[llvm] AMDGPU/GlobalISel: Import D16 load patterns and add combines for them (PR #153178)
LLVM Continuous Integration via llvm-commits
llvm-commits at lists.llvm.org
Thu Sep 11 03:30:21 PDT 2025
llvm-ci wrote:
LLVM Buildbot has detected a new failure on builder `llvm-clang-x86_64-expensive-checks-ubuntu` running on `as-builder-4` while building `llvm` at step 7 "test-check-all".
Full details are available at: https://lab.llvm.org/buildbot/#/builders/187/builds/11235
<details>
<summary>Here is the relevant piece of the build log for the reference</summary>
```
Step 7 (test-check-all) failure: Test just built components: check-all completed (failure)
******************** TEST 'LLVM :: CodeGen/AMDGPU/GlobalISel/atomic_load_global.ll' FAILED ********************
Exit Code: 1
Command Output (stdout):
--
# RUN: at line 2
/home/buildbot/worker/as-builder-4/ramdisk/expensive-checks/build/bin/llc -global-isel -new-reg-bank-select -mtriple=amdgcn-amd-amdhsa -mcpu=tahiti < /home/buildbot/worker/as-builder-4/ramdisk/expensive-checks/llvm-project/llvm/test/CodeGen/AMDGPU/GlobalISel/atomic_load_global.ll | /home/buildbot/worker/as-builder-4/ramdisk/expensive-checks/build/bin/FileCheck -check-prefixes=GCN,GFX6 /home/buildbot/worker/as-builder-4/ramdisk/expensive-checks/llvm-project/llvm/test/CodeGen/AMDGPU/GlobalISel/atomic_load_global.ll
# executed command: /home/buildbot/worker/as-builder-4/ramdisk/expensive-checks/build/bin/llc -global-isel -new-reg-bank-select -mtriple=amdgcn-amd-amdhsa -mcpu=tahiti
# executed command: /home/buildbot/worker/as-builder-4/ramdisk/expensive-checks/build/bin/FileCheck -check-prefixes=GCN,GFX6 /home/buildbot/worker/as-builder-4/ramdisk/expensive-checks/llvm-project/llvm/test/CodeGen/AMDGPU/GlobalISel/atomic_load_global.ll
# RUN: at line 3
/home/buildbot/worker/as-builder-4/ramdisk/expensive-checks/build/bin/llc -global-isel -new-reg-bank-select -mtriple=amdgcn-amd-amdhsa -mcpu=kaveri < /home/buildbot/worker/as-builder-4/ramdisk/expensive-checks/llvm-project/llvm/test/CodeGen/AMDGPU/GlobalISel/atomic_load_global.ll | /home/buildbot/worker/as-builder-4/ramdisk/expensive-checks/build/bin/FileCheck -check-prefixes=GCN,GFX7 /home/buildbot/worker/as-builder-4/ramdisk/expensive-checks/llvm-project/llvm/test/CodeGen/AMDGPU/GlobalISel/atomic_load_global.ll
# executed command: /home/buildbot/worker/as-builder-4/ramdisk/expensive-checks/build/bin/llc -global-isel -new-reg-bank-select -mtriple=amdgcn-amd-amdhsa -mcpu=kaveri
# executed command: /home/buildbot/worker/as-builder-4/ramdisk/expensive-checks/build/bin/FileCheck -check-prefixes=GCN,GFX7 /home/buildbot/worker/as-builder-4/ramdisk/expensive-checks/llvm-project/llvm/test/CodeGen/AMDGPU/GlobalISel/atomic_load_global.ll
# RUN: at line 4
/home/buildbot/worker/as-builder-4/ramdisk/expensive-checks/build/bin/llc -global-isel -new-reg-bank-select -mtriple=amdgcn-amd-amdhsa -mcpu=fiji < /home/buildbot/worker/as-builder-4/ramdisk/expensive-checks/llvm-project/llvm/test/CodeGen/AMDGPU/GlobalISel/atomic_load_global.ll | /home/buildbot/worker/as-builder-4/ramdisk/expensive-checks/build/bin/FileCheck -check-prefixes=GCN,GFX8 /home/buildbot/worker/as-builder-4/ramdisk/expensive-checks/llvm-project/llvm/test/CodeGen/AMDGPU/GlobalISel/atomic_load_global.ll
# executed command: /home/buildbot/worker/as-builder-4/ramdisk/expensive-checks/build/bin/llc -global-isel -new-reg-bank-select -mtriple=amdgcn-amd-amdhsa -mcpu=fiji
# executed command: /home/buildbot/worker/as-builder-4/ramdisk/expensive-checks/build/bin/FileCheck -check-prefixes=GCN,GFX8 /home/buildbot/worker/as-builder-4/ramdisk/expensive-checks/llvm-project/llvm/test/CodeGen/AMDGPU/GlobalISel/atomic_load_global.ll
# RUN: at line 5
/home/buildbot/worker/as-builder-4/ramdisk/expensive-checks/build/bin/llc -global-isel -new-reg-bank-select -mtriple=amdgcn-amd-amdhsa -mcpu=gfx900 < /home/buildbot/worker/as-builder-4/ramdisk/expensive-checks/llvm-project/llvm/test/CodeGen/AMDGPU/GlobalISel/atomic_load_global.ll | /home/buildbot/worker/as-builder-4/ramdisk/expensive-checks/build/bin/FileCheck -check-prefixes=GCN,GFX9 /home/buildbot/worker/as-builder-4/ramdisk/expensive-checks/llvm-project/llvm/test/CodeGen/AMDGPU/GlobalISel/atomic_load_global.ll
# executed command: /home/buildbot/worker/as-builder-4/ramdisk/expensive-checks/build/bin/llc -global-isel -new-reg-bank-select -mtriple=amdgcn-amd-amdhsa -mcpu=gfx900
# .---command stderr------------
# |
# | # After AMDGPURegBankCombiner
# | # Machine code for function atomic_load_global_monotonic_i16_d16_hi_vector_insert: IsSSA, TracksLiveness, Legalized, RegBankSelected
# | Function Live Ins: $sgpr4_sgpr5 in %2, $sgpr6_sgpr7 in %3, $sgpr8_sgpr9 in %4, $sgpr10_sgpr11 in %5, $sgpr12 in %6, $sgpr13 in %7, $sgpr14 in %8, $sgpr15 in %9
# |
# | bb.1 (%ir-block.0):
# | liveins: $vgpr0, $vgpr1, $vgpr2
# | %10:vgpr(s32) = COPY $vgpr0
# | %11:vgpr(s32) = COPY $vgpr1
# | %0:vgpr(p1) = G_MERGE_VALUES %10:vgpr(s32), %11:vgpr(s32)
# | %1:vgpr(<2 x s16>) = COPY $vgpr2
# | %19:vgpr(s32) = G_ZEXTLOAD %0:vgpr(p1) :: (load monotonic (s16) from %ir.ptr, addrspace 1)
# | %13:vgpr(<2 x s16>) = G_AMDGPU_LOAD_D16_HI %0:vgpr(p1), %1:vgpr(<2 x s16>) :: (load monotonic (s16) from %ir.ptr, addrspace 1)
# | $vgpr0 = COPY %13:vgpr(<2 x s16>)
# | SI_RETURN implicit $vgpr0
# |
# | # End machine code for function atomic_load_global_monotonic_i16_d16_hi_vector_insert.
# |
# | *** Bad machine code: Extra explicit operand on non-variadic instruction ***
# | - function: atomic_load_global_monotonic_i16_d16_hi_vector_insert
# | - basic block: %bb.1 (0x591483d12918)
# | - instruction: %13:vgpr(<2 x s16>) = G_AMDGPU_LOAD_D16_HI %0:vgpr(p1), %1:vgpr(<2 x s16>) :: (load monotonic (s16) from %ir.ptr, addrspace 1)
# | - operand 2: %1:vgpr
# | LLVM ERROR: Found 1 machine code errors.
# | PLEASE submit a bug report to https://github.com/llvm/llvm-project/issues/ and include the crash backtrace and instructions to reproduce the bug.
# | Stack dump:
# | 0. Program arguments: /home/buildbot/worker/as-builder-4/ramdisk/expensive-checks/build/bin/llc -global-isel -new-reg-bank-select -mtriple=amdgcn-amd-amdhsa -mcpu=gfx900
# | 1. Running pass 'CallGraph Pass Manager' on module '<stdin>'.
# | 2. Running pass 'Verify generated machine code' on function '@atomic_load_global_monotonic_i16_d16_hi_vector_insert'
...
```
</details>
https://github.com/llvm/llvm-project/pull/153178
More information about the llvm-commits
mailing list