fdo-mirrors/mesa
1
0
Fork 0
mirror of https://gitlab.freedesktop.org/mesa/mesa.git synced 2026-05-21 06:48:09 +02:00
Code Issues Projects Releases Packages Wiki Activity Actions 1
172929 commits 66 branches 936 tags 998 MiB
Commit graph

2575 commits

Author SHA1 Message Date
Georg Lehmann
49bdc3a1c2 aco/opcodes: move v_cndmask_b32 back to the VOP2 list 
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21882>
 2023-06-16 13:45:38 +00:00
Tony Wasserka
89c86af39e aco/spill: Use arena allocator for spills 
Gives a marginal speedup.

Co-authored-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/12000>
 2023-06-16 12:46:01 +00:00
Tony Wasserka
73835f4b09 aco/spill: Use arena allocator for next use distances 
Speedup of aco::spill(): ~50%

Co-authored-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/12000>
 2023-06-16 12:46:01 +00:00
Georg Lehmann
bbda9f7390 aco: validate ir for prologs and after lower_to_hw_instr 
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23507>
 2023-06-12 19:43:17 +00:00
Georg Lehmann
2028df8757 aco: don't validate p_constaddr_addlo/p_resumeaddr_addlo operands 
These can have two literals so validation would fail.

Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23507>
 2023-06-12 19:43:17 +00:00
Georg Lehmann
b9854a9097 aco: move cfg validation to its own function 
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23507>
 2023-06-12 19:43:17 +00:00
Georg Lehmann
e5df6ee605 aco: make validation work without SSA temps 
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23507>
 2023-06-12 19:43:17 +00:00
Friedrich Vock
9de8134410 aco: Fix assert in insert_exec_mask 
This assert would trigger on unconditional demotes, because the demotes
don't remove the mask_type_global flag from the exec mask.

Cc: mesa-stable
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23594>
 2023-06-12 14:20:28 +00:00
Friedrich Vock
3ea01b86f0 aco: Fix live_var_analysis assert 
Fixes: 3d4f6a00b ('aco/spill: allow for disconnected CFG')
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23586>
 2023-06-12 13:52:03 +00:00
Friedrich Vock
7f3cfcc96a aco: Reset scratch_rsrc on blocks without predecessors 
Fixes hangs with raytracing in Hellblade: Senua's Sacrifice.

Fixes: 3d4f6a00b ('aco/spill: allow for disconnected CFG')
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23586>
 2023-06-12 13:52:03 +00:00
Timur Kristóf
67a0f2532f aco: Mark exec write used when it writes other registers. 
When an exec write isn't used but writes other registers
besides exec, and also reads exec (such as s_and_saveexec),
we would mistakenly delete the previous instruction that
writes the exec value that this instruction uses.

No Fossil DB changes on Rembrandt (GFX10.3).

Fixes: 0211e66f65
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/9036
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Georg Lehmann <dadschoorse@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23576>
 2023-06-12 15:07:55 +02:00
Konstantin Seurer
51cd2965c7 aco/rt: Do not initialize the next shader addr 
The uniform one is already set and the raygen shader isn't guarded
anymore.

Reviewed-by: Friedrich Vock <friedrich.vock@gmx.de>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23545>
 2023-06-10 10:00:27 +00:00
Friedrich Vock
64fda091de aco: Lower divergent bool phis iteratively 
Avoids stack overflows with really large programs.

No fossil-db changes.

Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/8760
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/8701

Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>

Cc: mesa-stable
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23531>
 2023-06-09 12:39:55 +00:00
Rhys Perry
56eb831155 aco: run nir_lower_int64 after nir_opt_uniform_atomics 
nir_opt_uniform_atomics can create 64-bit ALU instructions which need to
be lowered.

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Georg Lehmann <dadschoorse@gmail.com>
Reviewed-by: Qiang Yu <yuq825@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23502>
 2023-06-09 11:18:33 +00:00
Rhys Perry
31c8c42f48 aco/tests: test that s_bfe bits is masked 
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Georg Lehmann <dadschoorse@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23464>
 2023-06-08 11:54:45 +01:00
Rhys Perry
08064a5542 aco: mask bits source of s_bfe 
The s_bfe instructions use 7 bits, not 5 like the NIR opcode requires.

No fossil-db changes (navi21).

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Georg Lehmann <dadschoorse@gmail.com>
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/9162
Cc: mesa-stable
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23464>
 2023-06-08 11:54:17 +01:00
Yonggang Luo
586391720b util: use uint32_t as the parameter of align function 
align on negative value doesn't make sense

Reviewed-by: Rohan Garg <rohan.garg@intel.com>
Signed-off-by: Yonggang Luo <luoyonggang@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23421>
 2023-06-08 06:41:21 +00:00
Daniel Schürmann
defdcd2058 aco: adjust RT prolog for shader functions [disables RT] 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22096>
 2023-06-08 00:37:03 +00:00
Daniel Schürmann
b33c01b00f radv/rt: set up RT shader args for separate compilation 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22096>
 2023-06-08 00:37:03 +00:00
Daniel Schürmann
393f3426b6 aco: implement select_program_rt() 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22096>
 2023-06-08 00:37:03 +00:00
Daniel Schürmann
f66f274304 aco: implement nir_intrinsic_load_resume_shader_address_amd 
Similar to p_constaddr but targeting BBs.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22096>
 2023-06-08 00:37:03 +00:00
Daniel Schürmann
03c4b5b0cc nir,amd: add nir_intrinsic_store_[scalar|vector]_arg_amd to overwrite inputs 
This intrinsic must only be used at top-level CF in order
to not break SSA properties.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22096>
 2023-06-08 00:37:03 +00:00
Daniel Schürmann
1be3a558f2 radv: add remaining RT shader args for separate compilation 
Also wrap RT args into struct {} rt for improved consistency
and remove some 'ray_' prefixes.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22096>
 2023-06-08 00:37:03 +00:00
Marek Olšák
6dc1ae1759 amd: drop support for LLVM 14 
Reviewed-by: Qiang Yu <yuq825@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23471>
 2023-06-07 19:56:55 +00:00
Marek Olšák
ab5662dc61 amd: drop support for LLVM 13 
We can remove the LLVM 13 Wave32 discard workaround and
SI_PROFILE_IGNORE_LLVM13_DISCARD_BUG that disabled the workaround.

Reviewed-by: Qiang Yu <yuq825@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23471>
 2023-06-07 19:56:55 +00:00
Rhys Perry
b5c31f14a2 aco: remove memory_barrier_buffer implementation 
This is no longer used.

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21624>
 2023-06-07 13:19:42 +00:00
Rhys Perry
417990b19e aco: consider position/primitive exports around memory barriers 
This is needed to create barriers which ensure stores finish before
position/primitive exports.

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21624>
 2023-06-07 13:19:41 +00:00
Georg Lehmann
dfb6d3e443 aco: use v_fma_mix for f2f32 and f2f16 on gfx11 if wave64 
v_fma_mix can be dual issued, trade some code size for throughput.

Foz-DB GFX1100:
Totals from 8204 (6.08% of 134864) affected shaders:
CodeSize: 89608584 -> 89693968 (+0.10%)
Latency: 160744811 -> 160699309 (-0.03%); split: -0.03%, +0.00%
InvThroughput: 19737977 -> 19678308 (-0.30%)

Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21402>
 2023-06-07 12:30:11 +00:00
Georg Lehmann
177dba62a1 aco: use v_add_f{16,32} with clamp for fsat 
v_add can be dual issued on gfx11, v_med3 cannot.
Don't use v_add directly to still optimize omod(fsat(x)).

Foz-DB GFX1100:
Totals from 32702 (24.24% of 134913) affected shaders:
Latency: 475008203 -> 474928037 (-0.02%); split: -0.02%, +0.00%
InvThroughput: 59226198 -> 59140787 (-0.14%); split: -0.14%, +0.00%

Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21402>
 2023-06-07 12:30:11 +00:00
Georg Lehmann
3a0bc8f007 aco/statistics: improve v_fma_mix dual issuing detection 
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21402>
 2023-06-07 12:30:11 +00:00
Georg Lehmann
0f023d90c0 aco/ir: return true in hasRegClass for Operand(reg, rc) 
Makes isOfType() usable after lower_to_hw_instr.

Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21402>
 2023-06-07 12:30:11 +00:00
Rhys Perry
53383fe8a5 aco: fix ds_sub_gs_reg_rtn validation 
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Qiang Yu <yuq825@gmail.com>
Fixes: 8d5cc23c18 ("aco: use gds reg when ordered xfb counter add")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23434>
 2023-06-06 16:09:28 +00:00
Qiang Yu
bd88c75d4c aco,radv: remove unused gs aco shader info 
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Signed-off-by: Qiang Yu <yuq825@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23432>
 2023-06-06 10:55:10 +00:00
Rhys Perry
9ae5c942da aco/tests: add discard export target tests 
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23371>
 2023-06-05 11:42:21 +00:00
Rhys Perry
21867b45c1 aco: fix has_color_exports=true for mrtz exports 
V_008DFC_SQ_EXP_NULL is after V_008DFC_SQ_EXP_MRTZ.

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Qiang Yu <yuq825@gmail.com>
Tested-by: Mikhail Gavrilov mikhail.v.gavrilov@gmail.com
Fixes: d3611af389 ("aco: support nir_export_amd with ps targets")
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/9135
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23371>
 2023-06-05 11:42:21 +00:00
Qiang Yu
2d0e8e0258 aco: use ac_get_image_dim for array check when image intrinsic 
This is to avoid missing array flag when <=GFX8 and 3D image
in which case is treated as 2D array image.

Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Signed-off-by: Qiang Yu <yuq825@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23094>
 2023-06-02 09:21:59 +00:00
Qiang Yu
ed97cd92dc aco: implement nir_bindless_image_fragment_mask_load_amd 
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Sigend-off-by: Qiang Yu <yuq825@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23094>
 2023-06-02 09:21:59 +00:00
Qiang Yu
50f9d644e8 aco: implement nir_xfb_counter_sub_amd 
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Signed-off-by: Qiang Yu <yuq825@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23094>
 2023-06-02 09:21:59 +00:00
Qiang Yu
8d5cc23c18 aco: use gds reg when ordered xfb counter add 
This is currently only used by radeonsi.

Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Signed-off-by: Qiang Yu <yuq825@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23094>
 2023-06-02 09:21:59 +00:00
Qiang Yu
438dcf6d0f aco/assembler: handle ds_(add|sub)_gs_reg_rtn encoding 
They are different than normal DS instructions, only
use DATA[0], not use ADDR.

Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Signed-off-by: Qiang Yu <yuq825@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23094>
 2023-06-02 09:21:59 +00:00
Qiang Yu
460b528c9e aco: implement load buffer with ACCESS_USES_FORMAT_AMD 
This is used by radeonsi for vs input load and cdna
image load emulation.

Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Signed-off-by: Qiang Yu <yuq825@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23094>
 2023-06-02 09:21:59 +00:00
Qiang Yu
12ee7eccf7 aco,radv: remove unused aco_shader_info fields 
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Signed-off-by: Qiang Yu <yuq825@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23094>
 2023-06-02 09:21:58 +00:00
Qiang Yu
89dab66561 aco: implement two load lds ngg intrininsic for radeonsi 
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Signed-off-by: Qiang Yu <yuq825@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23094>
 2023-06-02 09:21:58 +00:00
Georg Lehmann
79821d7afb aco: p_start_linear_vgpr doesn't always need exec mask 
Foz-DB Navi21:
Totals from 1605 (1.21% of 132657) affected shaders:
CodeSize: 14023700 -> 14020320 (-0.02%)
Instrs: 2589881 -> 2589052 (-0.03%)
Latency: 22478420 -> 22473359 (-0.02%)
InvThroughput: 3851237 -> 3851092 (-0.00%)
Copies: 215316 -> 215438 (+0.06%); split: -0.39%, +0.44%

Allows more vcmpx usage.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23302>
 2023-06-01 20:18:23 +00:00
Georg Lehmann
7836260af8 aco: cleanup v_cmp_class usage 
It's not the best way to check for NaN.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23310>
 2023-05-31 12:56:37 +00:00
Samuel Pitoiset
1947500208 aco: remove nir_intrinsic_load_barycentric_at_sample occurences 
This is lowered earlier and shouldn't get there.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23307>
 2023-05-31 07:25:46 +00:00
Chia-I Wu
750d641ca6 aco: fix alignment check in emit_load 
align_offset already takes const_offset into consideration.  Remove the
adjustment.

Fixes: 542733dbbf ("aco: add emit_load helper")
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/9097
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23242>
 2023-05-30 16:02:34 +00:00
Rhys Perry
94958e637d aco: improve printing of s_delay_alu 
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Georg Lehmann <dadschoorse@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23213>
 2023-05-30 12:42:00 +00:00
Rhys Perry
54c0088629 aco: insert s_delay_alu on the linear CFG 
fossil-db (gfx1100):
Totals from 10498 (7.87% of 133428) affected shaders:
Instrs: 22274711 -> 22277717 (+0.01%); split: -0.01%, +0.03%
CodeSize: 114557040 -> 114569064 (+0.01%); split: -0.01%, +0.02%
Latency: 236505186 -> 236497338 (-0.00%); split: -0.00%, +0.00%
InvThroughput: 33425052 -> 33423876 (-0.00%); split: -0.00%, +0.00%

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Georg Lehmann <dadschoorse@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23213>
 2023-05-30 12:42:00 +00:00
Rhys Perry
d7f48a61ec aco: use pass_flags to recover s_delay_alu cycles 
This is simpler and more accurate.

fossil-db (gfx1100):
Totals from 11678 (8.75% of 133428) affected shaders:
Instrs: 25448655 -> 25436028 (-0.05%)
CodeSize: 130364728 -> 130314220 (-0.04%)
Latency: 325247603 -> 325231064 (-0.01%); split: -0.01%, +0.00%
InvThroughput: 45901166 -> 45900022 (-0.00%); split: -0.00%, +0.00%

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Georg Lehmann <dadschoorse@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23213>
 2023-05-30 12:42:00 +00:00