The idea is to build a hardware command buffer for every CSO and memcpy
the buffer to a command stream at bind time (or dirty-state-emission time,
to be precise).
The loop emulation unrolls loops as may times as possbile while still
keeping the shader program below the maximum instruction limit. At this
point, there are no checks for constant conditionals. This is only enabled
for fragment shaders.
i965: Use the state base address to avoid relocations.
This makes the binding table code simpler, and is required for gen6,
which requires binding table addresses to be under 64k offset from the
surface state base addr.
No significant change in performance on firefox-talos-gfx.
It turns out that computing a 56 byte key to look up a 20-byte object
out of a hash table was some sort of a bad idea. Whoops.
before:
[ # ] backend test min(s) median(s) stddev. count
[ 0] gl firefox-talos-gfx 37.799 38.203 0.39% 6/6
after:
[ 0] gl firefox-talos-gfx 34.761 34.784 0.17% 5/6
i965: Convert the binding table to streamed indirect state.
This slightly reduces reduces cairo-gl firefox-talos-gfx runtime on my
Ironlake:
before:
[ # ] backend test min(s) median(s) stddev. count
[ 0] gl firefox-talos-gfx 38.236 38.383 0.43% 5/6
after:
[ 0] gl firefox-talos-gfx 37.799 38.203 0.39% 6/6
It turns out the cost of caching these objects and looking them up in
the cache again is greater than the cost of just computing the object
again, particularly when the overhead of having a separate BO to pin
is removed.
(Those that are paying close attention will note that this is a
reversal of the path I was moving the driver in a couple of years ago.
The major thing that has changed is that back then all state was
recomputed when we wrapped the streaming state buffer, including
recompiling our precious programs. Now, we're uncaching just the
objects that are cheap to compute, and retaining caching of expensive
objects)
i965: Set the CC VP state immediately on state change.
The cache lookup of these two little floats was .12% of total CPU time
on firefox-talos-gfx because we did it any time commonly-changed state
changed. On the other hand, updating the CC VP bo immediately whenver
CC VP state changes is a .07% overhead due to putting a driver hoook
in glEnable().