Same problem (hang+kernel oops) with libdrm2 downgraded to 2.4.11-1
Previous report had libdrm2 2.4.11+git+20090519+f355ad8-1
Changing AccelMethod to exa produces following gdb backtrace (now using
downgraded libdrm2)
Program received signal SIGINT, Interrupt.
0x00007fbfa7d1e087 in ioctl () from /lib/libc.so.6
#0 0x00007fbfa7d1e087 in ioctl () from /lib/libc.so.6
No symbol table info available.
#1 0x00007fbfa67d7623 in drmIoctl (fd=10, request=25668, arg=0x0) at
../../libdrm/xf86drm.c:187
ret = 15
#2 0x00007fbfa67d7926 in drmCommandNone (fd=10, drmCommandIndex=<value
optimized out>) at ../../libdrm/xf86drm.c:2313
No locals.
#3 0x00007fbfa5d09cf4 in RADEONDownloadFromScreenCP (pSrc=<value optimized
out>, x=0, y=15, w=21, h=0, dst=0x1da44c0 "P", dst_pitch=84) at
../../src/radeon_exa_funcs.c:411
oldhpass = <value optimized out>
i = 117
hpass = 0
scratch_pitch_offset = 12061448
indirect = {idx = 0, start = 0, end = 0, discard = 0}
__head = <value optimized out>
pScrn = 0x12a0820
src =
0x7fbf93580000
"\b\b\b<FF>\b\b\b<FF>\b\b\b<FF>\b\b\b<FF>hhh<FF>\215\220\220<FF>\215\220\220<FF>\215\220\220<FF><A0><A0><A0><FF><B2><B5><B5><FF><C5><CA><CA><FF><C5><CA><CA><FF><C8><D4><D8><FF><C8><D4><D8><FF><D0><D9><DD><FF><C8><D4><D8><FF><D0><D4><D8><FF><D0><D4><D8><FF><D0><D4><D8><FF><D0><D4><D8><FF><C5><CC><CD><FF><C5><CC><CD><FF><B2><B8><BA><FF><A0><A4><A8><FF>\210\216\215<FF>\210\216\215<FF>`d`<FF>`d`<FF>HIJ<FF>HIJ<FF>"
bpp = <value optimized out>
datatype = 6
src_pitch_offset = 11940080
scratch_pitch = 128
scratch_off = 0
scratch = 0x12e7a70
#4 0x00007fbfa544a2e9 in exaCopyDirty (migrate=0x7fff101aab60,
pValidDst=0x1da3f40, pValidSrc=<value optimized out>, transfer=0x7fbfa5d09750
<RADEONDownloadFromScreenCP>, fallback_src=
0x7fbf9e07c000
"\002\002\002\002\006\006\006\027\a\a\a\031\a\a\a\031\a\a\a\031\a\a\a\031\a\a\a\031\006\006\006\027\002\002\002\b\b\b\bl",
fallback_dst=0x1da44c0 "P", fallback_srcpitch=128,
fallback_dstpitch=84, fallback_index=1, sync=0x7fbfa5444490 <exaWaitSync>)
at ../../exa/exa_migration.c:210
pPixmap = 0x1da4480
pExaPixmap = 0x1da3f00
damage = <value optimized out>
CopyReg = {extents = {x1 = 0, y1 = 0, x2 = 21, y2 = 15}, data = 0x0}
save_offscreen = 1
save_pitch = 128
pBox = 0x7fff101aa9a0
nbox = 0
access_prepared = 0
need_sync = 0
#5 0x00007fbfa544a620 in exaDoMoveOutPixmap (migrate=0x7fff101aab60) at
../../exa/exa_migration.c:258
pPixmap = 0x1da4480
#6 0x00007fbfa544ad47 in exaDoMigration (pixmaps=0x7fff101aab60, npixmaps=1,
can_accel=0) at ../../exa/exa_migration.c:679
pExaScr = 0x12e74c0
i = 1
j = <value optimized out>
__func__ = "exaDoMigration"
#7 0x00007fbfa5446c8a in exaGetImage (pDrawable=0x1da4480, x=0, y=0, w=21,
h=15, format=2, planeMask=4294967295, d=0x7fbf806a7000 "") at
../../exa/exa_accel.c:1198
pixmaps = {{as_dst = 0, as_src = 1, pPix = 0x1da4480, pReg =
0x7fff101aab80}}
Reg = {extents = {x1 = 0, y1 = 0, x2 = 21, y2 = 15}, data = 0x0}
pPix = <value optimized out>
xoff = <value optimized out>
yoff = <value optimized out>
ok = <value optimized out>
#8 0x00000000004def9d in miSpriteGetImage (pDrawable=0x1da4480, sx=0, sy=0,
w=21, h=15, format=2, planemask=4294967295, pdstLine=0x7fbf806a7000 "") at
../../mi/misprite.c:354
pScreen = 0x12b3eb0
pDev = 0x0
pCursorInfo = <value optimized out>
#9 0x000000000050d844 in ProcShmGetImage (client=0x4b5e080) at
../../Xext/shm.c:969
pDraw = 0x1da4480
lenPer = 0
length = 140460469940224
plane = 0
xgi = {type = 1 '\001', depth = 24 '\030', sequenceNumber = 14619,
length = 0, visual = 0, size = 1260, pad0 = 91857776, pad1 = 0, pad2 =
19974240, pad3 = 0}
shmdesc = 0x1da3f70
rc = <value optimized out>
#10 0x000000000050e288 in ProcShmDispatch (client=0x4b5e080) at
../../Xext/shm.c:1125
No locals.
#11 0x000000000044d374 in Dispatch () at ../../dix/dispatch.c:437
result = <value optimized out>
client = 0x4b5e080
nready = 0
start_tick = 10660
#12 0x000000000043321d in main (argc=8, argv=0x7fff101aafb8, envp=<value
optimized out>) at ../../dix/main.c:397
i = 1
alwaysCheckForInput = {0, 1}
--
To UNSUBSCRIBE, email to debian-x-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
