Mesa (staging/21.1): aco: don't create 4 and 5 dword NSA instructions on GFX10

[GitLab Mirror]

Module: Mesa
Branch: staging/21.1
Commit: f74f86547b3cad14899d3eb4afcc755306a46888
URL:    http://cgit.freedesktop.org/mesa/mesa/commit/?id=f74f86547b3cad14899d3eb4afcc755306a46888

Author: Rhys Perry <pendingchaos02 at gmail.com>
Date:   Wed Jun  2 16:30:35 2021 +0100

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 at gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof at gmail.com>
Fixes: c353895c922 ("aco: use non-sequential addressing")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10898>
(cherry picked from commit 903f814b78f3bdb6f330889277ada147070bfd7b)

---

 .pick_status.json                              | 2 +-
 src/amd/compiler/README-ISA.md                 | 5 +++++
 src/amd/compiler/aco_instruction_selection.cpp | 6 +++++-
 3 files changed, 11 insertions(+), 2 deletions(-)

diff --git a/.pick_status.json b/.pick_status.json
index 1e5327ffa67..c58e91393ce 100644
--- a/.pick_status.json
+++ b/.pick_status.json
@@ -1084,7 +1084,7 @@
         "description": "aco: don't create 4 and 5 dword NSA instructions on GFX10",
         "nominated": true,
         "nomination_type": 1,
-        "resolution": 0,
+        "resolution": 1,
         "main_sha": null,
         "because_sha": "c353895c92270c0e2a6e2b849c24d558efae0d5e"
     },
diff --git a/src/amd/compiler/README-ISA.md b/src/amd/compiler/README-ISA.md
index a790522ba4f..cb9d8da0298 100644
--- a/src/amd/compiler/README-ISA.md
+++ b/src/amd/compiler/README-ISA.md
@@ -250,3 +250,8 @@ Only `s_waitcnt_vscnt null, 0`. Needed even if the first instruction is a load.
 ### NSAClauseBug
 
 "MIMG-NSA in a hard clause has unpredictable results on GFX10.1"
+
+### NSAMaxSize5
+
+NSA MIMG instructions should be limited to 3 dwords before GFX10.3 to avoid
+stability issues: https://reviews.llvm.org/D103348
diff --git a/src/amd/compiler/aco_instruction_selection.cpp b/src/amd/compiler/aco_instruction_selection.cpp
index 75ab73528a5..594d195374f 100644
--- a/src/amd/compiler/aco_instruction_selection.cpp
+++ b/src/amd/compiler/aco_instruction_selection.cpp
@@ -5555,7 +5555,11 @@ static MIMG_instruction *emit_mimg(Builder& bld, aco_opcode op,
                                    unsigned wqm_mask=0,
                                    Operand vdata=Operand(v1))
 {
-   if (bld.program->chip_class < GFX10) {
+   /* Limit NSA instructions to 3 dwords on GFX10 to avoid stability issues. */
+   unsigned max_nsa_size = bld.program->chip_class >= GFX10_3 ? 13 : 5;
+   bool use_nsa = bld.program->chip_class >= GFX10 && coords.size() <= max_nsa_size;
+
+   if (!use_nsa) {
       Temp coord = coords[0];
       if (coords.size() > 1) {
          coord = bld.tmp(RegType::vgpr, coords.size());