8259c97b2d
aco: propagate temporaries into expanded vectors
...
Gives a very slight decrease in code size:
Totals from affected shaders:
Code Size: 1708488 -> 1702768 (-0.33 %) bytes
Max Waves: 2858 -> 2855 (-0.10 %)
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com >
2019-12-07 11:23:11 +01:00
da7ff58835
aco: make 1/2*PI a literal constant on SI/CI
...
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com >
2019-12-07 11:23:11 +01:00
6a586a6006
aco: split read/writelane opcode into VOP2/VOP3 version for SI/CI
...
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com >
2019-12-07 11:23:11 +01:00
caea4bbfdc
aco: fix SMEM offsets for SI/CI
...
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com >
2019-12-07 11:23:11 +01:00
c0dbf42a03
aco/wave32: Change uniform bool optimization to work with wave32.
...
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com >
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com >
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev >
2019-12-04 10:36:01 +00:00
df645fa369
aco: implement VK_KHR_shader_float_controls
...
This actually supports more of the extension than the LLVM backend but we
can't enable it because ACO doesn't work with all stages yet.
With more of it enabled, some CTS tests fail because our 64-bit sqrt
is very imprecise. I can't find any precision requirements for it
anywhere, so I'm thinking it might be a CTS issue.
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com >
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev >
2019-11-15 17:36:21 +00:00
b062b92ab1
aco: don't combine literals into v_cndmask_b32/v_subb/v_addc
...
No pipeline-db changes
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com >
Fixes: 93c8ebfadaniel@schuermann.dev >
2019-11-15 17:36:21 +00:00
9b8dc6929e
aco: Optimize out trivial code from uniform bools.
...
This should remove most of the excess code size that was
introduced by making all booleans per-lane.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com >
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com >
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev >
2019-11-14 17:27:11 +01:00
94e355148f
aco: Make sure not to mistakenly propagate 64-bit constants.
...
ACO's optimizer would try to propagate 64-bit constants, but
does so in such a way that wouldn't work due to how the 64-bit
constants are handled in the IR.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com >
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com >
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev >
2019-11-14 17:27:10 +01:00
2c98d79d11
aco: don't propagate vgprs into v_readlane/v_writelane
...
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com >
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev >
Fixes: 93c8ebfa
2019-11-12 17:21:38 +00:00
78e3ea9a0f
aco: add Instruction::usesModifiers() and add more checks in the optimizer
...
No pipeline-db changes.
v2: use early-exit for VOP3
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev > (v1)
2019-11-08 00:14:06 +00:00
c96289a70e
aco: keep can_reorder/barrier when combining addition into SMEM
...
Affects 30 shaders in the pipeline-db (all youngblood).
Totals from affected shaders:
SGPRS: 2656 -> 2456 (-7.53 %)
VGPRS: 2260 -> 2260 (0.00 %)
Spilled SGPRs: 0 -> 0 (0.00 %)
Spilled VGPRs: 0 -> 0 (0.00 %)
Private memory VGPRs: 0 -> 0 (0.00 %)
Scratch size: 0 -> 0 (0.00 %) dwords per thread
Code Size: 240680 -> 240944 (0.11 %) bytes
LDS: 0 -> 0 (0.00 %) blocks
Max Waves: 90 -> 90 (0.00 %)
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com >
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev >
2019-10-22 18:52:29 +00:00
bdf47a1273
aco: properly combine additions into ds_write2_b64/ds_read2_b64
...
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com >
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev >
2019-10-22 18:52:29 +00:00
a400928f4a
aco: fix 64-bit p_extract_vector on 32-bit p_create_vector
...
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com >
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev >
2019-10-22 18:52:29 +00:00
4b458b3e8f
aco: don't combine minmax3 if there is a neg or abs modifier in between
...
This fixes a graphical corruption in HotS.
No pipelinedb changes other than that.
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com >
2019-10-17 16:21:19 +00:00
45d6c69b99
aco: use can_accept_constant in valu_can_accept_literal
...
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com >
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev >
2019-10-11 14:26:58 +00:00
b37857bcea
aco: don't apply sgprs/constants to read/write lane instructions
...
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com >
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev >
2019-10-11 14:26:58 +00:00
93c8ebfa78
aco: Initial commit of independent AMD compiler
...
ACO (short for AMD Compiler) is a new compiler backend with the goal to replace
LLVM for Radeon hardware for the RADV driver.
ACO currently supports only VS, PS and CS on VI and Vega.
There are some optimizations missing because of unmerged NIR changes
which may decrease performance.
Full commit history can be found at
https://github.com/daniel-schuermann/mesa/commits/backend
Co-authored-by: Daniel Schürmann <daniel@schuermann.dev >
Co-authored-by: Rhys Perry <pendingchaos02@gmail.com >
Co-authored-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl >
Co-authored-by: Connor Abbott <cwabbott0@gmail.com >
Co-authored-by: Michael Schellenberger Costa <mschellenbergercosta@googlemail.com >
Co-authored-by: Timur Kristóf <timur.kristof@gmail.com >
Acked-by: Samuel Pitoiset <samuel.pitoiset@gmail.com >
Acked-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl >
2019-09-19 12:10:00 +02:00
Daniel Schürmann