[llvm] [AMDGPU] Enable WMMA256bInsts + Wave32 for gfx1200/gfx1201 + SISchedule fix + TargetParser gfx1200 propagation (PR #202093)

via llvm-commits llvm-commits at lists.llvm.org
Wed Jun 10 12:10:00 PDT 2026


github-actions[bot] wrote:

<!--PREMERGE ADVISOR COMMENT: Windows-->
# :window: Windows x64 Test Results

* 135599 tests passed
* 3394 tests skipped
* 2 tests failed

## Failed Tests
(click on a test name to see its output)

### LLVM
<details>
<summary>LLVM.CodeGen/AMDGPU/llvm.amdgcn.sched.group.barrier.gfx12.ll</summary>

```
Exit Code: 1

Command Output (stdout):
--
# RUN: at line 2
c:\_work\llvm-project\llvm-project\build\bin\llc.exe -mtriple=amdgcn -mcpu=gfx1200 -misched-cluster=0 < C:\_work\llvm-project\llvm-project\llvm\test\CodeGen\AMDGPU\llvm.amdgcn.sched.group.barrier.gfx12.ll | c:\_work\llvm-project\llvm-project\build\bin\filecheck.exe -check-prefix=GCN C:\_work\llvm-project\llvm-project\llvm\test\CodeGen\AMDGPU\llvm.amdgcn.sched.group.barrier.gfx12.ll
# executed command: 'c:\_work\llvm-project\llvm-project\build\bin\llc.exe' -mtriple=amdgcn -mcpu=gfx1200 -misched-cluster=0
# note: command had no output on stdout or stderr
# executed command: 'c:\_work\llvm-project\llvm-project\build\bin\filecheck.exe' -check-prefix=GCN 'C:\_work\llvm-project\llvm-project\llvm\test\CodeGen\AMDGPU\llvm.amdgcn.sched.group.barrier.gfx12.ll'
# .---command stderr------------
# | C:\_work\llvm-project\llvm-project\llvm\test\CodeGen\AMDGPU\llvm.amdgcn.sched.group.barrier.gfx12.ll:12:13: error: GCN-NEXT: expected string not found in input
# | ; GCN-NEXT: v_lshlrev_b32_e32 v0, 4, v0
# |             ^
# | <stdin>:18:33: note: scanning from here
# |  s_load_b64 s[0:1], s[4:5], 0x24
# |                                 ^
# | <stdin>:19:32: note: possible intended match here
# |  v_dual_mov_b32 v33, 0 :: v_dual_lshlrev_b32 v0, 4, v0
# |                                ^
# | 
# | Input file: <stdin>
# | Check file: C:\_work\llvm-project\llvm-project\llvm\test\CodeGen\AMDGPU\llvm.amdgcn.sched.group.barrier.gfx12.ll
# | 
# | -dump-input=help explains the following input dump.
# | 
# | Input was:
# | <<<<<<
# |            .
# |            .
# |            .
# |           13:  .globl test_sched_group_barrier_pipeline_SWMMAC_cluster ; -- Begin function test_sched_group_barrier_pipeline_SWMMAC_cluster 
# |           14:  .p2align 8 
# |           15:  .type test_sched_group_barrier_pipeline_SWMMAC_cluster, at function 
# |           16: test_sched_group_barrier_pipeline_SWMMAC_cluster: ; @test_sched_group_barrier_pipeline_SWMMAC_cluster 
# |           17: ; %bb.0: ; %entry 
# |           18:  s_load_b64 s[0:1], s[4:5], 0x24 
# | next:12'0                                     X error: no match found
# |           19:  v_dual_mov_b32 v33, 0 :: v_dual_lshlrev_b32 v0, 4, v0 
# | next:12'0     ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
# | next:12'1                                    ?                        possible intended match
# |           20:  s_delay_alu instid0(VALU_DEP_1) | instskip(SKIP_1) | instid1(VALU_DEP_1) 
# | next:12'0     ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
# |           21:  v_and_b32_e32 v32, 0x3ff0, v0 
# | next:12'0     ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
# |           22:  s_wait_kmcnt 0x0 
# | next:12'0     ~~~~~~~~~~~~~~~~~~
# |           23:  v_add_nc_u32_e32 v28, s0, v32 
# | next:12'0     ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
# |           24:  v_add_nc_u32_e32 v32, s1, v32 
# | next:12'0     ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
# |            .
# |            .
# |            .
# | >>>>>>
# `-----------------------------
# error: command failed with exit status: 1

--

```
</details>
<details>
<summary>LLVM.tools/llvm-mca/AMDGPU/gfx12-pseudo-scalar-trans.s</summary>

```
Exit Code: 1

Command Output (stdout):
--
# RUN: at line 2
c:\_work\llvm-project\llvm-project\build\bin\llvm-mca.exe -mtriple=amdgcn -mcpu=gfx1200 --timeline --iterations=1 --timeline-max-cycles=0 < C:\_work\llvm-project\llvm-project\llvm\test\tools\llvm-mca\AMDGPU\gfx12-pseudo-scalar-trans.s | c:\_work\llvm-project\llvm-project\build\bin\filecheck.exe C:\_work\llvm-project\llvm-project\llvm\test\tools\llvm-mca\AMDGPU\gfx12-pseudo-scalar-trans.s
# executed command: 'c:\_work\llvm-project\llvm-project\build\bin\llvm-mca.exe' -mtriple=amdgcn -mcpu=gfx1200 --timeline --iterations=1 --timeline-max-cycles=0
# note: command had no output on stdout or stderr
# executed command: 'c:\_work\llvm-project\llvm-project\build\bin\filecheck.exe' 'C:\_work\llvm-project\llvm-project\llvm\test\tools\llvm-mca\AMDGPU\gfx12-pseudo-scalar-trans.s'
# .---command stderr------------
# | C:\_work\llvm-project\llvm-project\llvm\test\tools\llvm-mca\AMDGPU\gfx12-pseudo-scalar-trans.s:59:15: error: CHECK-NEXT: expected string not found in input
# | # CHECK-NEXT: [0] [1] [2] [3] [4] [5] [6] Instructions:
# |               ^
# | <stdin>:48:34: note: scanning from here
# | Resource pressure by instruction:
# |                                  ^
# | <stdin>:49:5: note: possible intended match here
# | [0] [1] [2] [3] [4] [5] [6] [7] Instructions:
# |     ^
# | 
# | Input file: <stdin>
# | Check file: C:\_work\llvm-project\llvm-project\llvm\test\tools\llvm-mca\AMDGPU\gfx12-pseudo-scalar-trans.s
# | 
# | -dump-input=help explains the following input dump.
# | 
# | Input was:
# | <<<<<<
# |            .
# |            .
# |            .
# |           43:  
# |           44: Resource pressure per iteration: 
# |           45: [0] [1] [2] [3] [4] [5] [6] [7]  
# |           46:  - - - 10.00 - 10.00 - -  
# |           47:  
# |           48: Resource pressure by instruction: 
# | next:59'0                                      X error: no match found
# |           49: [0] [1] [2] [3] [4] [5] [6] [7] Instructions: 
# | next:59'0     ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
# | next:59'1         ?                                          possible intended match
# |           50:  - - - 1.00 - 1.00 - - v_s_exp_f32 s0, s0 
# | next:59'0     ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
# |           51:  - - - 1.00 - 1.00 - - v_s_log_f32 s0, s0 
# | next:59'0     ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
# |           52:  - - - 1.00 - 1.00 - - v_s_rcp_f32 s1, s1 
# | next:59'0     ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
# |           53:  - - - 1.00 - 1.00 - - v_s_rsq_f32 s1, s0 
# | next:59'0     ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
# |           54:  - - - 1.00 - 1.00 - - v_s_sqrt_f32 s2, s1 
# | next:59'0     ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
# |            .
# |            .
# |            .
# | >>>>>>
# `-----------------------------
# error: command failed with exit status: 1

--

```
</details>

If these failures are unrelated to your changes (for example tests are broken or flaky at HEAD), please open an issue at https://github.com/llvm/llvm-project/issues and add the `infrastructure` label.

https://github.com/llvm/llvm-project/pull/202093


More information about the llvm-commits mailing list