Instead of polling the fences regularly, have a thread
that blocks for a single fence using a separate shared
context, then uses eventfd to wake up the main thread
when something happens.
Inside the guest, glmark2 typicially runs twice as fast with the thread
sync. Although in general, the performances seems to be about +30%. The
benefits is mostly for CPU-bounds tasks (when main the thread hits 100%)
A naive perf stat of the vtest renderer with glmark2 "build" test with a
fixed number of frames (500) results in the following stats data:
(do not value timing related informations, since the renderer is ran and
stopped manually)
without thread:
3032.282265 task-clock (msec) # 0.420 CPUs utilized
4,277 context-switches # 0.001 M/sec
102 cpu-migrations # 0.034 K/sec
9,020 page-faults # 0.003 M/sec
7,884,098,254 cycles # 2.600 GHz
4,440,126,451 stalled-cycles-frontend # 56.32% frontend cycles idle
<not supported> stalled-cycles-backend
11,024,091,578 instructions # 1.40 insns per cycle
# 0.40 stalled
# cycles per insn
1,091,831,588 branches # 360.069 M/sec
5,426,846 branch-misses # 0.50% of all branches
with thread:
3403.592921 task-clock (msec) # 0.452 CPUs utilized
7,145 context-switches # 0.002 M/sec
410 cpu-migrations # 0.120 K/sec
6,191 page-faults # 0.002 M/sec
7,475,038,064 cycles # 2.196 GHz
4,487,043,071 stalled-cycles-frontend # 60.03% frontend cycles idle
<not supported> stalled-cycles-backend
9,925,205,494 instructions # 1.33 insns per cycle
# 0.45 stalled
# cycles per insn
834,375,503 branches # 245.146 M/sec
4,919,995 branch-misses # 0.59% of all branches
Signed-off-by: Marc-André Lureau <marcandre.lureau@gmail.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
the protocol failed to handle larger shaders, this allow
the renderer to reassemble large shaders and recombined
the chunks before passing them to the GLSL translation.
This also enhances the renderer protocol to allow
for some more info in the shader object, and removes
the separate vs/gs/fs variants in favour of a type field
in the shader.
This code ended up in the other file and really wasn't necessary
there.
Remove the transfer code from virglrenderer.c, move into main
renderer file, and match it with the corresponding transfer
reader.
This should at least fix the crash in compressed textures
with ARB_get_texture_sub_image
This fixes a number of issues with how transform feedback works
it does requires ARB_transform_feedback3 to work at all, but
hopefully this extension is widespread enough, if not we can
revisit later.
It uses transform feedback objects to store the stream out
state.
This merges the error/bounds checking on the transfer
code, but keeps the same API, it also uses a struct
to pass through the transfer info.
this also passes a return value out to make testing easier.
Marc-André Lureau
Dave Airlie
Dave Airlie