[Bug 62244] New: SIGFPE with cogl GL client in wayland

bugzilla-daemon at freedesktop.org bugzilla-daemon at freedesktop.org
Tue Mar 12 09:13:42 PDT 2013


https://bugs.freedesktop.org/show_bug.cgi?id=62244

          Priority: medium
            Bug ID: 62244
          Assignee: dri-devel at lists.freedesktop.org
           Summary: SIGFPE with cogl GL client in wayland
          Severity: normal
    Classification: Unclassified
                OS: All
          Reporter: bugzilla at hadess.net
          Hardware: Other
            Status: NEW
           Version: unspecified
         Component: Drivers/DRI/R600
           Product: Mesa

I'm working with (mostly) the standard Fedora 18 packages for wayland/weston.

mesa-dri-drivers-9.0.3-1.fc18.x86_64
mesa-libEGL-9.0.3-1.fc18.x86_64
mesa-libwayland-egl-9.0.3-1.fc18.x86_64

1. Rebuild wayland (to get an updated libwayland-client, otherwise it crashes
when connecting to the display, not necessary on rawhide apparently)
2. Rebuild cogl with wayland support (add --enable-wayland-egl-platform to the
configure args)
3. Run weston &
4. Run this simple cogl example:
COGL_RENDERER=egl_wayland libtool --mode=execute gdb --args
./examples/cogl-crate

Program received signal SIGFPE, Arithmetic exception.
0x00007ffff4952a7c in eg_surface_init_2d (surf_man=surf_man at entry=0x6211f0,
surf=surf at entry=0x7d00c0, level=level at entry=0x7fffffffb870, bpe=bpe at entry=1,
tile_split=0, offset=1310720, start_level=0) at radeon_surface.c:646
646            slice_pt = tileb / tile_split;
(gdb) bt
#0  0x00007ffff4952a7c in eg_surface_init_2d (surf_man=surf_man at entry=0x6211f0,
surf=surf at entry=0x7d00c0, level=level at entry=0x7fffffffb870, bpe=bpe at entry=1,
tile_split=0, offset=1310720, start_level=0) at radeon_surface.c:646
#1  0x00007ffff4952dfb in eg_surface_init_2d_miptrees (surf_man=0x6211f0,
surf=0x7d00c0) at radeon_surface.c:802
#2  0x00007ffff4bc1cd8 in r600_setup_surface (pitch_in_bytes_override=2560,
rtex=0x7d0000, screen=0x621250) at r600_texture.c:169
#3  r600_texture_create_object (screen=screen at entry=0x621250,
base=base at entry=0x7fffffffcc20, pitch_in_bytes_override=2560,
buf=buf at entry=0x7cff70, alloc_bo=alloc_bo at entry=0 '\000',
surface=surface at entry=0x7fffffffbf20) at r600_texture.c:406
#4  0x00007ffff4bc21ae in r600_texture_from_handle (screen=0x621250,
templ=0x7fffffffcc20, whandle=<optimized out>) at r600_texture.c:556
#5  0x00007ffff4be6ca3 in dri2_drawable_process_buffers (count=2,
buffers=0x74f140, drawable=0x757980) at dri2.c:280
#6  dri2_allocate_textures (drawable=0x757980, statts=<optimized out>,
count=<optimized out>) at dri2.c:391
#7  0x00007ffff4be7f80 in dri_st_framebuffer_validate (stfbi=<optimized out>,
statts=0x7cc930, count=2, out=0x7fffffffcd20) at dri_drawable.c:78
#8  0x00007ffff4bec776 in st_framebuffer_validate (stfb=0x7cc4f0, st=0x6dfb10,
st=0x6dfb10) at ../../src/mesa/state_tracker/st_manager.c:195
#9  0x00007ffff4bed9d3 in st_api_make_current (stapi=<optimized out>,
stctxi=0x6dfb10, stdrawi=<optimized out>, streadi=<optimized out>) at
../../src/mesa/state_tracker/st_manager.c:734
#10 0x00007ffff4be775a in dri_make_current (cPriv=<optimized out>,
driDrawPriv=0x74f2e0, driReadPriv=0x74f2e0) at dri_context.c:243
#11 0x00007ffff4bb4ec6 in driBindContext (pcp=<optimized out>, pdp=<optimized
out>, prp=<optimized out>) at
../../../../src/mesa/drivers/dri/common/dri_util.c:381
#12 0x00007ffff6d0e29c in dri2_make_current (drv=0x619750, disp=0x618a10,
dsurf=0x74f0c0, rsurf=0x74f0c0, ctx=0x624f20) at egl_dri2.c:916
#13 0x00007ffff6d04bf9 in eglMakeCurrent (dpy=0x618a10, draw=0x74f0c0,
read=0x74f0c0, ctx=0x624f20) at eglapi.c:498
#14 0x00007ffff7dc27ec in _cogl_winsys_egl_make_current (display=0x624e30,
draw=0x74f0c0, read=0x74f0c0, context=0x624f20) at
./winsys/cogl-winsys-egl.c:279
#15 0x00007ffff7dc340a in bind_onscreen_with_context (onscreen=0x75a030,
egl_context=0x624f20) at ./winsys/cogl-winsys-egl.c:673
#16 0x00007ffff7dc34b4 in bind_onscreen (onscreen=0x75a030) at
./winsys/cogl-winsys-egl.c:698
#17 0x00007ffff7dc34ce in _cogl_winsys_onscreen_bind (onscreen=0x75a030) at
./winsys/cogl-winsys-egl.c:704
#18 0x00007ffff7d3e90e in _cogl_framebuffer_gl_bind (framebuffer=0x75a030,
target=36160) at ./driver/gl/cogl-framebuffer-gl.c:246
#19 0x00007ffff7d3eb0b in _cogl_framebuffer_gl_flush_state
(draw_buffer=0x75a030, read_buffer=0x75a030, 
    state=(COGL_FRAMEBUFFER_STATE_BIND | COGL_FRAMEBUFFER_STATE_VIEWPORT |
COGL_FRAMEBUFFER_STATE_CLIP | COGL_FRAMEBUFFER_STATE_DITHER |
COGL_FRAMEBUFFER_STATE_MODELVIEW | COGL_FRAMEBUFFER_STATE_PROJECTION |
COGL_FRAMEBUFFER_STATE_COLOR_MASK | COGL_FRAMEBUFFER_STATE_FRONT_FACE_WINDING))
at ./driver/gl/cogl-framebuffer-gl.c:321
#20 0x00007ffff7dacb04 in _cogl_framebuffer_flush_state (draw_buffer=0x75a030,
read_buffer=0x75a030, 
    state=(COGL_FRAMEBUFFER_STATE_BIND | COGL_FRAMEBUFFER_STATE_VIEWPORT |
COGL_FRAMEBUFFER_STATE_CLIP | COGL_FRAMEBUFFER_STATE_DITHER |
COGL_FRAMEBUFFER_STATE_MODELVIEW | COGL_FRAMEBUFFER_STATE_PROJECTION |
COGL_FRAMEBUFFER_STATE_COLOR_MASK | COGL_FRAMEBUFFER_STATE_FRONT_FACE_WINDING))
at ./cogl-framebuffer.c:1144
#21 0x00007ffff7dab4eb in cogl_framebuffer_clear4f (framebuffer=0x75a030,
buffers=3, red=0, green=0, blue=0, alpha=1) at ./cogl-framebuffer.c:369
#22 0x0000000000401b05 in paint (data=0x7fffffffd260) at cogl-crate.c:87
#23 0x0000000000402172 in main (argc=1, argv=0x7fffffffd498) at
cogl-crate.c:285
(gdb) list
641        tileh = 8;
642        tileb = tilew * tileh * bpe * surf->nsamples;
643        /* slices per tile */
644        slice_pt = 1;
645        if (tileb > tile_split) {
646            slice_pt = tileb / tile_split;
647        }
648        tileb = tileb / slice_pt;
649    
650        /* macro tile width & height */

Device is:
01:00.0 VGA compatible controller [0300]: Advanced Micro Devices [AMD] nee ATI
Madison [Radeon HD 5000M Series] [1002:68c1]

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.freedesktop.org/archives/dri-devel/attachments/20130312/3df94c27/attachment.html>


More information about the dri-devel mailing list