Skip to content

[MemCpyOpt][profcheck] Set unknown branch weights for certain selects #167597

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

Merged
merged 1 commit into from
Nov 14, 2025
+10 −3
Merged

[MemCpyOpt][profcheck] Set unknown branch weights for certain selects #167597

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
5 changes: 5 additions & 0 deletions llvm/lib/Transforms/Scalar/MemCpyOptimizer.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -47,6 +47,7 @@
#include "llvm/IR/LLVMContext.h"
#include "llvm/IR/Module.h"
#include "llvm/IR/PassManager.h"
#include "llvm/IR/ProfDataUtils.h"
#include "llvm/IR/Type.h"
#include "llvm/IR/User.h"
#include "llvm/IR/Value.h"
Expand Down Expand Up @@ -1366,6 +1367,10 @@ bool MemCpyOptPass::processMemSetMemCpyDependence(MemCpyInst *MemCpy,
Value *SizeDiff = Builder.CreateSub(DestSize, SrcSize);
Value *MemsetLen = Builder.CreateSelect(
Ule, ConstantInt::getNullValue(DestSize->getType()), SizeDiff);
// FIXME (#167968): we could explore estimating the branch_weights based on
// value profiling data about the 2 sizes.
if (auto *SI = dyn_cast<SelectInst>(MemsetLen))
Copy link
Contributor

@teresajohnson teresajohnson Nov 13, 2025

Choose a reason for hiding this comment

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

Should this have a TODO or FIXME? I.e. is there a hope that we would be able to put something other than unknown here?

Copy link
Member Author

@mtrofin mtrofin Nov 13, 2025

Choose a reason for hiding this comment

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

Chatted offline with @xur-llvm, actually we do record a histogram of frequency of various kinds of sizes, so some kind of analysis would be possible. Let's see how many of these selects make it in CodeGenPrepare (we have the DEBUG_TYPE to help with that). My suspicion is that we (1) don't do a good VP propagation for memops in the first place, and (2) some of these memops are synthetic (from intrinsics). Currently the profcheck work is about branches (br, switch, select), but we should get back to this.

setExplicitlyUnknownBranchWeightsIfProfiled(*SI, DEBUG_TYPE);
Instruction *NewMemSet =
Builder.CreateMemSet(Builder.CreatePtrAdd(Dest, SrcSize),
MemSet->getOperand(1), MemsetLen, Alignment);
Expand Down
8 changes: 5 additions & 3 deletions llvm/test/Transforms/MemCpyOpt/memset-memcpy-dbgloc.ll
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -7,14 +7,14 @@ target datalayout = "e-m:o-i64:64-f80:128-n8:16:32:64-S128"
declare void @llvm.memset.p0.i64(ptr nocapture, i8, i64, i1)
declare void @llvm.memcpy.p0.p0.i64(ptr nocapture, ptr nocapture readonly, i64, i1)

define void @test_constant(i64 %src_size, ptr %dst, i64 %dst_size, i8 %c) !dbg !5 {
define void @test_constant(i64 %src_size, ptr %dst, i64 %dst_size, i8 %c) !dbg !5 !prof !14 {
; CHECK-LABEL: define void @test_constant(
; CHECK-SAME: i64 [[SRC_SIZE:%.*]], ptr [[DST:%.*]], i64 [[DST_SIZE:%.*]], i8 [[C:%.*]]) !dbg [[DBG5:![0-9]+]] {
; CHECK-SAME: i64 [[SRC_SIZE:%.*]], ptr [[DST:%.*]], i64 [[DST_SIZE:%.*]], i8 [[C:%.*]]) !dbg [[DBG5:![0-9]+]] !prof {{.*}} {
; CHECK-NEXT: [[NON_ZERO:%.*]] = icmp ne i64 [[SRC_SIZE]], 0
; CHECK-NEXT: call void @llvm.assume(i1 [[NON_ZERO]])
; CHECK-NEXT: [[TMP1:%.*]] = icmp ule i64 [[DST_SIZE]], [[SRC_SIZE]], !dbg [[DBG11:![0-9]+]]
; CHECK-NEXT: [[TMP2:%.*]] = sub i64 [[DST_SIZE]], [[SRC_SIZE]], !dbg [[DBG11]]
; CHECK-NEXT: [[TMP3:%.*]] = select i1 [[TMP1]], i64 0, i64 [[TMP2]], !dbg [[DBG11]]
; CHECK-NEXT: [[TMP3:%.*]] = select i1 [[TMP1]], i64 0, i64 [[TMP2]], !dbg [[DBG11]], !prof [[SELPROF:![0-9]+]]
; CHECK-NEXT: [[TMP4:%.*]] = getelementptr i8, ptr [[DST]], i64 [[SRC_SIZE]], !dbg [[DBG11]]
; CHECK-NEXT: call void @llvm.memset.p0.i64(ptr align 1 [[TMP4]], i8 [[C]], i64 [[TMP3]], i1 false), !dbg [[DBG11]]
; CHECK-NEXT: call void @llvm.memcpy.p0.p0.i64(ptr [[DST]], ptr @C, i64 [[SRC_SIZE]], i1 false), !dbg [[DBG12:![0-9]+]]
Expand All @@ -29,6 +29,7 @@ define void @test_constant(i64 %src_size, ptr %dst, i64 %dst_size, i8 %c) !dbg !

; Validate that the memset is mapped to DILocation for the original memset.
; CHECK: [[DBG11]] = !DILocation(line: 1,
; CHECK: [[SELPROF]] = !{!"unknown", !"memcpyopt"}
; CHECK: [[DBG12]] = !DILocation(line: 2,
; CHECK: [[DBG13]] = !DILocation(line: 3,

Expand All @@ -50,3 +51,4 @@ define void @test_constant(i64 %src_size, ptr %dst, i64 %dst_size, i8 %c) !dbg !
!11 = !DILocation(line: 1, column: 1, scope: !5)
!12 = !DILocation(line: 2, column: 1, scope: !5)
!13 = !DILocation(line: 3, column: 1, scope: !5)
!14 = !{!"function_entry_count", i32 10}
Loading