[clang] [llvm] Revert "[InstCombine] Infer nuw for gep inbounds from base of object" (PR #120460)

LLVM Continuous Integration via cfe-commits cfe-commits at lists.llvm.org
Wed Dec 18 10:14:52 PST 2024


llvm-ci wrote:

LLVM Buildbot has detected a new failure on builder `llvm-clang-x86_64-sie-ubuntu-fast` running on `sie-linux-worker` while building `clang,llvm` at step 6 "test-build-unified-tree-check-all".

Full details are available at: https://lab.llvm.org/buildbot/#/builders/144/builds/14271

<details>
<summary>Here is the relevant piece of the build log for the reference</summary>

```
Step 6 (test-build-unified-tree-check-all) failure: test (failure)
******************** TEST 'LLVM :: Transforms/PhaseOrdering/X86/excessive-unrolling.ll' FAILED ********************
Exit Code: 1

Command Output (stderr):
--
RUN: at line 2: /home/buildbot/buildbot-root/llvm-clang-x86_64-sie-ubuntu-fast/build/bin/opt -passes='default<O3>' -unroll-runtime -S /home/buildbot/buildbot-root/llvm-clang-x86_64-sie-ubuntu-fast/llvm-project/llvm/test/Transforms/PhaseOrdering/X86/excessive-unrolling.ll | /home/buildbot/buildbot-root/llvm-clang-x86_64-sie-ubuntu-fast/build/bin/FileCheck /home/buildbot/buildbot-root/llvm-clang-x86_64-sie-ubuntu-fast/llvm-project/llvm/test/Transforms/PhaseOrdering/X86/excessive-unrolling.ll
+ /home/buildbot/buildbot-root/llvm-clang-x86_64-sie-ubuntu-fast/build/bin/FileCheck /home/buildbot/buildbot-root/llvm-clang-x86_64-sie-ubuntu-fast/llvm-project/llvm/test/Transforms/PhaseOrdering/X86/excessive-unrolling.ll
+ /home/buildbot/buildbot-root/llvm-clang-x86_64-sie-ubuntu-fast/build/bin/opt '-passes=default<O3>' -unroll-runtime -S /home/buildbot/buildbot-root/llvm-clang-x86_64-sie-ubuntu-fast/llvm-project/llvm/test/Transforms/PhaseOrdering/X86/excessive-unrolling.ll
/home/buildbot/buildbot-root/llvm-clang-x86_64-sie-ubuntu-fast/llvm-project/llvm/test/Transforms/PhaseOrdering/X86/excessive-unrolling.ll:182:15: error: CHECK-NEXT: expected string not found in input
; CHECK-NEXT: [[TMP0:%.*]] = getelementptr inbounds [58 x double], ptr @b, i64 0, i64 [[INDEX]]
              ^
<stdin>:156:67: note: scanning from here
 %index = phi i64 [ 0, %vector.ph ], [ %index.next, %vector.body ]
                                                                  ^
<stdin>:156:67: note: with "INDEX" equal to "%index"
 %index = phi i64 [ 0, %vector.ph ], [ %index.next, %vector.body ]
                                                                  ^
<stdin>:157:2: note: possible intended match here
 %0 = getelementptr inbounds nuw [58 x double], ptr @b, i64 0, i64 %index
 ^

Input file: <stdin>
Check file: /home/buildbot/buildbot-root/llvm-clang-x86_64-sie-ubuntu-fast/llvm-project/llvm/test/Transforms/PhaseOrdering/X86/excessive-unrolling.ll

-dump-input=help explains the following input dump.

Input was:
<<<<<<
             1: ; ModuleID = '/home/buildbot/buildbot-root/llvm-clang-x86_64-sie-ubuntu-fast/llvm-project/llvm/test/Transforms/PhaseOrdering/X86/excessive-unrolling.ll' 
             2: source_filename = "/home/buildbot/buildbot-root/llvm-clang-x86_64-sie-ubuntu-fast/llvm-project/llvm/test/Transforms/PhaseOrdering/X86/excessive-unrolling.ll" 
             3: target datalayout = "e-m:o-p270:32:32-p271:32:32-p272:64:64-i64:64-f80:128-n8:16:32:64-S128" 
             4: target triple = "x86_64-apple-macosx" 
             5:  
             6:  at b = local_unnamed_addr global [58 x double] zeroinitializer, align 16 
             7:  at c = local_unnamed_addr global [58 x double] zeroinitializer, align 16 
             8:  at a = local_unnamed_addr global [58 x double] zeroinitializer, align 16 
             9:  
            10: ; Function Attrs: nofree norecurse nosync nounwind memory(readwrite, argmem: none, inaccessiblemem: none) 
            11: define void  at test_known_trip_count() local_unnamed_addr #0 { 
label:13'0                  ^~~~~~~~~~~~~~~~~~~~~~~
label:13'1                  ^~~~~~~~~~~~~~~~~~~~~~~
            12: entry: 
next:14         ^~~~~~
            13:  %wide.load = load <2 x double>, ptr @b, align 16 
next:15'0        ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
next:15'1        ^~~~~~~~~~                                        captured var "WIDE_LOAD"
            14:  %wide.load3 = load <2 x double>, ptr getelementptr inbounds nuw (i8, ptr @b, i64 16), align 16 
next:16'0        ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
next:16'1        ^~~~~~~~~~~                                                                                     captured var "WIDE_LOAD3"
            15:  %wide.load4 = load <2 x double>, ptr @c, align 16 
...

```

</details>

https://github.com/llvm/llvm-project/pull/120460


More information about the cfe-commits mailing list