AlexIndustrial/mesa
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
140,639 Commits 1 Branch 0 Tags
1f2518ef9fce84f23a2c42e8f76805d9610176d2
Commit Graph

572 Commits

Author SHA1 Message Date
Rhys Perry 1f2518ef9f aco: implement nir_op_extract/nir_op_insert 
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/3151>
 2021-06-08 08:57:43 +00:00
Caio Marcelo de Oliveira Filho c8a7bd0dc8 nir: Rename WORK_GROUP (and similar) to WORKGROUP 
Be consistent with other usages in Vulkan and SPIR-V, and the recently
added workgroup_size field.

Acked-by: Emma Anholt <emma@anholt.net>
Acked-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Acked-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11190>
 2021-06-07 22:34:42 +00:00
Daniel Schürmann dc807dff3e radv,aco: scalarize all phis via nir_lower_phis_to_scalar() 
This allows to remove some ACO code which did so previously.

Totals from 93 (0.06% of 149839) affected shaders (Navi2):
CodeSize: 582424 -> 582348 (-0.01%); split: -0.10%, +0.08%
Instrs: 107083 -> 107011 (-0.07%); split: -0.08%, +0.01%
Latency: 483338 -> 484881 (+0.32%); split: -0.09%, +0.40%
InvThroughput: 101129 -> 101532 (+0.40%); split: -0.03%, +0.42%
Copies: 9893 -> 9774 (-1.20%); split: -1.28%, +0.08%
Branches: 2862 -> 2858 (-0.14%)
PreSGPRs: 3342 -> 3339 (-0.09%)
PreVGPRs: 4567 -> 4565 (-0.04%)

Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11181>
 2021-06-04 16:47:01 +00:00
Rhys Perry 903f814b78 aco: don't create 4 and 5 dword NSA instructions on GFX10 
"stability issues", apparently: https://reviews.llvm.org/D103348

fossil-db (Navi10):
Totals from 4512 (3.01% of 149839) affected shaders:
VGPRs: 221516 -> 223308 (+0.81%); split: -0.07%, +0.88%
CodeSize: 23000080 -> 23070672 (+0.31%); split: -0.08%, +0.39%
MaxWaves: 107718 -> 107496 (-0.21%); split: +0.11%, -0.32%
Instrs: 4321890 -> 4362822 (+0.95%); split: -0.00%, +0.95%
Latency: 71495710 -> 71581476 (+0.12%); split: -0.07%, +0.19%
InvThroughput: 11858568 -> 11938960 (+0.68%); split: -0.00%, +0.68%
VClause: 76575 -> 76585 (+0.01%); split: -0.05%, +0.07%
SClause: 168771 -> 168709 (-0.04%); split: -0.06%, +0.02%
Copies: 182305 -> 221948 (+21.75%); split: -0.00%, +21.75%
PreVGPRs: 194657 -> 195635 (+0.50%); split: -0.00%, +0.50%

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Fixes: c353895c92 ("aco: use non-sequential addressing")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10898>
 2021-06-03 03:49:07 +00:00
Timur Kristóf a93092d0ed aco: Use s_cbranch_vccz/nz in post-RA optimization. 
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>
 2021-05-28 12:14:53 +00:00
Samuel Pitoiset 729ebe4b17 aco: fix emitting discard when the program just ends 
For fragment shaders that only contain a discard, the exec mask has
to be zero'd and everything discarded.

It seems unnecessary to emit an export here because if the FS has no
exports, the compiler already emits a null export at the end.

Fixes incorrect hair rendering in Detroit: Become Human.

fossil-db (Sienna Cichlid):
Totals from 3 (0.00% of 149839) affected shaders:
CodeSize: 2896 -> 2872 (-0.83%)
Instrs: 556 -> 553 (-0.54%)
Latency: 29266 -> 29214 (-0.18%)
InvThroughput: 3374 -> 3372 (-0.06%)

Cc: 21.1 mesa-stable
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10955>
 2021-05-26 10:32:59 +00:00
Daniel Schürmann 32c7d17120 aco: remove condition operand from branch in invert block 
As value numbering only handles logical blocks, this
could lead to invalid IR until insert_exec_mask().
No fossil-db changes.

Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10894>
 2021-05-20 17:44:20 +00:00
Samuel Pitoiset fe2a5716ee aco: fix derivatives/intrinsics with SGPR sources 
ds_swizzle_b32 requires a VGPR and DPP can't encode SGPR sources.

Fixes
dEQP-VK.graphicsfuzz.cov-derivative-uniform-vector-global-loop-count.

Cc: 21.1 mesa-stable
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10840>
 2021-05-20 13:24:31 +00:00
Bas Nieuwenhuizen c7904b5b9b aco: Implement bvh64_intersect_ray_amd intrinsic. 
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10818>
 2021-05-18 23:02:25 +02:00
Bas Nieuwenhuizen bfe2802188 aco: Add load_sbt_amd intrinsic implementation. 
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9767>
 2021-05-18 18:29:36 +00:00
Timur Kristóf bb127c2130 radv: Use new NIR lowering of NGG GS when ACO is used. 
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/10740>
 2021-05-12 13:47:04 +00:00
Timur Kristóf 9732881729 radv: Use new NGG NIR lowering for VS/TES when ACO is used. 
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/10740>
 2021-05-12 13:47:04 +00:00
Timur Kristóf 89a76ff786 aco: Implement new NGG specific NIR intrinsics. 
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/10740>
 2021-05-12 13:47:04 +00:00
Timur Kristóf 75a002f809 aco: Split ngg_emit_sendmsg_gs_alloc_req from the wave0 check. 
This allows us to emit the gs_alloc_req independently of the
wave ID check, which is what the NIR lowering will need.

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/10740>
 2021-05-12 13:47:04 +00:00
Timur Kristóf 00fd087f0a aco: Allow workgroup barrier and shared scope for NGG shaders. 
NGG already needs to use workgroup barriers, but this commit allows
them to come from NIR as opposed to just emitting it in ACO
instruction selection.

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/10740>
 2021-05-12 13:47:04 +00:00
Rhys Perry a54f111831 radv,aco: compact vertex buffer descriptors 
It seems common for there to be holes.

fossil-db (GFX10.3, robustBufferAccess enabled):
Totals from 33791 (23.10% of 146267) affected shaders:
(no statistics changed)

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/7871>
 2021-05-10 12:09:14 +00:00
Rhys Perry 20a0744e22 Revert "radv,aco: don't use MUBUF for multi-channel loads on GFX8 with robustness2" 
This reverts commit a8a6b9fb2fdcb1bea55707fa0c2b8e96f03c6b5b.

This is no longer necessary now that we fixup the size when creating the
descriptors.

Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7871>
 2021-05-10 12:09:14 +00:00
Rhys Perry 157c6b0f33 radv,aco: use per-attribute vertex descriptors for robustness 
We have to use a different num_records for each attribute to correctly
implement robust buffer access.

fossil-db (GFX10.3, robustBufferAccess enabled):
Totals from 60059 (41.06% of 146267) affected shaders:
VGPRs: 2169040 -> 2169024 (-0.00%); split: -0.02%, +0.02%
CodeSize: 79473128 -> 81156016 (+2.12%); split: -0.00%, +2.12%
MaxWaves: 1635360 -> 1635258 (-0.01%); split: +0.00%, -0.01%
Instrs: 15559040 -> 15793205 (+1.51%); split: -0.01%, +1.52%
Latency: 90954792 -> 91308768 (+0.39%); split: -0.30%, +0.69%
InvThroughput: 14937873 -> 14958761 (+0.14%); split: -0.04%, +0.18%
VClause: 444280 -> 412074 (-7.25%); split: -9.22%, +1.97%
SClause: 588545 -> 644141 (+9.45%); split: -0.54%, +9.99%
Copies: 1010395 -> 1011232 (+0.08%); split: -0.44%, +0.53%
Branches: 274279 -> 274282 (+0.00%); split: -0.00%, +0.00%
PreSGPRs: 1431171 -> 1405056 (-1.82%); split: -2.89%, +1.07%
PreVGPRs: 1575253 -> 1575259 (+0.00%)

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/7871>
 2021-05-10 12:09:14 +00:00
Rhys Perry dfa38fa0c7 aco: group loads from the same vertex binding into the same clause 
In the future, we might have vertex attribute loads from the same binding
but with different descriptors. Since they will be loading from the same
buffer, we should continue grouping them into clauses.

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/7871>
 2021-05-10 12:09:14 +00:00
Rhys Perry ee9b744cb5 radv,aco: use nir_address_format_vec2_index_32bit_offset 
The vec2 index helps the compiler make use of SMEM's SOFFSET field when
loading descriptors.

fossil-db (GFX10.3):
Totals from 126326 (86.37% of 146267) affected shaders:
VGPRs: 4898704 -> 4899088 (+0.01%); split: -0.02%, +0.03%
SpillSGPRs: 13490 -> 14404 (+6.78%); split: -1.10%, +7.87%
CodeSize: 306442996 -> 302277700 (-1.36%); split: -1.36%, +0.01%
MaxWaves: 3277108 -> 3276624 (-0.01%); split: +0.01%, -0.02%
Instrs: 58301101 -> 57469370 (-1.43%); split: -1.43%, +0.01%
VClause: 1208270 -> 1199264 (-0.75%); split: -1.02%, +0.28%
SClause: 2517691 -> 2432744 (-3.37%); split: -3.75%, +0.38%
Copies: 3518643 -> 3161097 (-10.16%); split: -10.45%, +0.29%
Branches: 1228383 -> 1228254 (-0.01%); split: -0.12%, +0.11%
PreSGPRs: 3973880 -> 4031099 (+1.44%); split: -0.19%, +1.63%
PreVGPRs: 3831599 -> 3831707 (+0.00%)
Cycles: 1785250712 -> 1778222316 (-0.39%); split: -0.42%, +0.03%
VMEM: 52873776 -> 50663317 (-4.18%); split: +0.18%, -4.36%
SMEM: 8534270 -> 8361666 (-2.02%); split: +1.79%, -3.82%

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/9523>
 2021-04-27 15:56:07 +00:00
Samuel Pitoiset 4c2add8cba aco: adjust NGG if provoking vertex mode is last 
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Tested-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10449>
 2021-04-27 07:31:03 +00:00
James Park 1351fcf3c3 amd: Fix warnings around variable sizes 
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6162>
 2021-04-23 10:37:22 +00:00
Timur Kristóf 74c467d988 aco: Mark VCC clobbered for iadd8 and iadd16 reductions on GFX6-7. 
On GFX6-7, the 8 and 16-bit integer add reductions use the 32-bit v_add
instruction, which clobbers the VCC register.

Cc: mesa-stable
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10346>
 2021-04-22 11:29:49 +00:00
Rhys Perry 0eaa5dfac0 aco: remove image parameter from get_sampler_desc() 
We can just check whether tex_instr is NULL instead.

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10036>
 2021-04-20 17:42:21 +00:00
Rhys Perry 3cbe9894f7 aco: set TRUNC_COORD=0 for nir_texop_tg4 
Fixes black squares in Assassin's Creed: Valhalla and rendering of
FidelityFX-CACAO demo.

fossil-db (sienna cichlid):
Totals from 3052 (2.09% of 146267) affected shaders:
SpillSGPRs: 8437 -> 8646 (+2.48%)
CodeSize: 30993832 -> 31116916 (+0.40%); split: -0.00%, +0.40%
Instrs: 5869934 -> 5886783 (+0.29%); split: -0.00%, +0.29%
Latency: 250330521 -> 250463770 (+0.05%); split: -0.00%, +0.05%
InvThroughput: 59797617 -> 59814584 (+0.03%); split: -0.00%, +0.03%
VClause: 92114 -> 92132 (+0.02%)
SClause: 197373 -> 197338 (-0.02%); split: -0.02%, +0.01%
Copies: 479482 -> 482394 (+0.61%); split: -0.01%, +0.61%
Branches: 219629 -> 219635 (+0.00%)
PreSGPRs: 248970 -> 249366 (+0.16%)

fossil-db (polaris10):
Totals from 3050 (2.06% of 147787) affected shaders:
SGPRs: 282864 -> 282912 (+0.02%); split: -0.01%, +0.02%
VGPRs: 242572 -> 242612 (+0.02%)
SpillSGPRs: 10387 -> 10675 (+2.77%)
CodeSize: 31872460 -> 31996128 (+0.39%)
MaxWaves: 10924 -> 10925 (+0.01%)
Instrs: 6222217 -> 6239072 (+0.27%)
Latency: 317482545 -> 317773685 (+0.09%); split: -0.00%, +0.09%
InvThroughput: 156149624 -> 156242072 (+0.06%); split: -0.00%, +0.06%
VClause: 92295 -> 92254 (-0.04%); split: -0.05%, +0.01%
SClause: 243342 -> 243321 (-0.01%); split: -0.01%, +0.00%
Copies: 678902 -> 681700 (+0.41%); split: -0.00%, +0.41%
Branches: 219698 -> 219703 (+0.00%)
PreSGPRs: 244251 -> 244644 (+0.16%)

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Fixes: 58f25098a0 ("radv: Use TRUNC_COORD on samplers")
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/3110
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10036>
 2021-04-20 17:42:21 +00:00
Samuel Pitoiset 9434675d60 aco: fix opquantize2f16 on GFX6-7 
Make sure to preserve signed zeroes.

Fixes dEQP-VK.spirv_assembly.instruction.compute.opquantize.flush_to_zero
on GFX6 (Pitcairn). Untested on GFX7.

Fixes: 54a09545ec ("aco: optimize a*0.0")
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10319>
 2021-04-19 16:33:37 +00:00
Marek Olšák ec1ddb976a amd/registers: rename IMG_FORMAT to GFX10_FORMAT to disambiguate the meaning 
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10261>
 2021-04-17 02:37:49 +00:00
Timur Kristóf 5dbab03a80 aco: Emit fewer branches for NGG VS/TES with late primitive export. 
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/10106>
 2021-04-14 14:25:10 +00:00
Timur Kristóf af7d5f5b86 aco: Set block_kind_export_end in create_vs/fs_exports. 
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/10106>
 2021-04-14 14:25:10 +00:00
Timur Kristóf 2b312a4fd7 aco: Extract ngg_nogs_export_prim_id to a separate function. 
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/10106>
 2021-04-14 14:25:10 +00:00
Timur Kristóf 231ef14b3d aco: Use s_setprio 3 at the beginning of every VS and TES. 
The user-set priority of shaders matters very little, but we hope
this might still help speed up VS input loads especially.

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/10106>
 2021-04-14 14:25:10 +00:00
Timur Kristóf 4c86c7aa15 aco: Remove useless s_setprio near gs_alloc_req. 
We learned that the gs_alloc_req is not actually when the export
space allocation happens. So it makes no sense to prioritize it.

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/10106>
 2021-04-14 14:25:10 +00:00
Timur Kristóf 75cd43741a aco: Align NGG scratch size to 16 so a single ds_read can always read it. 
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/10155>
 2021-04-14 14:05:24 +00:00
Timur Kristóf c1346e5c22 aco: Optimize workgroup exclusive scan to better avoid bank conflicts. 
Previously, every wave had multiple active lanes read the LDS, and
the data was processed by VALU DPP instructions.

Now, only the first lane reads the LDS in order to avoid bank
conflicts, and the results are processed by SALU.

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/10155>
 2021-04-14 14:05:24 +00:00
Rhys Perry d8f12fd421 aco: fix 16-bit f2{u8,i8} on GFX6/7 
Not really tested.

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Tony Wasserka <tony.wasserka@gmx.de>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10081>
 2021-04-12 16:19:46 +00:00
Rhys Perry d0e15b8c22 aco: fix 16-bit u2f32 
This shouldn't sign-extend.

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Tony Wasserka <tony.wasserka@gmx.de>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10081>
 2021-04-12 16:19:46 +00:00
Samuel Pitoiset 1ad295ed6f radv: allow to force VRS rates on GFX10.3 with RADV_FORCE_VRS 
This allows to force the VRS rates via RADV_FORCE_VRS, the supported
values are 2x2, 1x2 and 2x1. This supports the primitive shading rate
mode for non GUI elements.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7794>
 2021-04-09 14:47:53 +02:00
Rhys Perry 835c5b7ebf aco: fix integer tg4 workaround with unnormalized coordinates 
Same as LLVM from 2abf62d348.

fossil-db (GFX8):
Totals from 15 (0.01% of 147787) affected shaders:
VGPRs: 744 -> 748 (+0.54%)
CodeSize: 100472 -> 100732 (+0.26%)
Instrs: 19995 -> 20059 (+0.32%)
Latency: 1001530 -> 1001859 (+0.03%)
InvThroughput: 378508 -> 378747 (+0.06%)
SClause: 676 -> 675 (-0.15%)
Copies: 1655 -> 1654 (-0.06%)
PreSGPRs: 735 -> 742 (+0.95%)

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/10053>
 2021-04-07 15:21:51 +00:00
Samuel Pitoiset 65bca137bd aco: implement a workaround for the image load DCC hw bug on GFX10.3 
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9919>
 2021-04-05 08:54:55 +00:00
Samuel Pitoiset 3dfb453626 aco: fix get_sampler_desc() for image loads 
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9919>
 2021-04-05 08:54:55 +00:00
Tony Wasserka 8557ac9a12 aco/isel: Add documentation for (u)int64->f16 conversion 
The upper 32 bits are truncated before converting, which still produces
correct results since they never meaningfully contribute to the result.

Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9597>
 2021-03-26 14:39:23 +00:00
Tony Wasserka b5be03f39f aco/isel: Fix large inputs being truncated in int32->f16 conversions 
The previous code produced incorrect results for inputs outside the
range [INT16_MIN, INT16_MAX].

A problematic case is e.g. i2f16 32768, which previously would be
converted to -32768.0 instead of returning the exactly representable
floating point result.

Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9597>
 2021-03-26 14:39:23 +00:00
Tony Wasserka 4ce8e422e3 aco/isel: Add documentation and asserts for convert_int 
This function has evolved to be a generic helper function used throughout
the file, so having those assumptions written down explicitly and document
unsupported edge cases should help prevent incorrect use.

Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9597>
 2021-03-26 14:39:23 +00:00
Tony Wasserka 1e03796fa4 aco/isel: Don't request sign extension when truncating signed integers 
This doesn't change semantics but allows us to reject this potentially
ambiguous configuration in convert_int in a later change.

Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9597>
 2021-03-26 14:39:23 +00:00
Tony Wasserka 3a2b055726 aco/isel: Fix i64/u64->float32 conversion for large inputs 
Previously, inputs such as 0x100000000 would have their upper 32-bits
ignored despite being representable by 32-bit floats.

Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9597>
 2021-03-26 14:39:23 +00:00
Tony Wasserka 436922c84a aco/isel: Don't emit unsupported i16<->f16 conversion opcodes on GFX6/7 
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Fixes: b86305bb57 ("nir/algebraic: collapse conversion opcodes (many patterns)")
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/4357
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9597>
 2021-03-26 14:39:23 +00:00
Rhys Perry e3c283e0bc aco: use -1.0*x and 1.0*|x| for fneg/fabs 
Besides -1.0*x being 1 dword smaller than x^0x80000000, this commit also
improves generated code when the application requires that denormals are
flushed.

Future versions of DXVK will require that 32-bit denormals are flushed.

fossil-db (GFX8):
Totals from 21021 (14.22% of 147787) affected shaders:
SGPRs: 1288960 -> 1288944 (-0.00%); split: -0.01%, +0.01%
VGPRs: 792672 -> 792848 (+0.02%); split: -0.01%, +0.03%
CodeSize: 62439228 -> 62403552 (-0.06%); split: -0.11%, +0.05%
MaxWaves: 136182 -> 136181 (-0.00%); split: +0.00%, -0.00%
Instrs: 12230882 -> 12239927 (+0.07%); split: -0.01%, +0.08%

fossil-db (GFX10.3):
Totals from 20191 (13.80% of 146267) affected shaders:
VGPRs: 799992 -> 800032 (+0.01%)
CodeSize: 59763656 -> 59715484 (-0.08%); split: -0.12%, +0.03%
MaxWaves: 525378 -> 525376 (-0.00%)
Instrs: 11511082 -> 11517419 (+0.06%); split: -0.00%, +0.06%

fossil-db (GFX8, d3d float controls):
Totals from 87160 (58.98% of 147787) affected shaders:
SGPRs: 5395072 -> 5408480 (+0.25%); split: -0.06%, +0.31%
VGPRs: 3596716 -> 3581592 (-0.42%); split: -0.55%, +0.13%
CodeSize: 271347396 -> 266814460 (-1.67%); split: -1.67%, +0.00%
MaxWaves: 539669 -> 540400 (+0.14%); split: +0.15%, -0.02%
Instrs: 53395194 -> 52257505 (-2.13%); split: -2.13%, +0.00%

fossil-db (GFX10.3, d3d float controls):
Totals from 82306 (56.27% of 146267) affected shaders:
VGPRs: 3572312 -> 3558848 (-0.38%); split: -0.44%, +0.06%
CodeSize: 273494748 -> 269648968 (-1.41%); split: -1.41%, +0.00%
MaxWaves: 2007156 -> 2009950 (+0.14%); split: +0.15%, -0.01%
Instrs: 52251568 -> 51356424 (-1.71%); split: -1.71%, +0.00%

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/9079>
 2021-03-24 14:02:41 +00:00
Rhys Perry 27e2f82f17 aco: implement image_deref_samples 
It used to be that this intrinsic was never created and texture
instructions were always used.

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Fixes: 50881d59e6 ("compiler/spirv: fix image sample queries")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9686>
 2021-03-19 10:31:46 +00:00
Timur Kristóf 89c8e22cc6 aco: Fix constant address offset calculation for ds_read2 instructions. 
Cc: mesa-stable
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/9678>
 2021-03-18 10:43:41 +00:00
Rhys Perry 5bc100eb2d aco: use a single instruction for uadd32_sat() on GFX8 
fossil-db (GFX8):
Totals from 8 (0.01% of 147787) affected shaders:
SGPRs: 352 -> 368 (+4.55%)
CodeSize: 49576 -> 48788 (-1.59%)
Instrs: 9487 -> 9318 (-1.78%)
Latency: 49935 -> 49607 (-0.66%)
InvThroughput: 138493 -> 137443 (-0.76%)

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/9598>
 2021-03-17 15:33:34 +00:00