Skip to content

[X86] For minsize memset/memcpy, use byte or double-word accesses #87003

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom
Merged
merged 3 commits into from
Oct 6, 2024
Merged

[X86] For minsize memset/memcpy, use byte or double-word accesses #87003

merged 3 commits into from
Oct 6, 2024

Conversation

AZero13
Copy link
Contributor

@AZero13 AZero13 commented Mar 28, 2024 •
edited
Loading

repstosb and repstosd are the same size, but stosd is only done for 0 because the process of multiplying the constant so that it is copied across the bytes of the 32-bit number adds extra instructions that cause the size to increase. For 0, repstosb and repstosd are the same size, but stosd is only done for 0 because the process of multiplying the constant so that it is copied across the bytes of the 32-bit number adds extra instructions that cause the size to increase. For 0, we do not need to do that at all.

For memcpy, the same goes, and as a result the minsize check was moved ahead because a jmp to memcpy encoded takes more bytes than repmovsb.

@AZero13 AZero13 changed the title [X86] Use AlwaysInline to determine whether to emit code or bail when… [X86] Use AlwaysInline to determine whether to emit code or bail when inlining is determined to be unprofitable Mar 28, 2024
@llvmbot
Copy link
Member

llvmbot commented Mar 28, 2024

@llvm/pr-subscribers-backend-x86

Author: AtariDreams (AtariDreams)

Changes

Assume true when we get to the getMemset code, as it ought to be profitable by then.

Just like how getMemcpy works.

Full diff: https://github.com/llvm/llvm-project/pull/87003.diff

2 Files Affected:

  • (modified) llvm/lib/Target/X86/X86SelectionDAGInfo.cpp (+5-3)
  • (modified) llvm/test/CodeGen/X86/memset-vs-memset-inline.ll (+4-131)
diff --git a/llvm/lib/Target/X86/X86SelectionDAGInfo.cpp b/llvm/lib/Target/X86/X86SelectionDAGInfo.cpp
index 7c630a2b0da080..50d273e69ada44 100644
--- a/llvm/lib/Target/X86/X86SelectionDAGInfo.cpp
+++ b/llvm/lib/Target/X86/X86SelectionDAGInfo.cpp
@@ -66,8 +66,10 @@ SDValue X86SelectionDAGInfo::EmitTargetCodeForMemset(
   // If not DWORD aligned or size is more than the threshold, call the library.
   // The libc version is likely to be faster for these cases. It can use the
   // address value and run time information about the CPU.
-  if (Alignment < Align(4) || !ConstantSize ||
-      ConstantSize->getZExtValue() > Subtarget.getMaxInlineSizeThreshold()) 
+  if (!ConstantSize ||
+      (!AlwaysInline &&
+       (Alignment < Align(4) ||
+        ConstantSize->getZExtValue() > Subtarget.getMaxInlineSizeThreshold())))
     return SDValue();
 
   uint64_t SizeVal = ConstantSize->getZExtValue();
@@ -142,7 +144,7 @@ SDValue X86SelectionDAGInfo::EmitTargetCodeForMemset(
                       DAG.getNode(ISD::ADD, dl, AddrVT, Dst,
                                   DAG.getConstant(Offset, dl, AddrVT)),
                       Val, DAG.getConstant(BytesLeft, dl, SizeVT), Alignment,
-                      isVolatile, AlwaysInline,
+                      isVolatile, /* AlwaysInline */ true,
                       /* isTailCall */ false, DstPtrInfo.getWithOffset(Offset));
   }
 
diff --git a/llvm/test/CodeGen/X86/memset-vs-memset-inline.ll b/llvm/test/CodeGen/X86/memset-vs-memset-inline.ll
index b8fdd936b43895..16022c6cbb3934 100644
--- a/llvm/test/CodeGen/X86/memset-vs-memset-inline.ll
+++ b/llvm/test/CodeGen/X86/memset-vs-memset-inline.ll
@@ -28,137 +28,10 @@ define void @regular_memset_calls_external_function(ptr %a, i8 %value) nounwind
 define void @inlined_set_doesnt_call_external_function(ptr %a, i8 %value) nounwind {
 ; CHECK-LABEL: inlined_set_doesnt_call_external_function:
 ; CHECK:       # %bb.0:
-; CHECK-NEXT:    movzbl %sil, %ecx
-; CHECK-NEXT:    movabsq $72340172838076673, %rax # imm = 0x101010101010101
-; CHECK-NEXT:    imulq %rcx, %rax
-; CHECK-NEXT:    movq %rax, 1016(%rdi)
-; CHECK-NEXT:    movq %rax, 1008(%rdi)
-; CHECK-NEXT:    movq %rax, 1000(%rdi)
-; CHECK-NEXT:    movq %rax, 992(%rdi)
-; CHECK-NEXT:    movq %rax, 984(%rdi)
-; CHECK-NEXT:    movq %rax, 976(%rdi)
-; CHECK-NEXT:    movq %rax, 968(%rdi)
-; CHECK-NEXT:    movq %rax, 960(%rdi)
-; CHECK-NEXT:    movq %rax, 952(%rdi)
-; CHECK-NEXT:    movq %rax, 944(%rdi)
-; CHECK-NEXT:    movq %rax, 936(%rdi)
-; CHECK-NEXT:    movq %rax, 928(%rdi)
-; CHECK-NEXT:    movq %rax, 920(%rdi)
-; CHECK-NEXT:    movq %rax, 912(%rdi)
-; CHECK-NEXT:    movq %rax, 904(%rdi)
-; CHECK-NEXT:    movq %rax, 896(%rdi)
-; CHECK-NEXT:    movq %rax, 888(%rdi)
-; CHECK-NEXT:    movq %rax, 880(%rdi)
-; CHECK-NEXT:    movq %rax, 872(%rdi)
-; CHECK-NEXT:    movq %rax, 864(%rdi)
-; CHECK-NEXT:    movq %rax, 856(%rdi)
-; CHECK-NEXT:    movq %rax, 848(%rdi)
-; CHECK-NEXT:    movq %rax, 840(%rdi)
-; CHECK-NEXT:    movq %rax, 832(%rdi)
-; CHECK-NEXT:    movq %rax, 824(%rdi)
-; CHECK-NEXT:    movq %rax, 816(%rdi)
-; CHECK-NEXT:    movq %rax, 808(%rdi)
-; CHECK-NEXT:    movq %rax, 800(%rdi)
-; CHECK-NEXT:    movq %rax, 792(%rdi)
-; CHECK-NEXT:    movq %rax, 784(%rdi)
-; CHECK-NEXT:    movq %rax, 776(%rdi)
-; CHECK-NEXT:    movq %rax, 768(%rdi)
-; CHECK-NEXT:    movq %rax, 760(%rdi)
-; CHECK-NEXT:    movq %rax, 752(%rdi)
-; CHECK-NEXT:    movq %rax, 744(%rdi)
-; CHECK-NEXT:    movq %rax, 736(%rdi)
-; CHECK-NEXT:    movq %rax, 728(%rdi)
-; CHECK-NEXT:    movq %rax, 720(%rdi)
-; CHECK-NEXT:    movq %rax, 712(%rdi)
-; CHECK-NEXT:    movq %rax, 704(%rdi)
-; CHECK-NEXT:    movq %rax, 696(%rdi)
-; CHECK-NEXT:    movq %rax, 688(%rdi)
-; CHECK-NEXT:    movq %rax, 680(%rdi)
-; CHECK-NEXT:    movq %rax, 672(%rdi)
-; CHECK-NEXT:    movq %rax, 664(%rdi)
-; CHECK-NEXT:    movq %rax, 656(%rdi)
-; CHECK-NEXT:    movq %rax, 648(%rdi)
-; CHECK-NEXT:    movq %rax, 640(%rdi)
-; CHECK-NEXT:    movq %rax, 632(%rdi)
-; CHECK-NEXT:    movq %rax, 624(%rdi)
-; CHECK-NEXT:    movq %rax, 616(%rdi)
-; CHECK-NEXT:    movq %rax, 608(%rdi)
-; CHECK-NEXT:    movq %rax, 600(%rdi)
-; CHECK-NEXT:    movq %rax, 592(%rdi)
-; CHECK-NEXT:    movq %rax, 584(%rdi)
-; CHECK-NEXT:    movq %rax, 576(%rdi)
-; CHECK-NEXT:    movq %rax, 568(%rdi)
-; CHECK-NEXT:    movq %rax, 560(%rdi)
-; CHECK-NEXT:    movq %rax, 552(%rdi)
-; CHECK-NEXT:    movq %rax, 544(%rdi)
-; CHECK-NEXT:    movq %rax, 536(%rdi)
-; CHECK-NEXT:    movq %rax, 528(%rdi)
-; CHECK-NEXT:    movq %rax, 520(%rdi)
-; CHECK-NEXT:    movq %rax, 512(%rdi)
-; CHECK-NEXT:    movq %rax, 504(%rdi)
-; CHECK-NEXT:    movq %rax, 496(%rdi)
-; CHECK-NEXT:    movq %rax, 488(%rdi)
-; CHECK-NEXT:    movq %rax, 480(%rdi)
-; CHECK-NEXT:    movq %rax, 472(%rdi)
-; CHECK-NEXT:    movq %rax, 464(%rdi)
-; CHECK-NEXT:    movq %rax, 456(%rdi)
-; CHECK-NEXT:    movq %rax, 448(%rdi)
-; CHECK-NEXT:    movq %rax, 440(%rdi)
-; CHECK-NEXT:    movq %rax, 432(%rdi)
-; CHECK-NEXT:    movq %rax, 424(%rdi)
-; CHECK-NEXT:    movq %rax, 416(%rdi)
-; CHECK-NEXT:    movq %rax, 408(%rdi)
-; CHECK-NEXT:    movq %rax, 400(%rdi)
-; CHECK-NEXT:    movq %rax, 392(%rdi)
-; CHECK-NEXT:    movq %rax, 384(%rdi)
-; CHECK-NEXT:    movq %rax, 376(%rdi)
-; CHECK-NEXT:    movq %rax, 368(%rdi)
-; CHECK-NEXT:    movq %rax, 360(%rdi)
-; CHECK-NEXT:    movq %rax, 352(%rdi)
-; CHECK-NEXT:    movq %rax, 344(%rdi)
-; CHECK-NEXT:    movq %rax, 336(%rdi)
-; CHECK-NEXT:    movq %rax, 328(%rdi)
-; CHECK-NEXT:    movq %rax, 320(%rdi)
-; CHECK-NEXT:    movq %rax, 312(%rdi)
-; CHECK-NEXT:    movq %rax, 304(%rdi)
-; CHECK-NEXT:    movq %rax, 296(%rdi)
-; CHECK-NEXT:    movq %rax, 288(%rdi)
-; CHECK-NEXT:    movq %rax, 280(%rdi)
-; CHECK-NEXT:    movq %rax, 272(%rdi)
-; CHECK-NEXT:    movq %rax, 264(%rdi)
-; CHECK-NEXT:    movq %rax, 256(%rdi)
-; CHECK-NEXT:    movq %rax, 248(%rdi)
-; CHECK-NEXT:    movq %rax, 240(%rdi)
-; CHECK-NEXT:    movq %rax, 232(%rdi)
-; CHECK-NEXT:    movq %rax, 224(%rdi)
-; CHECK-NEXT:    movq %rax, 216(%rdi)
-; CHECK-NEXT:    movq %rax, 208(%rdi)
-; CHECK-NEXT:    movq %rax, 200(%rdi)
-; CHECK-NEXT:    movq %rax, 192(%rdi)
-; CHECK-NEXT:    movq %rax, 184(%rdi)
-; CHECK-NEXT:    movq %rax, 176(%rdi)
-; CHECK-NEXT:    movq %rax, 168(%rdi)
-; CHECK-NEXT:    movq %rax, 160(%rdi)
-; CHECK-NEXT:    movq %rax, 152(%rdi)
-; CHECK-NEXT:    movq %rax, 144(%rdi)
-; CHECK-NEXT:    movq %rax, 136(%rdi)
-; CHECK-NEXT:    movq %rax, 128(%rdi)
-; CHECK-NEXT:    movq %rax, 120(%rdi)
-; CHECK-NEXT:    movq %rax, 112(%rdi)
-; CHECK-NEXT:    movq %rax, 104(%rdi)
-; CHECK-NEXT:    movq %rax, 96(%rdi)
-; CHECK-NEXT:    movq %rax, 88(%rdi)
-; CHECK-NEXT:    movq %rax, 80(%rdi)
-; CHECK-NEXT:    movq %rax, 72(%rdi)
-; CHECK-NEXT:    movq %rax, 64(%rdi)
-; CHECK-NEXT:    movq %rax, 56(%rdi)
-; CHECK-NEXT:    movq %rax, 48(%rdi)
-; CHECK-NEXT:    movq %rax, 40(%rdi)
-; CHECK-NEXT:    movq %rax, 32(%rdi)
-; CHECK-NEXT:    movq %rax, 24(%rdi)
-; CHECK-NEXT:    movq %rax, 16(%rdi)
-; CHECK-NEXT:    movq %rax, 8(%rdi)
-; CHECK-NEXT:    movq %rax, (%rdi)
+; CHECK-NEXT:    movl %esi, %eax
+; CHECK-NEXT:    movl $1024, %ecx # imm = 0x400
+; CHECK-NEXT:    # kill: def $al killed $al killed $eax
+; CHECK-NEXT:    rep;stosb %al, %es:(%rdi)
 ; CHECK-NEXT:    retq
   tail call void @llvm.memset.inline.p0.i64(ptr %a, i8 %value, i64 1024, i1 0)
   ret void

@AZero13 AZero13 force-pushed the memset branch 3 times, most recently from d2acc03 to 71657df Compare March 28, 2024 21:19
@AZero13 AZero13 force-pushed the memset branch 2 times, most recently from e88fbe6 to 652ee72 Compare April 5, 2024 17:34
@AZero13 AZero13 changed the title [X86] Use AlwaysInline to determine whether to emit code or bail when inlining is determined to be unprofitable [X86] For minsize, use size for alignment, rather than actual alignment Apr 5, 2024
@AZero13 AZero13 force-pushed the memset branch 6 times, most recently from d6285f9 to 267bb33 Compare April 5, 2024 18:07
@topperc
Copy link
Collaborator

topperc commented Apr 5, 2024

Title should mention "memset". Otherwise its unclear what case you're talking about.

@AZero13 AZero13 force-pushed the memset branch 3 times, most recently from a265b4d to 9bf30fe Compare April 5, 2024 19:07
@AZero13 AZero13 force-pushed the memset branch 5 times, most recently from ccf10bd to 5e36b38 Compare April 6, 2024 13:24
@AZero13 AZero13 requested review from RKSimon and topperc April 6, 2024 13:24
@AZero13 AZero13 changed the title [X86] For inline memset/memcpy with minsize, use stosb or stosd [X86] For inline memset/memcpy with minsize, use byte or double-word accesses Sep 19, 2024
@AZero13 AZero13 changed the title [X86] For inline memset/memcpy with minsize, use byte or double-word accesses [X86] For minsize memset/memcpy, use byte or double-word accesses Sep 19, 2024
@AZero13
Copy link
Contributor Author

AZero13 commented Sep 20, 2024

@phoebewang is this adequate?

@phoebewang
Copy link
Contributor

@phoebewang is this adequate?

Can you split the 2nd commit into one NFC for refactor and one for the change?

AZero13 added 3 commits October 3, 2024 12:18
@AZero13
Refactor to match memset (NFC)
ddcb813 
Refactored the memset and memcpy codegen to share the alignment-determining code.
@AZero13
[X86] Pre-commit test (NFC)
5b2b92e
@AZero13
[X86] For minsize memset/memcpy, use byte or double-word accesses
bca3456 
repstosb and repstosd are the same size, but stosd is only done for 0 because the process of multiplying the constant so that it is copied across the bytes of the 32-bit number adds extra instructions that cause the size to increase. For 0, repstosb and repstosd are the same size, but stosd is only done for 0 because the process of multiplying the constant so that it is copied across the bytes of the 32-bit number adds extra instructions that cause the size to increase. For 0, we do not need to do that at all.

For memcpy, the same goes, and as a result the minsize check was moved ahead because a jmp to memcpy encoded takes more bytes than repmovsb.
@AZero13
Copy link
Contributor Author

AZero13 commented Oct 3, 2024

repstosb and repstosd are the same size, but stosd is only done for 0 because the process of multiplying the constant so that it is copied across the bytes of the 32-bit number adds extra instructions that cause the size to increase. For 0, repstosb and repstosd are the same size, but stosd is only done for 0 because the process of multiplying the constant so that it is copied across the bytes of the 32-bit number adds extra instructions that cause the size to increase. For 0, we do not need to do that at all.

For memcpy, the same goes, and as a result the minsize check was moved ahead because a jmp to memcpy encoded takes more bytes than repmovsb.

Done!

@AZero13
Copy link
Contributor Author

AZero13 commented Oct 3, 2024

@phoebewang @topperc Is this good now?

phoebewang
phoebewang approved these changes Oct 4, 2024
View reviewed changes
Copy link
Contributor

@phoebewang phoebewang left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LGTM.

@AZero13
Copy link
Contributor Author

AZero13 commented Oct 4, 2024

Thank you @phoebewang. Can we please merge?

@phoebewang phoebewang merged commit 6de5305 into llvm:main Oct 6, 2024
8 checks passed
@llvm-ci
Copy link
Collaborator

llvm-ci commented Oct 6, 2024

LLVM Buildbot has detected a new failure on builder sanitizer-aarch64-linux running on sanitizer-buildbot8 while building llvm at step 2 "annotate".

Full details are available at: https://lab.llvm.org/buildbot/#/builders/51/builds/4778

Here is the relevant piece of the build log for the reference 
Step 2 (annotate) failure: 'python ../sanitizer_buildbot/sanitizers/zorg/buildbot/builders/sanitizers/buildbot_selector.py' (failure)
...
[179/183] Generating MSAN_INST_TEST_OBJECTS.msan_test.cpp.aarch64-with-call.o
[180/183] Generating Msan-aarch64-with-call-Test
[181/183] Generating MSAN_INST_TEST_OBJECTS.msan_test.cpp.aarch64.o
[182/183] Generating Msan-aarch64-Test
[182/183] Running compiler_rt regression tests
llvm-lit: /home/b/sanitizer-aarch64-linux/build/llvm-project/llvm/utils/lit/lit/discovery.py:276: warning: input '/home/b/sanitizer-aarch64-linux/build/build_default/runtimes/runtimes-bins/compiler-rt/test/interception/Unit' contained no tests
llvm-lit: /home/b/sanitizer-aarch64-linux/build/llvm-project/llvm/utils/lit/lit/discovery.py:276: warning: input '/home/b/sanitizer-aarch64-linux/build/build_default/runtimes/runtimes-bins/compiler-rt/test/sanitizer_common/Unit' contained no tests
llvm-lit: /home/b/sanitizer-aarch64-linux/build/llvm-project/llvm/utils/lit/lit/main.py:72: note: The test suite configuration requested an individual test timeout of 0 seconds but a timeout of 900 seconds was requested on the command line. Forcing timeout to be 900 seconds.
-- Testing: 2814 of 5568 tests, 48 workers --
Testing:  0.. 10.. 20.. 30.. 40.. 50.. 60.. 70.. 80.. 90.
FAIL: libFuzzer-aarch64-static-libcxx-Linux :: reduce_inputs.test (2744 of 2814)
******************** TEST 'libFuzzer-aarch64-static-libcxx-Linux :: reduce_inputs.test' FAILED ********************
Exit Code: 1

Command Output (stderr):
--
RUN: at line 3: rm -rf /home/b/sanitizer-aarch64-linux/build/build_default/runtimes/runtimes-bins/compiler-rt/test/fuzzer/AARCH64StaticLibcxxLinuxConfig/Output/reduce_inputs.test.tmp/C
+ rm -rf /home/b/sanitizer-aarch64-linux/build/build_default/runtimes/runtimes-bins/compiler-rt/test/fuzzer/AARCH64StaticLibcxxLinuxConfig/Output/reduce_inputs.test.tmp/C
RUN: at line 4: mkdir -p /home/b/sanitizer-aarch64-linux/build/build_default/runtimes/runtimes-bins/compiler-rt/test/fuzzer/AARCH64StaticLibcxxLinuxConfig/Output/reduce_inputs.test.tmp/C
+ mkdir -p /home/b/sanitizer-aarch64-linux/build/build_default/runtimes/runtimes-bins/compiler-rt/test/fuzzer/AARCH64StaticLibcxxLinuxConfig/Output/reduce_inputs.test.tmp/C
RUN: at line 5: /home/b/sanitizer-aarch64-linux/build/build_default/./bin/clang    -Wthread-safety -Wthread-safety-reference -Wthread-safety-beta   --driver-mode=g++ -O2 -gline-tables-only -fsanitize=address,fuzzer -I/home/b/sanitizer-aarch64-linux/build/llvm-project/compiler-rt/lib/fuzzer  -Wthread-safety -Wthread-safety-reference -Wthread-safety-beta  /home/b/sanitizer-aarch64-linux/build/llvm-project/compiler-rt/test/fuzzer/ShrinkControlFlowSimpleTest.cpp -o /home/b/sanitizer-aarch64-linux/build/build_default/runtimes/runtimes-bins/compiler-rt/test/fuzzer/AARCH64StaticLibcxxLinuxConfig/Output/reduce_inputs.test.tmp-ShrinkControlFlowSimpleTest
+ /home/b/sanitizer-aarch64-linux/build/build_default/./bin/clang -Wthread-safety -Wthread-safety-reference -Wthread-safety-beta --driver-mode=g++ -O2 -gline-tables-only -fsanitize=address,fuzzer -I/home/b/sanitizer-aarch64-linux/build/llvm-project/compiler-rt/lib/fuzzer -Wthread-safety -Wthread-safety-reference -Wthread-safety-beta /home/b/sanitizer-aarch64-linux/build/llvm-project/compiler-rt/test/fuzzer/ShrinkControlFlowSimpleTest.cpp -o /home/b/sanitizer-aarch64-linux/build/build_default/runtimes/runtimes-bins/compiler-rt/test/fuzzer/AARCH64StaticLibcxxLinuxConfig/Output/reduce_inputs.test.tmp-ShrinkControlFlowSimpleTest
RUN: at line 6: /home/b/sanitizer-aarch64-linux/build/build_default/./bin/clang    -Wthread-safety -Wthread-safety-reference -Wthread-safety-beta   --driver-mode=g++ -O2 -gline-tables-only -fsanitize=address,fuzzer -I/home/b/sanitizer-aarch64-linux/build/llvm-project/compiler-rt/lib/fuzzer  -Wthread-safety -Wthread-safety-reference -Wthread-safety-beta  /home/b/sanitizer-aarch64-linux/build/llvm-project/compiler-rt/test/fuzzer/ShrinkControlFlowTest.cpp -o /home/b/sanitizer-aarch64-linux/build/build_default/runtimes/runtimes-bins/compiler-rt/test/fuzzer/AARCH64StaticLibcxxLinuxConfig/Output/reduce_inputs.test.tmp-ShrinkControlFlowTest
+ /home/b/sanitizer-aarch64-linux/build/build_default/./bin/clang -Wthread-safety -Wthread-safety-reference -Wthread-safety-beta --driver-mode=g++ -O2 -gline-tables-only -fsanitize=address,fuzzer -I/home/b/sanitizer-aarch64-linux/build/llvm-project/compiler-rt/lib/fuzzer -Wthread-safety -Wthread-safety-reference -Wthread-safety-beta /home/b/sanitizer-aarch64-linux/build/llvm-project/compiler-rt/test/fuzzer/ShrinkControlFlowTest.cpp -o /home/b/sanitizer-aarch64-linux/build/build_default/runtimes/runtimes-bins/compiler-rt/test/fuzzer/AARCH64StaticLibcxxLinuxConfig/Output/reduce_inputs.test.tmp-ShrinkControlFlowTest
RUN: at line 7: /home/b/sanitizer-aarch64-linux/build/build_default/runtimes/runtimes-bins/compiler-rt/test/fuzzer/AARCH64StaticLibcxxLinuxConfig/Output/reduce_inputs.test.tmp-ShrinkControlFlowSimpleTest  -exit_on_item=0eb8e4ed029b774d80f2b66408203801cb982a60   -runs=1000000 /home/b/sanitizer-aarch64-linux/build/build_default/runtimes/runtimes-bins/compiler-rt/test/fuzzer/AARCH64StaticLibcxxLinuxConfig/Output/reduce_inputs.test.tmp/C 2>&1 | FileCheck /home/b/sanitizer-aarch64-linux/build/llvm-project/compiler-rt/test/fuzzer/reduce_inputs.test
+ FileCheck /home/b/sanitizer-aarch64-linux/build/llvm-project/compiler-rt/test/fuzzer/reduce_inputs.test
+ /home/b/sanitizer-aarch64-linux/build/build_default/runtimes/runtimes-bins/compiler-rt/test/fuzzer/AARCH64StaticLibcxxLinuxConfig/Output/reduce_inputs.test.tmp-ShrinkControlFlowSimpleTest -exit_on_item=0eb8e4ed029b774d80f2b66408203801cb982a60 -runs=1000000 /home/b/sanitizer-aarch64-linux/build/build_default/runtimes/runtimes-bins/compiler-rt/test/fuzzer/AARCH64StaticLibcxxLinuxConfig/Output/reduce_inputs.test.tmp/C
RUN: at line 11: /home/b/sanitizer-aarch64-linux/build/build_default/runtimes/runtimes-bins/compiler-rt/test/fuzzer/AARCH64StaticLibcxxLinuxConfig/Output/reduce_inputs.test.tmp-ShrinkControlFlowSimpleTest -runs=0 /home/b/sanitizer-aarch64-linux/build/build_default/runtimes/runtimes-bins/compiler-rt/test/fuzzer/AARCH64StaticLibcxxLinuxConfig/Output/reduce_inputs.test.tmp/C 2>&1 | FileCheck /home/b/sanitizer-aarch64-linux/build/llvm-project/compiler-rt/test/fuzzer/reduce_inputs.test --check-prefix=COUNT
+ /home/b/sanitizer-aarch64-linux/build/build_default/runtimes/runtimes-bins/compiler-rt/test/fuzzer/AARCH64StaticLibcxxLinuxConfig/Output/reduce_inputs.test.tmp-ShrinkControlFlowSimpleTest -runs=0 /home/b/sanitizer-aarch64-linux/build/build_default/runtimes/runtimes-bins/compiler-rt/test/fuzzer/AARCH64StaticLibcxxLinuxConfig/Output/reduce_inputs.test.tmp/C
+ FileCheck /home/b/sanitizer-aarch64-linux/build/llvm-project/compiler-rt/test/fuzzer/reduce_inputs.test --check-prefix=COUNT
/home/b/sanitizer-aarch64-linux/build/llvm-project/compiler-rt/test/fuzzer/reduce_inputs.test:12:8: error: COUNT: expected string not found in input
COUNT: seed corpus: files: 4
       ^
<stdin>:1:1: note: scanning from here
INFO: Running with entropic power schedule (0xFF, 100).
^
<stdin>:7:7: note: possible intended match here
INFO: seed corpus: files: 3 min: 2b max: 3b total: 7b rss: 31Mb
      ^

Input file: <stdin>
Check file: /home/b/sanitizer-aarch64-linux/build/llvm-project/compiler-rt/test/fuzzer/reduce_inputs.test

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

Input was:
<<<<<<
            1: INFO: Running with entropic power schedule (0xFF, 100). 
check:12'0     X~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ error: no match found
Step 11 (test compiler-rt debug) failure: test compiler-rt debug (failure)
...
[179/183] Generating MSAN_INST_TEST_OBJECTS.msan_test.cpp.aarch64-with-call.o
[180/183] Generating Msan-aarch64-with-call-Test
[181/183] Generating MSAN_INST_TEST_OBJECTS.msan_test.cpp.aarch64.o
[182/183] Generating Msan-aarch64-Test
[182/183] Running compiler_rt regression tests
llvm-lit: /home/b/sanitizer-aarch64-linux/build/llvm-project/llvm/utils/lit/lit/discovery.py:276: warning: input '/home/b/sanitizer-aarch64-linux/build/build_default/runtimes/runtimes-bins/compiler-rt/test/interception/Unit' contained no tests
llvm-lit: /home/b/sanitizer-aarch64-linux/build/llvm-project/llvm/utils/lit/lit/discovery.py:276: warning: input '/home/b/sanitizer-aarch64-linux/build/build_default/runtimes/runtimes-bins/compiler-rt/test/sanitizer_common/Unit' contained no tests
llvm-lit: /home/b/sanitizer-aarch64-linux/build/llvm-project/llvm/utils/lit/lit/main.py:72: note: The test suite configuration requested an individual test timeout of 0 seconds but a timeout of 900 seconds was requested on the command line. Forcing timeout to be 900 seconds.
-- Testing: 2814 of 5568 tests, 48 workers --
Testing:  0.. 10.. 20.. 30.. 40.. 50.. 60.. 70.. 80.. 90.
FAIL: libFuzzer-aarch64-static-libcxx-Linux :: reduce_inputs.test (2744 of 2814)
******************** TEST 'libFuzzer-aarch64-static-libcxx-Linux :: reduce_inputs.test' FAILED ********************
Exit Code: 1

Command Output (stderr):
--
RUN: at line 3: rm -rf /home/b/sanitizer-aarch64-linux/build/build_default/runtimes/runtimes-bins/compiler-rt/test/fuzzer/AARCH64StaticLibcxxLinuxConfig/Output/reduce_inputs.test.tmp/C
+ rm -rf /home/b/sanitizer-aarch64-linux/build/build_default/runtimes/runtimes-bins/compiler-rt/test/fuzzer/AARCH64StaticLibcxxLinuxConfig/Output/reduce_inputs.test.tmp/C
RUN: at line 4: mkdir -p /home/b/sanitizer-aarch64-linux/build/build_default/runtimes/runtimes-bins/compiler-rt/test/fuzzer/AARCH64StaticLibcxxLinuxConfig/Output/reduce_inputs.test.tmp/C
+ mkdir -p /home/b/sanitizer-aarch64-linux/build/build_default/runtimes/runtimes-bins/compiler-rt/test/fuzzer/AARCH64StaticLibcxxLinuxConfig/Output/reduce_inputs.test.tmp/C
RUN: at line 5: /home/b/sanitizer-aarch64-linux/build/build_default/./bin/clang    -Wthread-safety -Wthread-safety-reference -Wthread-safety-beta   --driver-mode=g++ -O2 -gline-tables-only -fsanitize=address,fuzzer -I/home/b/sanitizer-aarch64-linux/build/llvm-project/compiler-rt/lib/fuzzer  -Wthread-safety -Wthread-safety-reference -Wthread-safety-beta  /home/b/sanitizer-aarch64-linux/build/llvm-project/compiler-rt/test/fuzzer/ShrinkControlFlowSimpleTest.cpp -o /home/b/sanitizer-aarch64-linux/build/build_default/runtimes/runtimes-bins/compiler-rt/test/fuzzer/AARCH64StaticLibcxxLinuxConfig/Output/reduce_inputs.test.tmp-ShrinkControlFlowSimpleTest
+ /home/b/sanitizer-aarch64-linux/build/build_default/./bin/clang -Wthread-safety -Wthread-safety-reference -Wthread-safety-beta --driver-mode=g++ -O2 -gline-tables-only -fsanitize=address,fuzzer -I/home/b/sanitizer-aarch64-linux/build/llvm-project/compiler-rt/lib/fuzzer -Wthread-safety -Wthread-safety-reference -Wthread-safety-beta /home/b/sanitizer-aarch64-linux/build/llvm-project/compiler-rt/test/fuzzer/ShrinkControlFlowSimpleTest.cpp -o /home/b/sanitizer-aarch64-linux/build/build_default/runtimes/runtimes-bins/compiler-rt/test/fuzzer/AARCH64StaticLibcxxLinuxConfig/Output/reduce_inputs.test.tmp-ShrinkControlFlowSimpleTest
RUN: at line 6: /home/b/sanitizer-aarch64-linux/build/build_default/./bin/clang    -Wthread-safety -Wthread-safety-reference -Wthread-safety-beta   --driver-mode=g++ -O2 -gline-tables-only -fsanitize=address,fuzzer -I/home/b/sanitizer-aarch64-linux/build/llvm-project/compiler-rt/lib/fuzzer  -Wthread-safety -Wthread-safety-reference -Wthread-safety-beta  /home/b/sanitizer-aarch64-linux/build/llvm-project/compiler-rt/test/fuzzer/ShrinkControlFlowTest.cpp -o /home/b/sanitizer-aarch64-linux/build/build_default/runtimes/runtimes-bins/compiler-rt/test/fuzzer/AARCH64StaticLibcxxLinuxConfig/Output/reduce_inputs.test.tmp-ShrinkControlFlowTest
+ /home/b/sanitizer-aarch64-linux/build/build_default/./bin/clang -Wthread-safety -Wthread-safety-reference -Wthread-safety-beta --driver-mode=g++ -O2 -gline-tables-only -fsanitize=address,fuzzer -I/home/b/sanitizer-aarch64-linux/build/llvm-project/compiler-rt/lib/fuzzer -Wthread-safety -Wthread-safety-reference -Wthread-safety-beta /home/b/sanitizer-aarch64-linux/build/llvm-project/compiler-rt/test/fuzzer/ShrinkControlFlowTest.cpp -o /home/b/sanitizer-aarch64-linux/build/build_default/runtimes/runtimes-bins/compiler-rt/test/fuzzer/AARCH64StaticLibcxxLinuxConfig/Output/reduce_inputs.test.tmp-ShrinkControlFlowTest
RUN: at line 7: /home/b/sanitizer-aarch64-linux/build/build_default/runtimes/runtimes-bins/compiler-rt/test/fuzzer/AARCH64StaticLibcxxLinuxConfig/Output/reduce_inputs.test.tmp-ShrinkControlFlowSimpleTest  -exit_on_item=0eb8e4ed029b774d80f2b66408203801cb982a60   -runs=1000000 /home/b/sanitizer-aarch64-linux/build/build_default/runtimes/runtimes-bins/compiler-rt/test/fuzzer/AARCH64StaticLibcxxLinuxConfig/Output/reduce_inputs.test.tmp/C 2>&1 | FileCheck /home/b/sanitizer-aarch64-linux/build/llvm-project/compiler-rt/test/fuzzer/reduce_inputs.test
+ FileCheck /home/b/sanitizer-aarch64-linux/build/llvm-project/compiler-rt/test/fuzzer/reduce_inputs.test
+ /home/b/sanitizer-aarch64-linux/build/build_default/runtimes/runtimes-bins/compiler-rt/test/fuzzer/AARCH64StaticLibcxxLinuxConfig/Output/reduce_inputs.test.tmp-ShrinkControlFlowSimpleTest -exit_on_item=0eb8e4ed029b774d80f2b66408203801cb982a60 -runs=1000000 /home/b/sanitizer-aarch64-linux/build/build_default/runtimes/runtimes-bins/compiler-rt/test/fuzzer/AARCH64StaticLibcxxLinuxConfig/Output/reduce_inputs.test.tmp/C
RUN: at line 11: /home/b/sanitizer-aarch64-linux/build/build_default/runtimes/runtimes-bins/compiler-rt/test/fuzzer/AARCH64StaticLibcxxLinuxConfig/Output/reduce_inputs.test.tmp-ShrinkControlFlowSimpleTest -runs=0 /home/b/sanitizer-aarch64-linux/build/build_default/runtimes/runtimes-bins/compiler-rt/test/fuzzer/AARCH64StaticLibcxxLinuxConfig/Output/reduce_inputs.test.tmp/C 2>&1 | FileCheck /home/b/sanitizer-aarch64-linux/build/llvm-project/compiler-rt/test/fuzzer/reduce_inputs.test --check-prefix=COUNT
+ /home/b/sanitizer-aarch64-linux/build/build_default/runtimes/runtimes-bins/compiler-rt/test/fuzzer/AARCH64StaticLibcxxLinuxConfig/Output/reduce_inputs.test.tmp-ShrinkControlFlowSimpleTest -runs=0 /home/b/sanitizer-aarch64-linux/build/build_default/runtimes/runtimes-bins/compiler-rt/test/fuzzer/AARCH64StaticLibcxxLinuxConfig/Output/reduce_inputs.test.tmp/C
+ FileCheck /home/b/sanitizer-aarch64-linux/build/llvm-project/compiler-rt/test/fuzzer/reduce_inputs.test --check-prefix=COUNT
/home/b/sanitizer-aarch64-linux/build/llvm-project/compiler-rt/test/fuzzer/reduce_inputs.test:12:8: error: COUNT: expected string not found in input
COUNT: seed corpus: files: 4
       ^
<stdin>:1:1: note: scanning from here
INFO: Running with entropic power schedule (0xFF, 100).
^
<stdin>:7:7: note: possible intended match here
INFO: seed corpus: files: 3 min: 2b max: 3b total: 7b rss: 31Mb
      ^

Input file: <stdin>
Check file: /home/b/sanitizer-aarch64-linux/build/llvm-project/compiler-rt/test/fuzzer/reduce_inputs.test

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

Input was:
<<<<<<
            1: INFO: Running with entropic power schedule (0xFF, 100). 
check:12'0     X~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ error: no match found

@AZero13 AZero13 deleted the memset branch October 6, 2024 17:13
Kyvangka1610 added a commit to Kyvangka1610/llvm-project that referenced this pull request Oct 6, 2024
@Kyvangka1610
Merge commit 'FETCH_HEAD'
3943af5 
* commit 'FETCH_HEAD':
  [X86] combineAndLoadToBZHI - don't do an return early return if we fail to match a load
  [X86] replace-load-and-with-bzhi.ll - add commuted test cases to show failure to fold
  [X86] replace-load-and-with-bzhi.ll - cleanup check-prefixes to use X86/X64 for 32/64-bit targets
  [ExecutionEngine] Avoid repeated hash lookups (NFC) (llvm#111275)
  [ByteCode] Avoid repeated hash lookups (NFC) (llvm#111273)
  [StaticAnalyzer] Avoid repeated hash lookups (NFC) (llvm#111272)
  [CodeGen] Avoid repeated hash lookups (NFC) (llvm#111274)
  [RISCV] Simplify fixed-vector-fp.ll run lines. NFC
  [libc++][format][1/3] Adds more benchmarks. (llvm#101803)
  [X86] combineOrXorWithSETCC - avoid duplicate SDLoc/operands code. NFC.
  [X86] convertIntLogicToFPLogic - avoid duplicate SDLoc/operands code. NFC.
  [libc] Clean up some include in `libc`. (llvm#110980)
  [X86] combineBitOpWithPACK - avoid duplicate SDLoc/operands code. NFC.
  [X86] combineBitOpWithMOVMSK - avoid duplicate SDLoc/operands code. NFC.
  [X86] combineBitOpWithShift - avoid duplicate SDLoc/operands code. NFC.
  [x86] combineMul - use computeKnownBits directly to find MUL_IMM constant splat.
  [X86] combineSubABS - avoid duplicate SDLoc. NFC.
  [ValueTypes][RISCV] Add v1bf16 type (llvm#111112)
  [VPlan] Add additional FOR hoisting test.
  [clang-tidy] Create bugprone-bitwise-pointer-cast check (llvm#108083)
  [InstCombine] Canonicalize more geps with constant gep bases and constant offsets. (llvm#110033)
  [LV] Honor uniform-after-vectorization in setVectorizedCallDecision.
  [ELF] Pass Ctx & to Arch/
  [ELF] Pass Ctx & to Arch/
  [libc++] Fix a typo (llvm#111239)
  [X86] For minsize memset/memcpy, use byte or double-word accesses (llvm#87003)
  [RISCV] Unify RVBShift_ri and RVBShiftW_ri with Shift_ri and ShiftW_ri. NFC (llvm#111263)
  Revert "Reapply "[AMDGPU][GlobalISel] Fix load/store of pointer vectors, buffer.*.pN (llvm#110714)" (llvm#111059)"
  [libc] Add missing include to __support/StringUtil/tables/stdc_errors.h. (llvm#111271)
  [libc] remove errno.h includes (llvm#110934)
  [NFC][rtsan] Update docs to include [[clang::blocking]] (llvm#111249)
  [RISCV] Give ZEXT_H_RV32 and ZEXT_H_RV64 R-type format to match PACK. NFC
  [mlir][SPIRV] Fix build (2) (llvm#111265)
  [mlir][SPIRV] Fix build error (llvm#111264)
  [mlir][NFC] Mark type converter in `populate...` functions as `const` (llvm#111250)
  [Basic] Avoid repeated hash lookups (NFC) (llvm#111228)
  [RISCV] Use THShift_ri class instead of RVBShift_ri for TH_TST instruction. NFC
  [VPlan] Only generate first lane for VPPredInstPHI if no others used.
  [ELF] Don't call getPPC64TargetInfo outside Driver. NFC
  [GISel] Don't preserve NSW flag when converting G_MUL of INT_MIN to G_SHL. (llvm#111230)
  [APInt] Slightly simplify APInt::ashrSlowCase. NFC (llvm#111220)
  [Sema] Avoid repeated hash lookups (NFC) (llvm#111227)
  [Affine] Avoid repeated hash lookups (NFC) (llvm#111226)
  [Driver] Avoid repeated hash lookups (NFC) (llvm#111225)
  [clang][test] Remove a broken bytecode test
  [ELF] Pass Ctx &
  [ELF] Pass Ctx & to Relocations

Signed-off-by: kyvangka1610 <[email protected]>
@zmodem
Copy link
Collaborator

zmodem commented Oct 7, 2024

We're hitting an assert after this change:

$ cat /tmp/bug.ll
target triple = "i686-pc-windows-msvc"

declare void @llvm.memset.p0.i32(ptr nocapture writeonly, i8, i32, i1 immarg)

define dso_local void @foo(ptr %0) {
  call void @llvm.memset.p0.i32(ptr noundef nonnull align 4 dereferenceable(90) %0, i8 -1, i32 90, i1 false)
  ret void
}

$ build/bin/llc /tmp/bug.ll
llc: /work/llvm-project/llvm/lib/CodeGen/SelectionDAG/SelectionDAG.cpp:7736: SDValue getMemsetValue(SDValue, EVT, SelectionDAG &, const SDLoc &): Assertion `C->getAPIntValue().getBitWidth() == 8' failed.
PLEASE submit a bug report to https://github.com/llvm/llvm-project/issues/ and include the crash backtrace.
Stack dump:
0.      Program arguments: build/bin/llc /tmp/bug.ll
1.      Running pass 'Function Pass Manager' on module '/tmp/bug.ll'.
2.      Running pass 'X86 DAG->DAG Instruction Selection' on function '@foo'
 #0 0x00005557413cadb8 llvm::sys::PrintStackTrace(llvm::raw_ostream&, int) (build/bin/llc+0x367adb8)
 #1 0x00005557413c896e llvm::sys::RunSignalHandlers() (build/bin/llc+0x367896e)
 #2 0x00005557413cb448 SignalHandler(int) Signals.cpp:0:0
 #3 0x00007f5003c591a0 (/lib/x86_64-linux-gnu/libc.so.6+0x3d1a0)
 #4 0x00007f5003ca70ec __pthread_kill_implementation ./nptl/pthread_kill.c:44:76
 #5 0x00007f5003c59102 gsignal ./signal/../sysdeps/posix/raise.c:27:6
 #6 0x00007f5003c424f2 abort ./stdlib/abort.c:81:7
 #7 0x00007f5003c42415 _nl_load_domain ./intl/loadmsgcat.c:1177:9
 #8 0x00007f5003c51d32 (/lib/x86_64-linux-gnu/libc.so.6+0x35d32)
 #9 0x00005557411b5e69 getMemsetValue(llvm::SDValue, llvm::EVT, llvm::SelectionDAG&, llvm::SDLoc const&) SelectionDAG.cpp:0:0
#10 0x0000555741193faa getMemsetStores(llvm::SelectionDAG&, llvm::SDLoc const&, llvm::SDValue, llvm::SDValue, llvm::SDValue, unsigned long, llvm::Align, bool, bool, llvm::MachinePointerInfo, llvm::AAMDNodes const&) SelectionDAG.cpp:0:0
#11 0x0000555741192f43 llvm::SelectionDAG::getMemset(llvm::SDValue, llvm::SDLoc const&, llvm::SDValue, llvm::SDValue, llvm::SDValue, llvm::Align, bool, bool, llvm::CallInst const*, llvm::MachinePointerInfo, llvm::AAMDNodes const&) (build/bin/llc+0x3442f43)
#12 0x000055573fe2fff6 llvm::X86SelectionDAGInfo::EmitTargetCodeForMemset(llvm::SelectionDAG&, llvm::SDLoc const&, llvm::SDValue, llvm::SDValue, llvm::SDValue, llvm::SDValue, llvm::Align, bool, bool, llvm::MachinePointerInfo) const (build/bin/llc+0x20dfff6)
#13 0x000055574119301d llvm::SelectionDAG::getMemset(llvm::SDValue, llvm::SDLoc const&, llvm::SDValue, llvm::SDValue, llvm::SDValue, llvm::Align, bool, bool, llvm::CallInst const*, llvm::MachinePointerInfo, llvm::AAMDNodes const&) (build/bin/llc+0x344301d)
#14 0x000055574113bb51 llvm::SelectionDAGBuilder::visitIntrinsicCall(llvm::CallInst const&, unsigned int) (build/bin/llc+0x33ebb51)
#15 0x0000555741107770 llvm::SelectionDAGBuilder::visit(llvm::Instruction const&) (build/bin/llc+0x33b7770)
#16 0x00005557411c7715 llvm::SelectionDAGISel::SelectBasicBlock(llvm::ilist_iterator_w_bits<llvm::ilist_detail::node_options<llvm::Instruction, true, false, void, true, llvm::BasicBlock>, false, true>, llvm::ilist_iterator_w_bits<llvm::ilist_detail::node_options<llvm::Instruction, true, false, void, true, llvm::BasicBlock>, false, true>, bool&) (build/bin/llc+0x3477715)
#17 0x00005557411c69cd llvm::SelectionDAGISel::SelectAllBasicBlocks(llvm::Function const&) (build/bin/llc+0x34769cd)
#18 0x00005557411c3e41 llvm::SelectionDAGISel::runOnMachineFunction(llvm::MachineFunction&) (build/bin/llc+0x3473e41)
#19 0x00005557411c1706 llvm::SelectionDAGISelLegacy::runOnMachineFunction(llvm::MachineFunction&) (build/bin/llc+0x3471706)
#20 0x0000555740557da6 llvm::MachineFunctionPass::runOnFunction(llvm::Function&) (build/bin/llc+0x2807da6)
#21 0x0000555740a9ee69 llvm::FPPassManager::runOnFunction(llvm::Function&) (build/bin/llc+0x2d4ee69)
#22 0x0000555740aa7362 llvm::FPPassManager::runOnModule(llvm::Module&) (build/bin/llc+0x2d57362)
#23 0x0000555740a9f93c llvm::legacy::PassManagerImpl::run(llvm::Module&) (build/bin/llc+0x2d4f93c)
#24 0x000055573e8901c7 main (build/bin/llc+0xb401c7)
#25 0x00007f5003c43b8a __libc_start_call_main ./csu/../sysdeps/nptl/libc_start_call_main.h:74:3
#26 0x00007f5003c43c45 call_init ./csu/../csu/libc-start.c:128:20
#27 0x00007f5003c43c45 __libc_start_main ./csu/../csu/libc-start.c:347:5
#28 0x000055573e889ec1 _start (build/bin/llc+0xb39ec1)
Aborted

I'll revert until this gets fixed.

zmodem added a commit that referenced this pull request Oct 7, 2024
@zmodem
Revert "[X86] For minsize memset/memcpy, use byte or double-word acce…
b2784ec 
…sses (#87003)"

This caused assertion failures:

  llvm/lib/CodeGen/SelectionDAG/SelectionDAG.cpp:7736:
  SDValue getMemsetValue(SDValue, EVT, SelectionDAG &, const SDLoc &):
  Assertion `C->getAPIntValue().getBitWidth() == 8' failed.

See comment on the PR for a reproducer.

> repstosb and repstosd are the same size, but stosd is only done for 0
> because the process of multiplying the constant so that it is copied
> across the bytes of the 32-bit number adds extra instructions that cause
> the size to increase. For 0, repstosb and repstosd are the same size,
> but stosd is only done for 0 because the process of multiplying the
> constant so that it is copied across the bytes of the 32-bit number adds
> extra instructions that cause the size to increase. For 0, we do not
> need to do that at all.
>
> For memcpy, the same goes, and as a result the minsize check was moved
> ahead because a jmp to memcpy encoded takes more bytes than repmovsb.

This reverts commit 6de5305.
AZero13 added a commit to AZero13/llvm-project that referenced this pull request Oct 7, 2024
@AZero13
Reapply "[X86] For minsize memset/memcpy, use byte or double-word acc…
6476cf8 
…esses (llvm#87003)"

Restore old Val if bytes are left over.
AZero13 added a commit to AZero13/llvm-project that referenced this pull request Oct 7, 2024
@AZero13
Reapply "[X86] For minsize memset/memcpy, use byte or double-word acc…
29b749e 
…esses (llvm#87003)"

Restore old Val if bytes are left over.
AZero13 added a commit to AZero13/llvm-project that referenced this pull request Oct 7, 2024
@AZero13
Reapply "[X86] For minsize memset/memcpy, use byte or double-word acc…
bc85b96 
…esses (llvm#87003)"

Restore old Val if bytes are left over.
@AZero13
Copy link
Contributor Author

AZero13 commented Oct 8, 2024

@zmodem Addressed! #111393

AZero13 added a commit to AZero13/llvm-project that referenced this pull request Oct 8, 2024
@AZero13
Reapply "[X86] For minsize memset/memcpy, use byte or double-word acc…
8812a54 
…esses (llvm#87003)"

Restore old Val if bytes are left over to prevent an assertion failure.
phoebewang pushed a commit that referenced this pull request Oct 8, 2024
@AZero13
Reapply "[X86] For minsize memset/memcpy, use byte or double-word acc…
adc6a9e 
…esses (#87003)" (#111393)

Restore old Val if bytes are left over to prevent an assertion failure.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@phoebewang phoebewang phoebewang approved these changes

@topperc topperc Awaiting requested review from topperc

@KanRobert KanRobert Awaiting requested review from KanRobert

@goldsteinn goldsteinn Awaiting requested review from goldsteinn

@RKSimon RKSimon Awaiting requested review from RKSimon

Assignees
No one assigned
Labels
backend:X86
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
8 participants
@AZero13 @llvmbot @topperc @phoebewang @llvm-ci @zmodem @RKSimon @goldsteinn