[Mesa-dev] questions about assertion hit in vmwgfx/svga driver
Brian Paul
brianp at vmware.com
Wed Oct 10 15:14:53 PDT 2012
On 10/10/2012 04:03 PM, Chris Fester wrote:
> Hi all,
>
> I'm working on a project involving virtualizing a rather old Linux
> distribution with 3D support. Some of the details of the setup:
>
> Host: Ubuntu 12.04 AMD64, fglrx-updates package installed
> -- Vmware guest: Ubuntu 12.04 32 bit, stock packages
> ---- chrooted sandbox environment: Fedora Core 2 with gcc 3.3.3 (many
> modifications)
>
> The chrooted environment properly mounts up /proc, /dev/shm, /dev/pts,
> /sys, etc. It also bind mounts /tmp so we get the socket for X11
> communication, as well as /dev/dri.
>
> It has been a long process getting all the dependencies to compile,
> writing a little assembly to mimic compiler built-ins, using
> libatomicops instead of compiler built-ins, and finally getting mesa
> to build. But I am now able to run glxgears from within the chrooted
> sandbox at about 100fps. Also glxinfo reports that I do have direct
> rendering. Woohoo!
>
> All of the source that I have built has been obtained with apt-get
> source<ubuntu package name>. I wanted to be sure that everything
> being compiled in the chroot was exactly the same version as what I
> was running in the guest OS. Specifically, the Ubuntu version of the
> mesa source package: mesa_8.0.3+8.0.2-0ubuntu3.2.
>
> So the down side... I have hit an assertion in
> src/gallium/drivers/svga/svga_resource_buffer_upload.c while running
> my company's GL application. The assertion details:
>
> src/gallium/drivers/svga/svga_resource_buffer_upload.c:591
> assert(!sbuf->head.prev&& !sbuf->head.next);
>
> The prev and next pointers happen to be the same value. They're also
> identical to the prev and next pointers in svga->dirty_buffers. And
> topping it all off, the pointers all point to the location of
> &svga->dirty_buffers.
>
> Below I've pasted more details from GDB. My question is regarding the
> linked list handling. I've noticed in the Linux doubly-linked-list
> implementation, the following happens on a deletion (from
> /usr/include/linux/list.h):
>
> static __inline__ void list_del(struct list_head *entry)
> {
> __list_del(entry->prev, entry->next);
> entry->next = entry->prev = 0; //<------------------- set to 0!
> }
>
> And the code from src/gallium/auxiliary/util/u_double_list.h:
>
> static INLINE void list_del(struct list_head *item)
> {
> item->prev->next = item->next;
> item->next->prev = item->prev;
> }
>
> Note that next and prev are not zero-ed out here.
>
> I also noticed this bit of code:
> svga_resource_buffer_upload.c:290 (in svga_buffer_upload_flush()):
> LIST_DEL(&sbuf->head);
> #ifdef DEBUG
> sbuf->head.next = sbuf->head.prev = NULL;
> #endif
>
> Why the #ifdef DEBUG here? Should the setting of next and prev to
> NULL take place all the time? Or should they be set to NULL in the
> u_double_list.h header's list_del()?
>
> I realize that my setup is pretty unique and this assertion may be my
> responsibility. If the svga/list code as it stands is correct, then
> I'd appreciate any advice on what may be incorrect in my setup that
> could cause the error. If I remove the assert, the application
> continues on happily, although I need to give it more soak time.
>
> Thanks for your efforts and your time!
> Chris Fester
>
> GDB information (email me directly if you require more context)
> follows. Note that I had to replace the assert with code that would
> generate a segfault, due to the application I am running replaces the
> default assert mechanism and stomps on the stack :(.
>
> (gdb) p svga->dirty_buffers
> $19 = {prev = 0x9563728, next = 0x9563728}
> (gdb) p sbuf->head
> $20 = {prev = 0x9563728, next = 0x9563728}
> (gdb) p&svga->dirty_buffers
> $21 = (struct list_head *) 0x9563728
>
> (gdb) bt
> #0 0x413e56b8 in svga_buffer_handle (svga=0x955f490, buf=0x1)
> at svga_resource_buffer_upload.c:597
> #1 0x413e78d8 in svga_hwtnl_flush (hwtnl=0x9538b18) at svga_draw.c:184
> #2 0x413eb6ed in update_state (svga=0x955f490, atoms=0x416f49e0,
> state=0x9563710) at svga_state.c:131
> #3 0x413eb85f in svga_update_state (svga=0x955f490, max_level=0)
> at svga_state.c:209
> #4 0x413eb8cc in svga_update_state_retry (svga=0x955f490, max_level=0)
> at svga_state.c:234
> #5 0x413e96c3 in svga_draw_vbo (pipe=0x955f490, info=0xbfcc4050)
> at svga_pipe_draw.c:348
> #6 0x41481521 in st_draw_vbo (ctx=0x95af2b8, arrays=0x960cc38,
> prims=0x960b520, nr_prims=1, ib=0x0, index_bounds_valid=0 '\0',
> min_index=0, max_index=4, tfb_vertcount=0x0)
> at state_tracker/st_draw.c:1112
> #7 0x4152ead3 in vbo_exec_vtx_flush (exec=0x960b100, keepUnmapped=1 '\001')
> at vbo/vbo_context.h:91
> #8 0x4152a9b2 in vbo_exec_FlushVertices_internal (exec=0x960b100,
> unmap=0 '\0') at vbo/vbo_exec_api.c:444
> #9 0x4152bf8c in vbo_exec_FlushVertices (ctx=0x95af2b8, flags=4294967292)
> at vbo/vbo_exec_api.c:1195
> #10 0x414fea84 in _mesa_set_enable (ctx=0x95af2b8, cap=2929, state=1 '\001')
> at main/enable.c:350
>
> (frame 0 - buf was optimized out, but sbuf is as follows):
> (gdb) p *sbuf
> $23 = {b = {b = {reference = {count = 5}, screen = 0x95488e0,
> target = PIPE_BUFFER, format = PIPE_FORMAT_R8_UNORM, width0 = 65536,
> height0 = 1, depth0 = 1, array_size = 1, last_level = 0, nr_samples = 0,
> usage = 4, bind = 16, flags = 0}, vtbl = 0x416f49a0}, swbuf = 0x0,
> user = 0 '\0', key = {flags = 20, format = SVGA3D_BUFFER, size = {
> width = 65536, height = 1, depth = 1}, numFaces = 1, numMipLevels = 1,
> cachable = 1}, handle = 0x95f4528, map = {count = 0, ranges = {{
> start = 160, end = 240}, {start = 0, end = 0}<repeats 31 times>},
> num_ranges = 1}, uploaded = {buffer = 0x0, offset = 0, start = 0,
> end = 0}, hwbuf = 0x960f6e8, dma = {pending = 1 '\001', flags = {
> discard = 0, unsynchronized = 1, reserved = 0}, boxes = 0x9563824,
> svga = 0x955f490}, head = {prev = 0x9563728, next = 0x9563728}}
>
> (gdb) p *svga
> $24 = {pipe = {winsys = 0x0, screen = 0x95488e0, priv = 0x0, draw = 0x9539800,
> <function pointers here>
> texture_barrier = 0, create_video_decoder = 0, create_video_buffer = 0},
> swc = 0x9563738, debug = {no_swtnl = 0 '\0', force_swtnl = 0 '\0',
> use_min_mipmap = 0 '\0', shader_id = 4, disable_shader = 4294967295,
> no_line_width = 0 '\0', force_hw_line_stipple = 0 '\0'}, swtnl = {
> draw = 0x9539800, backend = 0x95356e8, hw_prim = 0, new_vbuf = 0 '\0',
> new_vdecl = 0 '\0'}, fs_bm = 0x95478d0, vs_bm = 0x95487d0, state = {
> dirty = {0, 0, 0, 4294967295}, texture_timestamp = 9, sw = {ve_format = {
> 1, 0<repeats 31 times>}, need_swvfetch = 0 '\0',
> need_pipeline = 0 '\0', need_swtnl = 0 '\0', in_swtnl_draw = 0 '\0'},
> hw_draw = {rs = {3452816845, 0, 3452816845, 0, 3452816845, 1, 3452816845,
> 3452816845, 0, 3452816845, 3452816845, 0, 3452816845, 0, 3452816845,
> 3452816845, 3452816845, 3452816845, 3452816845, 1065353216, 0,
> 1117782016, 3452816845, 3452816845, 3452816845, 3452816845,
> 3452816845, 0, 3452816845, 3452816845, 2, 0, 5, 6, 1, 1,
> 3452816845<repeats 11 times>, 15, 3452816845, 3452816845, 3452816845,
> 3452816845, 3452816845, 3452816845, 3452816845, 0, 0, 0, 3452816845,
> 3452816845, 3452816845, 3452816845, 3452816845, 0, 0, 3452816845,
> 3452816845, 0, 3452816845<repeats 17 times>, 0, 3452816845,
> 3452816845, 3452816845, 3452816845, 3452816845, 3452816845,
> 3452816845, 0, 3452816845, 3452816845, 3452816845, 3452816845,
> 3452816845, 3452816845}, ts = {{
> 3452816845<repeats 30 times>}<repeats 16 times>}, cb = {{{
> 0.00249999994, 0, 0, 0}, {0, -0.00333333341, 0, 0}, {0, 0, -1, 0},
> {-1, 1, 0, 1}, {1, 1, 0.5, 1}, {-0.00124999997, 0.00166666671, 0.5,
> 0}, {-431602080, -431602080, -431602080,
> -431602080}<repeats 250 times>}, {{0, 0, 0, 1}, {1, 0, 0, 0}, {0,
> 0, 1, 1}, {0, 0, 0, 0.999599993}, {-431602080, -431602080,
> -431602080, -431602080}<repeats 252 times>}, {{-431602080,
> -431602080, -431602080, -431602080}<repeats 256 times>}},
> fs = 0x985aa68, vs = 0x985c4e8, views = {{texture = 0x0, v = 0x0,
> min_lod = 0, max_lod = 0, dirty = 0}<repeats 16 times>},
> num_views = 0}, hw_clear = {viewport = {x = 0, y = 0, w = 800, h = 600},
> depthrange = {zmin = 0, zmax = 1}, framebuffer = {width = 0, height = 0,
> nr_cbufs = 0, cbufs = {0x985de10, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0},
> zsbuf = 0x985de68}, prescale = {translate = {-0.00124999997,
> 0.00166666671, 0.5, 0}, scale = {1, 1, 0.5, 1},
> enabled = 1 '\001'}}}, curr = {blend = 0x985a860, depth = 0x9860038,
> rast = 0x985a7e0, sampler = {0x0<repeats 16 times>}, velems = 0x985bad0,
> sampler_views = {0x0<repeats 16 times>}, fs = 0x9859c00, vs = 0x985ac88,
> vb = {{stride = 8, buffer_offset = 200, buffer = 0x960ed90}, {stride = 0,
> buffer_offset = 0, buffer = 0x0}<repeats 31 times>}, ib = {
> index_size = 0, offset = 0, buffer = 0x0}, cb = {0x97f7450, 0x960f740,
> 0x0}, framebuffer = {width = 800, height = 600, nr_cbufs = 1, cbufs = {
> 0x985de10, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0}, zsbuf = 0x985de68},
> depthscale = 1.1920929e-07, nr_fbs = 1, poly_stipple = {stipple = {
> 0<repeats 32 times>}}, scissor = {minx = 0, miny = 0, maxx = 800,
> maxy = 600}, blend_color = {color = {0, 0, 0, 0}}, stencil_ref = {
> ref_value = "\000"}, clip = {ucp = {{0, 0, 0, 0}, {0, 0, 0, 0}, {0, 0,
> 0, 0}, {0, 0, 0, 0}, {0, 0, 0, 0}, {0, 0, 0, 0}, {0, 0, 0, 0}, {0,
> 0, 0, 0}}}, viewport = {scale = {400, -300, 0.5, 1}, translate = {
> 400, 300, 0.5, 0}}, num_samplers = 0, num_sampler_views = 0,
> num_vertex_buffers = 1, reduced_prim = 4, tex_flags = {flag_1d = 0,
> flag_srgb = 0}, any_user_vertex_buffers = 0 '\0',
> zero_stride_vertex_elements = 0, num_zero_stride_vertex_elements = 0,
> zero_stride_constants = {0<repeats 128 times>}}, dirty = 1568, rebind = {
> rendertargets = 0, texture_samplers = 1}, upload_ib = 0x9548828,
> upload_vb = 0x9548850, hwtnl = 0x9538b18, sq = 0x0, dirty_buffers = {
> prev = 0x9563728, next = 0x9563728}}
>
> This all was initiated by our application doing a glEnable( GL_DEPTH_TEST ).
You might want to try the latest Mesa sources from git. It should be
easy to build/install in your guest if you've got the other pieces in
place.
Otherwise, if your app is proprietary, maybe you could make an
apitrace of the GL calls and send it to me for testing.
https://github.com/apitrace/apitrace
-Brian
More information about the mesa-dev
mailing list