Now that we added an index src to the NIR intrinsic, it can
happen that these generate MUBUF instructions which have both
an index and an offset.
Extend this ACO optimization to the case when idxen is used.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Georg Lehmann <dadschoorse@gmail.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17551>
Also modify all existing uses to pass a zero to this new src.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Georg Lehmann <dadschoorse@gmail.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Alyssa Rosenzweig <alyssa@collabora.com> (nir)
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17551>
Remove unused arguments, clean up allow_combining vs. swizzled etc.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Georg Lehmann <dadschoorse@gmail.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17551>
Previously, we always treated these as coherent, but now let's make
this configurable. Also set all current users to ACCESS_COHERENT.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Georg Lehmann <dadschoorse@gmail.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Alyssa Rosenzweig <alyssa@collabora.com> (nir)
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17551>
Avoids later moves with extractions from the vector.
Reduces VALU operation in the raytrace loop by ~6%, increasing
the RT performance in Q2RTX on a 6800 XT by about ~1.3%.
Suggested by Georg.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19148>
GFX11 has a lot of complicated data dependency hazards.
For debugging GFX10+ data dependency hazards. This creates an excessive
amount of s_waitcnt_depctr.
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18273>
fossil-db (gfx1100):
Totals from 57858 (42.85% of 135032) affected shaders:
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18273>
Same as GFX10, but in a separate pass because it's the only hazard that's
shared.
No fossil-db changes.
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18273>
"Removing jumps" in ACO means skipping the jump instruction
at the beginning of a divergent branch (but still modify exec).
ACO already supports implicitly removing jumps when it decides
that executing a branch with empty exec mask is more beneficial
than a jump.
This commit adds the possibility to use this explicitly
through nir_selection_control. ACO will respect this
setting and remove the branch instructions when this is specified,
unless it decides that this would cause bugs (eg. exp instruction).
There are two cases that benefit from the new change:
1. When the application requests to "flatten" a branch (ie.
remove control flow), we now respect that.
2. When the compiler stack determines that a divergent branch
is always taken.
v2 by Georg Lehmann: fixed applying sel_ctrl to else blocks
Fossil DB stats on Navi 21:
Totals from 13 (0.01% of 134906) affected shaders:
CodeSize: 136616 -> 136496 (-0.09%)
Instrs: 26196 -> 26166 (-0.11%)
Latency: 417928 -> 417889 (-0.01%)
Branches: 1241 -> 1211 (-2.42%)
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Reviewed-By: Georg Lehmann <dadschoorse@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17921>
This was missed. I guess CI doesn't have a recent enough LLVM for these
tests.
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17710>
Add GFX11 support and use wait_imm::unset_counter. Looping in the waitcnt
pass was probably broken on GFX11 because of this.
fossil-db (gfx1100):
Totals from 899 (0.56% of 161689) affected shaders:
Instrs: 1319368 -> 1319179 (-0.01%)
CodeSize: 7124640 -> 7123884 (-0.01%)
Latency: 26554304 -> 26404606 (-0.56%)
InvThroughput: 9032485 -> 8978773 (-0.59%); split: -0.59%, +0.00%
No navi10 fossil-db changes.
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17710>
According to an LLVM comment, these are swapped in GFX11.
No fossil-db changes.
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17710>
It doesn't make sense for a "s_waitcnt vmcnt(0)" to affect a store or DS
instruction.
LLVM checks for "s_waitcnt vmcnt(0) lgkmcnt(0) expcnt(0)" but ignores
s_waitcnt_vscnt (which I assume is a bug).
No fossil-db changes.
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Fixes: bcf94bb933 ("aco: properly recognize that s_waitcnt mitigates VMEMtoScalarWriteHazard")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18270>
As monotonic_buffer_resource is not thread-safe,
we use a thread_local instance which gets allocated once.
This change reduces the compile time spent in ACO by
approximately 10%.
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18112>
This basic allocator implements an arena allocation strategy
and cannot free individual allocations.
It is intended for very fast memory allocations in situations
where memory is used to build up a few objects and then is
released all at once.
This class mimics std::pmr::monotonic_buffer_resource.
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18112>
Quad-divergent CF and vertex selection doesn't work, but should at least
prevent crashes.
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17333>
The error is:
../mesa/src/amd/compiler/aco_register_allocation.cpp:382:7: error: no matching function for call to 'printf'
printf(u8"☐");
Fixes: 209a89e51d ("aco: Convert to use u8 literal for Unicode character to fixes msvc warning")
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/7318
Signed-off-by: Yonggang Luo <luoyonggang@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Tested-by: Marcus Seyfarth <m.seyfarth@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18796>
