[llvm-exegesis] Exclude loads/stores from aliasing instruction set #156300
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
In the serial snippet generator and function that computes the aliasing instructions, I don't think we want to include load/store instructions to create a chain as that could make the results more unreliable. There is a hasMemoryOperands() check, but I think that's an X86 way for checking for loads/stores. For AArch64, we should check mayLoad() and mayStore(), and probably for other architectures too.
|
@llvm/pr-subscribers-tools-llvm-exegesis Author: Sjoerd Meijer (sjoerdmeijer) ChangesIn the serial snippet generator and function that computes the aliasing instructions, I don't think we want to include load/store instructions to create a chain as that could make the results more unreliable. There is a hasMemoryOperands() check, but I think that's an X86 way for checking for loads/stores. For AArch64, we should check mayLoad() and mayStore(), and probably for other architectures too. Full diff: https://github.com/llvm/llvm-project/pull/156300.diff 2 Files Affected:
diff --git a/llvm/test/tools/llvm-exegesis/AArch64/no-aliasing-ld-str.s b/llvm/test/tools/llvm-exegesis/AArch64/no-aliasing-ld-str.s
new file mode 100644
index 0000000000000..65e1203bb275d
--- /dev/null
+++ b/llvm/test/tools/llvm-exegesis/AArch64/no-aliasing-ld-str.s
@@ -0,0 +1,8 @@
+REQUIRES: aarch64-registered-target
+
+RUN: llvm-exegesis -mtriple=aarch64 -mcpu=neoverse-v2 -mode=latency --dump-object-to-disk=%d --opcode-name=FMOVWSr --benchmark-phase=assemble-measured-code 2>&1
+RUN: llvm-objdump -d %d > %t.s
+RUN: FileCheck %s < %t.s
+
+CHECK-NOT: ld{{[1-4]}}
+CHECK-NOT: st{{[1-4]}}
diff --git a/llvm/tools/llvm-exegesis/lib/SerialSnippetGenerator.cpp b/llvm/tools/llvm-exegesis/lib/SerialSnippetGenerator.cpp
index bdfc93e22273b..551da570685d2 100644
--- a/llvm/tools/llvm-exegesis/lib/SerialSnippetGenerator.cpp
+++ b/llvm/tools/llvm-exegesis/lib/SerialSnippetGenerator.cpp
@@ -57,6 +57,8 @@ computeAliasingInstructions(const LLVMState &State, const Instruction *Instr,
continue;
if (OtherInstr.hasMemoryOperands())
continue;
+ if (OtherInstr.Description.mayLoad() || OtherInstr.Description.mayStore())
+ continue;
if (!ET.allowAsBackToBack(OtherInstr))
continue;
if (Instr->hasAliasingRegistersThrough(OtherInstr, ForbiddenRegisters))
|
There was a problem hiding this comment.
hasMemoryOperands (inside llvm-exegesis in
09c2839 changed it from just requiring a memory operand to also requiring that the operand by explicit and a register. Not sure why it was changed. I'm thinking updating the condition there is the better option.
Thanks for the suggestion, have move the checks to that place. |
boomanaiden154
approved these changes
Sep 3, 2025
|
LLVM Buildbot has detected a new failure on builder Full details are available at: https://lab.llvm.org/buildbot/#/builders/33/builds/22310 Here is the relevant piece of the build log for the reference |
|
LLVM Buildbot has detected a new failure on builder Full details are available at: https://lab.llvm.org/buildbot/#/builders/11/builds/23173 Here is the relevant piece of the build log for the reference |
|
LLVM Buildbot has detected a new failure on builder Full details are available at: https://lab.llvm.org/buildbot/#/builders/137/builds/24800 Here is the relevant piece of the build log for the reference |
|
LLVM Buildbot has detected a new failure on builder Full details are available at: https://lab.llvm.org/buildbot/#/builders/175/builds/24640 Here is the relevant piece of the build log for the reference |
|
LLVM Buildbot has detected a new failure on builder Full details are available at: https://lab.llvm.org/buildbot/#/builders/185/builds/24650 Here is the relevant piece of the build log for the reference |
|
LLVM Buildbot has detected a new failure on builder Full details are available at: https://lab.llvm.org/buildbot/#/builders/94/builds/10479 Here is the relevant piece of the build log for the reference |
llvm-sync bot
pushed a commit
to arm/arm-toolchain
that referenced
this pull request
Sep 3, 2025
… instruction set" (#156735) Reverts llvm/llvm-project#156300 Need to look at the X86 test failures.
searlmc1
pushed a commit
to ROCm/llvm-project
that referenced
this pull request
Sep 4, 2025
…n set" (llvm#156735) Reverts llvm#156300 Need to look at the X86 test failures.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
In the serial snippet generator and function that computes the aliasing instructions, I don't think we want to include load/store instructions to create a chain as that could make the results more unreliable.
There is a hasMemoryOperands() check, but I think that's an X86 way for checking for loads/stores. For AArch64, we should check mayLoad() and mayStore(), and probably for other architectures too.