After a few tries it works sometimes, but sometimes it freezes completely. Glxgears work fine when having desktop effects, but other 3d application cause the crash. :(

Comment 7 Jérôme Glisse 2009-10-14 11:26:53 UTC
Can you test with fedora 12 livecd and report if it works with it.

Comment 8 Yul Rottmann 2009-10-22 17:58:03 UTC
At least I cannot reproduce this bug any more. Thanks.

Comment 9 Jérôme Glisse 2009-10-23 07:38:05 UTC
So i am closing it, reopen if it happens again with newer version of fedora.