[llvm-branch-commits] [llvm] [CodeGen][AMDGPU] Improve modelling for COPY scheduling (PR #227154)
via llvm-branch-commits
llvm-branch-commits at lists.llvm.org
Mon Sep 28 20:58:24 PDT 2026
github-actions[bot] wrote:
<!--PREMERGE ADVISOR COMMENT: Windows-->
# :window: Windows x64 Test Results
* 142872 tests passed
* 3383 tests skipped
* 1 test failed
## Failed Tests
(click on a test name to see its output)
### LLVM
<details>
<summary>LLVM.CodeGen/AMDGPU/llvm.amdgcn.cvt.scale.pk32.gfx13.ll</summary>
```
Exit Code: 1
Command Output (stdout):
--
# RUN: at line 2
c:\_work\llvm-project\llvm-project\build\bin\llc.exe -global-isel=0 -mtriple=amdgpu13.10 < C:\_work\llvm-project\llvm-project\llvm\test\CodeGen\AMDGPU\llvm.amdgcn.cvt.scale.pk32.gfx13.ll | c:\_work\llvm-project\llvm-project\build\bin\filecheck.exe -check-prefix=GFX13-SDAG C:\_work\llvm-project\llvm-project\llvm\test\CodeGen\AMDGPU\llvm.amdgcn.cvt.scale.pk32.gfx13.ll
# executed command: 'c:\_work\llvm-project\llvm-project\build\bin\llc.exe' -global-isel=0 -mtriple=amdgpu13.10
# note: command had no output on stdout or stderr
# executed command: 'c:\_work\llvm-project\llvm-project\build\bin\filecheck.exe' -check-prefix=GFX13-SDAG 'C:\_work\llvm-project\llvm-project\llvm\test\CodeGen\AMDGPU\llvm.amdgcn.cvt.scale.pk32.gfx13.ll'
# note: command had no output on stdout or stderr
# RUN: at line 3
c:\_work\llvm-project\llvm-project\build\bin\llc.exe -global-isel=1 -mtriple=amdgpu13.10 < C:\_work\llvm-project\llvm-project\llvm\test\CodeGen\AMDGPU\llvm.amdgcn.cvt.scale.pk32.gfx13.ll | c:\_work\llvm-project\llvm-project\build\bin\filecheck.exe -check-prefix=GFX13-GISEL C:\_work\llvm-project\llvm-project\llvm\test\CodeGen\AMDGPU\llvm.amdgcn.cvt.scale.pk32.gfx13.ll
# executed command: 'c:\_work\llvm-project\llvm-project\build\bin\llc.exe' -global-isel=1 -mtriple=amdgpu13.10
# note: command had no output on stdout or stderr
# executed command: 'c:\_work\llvm-project\llvm-project\build\bin\filecheck.exe' -check-prefix=GFX13-GISEL 'C:\_work\llvm-project\llvm-project\llvm\test\CodeGen\AMDGPU\llvm.amdgcn.cvt.scale.pk32.gfx13.ll'
# .---command stderr------------
# | C:\_work\llvm-project\llvm-project\llvm\test\CodeGen\AMDGPU\llvm.amdgcn.cvt.scale.pk32.gfx13.ll:913:21: error: GFX13-GISEL-NEXT: expected string not found in input
# | ; GFX13-GISEL-NEXT: v_readfirstlane_b32 s0, v2
# | ^
# | <stdin>:1933:62: note: scanning from here
# | v_cvt_scale_pk32_f32_bf6 v[2:33], v[34:39], 0x64 scale_sel:2
# | ^
# | <stdin>:1934:2: note: possible intended match here
# | v_readfirstlane_b32 s4, v2
# | ^
# | C:\_work\llvm-project\llvm-project\llvm\test\CodeGen\AMDGPU\llvm.amdgcn.cvt.scale.pk32.gfx13.ll:1178:21: error: GFX13-GISEL-NEXT: expected string not found in input
# | ; GFX13-GISEL-NEXT: v_readfirstlane_b32 s0, v2
# | ^
# | <stdin>:2406:62: note: scanning from here
# | v_cvt_scale_pk32_f32_fp6 v[2:33], v[34:39], 0x64 scale_sel:2
# | ^
# | <stdin>:2407:2: note: possible intended match here
# | v_readfirstlane_b32 s4, v2
# | ^
# |
# | Input file: <stdin>
# | Check file: C:\_work\llvm-project\llvm-project\llvm\test\CodeGen\AMDGPU\llvm.amdgcn.cvt.scale.pk32.gfx13.ll
# |
# | -dump-input=help explains the following input dump.
# |
# | Input was:
# | <<<<<<
# | .
# | .
# | .
# | 1928: ; %bb.0:
# | 1929: v_dual_mov_b32 v39, s5 :: v_dual_mov_b32 v38, s4
# | 1930: v_dual_mov_b32 v37, s3 :: v_dual_mov_b32 v36, s2
# | 1931: v_dual_mov_b32 v35, s1 :: v_dual_mov_b32 v34, s0
# | 1932: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
# | 1933: v_cvt_scale_pk32_f32_bf6 v[2:33], v[34:39], 0x64 scale_sel:2
# | next:913'0 { search range start (exclusive)
# | next:913'1 error: no match found in search range
# | 1934: v_readfirstlane_b32 s4, v2
# | next:913'2 ? possible intended match
# | 1935: s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(VALU_DEP_3)
# | 1936: v_readfirstlane_b32 s5, v3
# | 1937: v_readfirstlane_b32 s6, v4
# | 1938: s_delay_alu instid0(VALU_DEP_4)
# | 1939: v_readfirstlane_b32 s7, v5
# | .
# | .
# | .
# | 2046: .long 0
# | 2047: .text
# | 2048: .globl test_cvt_scale_pk32_f32_bf6_vs ; -- Begin function test_cvt_scale_pk32_f32_bf6_vs
# | 2049: .p2align 8
# | 2050: .type test_cvt_scale_pk32_f32_bf6_vs, at function
# | 2051: test_cvt_scale_pk32_f32_bf6_vs: ; @test_cvt_scale_pk32_f32_bf6_vs
# | next:913'3 } search range end (exclusive)
# | 2052: ; %bb.0:
# | 2053: v_cvt_scale_pk32_f32_bf6 v[8:39], v[2:7], s0
# | 2054: s_clause 0x7
# | 2055: global_store_b128 v[0:1], v[8:11], off
# | 2056: global_store_b128 v[0:1], v[12:15], off offset:16
# | .
# | .
# | .
# | 2401: ; %bb.0:
# | 2402: v_dual_mov_b32 v39, s5 :: v_dual_mov_b32 v38, s4
# | 2403: v_dual_mov_b32 v37, s3 :: v_dual_mov_b32 v36, s2
# | 2404: v_dual_mov_b32 v35, s1 :: v_dual_mov_b32 v34, s0
# | 2405: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
# | 2406: v_cvt_scale_pk32_f32_fp6 v[2:33], v[34:39], 0x64 scale_sel:2
# | next:1178'0 { search range start (exclusive)
# | next:1178'1 error: no match found in search range
# | 2407: v_readfirstlane_b32 s4, v2
# | next:1178'2 ? possible intended match
# | 2408: s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(VALU_DEP_3)
# | 2409: v_readfirstlane_b32 s5, v3
# | 2410: v_readfirstlane_b32 s6, v4
# | 2411: s_delay_alu instid0(VALU_DEP_4)
# | 2412: v_readfirstlane_b32 s7, v5
# | .
# | .
# | .
# | 2519: .long 0
# | 2520: .text
# | 2521: .globl test_cvt_scale_pk32_f32_fp6_vs ; -- Begin function test_cvt_scale_pk32_f32_fp6_vs
# | 2522: .p2align 8
# | 2523: .type test_cvt_scale_pk32_f32_fp6_vs, at function
# | 2524: test_cvt_scale_pk32_f32_fp6_vs: ; @test_cvt_scale_pk32_f32_fp6_vs
# | next:1178'3 } search range end (exclusive)
# | 2525: ; %bb.0:
# | 2526: v_cvt_scale_pk32_f32_fp6 v[8:39], v[2:7], s0
# | 2527: s_clause 0x7
# | 2528: global_store_b128 v[0:1], v[8:11], off
# | 2529: global_store_b128 v[0:1], v[12:15], off offset:16
# | .
# | .
# | .
# | >>>>>>
# `-----------------------------
# 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/227154
More information about the llvm-branch-commits
mailing list