[llvm] [MachineScheduler] Keep live-in physreg copies ahead of inline asm (PR #219877)

via llvm-commits llvm-commits at lists.llvm.org
Sun Aug 30 22:47:14 PDT 2026


github-actions[bot] wrote:

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

* 140639 tests passed
* 3770 tests skipped
* 1 test failed

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

### LLVM
<details>
<summary>LLVM.CodeGen/AMDGPU/spill-vgpr-block.ll</summary>

```
Exit Code: 1

Command Output (stdout):
--
# RUN: at line 2
c:\_work\llvm-project\llvm-project\build\bin\llc.exe -global-isel=1 -mtriple=amdgpu12.00--amdpal -mattr=+block-vgpr-csr < C:\_work\llvm-project\llvm-project\llvm\test\CodeGen\AMDGPU\spill-vgpr-block.ll | c:\_work\llvm-project\llvm-project\build\bin\filecheck.exe -check-prefixes=CHECK,GISEL C:\_work\llvm-project\llvm-project\llvm\test\CodeGen\AMDGPU\spill-vgpr-block.ll
# executed command: 'c:\_work\llvm-project\llvm-project\build\bin\llc.exe' -global-isel=1 -mtriple=amdgpu12.00--amdpal -mattr=+block-vgpr-csr
# note: command had no output on stdout or stderr
# executed command: 'c:\_work\llvm-project\llvm-project\build\bin\filecheck.exe' -check-prefixes=CHECK,GISEL 'C:\_work\llvm-project\llvm-project\llvm\test\CodeGen\AMDGPU\spill-vgpr-block.ll'
# .---command stderr------------
# | C:\_work\llvm-project\llvm-project\llvm\test\CodeGen\AMDGPU\spill-vgpr-block.ll:53:15: error: GISEL-NEXT: is not on the line after the previous match
# | ; GISEL-NEXT: s_mov_b64 s[10:11], s[6:7]
# |               ^
# | <stdin>:72:2: note: 'next' match was here
# |  s_mov_b64 s[10:11], s[6:7]
# |  ^
# | <stdin>:69:9: note: previous match ended here
# | ; %bb.0:
# |         ^
# | <stdin>:70:1: note: non-matching line after previous match is here
# |  s_load_b32 s12, s[4:5], 0x0
# | ^
# | 
# | Input file: <stdin>
# | Check file: C:\_work\llvm-project\llvm-project\llvm\test\CodeGen\AMDGPU\spill-vgpr-block.ll
# | 
# | -dump-input=help explains the following input dump.
# | 
# | Input was:
# | <<<<<<
# |            .
# |            .
# |            .
# |           64:  .text 
# |           65:  .globl entry_func ; -- Begin function entry_func 
# |           66:  .p2align 8 
# |           67:  .type entry_func, at function 
# |           68: entry_func: ; @entry_func 
# |           69: ; %bb.0: 
# | next:53'0            {   search range start (exclusive)
# |           70:  s_load_b32 s12, s[4:5], 0x0 
# |           71:  v_mov_b32_e32 v31, v0 
# |           72:  s_mov_b64 s[10:11], s[6:7] 
# | next:53'1      !~~~~~~~~~~~~~~~~~~~~~~~~~   error: match on wrong line
# |           73:  s_mov_b64 s[6:7], s[0:1] 
# |           74:  ;;#ASMSTART 
# |           75:  s_nop 
# |           76:  ;;#ASMEND 
# |           77:  s_add_co_u32 s8, s4, 4 
# |            .
# |            .
# |            .
# |          144:  .vgpr_count: 0x98 
# |          145:  .registers: 
# |          146:  '0x2e12 (COMPUTE_PGM_RSRC1)': 0xe00f0012 
# |          147:  '0x2e13 (COMPUTE_PGM_RSRC2)': 0x1391 
# |          148: ... 
# |          149:  .end_amdgpu_pal_metadata 
# | next:53'2                               } search range end (exclusive)
# | >>>>>>
# `-----------------------------
# 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/219877


More information about the llvm-commits mailing list