Comme=
nt # 58
on bug 11078=
1
from =
Richard Thier
The output from gdb glxgears on the segfault:
#0 0xb75a8892 in radeon_winsys_bo_create (rws=3D0x45e100, size=3D4096,
alignment=3D4096, domain=3DRADEON_DOMAIN_GTT, flags=3D(unknown: 0))
at radeon_drm_bo.c:993
#1 0xb75a972b in radeon_cs_create_fence (rcs=3Drcs@entry=3D0xb06a4010)=
at
radeon_drm_cs.c:753
#2 0xb75aa327 in radeon_drm_cs_flush (rcs=3D0xb06a4010, flags=3D2,
pfence=3D0xbffff7d8) at radeon_drm_cs.c:593
#3 0xb757845a in r300_flush_and_cleanup (r300=3Dr300@entry=3D0x446090,
flags=3Dflags@entry=3D2, fence=3Dfence@entry=3D0xbffff7d8) at r300_=
flush.c:56
#4 0xb757860f in r300_flush (pipe=3D0x446090, flags=3D2, fence=3D0xbffff7d=
8) at
r300_flush.c:82
#5 0xb731508f in st_context_flush (stctxi=3D0x55ed20, flags=3D3, fence=3D0=
xbffff7d8)
at state_tracker/st_manager.c:635
#6 0xb74511f8 in dri_flush (cPriv=3D0x445fd0, dPriv=3D0x5760d0, flags=3D3,
reason=3D__DRI2_THROTTLE_SWAPBUFFER) at dri_drawable.c:568
#7 0xb7f9dbe8 in dri2Flush (psc=3Dpsc@entry=3D0x4165d0, ctx=3D<opti=
mized out>,
draw=3Ddraw@entry=3D0x5590e0, flags=3D3,=20
throttle_reason=3D__DRI2_THROTTLE_SWAPBUFFER) at dri2_glx.c:553
#8 0xb7f9e09d in dri2SwapBuffers (pdraw=3D0x5590e0, target_msc=3D0, diviso=
r=3D0,
remainder=3D0, flush=3D1) at dri2_glx.c:845
#9 0xb7f73049 in glXSwapBuffers (dpy=3D0x407160, drawable=3D18874370) at
glxcmds.c:843
#10 0x00401742 in ?? ()
#11 0xb7b5c669 in __libc_start_main () from /usr/lib/libc.so.6
#12 0x00401cd5 in ?? ()
I have changed this relevant part with the added flag:
/* Create a fence, which is a dummy BO. */
fence =3D cs->ws->base.buffer_create(&cs->ws->base, 1, =
1,
RADEON_DOMAIN_GTT,
RADEON_FLAG_NO_SUBALLOC
| RADEON_FLAG_NO_INTERPROCESS_SHARIN=
G);
Currently I am testing this, in latest but the build is still going. Will
upload patch for it only if it works. I think it does not harm, but this is
happening when the bo is a "fence". I have no idea what a fence i=
s in this
terminology, but surely it sounds some synchronization primitive so I am un=
sure
that I can add the flag safely, but knowing earlier it was not in the code
likely I can just add it.
Will see how this shrinks the GEM_CREATE call numbers.