abdcde9bbc
Currently, SLP vectorizer do not care about loops and their trip count. It may lead to inefficient vectorization in some cases. Patch adds loop nest-aware tree building and cost estimation. When it comes to tree building, it now checks that tree do not span across different loop nests. The nodes from other loop nests are immediate buildvector nodes. The cost model adds the knowledge about loop trip count. If it is unknown, the default value is used, controlled by the -slp-cost-loop-min-trip-count=<value> option. The cost of the vector nodes in the loop is multiplied by the number of iteration (trip count), because each vector node will be executed the trip count number of times. This allows better cost estimation. Original Reviewers: jdenny-ornl, vporpo, hiraditya, RKSimon Original PR: https://github.com/llvm/llvm-project/pull/150450 Recommit after revert in c7bd3062f1dac975cf9b706f457b3c55b4bf57ff and in 4e500bd0015042b0cd4b7c87b81caeea06072d24 Reviewers: Pull Request: https://github.com/llvm/llvm-project/pull/187391
67 lines
3.1 KiB
LLVM
67 lines
3.1 KiB
LLVM
; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
|
|
; RUN: opt < %s -passes=slp-vectorizer -S -mtriple=x86_64-unknown-linux -pass-remarks-output=%t | FileCheck %s
|
|
; RUN: FileCheck --input-file=%t --check-prefix=YAML %s
|
|
|
|
; YAML: --- !Passed
|
|
; YAML-NEXT: Pass: slp-vectorizer
|
|
; YAML-NEXT: Name: VectorizedList
|
|
; YAML-NEXT: Function: test
|
|
; YAML-NEXT: Args:
|
|
; YAML-NEXT: - String: 'SLP vectorized with cost '
|
|
; YAML-NEXT: - Cost: '-38'
|
|
; YAML-NEXT: - String: ' and with tree size '
|
|
; YAML-NEXT: - TreeSize: '7'
|
|
; YAML-NEXT: ...
|
|
;
|
|
define double @test(ptr %a, i32 %n) {
|
|
; CHECK-LABEL: @test(
|
|
; CHECK-NEXT: entry:
|
|
; CHECK-NEXT: [[CMP15:%.*]] = icmp eq i32 [[N:%.*]], 0
|
|
; CHECK-NEXT: br i1 [[CMP15]], label [[FOR_COND_CLEANUP:%.*]], label [[FOR_BODY:%.*]]
|
|
; CHECK: for.cond.cleanup:
|
|
; CHECK-NEXT: [[TMP0:%.*]] = phi <2 x double> [ zeroinitializer, [[ENTRY:%.*]] ], [ [[TMP5:%.*]], [[FOR_BODY]] ]
|
|
; CHECK-NEXT: [[TMP1:%.*]] = extractelement <2 x double> [[TMP0]], i32 0
|
|
; CHECK-NEXT: [[TMP2:%.*]] = extractelement <2 x double> [[TMP0]], i32 1
|
|
; CHECK-NEXT: [[MUL:%.*]] = fmul double [[TMP1]], [[TMP2]]
|
|
; CHECK-NEXT: ret double [[MUL]]
|
|
; CHECK: for.body:
|
|
; CHECK-NEXT: [[I_018:%.*]] = phi i32 [ [[ADD5:%.*]], [[FOR_BODY]] ], [ 0, [[ENTRY]] ]
|
|
; CHECK-NEXT: [[TMP3:%.*]] = phi <2 x double> [ [[TMP5]], [[FOR_BODY]] ], [ zeroinitializer, [[ENTRY]] ]
|
|
; CHECK-NEXT: [[IDXPROM:%.*]] = zext i32 [[I_018]] to i64
|
|
; CHECK-NEXT: [[ARRAYIDX:%.*]] = getelementptr inbounds double, ptr [[A:%.*]], i64 [[IDXPROM]]
|
|
; CHECK-NEXT: [[TMP4:%.*]] = load <2 x double>, ptr [[ARRAYIDX]], align 8
|
|
; CHECK-NEXT: [[TMP5]] = fadd <2 x double> [[TMP3]], [[TMP4]]
|
|
; CHECK-NEXT: [[ADD5]] = add i32 [[I_018]], 2
|
|
; CHECK-NEXT: [[CMP:%.*]] = icmp ult i32 [[ADD5]], [[N]]
|
|
; CHECK-NEXT: br i1 [[CMP]], label [[FOR_BODY]], label [[FOR_COND_CLEANUP]], !prof [[PROF0:![0-9]+]]
|
|
;
|
|
entry:
|
|
%cmp15 = icmp eq i32 %n, 0
|
|
br i1 %cmp15, label %for.cond.cleanup, label %for.body
|
|
|
|
for.cond.cleanup: ; preds = %for.body, %entry
|
|
%x.0.lcssa = phi double [ 0.000000e+00, %entry ], [ %add, %for.body ]
|
|
%y.0.lcssa = phi double [ 0.000000e+00, %entry ], [ %add4, %for.body ]
|
|
%mul = fmul double %x.0.lcssa, %y.0.lcssa
|
|
ret double %mul
|
|
|
|
for.body: ; preds = %entry, %for.body
|
|
%i.018 = phi i32 [ %add5, %for.body ], [ 0, %entry ]
|
|
%y.017 = phi double [ %add4, %for.body ], [ 0.000000e+00, %entry ]
|
|
%x.016 = phi double [ %add, %for.body ], [ 0.000000e+00, %entry ]
|
|
%idxprom = zext i32 %i.018 to i64
|
|
%arrayidx = getelementptr inbounds double, ptr %a, i64 %idxprom
|
|
%0 = load double, ptr %arrayidx, align 8
|
|
%add = fadd double %x.016, %0
|
|
%add1 = or disjoint i32 %i.018, 1
|
|
%idxprom2 = zext i32 %add1 to i64
|
|
%arrayidx3 = getelementptr inbounds double, ptr %a, i64 %idxprom2
|
|
%1 = load double, ptr %arrayidx3, align 8
|
|
%add4 = fadd double %y.017, %1
|
|
%add5 = add i32 %i.018, 2
|
|
%cmp = icmp ult i32 %add5, %n
|
|
br i1 %cmp, label %for.body, label %for.cond.cleanup, !prof !0
|
|
}
|
|
|
|
!0 = !{!"branch_weights", i32 12, i32 1}
|