This can not happen because the post-RA optimizer doesn't support sub dword
writes at the moment, but everytime I look at this I wonder if there might
be a bug here.
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22821>
It's just v_fma with fixed DPP8 and builtin s_waitcnt_expcnt, so it can mostly
be handled as a pure VALU instruction.
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21023>
Register writes from the pre-header might not be correct for any but
the first loop iteration because they can be clobbered inside the loop.
Foz-DB Navi21:
Totals from 18 (0.01% of 134913) affected shaders:
CodeSize: 251384 -> 251508 (+0.05%)
Instrs: 47644 -> 47664 (+0.04%)
Latency: 801801 -> 801852 (+0.01%)
InvThroughput: 177579 -> 177593 (+0.01%)
Copies: 4752 -> 4771 (+0.40%)
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/8376
Fixes: d3b0f78110 ("aco/optimizer_postRA: Initialize loop header with preheader information")
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21540>
Any unused split_vector definition can always use the same register
as the operand. This avoids creating unnecessary copies.
Fossil DB stats on Rembrandt (RDNA2):
Totals from 3904 (2.89% of 134906) affected shaders:
CodeSize: 18326692 -> 18271688 (-0.30%)
Instrs: 3386632 -> 3372888 (-0.41%)
Latency: 42337481 -> 42330085 (-0.02%); split: -0.02%, +0.00%
InvThroughput: 6566731 -> 6566424 (-0.00%); split: -0.01%, +0.00%
Copies: 224301 -> 210559 (-6.13%)
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16161>
Eliminate unnecessary copies when the operand registers of a
p_split_vector instruction are not clobbered between the p_split_vector
and the user of its definitions.
This happens when p_split_vector doesn't kill its operand and its
definitions have a shorter lifespan that the operand. It affects every
NGG culling shader among other things.
This optimization exists because it's too difficult to solve it
in RA, and should be removed after we solved this in RA.
v2 by Daniel Schürmann:
- Rearrange and simplify conditions for the new optimization
- Fix a few bugs
v3 by Daniel Schürmann:
- Check number of encoded ALU operands
Fossil DB stats on Rembrandt (RDNA2):
Totals from 64896 (48.10% of 134906) affected shaders:
CodeSize: 175693348 -> 175434944 (-0.15%)
Instrs: 33333912 -> 33269388 (-0.19%)
Latency: 183766084 -> 183763432 (-0.00%); split: -0.00%, +0.00%
InvThroughput: 28589651 -> 28589340 (-0.00%); split: -0.00%, +0.00%
Copies: 2806550 -> 2742038 (-2.30%)
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16161>
This allows is_overwritten_since to return false when the last
writer instruction of a register can't be tracked but we know it wasn't
written in the current block.
Fossil DB stats on Rembrandt (RDNA2):
Totals from 1163 (0.86% of 134906) affected shaders:
CodeSize: 9815920 -> 9805016 (-0.11%)
Instrs: 1843688 -> 1840962 (-0.15%)
Latency: 19219153 -> 19209171 (-0.05%)
InvThroughput: 3354375 -> 3353852 (-0.02%)
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16161>
This works because of SSA and should be safer than just setting 'not_written_yet'.
No Fossil DB changes on Rembrandt (RDNA2).
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16161>
Copy the information from the first predecessor then check whether
it matches other predecessors and modify the data accordingly.
Marked for backporting to stable to make it possible to also
backport fixes based on this.
Cc: mesa-stable
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18103>
According to perf, this roughly halves the impact of the post-RA
optimizer in ACO's compile times.
Measurement was taken using a debug optimized build using
NIR_DEBUG=novalidate RADV_DEBUG=nocache and replaying the Fossil DB
from the Doom Eternal shaders.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18103>
Change the terminology around the post-RA optimizer, primarily this
changes the use of "clobbered" to "overwritten" to avoid confusion,
and it removes some redundant states.
Proposed for backporting to stable, to make sure it is easy to
backport further fixes (if any) on top of this.
Fossil DB stats unaffected on Navi 21.
Cc: mesa-stable
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18488>
Change reset_block() so it only considers the logical
predecessors for VGPRs. Relevant for some optimizations
across loops.
This commit fixes an assertion failure which was triggered
by Zink in a piglit test.
Fossil DB stats unaffected on Navi 21.
Fixes: 2e56e23420
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18488>
This assumption is no longer true since the post-RA optimizer
can work across blocks. It is now possible that some control
flow paths overwrite some but not all registers of an operand.
This commit may prevent invalid optimizations and/or assertion
failures (on debug builds).
Fossil DB stats unaffected on Navi 21.
Fixes: 0e4747d3fb
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18488>
Affects blocks whose some (but not all) predecessors overwrite a register.
This commit fixes glitches in some games which regressed because of the
improved SCC no-compare optimization.
Fossil DB stats on Navi 21:
Totals from 2816 (2.09% of 134906) affected shaders:
CodeSize: 24224276 -> 24241580 (+0.07%)
Instrs: 4570595 -> 4574921 (+0.09%)
Latency: 53680256 -> 53693655 (+0.02%); split: -0.00%, +0.02%
InvThroughput: 9829289 -> 9830573 (+0.01%)
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/7257
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/7305
Fixes: 2e56e23420
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18488>
When SCC is clobbered between s_cmp and its operand's writer,
the current optimization that eliminates s_cmp won't kick in.
However, when s_cmp is the only user of its operand temporary,
it is possible to "pull down" the instruction that wrote the operand.
Fossil DB stats on Navi 21:
Totals from 63302 (46.92% of 134906) affected shaders:
CodeSize: 176689272 -> 176418332 (-0.15%)
Instrs: 33552237 -> 33484502 (-0.20%)
Latency: 205847485 -> 205816205 (-0.02%); split: -0.02%, +0.00%
InvThroughput: 34321285 -> 34319908 (-0.00%); split: -0.00%, +0.00%
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16266>
This was simply left out by accident when I wrote this.
Fossil DB stats on Navi 21:
Totals from 70165 (52.01% of 134906) affected shaders:
CodeSize: 246375656 -> 245814396 (-0.23%)
Instrs: 46519773 -> 46379458 (-0.30%)
Latency: 385159303 -> 385089261 (-0.02%); split: -0.02%, +0.00%
InvThroughput: 66490172 -> 66487867 (-0.00%); split: -0.00%, +0.00%
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16266>
Also delete them when they are already dead in process_instruction().
No Fossil DB changes.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16266>
Also add a few more assertions to make sure the registers are
within the bounds of the array.
Cc: mesa-stable
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Tony Wasserka <tony.wasserka@gmx.de>
Reviewed-by: Joshua Ashton <joshua@froggi.es>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/12682>
In this case std::array doesn't behave like a regular array, therefore
it is NOT okay to index it outside the array, even though std::fill
needs us to do so.
Change the syntax to do the same thing slightly differently,
and add an assertion to make sure the registers are always within
the array's bounds.
Closes: #5289
Fixes: 0e4747d3fb
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Tony Wasserka <tony.wasserka@gmx.de>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/12664>
This migration was done with libclang-based automatic tooling, which
performed these replacements:
* Operand(uint8_t) -> Operand::c8
* Operand(uint16_t) -> Operand::c16
* Operand(uint32_t, false) -> Operand::c32
* Operand(uint32_t, bool) -> Operand::c32_or_c64
* Operand(uint64_t) -> Operand::c64
* Operand(0) -> Operand::zero(num_bytes)
Casts that were previously used for constructor selection have automatically
been removed (e.g. Operand((uint16_t)1) -> Operand::c16(1)).
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11653>
A simple post-RA optimization which takes advantage of the
s_cbranch_vccz and s_cbranch_vccnz instructions.
It works on the following pattern:
vcc = v_cmp ...
scc = s_and vcc, exec
p_cbranch scc
The result looks like this:
vcc = v_cmp ...
p_cbranch vcc
Fossil DB results on Sienna Cichlid:
Totals from 4814 (3.21% of 149839) affected shaders:
CodeSize: 15371176 -> 15345964 (-0.16%)
Instrs: 3028557 -> 3022254 (-0.21%)
Latency: 21872753 -> 21823476 (-0.23%); split: -0.23%, +0.00%
InvThroughput: 4470282 -> 4468691 (-0.04%); split: -0.04%, +0.00%
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7779>
This commit adds the skeleton of a new ACO post-RA optimizer,
which is intended to be a simple pass called after RA, and
is meant to do code changes which can only be done
after RA.
It is currently empty, the actual optimizations will be added
in their own commits. It only has a DCE pass, which deletes
some dead code generated by the spiller.
Fossil DB results on Sienna Cichlid:
Totals from 375 (0.25% of 149839) affected shaders:
CodeSize: 2933056 -> 2907192 (-0.88%)
Instrs: 534154 -> 530706 (-0.65%)
Latency: 12088064 -> 12084907 (-0.03%); split: -0.03%, +0.00%
InvThroughput: 4433454 -> 4432421 (-0.02%); split: -0.02%, +0.00%
Copies: 81649 -> 78203 (-4.22%)
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7779>
