Replace uses of brw_builder::at() with various more descriptive
variants. Use block pointer from instruction when possible.
A couple of special cases remained and will be handled in separate patches.
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34681>
Since brw_inst now has the block it belongs and the block can
reach the shader, the only necessary information to create a
builder is the brw_inst itself.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33815>
The SIMD splitting pass does not handle wide force_writemask_all
instructions correctly at the moment. For example, a SIMD32 TXF
on pre-Xe2 would get split to a pair of SIMD16. But it will set
the groups to operate on channels 15:0 and 31:16. That's not what
we want for a NoMask instruction - both should be 15:0, i.e.
bld.group(inst->exec_size, 0).
We could (and perhaps should) fix the SIMD splitting pass to handle
this, but the pass already has subtle complexity in which builders
are used. Or we could alter fs_builder::group(), but that has broader
implications. As a stop-gap, just make opt_combine_covergent_txfs stop
relying on SIMD splitting. It's trivial to do and fixes the issue
without risking other breakage.
Fixes: 6341b3cd87 ("brw: Combine convergent texture buffer fetches into fewer loads")
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/32714>
Borderlands 3 (both DX11 and DX12 renderers) have a common pattern
across many shaders:
con 32x4 %510 = (uint32)txf %2 (handle), %1191 (0x10) (coord), %1 (0x0) (lod), 0 (texture)
con 32x4 %512 = (uint32)txf %2 (handle), %1511 (0x11) (coord), %1 (0x0) (lod), 0 (texture)
...
con 32x4 %550 = (uint32)txf %2 (handle), %1549 (0x25) (coord), %1 (0x0) (lod), 0 (texture)
con 32x4 %552 = (uint32)txf %2 (handle), %1551 (0x26) (coord), %1 (0x0) (lod), 0 (texture)
A single basic block contains piles of texelFetches from a 1D buffer
texture, with constant coordinates. In most cases, only the .x channel
of the result is read. So we have something on the order of 28 sampler
messages, each asking for...a single uint32_t scalar value. Because our
sampler doesn't have any support for convergent block loads (like the
untyped LSC transpose messages for SSBOs)...this means we were emitting
SIMD8/16 (or SIMD16/32 on Xe2) sampler messages for every single scalar,
replicating what's effectively a SIMD1 value to the entire register.
This is hugely wasteful, both in terms of register pressure, and also in
back-and-forth sending and receiving memory messages.
The good news is we can take advantage of our explicit SIMD model to
handle this more efficiently. This patch adds a new optimization pass
that detects a series of SHADER_OPCODE_TXF_LOGICAL, in the same basic
block, with constant offsets, from the same texture. It constructs a
new divergent coordinate where each channel is one of the constants
(i.e <10, 11, 12, ..., 26> in the above example). It issues a new
NoMask divergent texel fetch which loads N useful channels in one go,
and replaces the rest with expansion MOVs that splat the SIMD1 result
back to the full SIMD width. (These get copy propagated away.)
We can pick the SIMD size of the load independently of the native shader
width as well. On Xe2, those 28 convergent loads become a single SIMD32
ld message. On earlier hardware, we use 2 SIMD16 messages. Or we can
use a smaller size when there aren't many to combine.
In fossil-db, this cuts 27% of send messages in affected shaders, 3-6%
of cycles, 2-3% of instructions, and 8-12% of live registers. On A770,
this improves performance of Borderlands 3 by roughly 2.5-3.5%.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/32573>
