shylie/llvm-project
1
0
Fork 0
Code Issues Pull Requests Actions 6 Packages Projects Releases Wiki Activity
llvm-project/llvm/test/Transforms/LoopVectorize/X86/conversion-cost.ll
Florian Hahn 77fb848894
Reapply "[LV] Simplify and unify resume value handling for epilogue vec." (#187504) 
This reverts commit cdaf29f84dd0abbd1f961982799059c92d76625b.

This version skips removeBranchOnConst when vectorizing the epilogue, as
it may trigger folds that remove the resume phi used as resume value
from the epilogue.

This fixes https://github.com/llvm/llvm-project/issues/187323.

Original message:
This patch tries to drastically simplify resume value handling for the
scalar loop when vectorizing the epilogue.

It uses a simpler, uniform approach for updating all resume values in
the scalar loop:

1. Create ResumeForEpilogue recipes for all scalar resume phis in the
main loop (the epilogue plan will have exactly the same scalar resume
phis, in exactly the same order)
2. Update ::execute for ResumeForEpilogue to set the underlying value
when executing. This is not super clean, but allows easy lookup of the
generated IR value when we update the resume phis in the epilogue. Once
we connect the 2 plans together explicitly, this can be removed.
3. Use the list of ResumeForEpilogue VPInstructions from the main loop
to update the resume/bypass values from the epilogue.

This simplifies the code quite a bit, makes it more robust (should fix
https://github.com/llvm/llvm-project/issues/179407) and also fixes a
mis-compile in the existing tests (see change in

llvm/test/Transforms/LoopVectorize/AArch64/partial-reduce-sub-epilogue-vec.ll,
where previously we would incorrectly resume using the start value when
the epilogue iteration check failed)

In some cases, we get simpler code, due to additional CSE, in some cases
the induction end value computations get moved from the epilogue
iteration check to the vector preheader. We could try to sink the
instructions as cleanup, but it is probably not worth the trouble.

Fixes https://github.com/llvm/llvm-project/issues/179407.

PR for recommit https://github.com/llvm/llvm-project/pull/188134
2026-03-23 22:09:40 +00:00

177 lines
11 KiB
LLVM
Raw Blame History

; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
; RUN: opt < %s -passes=loop-vectorize -mtriple=x86_64-apple-macosx10.8.0 -mcpu=corei7-avx -S | FileCheck %s
target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-v64:64:64-v128:128:128-a0:0:64-s0:64:64-f80:128:128-n8:16:32:64-S128"
target triple = "x86_64-apple-macosx10.8.0"
define void @conversion_cost1(i32 %n, ptr nocapture %A, ptr nocapture %B) nounwind uwtable ssp {
; CHECK-LABEL: @conversion_cost1(
; CHECK-NEXT: [[TMP1:%.*]] = icmp sgt i32 [[N:%.*]], 3
; CHECK-NEXT: br i1 [[TMP1]], label [[ITER_CHECK:%.*]], label [[DOT_CRIT_EDGE:%.*]]
; CHECK: iter.check:
; CHECK-NEXT: [[TMP2:%.*]] = add i32 [[N]], -3
; CHECK-NEXT: [[TMP3:%.*]] = zext i32 [[TMP2]] to i64
; CHECK-NEXT: [[MIN_ITERS_CHECK:%.*]] = icmp ult i64 [[TMP3]], 4
; CHECK-NEXT: br i1 [[MIN_ITERS_CHECK]], label [[VEC_EPILOG_SCALAR_PH:%.*]], label [[VECTOR_MAIN_LOOP_ITER_CHECK:%.*]]
; CHECK: vector.main.loop.iter.check:
; CHECK-NEXT: [[MIN_ITERS_CHECK1:%.*]] = icmp ult i64 [[TMP3]], 32
; CHECK-NEXT: br i1 [[MIN_ITERS_CHECK1]], label [[VEC_EPILOG_PH:%.*]], label [[VECTOR_PH:%.*]]
; CHECK: vector.ph:
; CHECK-NEXT: [[N_MOD_VF:%.*]] = urem i64 [[TMP3]], 32
; CHECK-NEXT: [[N_VEC:%.*]] = sub i64 [[TMP3]], [[N_MOD_VF]]
; CHECK-NEXT: [[IND_END:%.*]] = add i64 3, [[N_VEC]]
; CHECK-NEXT: br label [[VECTOR_BODY:%.*]]
; CHECK: vector.body:
; CHECK-NEXT: [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
; CHECK-NEXT: [[VEC_IND:%.*]] = phi <32 x i8> [ <i8 3, i8 4, i8 5, i8 6, i8 7, i8 8, i8 9, i8 10, i8 11, i8 12, i8 13, i8 14, i8 15, i8 16, i8 17, i8 18, i8 19, i8 20, i8 21, i8 22, i8 23, i8 24, i8 25, i8 26, i8 27, i8 28, i8 29, i8 30, i8 31, i8 32, i8 33, i8 34>, [[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], [[VECTOR_BODY]] ]
; CHECK-NEXT: [[OFFSET_IDX:%.*]] = add i64 3, [[INDEX]]
; CHECK-NEXT: [[TMP5:%.*]] = getelementptr inbounds i8, ptr [[A:%.*]], i64 [[OFFSET_IDX]]
; CHECK-NEXT: store <32 x i8> [[VEC_IND]], ptr [[TMP5]], align 1
; CHECK-NEXT: [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 32
; CHECK-NEXT: [[VEC_IND_NEXT]] = add <32 x i8> [[VEC_IND]], splat (i8 32)
; CHECK-NEXT: [[TMP7:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
; CHECK-NEXT: br i1 [[TMP7]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP0:![0-9]+]]
; CHECK: middle.block:
; CHECK-NEXT: [[CMP_N:%.*]] = icmp eq i64 [[TMP3]], [[N_VEC]]
; CHECK-NEXT: br i1 [[CMP_N]], label [[DOT_CRIT_EDGE_LOOPEXIT:%.*]], label [[VEC_EPILOG_ITER_CHECK:%.*]]
; CHECK: vec.epilog.iter.check:
; CHECK-NEXT: [[MIN_EPILOG_ITERS_CHECK:%.*]] = icmp ult i64 [[N_MOD_VF]], 4
; CHECK-NEXT: br i1 [[MIN_EPILOG_ITERS_CHECK]], label [[VEC_EPILOG_SCALAR_PH]], label [[VEC_EPILOG_PH]], !prof [[PROF3:![0-9]+]]
; CHECK: vec.epilog.ph:
; CHECK-NEXT: [[VEC_EPILOG_RESUME_VAL:%.*]] = phi i64 [ [[N_VEC]], [[VEC_EPILOG_ITER_CHECK]] ], [ 0, [[VECTOR_MAIN_LOOP_ITER_CHECK]] ]
; CHECK-NEXT: [[BC_RESUME_VAL:%.*]] = phi i64 [ [[IND_END]], [[VEC_EPILOG_ITER_CHECK]] ], [ 3, [[VECTOR_MAIN_LOOP_ITER_CHECK]] ]
; CHECK-NEXT: [[N_MOD_VF2:%.*]] = urem i64 [[TMP3]], 4
; CHECK-NEXT: [[N_VEC3:%.*]] = sub i64 [[TMP3]], [[N_MOD_VF2]]
; CHECK-NEXT: [[IND_END4:%.*]] = add i64 3, [[N_VEC3]]
; CHECK-NEXT: [[TMP8:%.*]] = trunc i64 [[BC_RESUME_VAL]] to i8
; CHECK-NEXT: [[DOTSPLATINSERT:%.*]] = insertelement <4 x i8> poison, i8 [[TMP8]], i64 0
; CHECK-NEXT: [[DOTSPLAT:%.*]] = shufflevector <4 x i8> [[DOTSPLATINSERT]], <4 x i8> poison, <4 x i32> zeroinitializer
; CHECK-NEXT: [[INDUCTION:%.*]] = add <4 x i8> [[DOTSPLAT]], <i8 0, i8 1, i8 2, i8 3>
; CHECK-NEXT: br label [[VEC_EPILOG_VECTOR_BODY:%.*]]
; CHECK: vec.epilog.vector.body:
; CHECK-NEXT: [[INDEX7:%.*]] = phi i64 [ [[VEC_EPILOG_RESUME_VAL]], [[VEC_EPILOG_PH]] ], [ [[INDEX_NEXT11:%.*]], [[VEC_EPILOG_VECTOR_BODY]] ]
; CHECK-NEXT: [[VEC_IND8:%.*]] = phi <4 x i8> [ [[INDUCTION]], [[VEC_EPILOG_PH]] ], [ [[VEC_IND_NEXT9:%.*]], [[VEC_EPILOG_VECTOR_BODY]] ]
; CHECK-NEXT: [[OFFSET_IDX10:%.*]] = add i64 3, [[INDEX7]]
; CHECK-NEXT: [[TMP10:%.*]] = getelementptr inbounds i8, ptr [[A]], i64 [[OFFSET_IDX10]]
; CHECK-NEXT: store <4 x i8> [[VEC_IND8]], ptr [[TMP10]], align 1
; CHECK-NEXT: [[INDEX_NEXT11]] = add nuw i64 [[INDEX7]], 4
; CHECK-NEXT: [[VEC_IND_NEXT9]] = add <4 x i8> [[VEC_IND8]], splat (i8 4)
; CHECK-NEXT: [[TMP12:%.*]] = icmp eq i64 [[INDEX_NEXT11]], [[N_VEC3]]
; CHECK-NEXT: br i1 [[TMP12]], label [[VEC_EPILOG_MIDDLE_BLOCK:%.*]], label [[VEC_EPILOG_VECTOR_BODY]], !llvm.loop [[LOOP4:![0-9]+]]
; CHECK: vec.epilog.middle.block:
; CHECK-NEXT: [[CMP_N12:%.*]] = icmp eq i64 [[TMP3]], [[N_VEC3]]
; CHECK-NEXT: br i1 [[CMP_N12]], label [[DOT_CRIT_EDGE_LOOPEXIT]], label [[VEC_EPILOG_SCALAR_PH]]
; CHECK: vec.epilog.scalar.ph:
; CHECK-NEXT: [[BC_RESUME_VAL6:%.*]] = phi i64 [ [[IND_END4]], [[VEC_EPILOG_MIDDLE_BLOCK]] ], [ [[IND_END]], [[VEC_EPILOG_ITER_CHECK]] ], [ 3, [[ITER_CHECK]] ]
; CHECK-NEXT: br label [[DOTLR_PH:%.*]]
; CHECK: .lr.ph:
; CHECK-NEXT: [[INDVARS_IV:%.*]] = phi i64 [ [[INDVARS_IV_NEXT:%.*]], [[DOTLR_PH]] ], [ [[BC_RESUME_VAL6]], [[VEC_EPILOG_SCALAR_PH]] ]
; CHECK-NEXT: [[TMP13:%.*]] = trunc i64 [[INDVARS_IV]] to i8
; CHECK-NEXT: [[TMP14:%.*]] = getelementptr inbounds i8, ptr [[A]], i64 [[INDVARS_IV]]
; CHECK-NEXT: store i8 [[TMP13]], ptr [[TMP14]], align 1
; CHECK-NEXT: [[INDVARS_IV_NEXT]] = add i64 [[INDVARS_IV]], 1
; CHECK-NEXT: [[LFTR_WIDEIV:%.*]] = trunc i64 [[INDVARS_IV_NEXT]] to i32
; CHECK-NEXT: [[EXITCOND:%.*]] = icmp eq i32 [[LFTR_WIDEIV]], [[N]]
; CHECK-NEXT: br i1 [[EXITCOND]], label [[DOT_CRIT_EDGE_LOOPEXIT]], label [[DOTLR_PH]], !llvm.loop [[LOOP5:![0-9]+]]
; CHECK: ._crit_edge.loopexit:
; CHECK-NEXT: br label [[DOT_CRIT_EDGE]]
; CHECK: ._crit_edge:
; CHECK-NEXT: ret void
;
%1 = icmp sgt i32 %n, 3
br i1 %1, label %.lr.ph, label %._crit_edge
.lr.ph: ; preds = %0, %.lr.ph
%indvars.iv = phi i64 [ %indvars.iv.next, %.lr.ph ], [ 3, %0 ]
%2 = trunc i64 %indvars.iv to i8
%3 = getelementptr inbounds i8, ptr %A, i64 %indvars.iv
store i8 %2, ptr %3, align 1
%indvars.iv.next = add i64 %indvars.iv, 1
%lftr.wideiv = trunc i64 %indvars.iv.next to i32
%exitcond = icmp eq i32 %lftr.wideiv, %n
br i1 %exitcond, label %._crit_edge, label %.lr.ph
._crit_edge: ; preds = %.lr.ph, %0
ret void
}
define void @conversion_cost2(i32 %n, ptr nocapture %A, ptr nocapture %B) nounwind uwtable ssp {
; CHECK-LABEL: @conversion_cost2(
; CHECK-NEXT: [[TMP1:%.*]] = icmp sgt i32 [[N:%.*]], 9
; CHECK-NEXT: br i1 [[TMP1]], label [[DOTLR_PH_PREHEADER:%.*]], label [[DOT_CRIT_EDGE:%.*]]
; CHECK: .lr.ph.preheader:
; CHECK-NEXT: [[TMP2:%.*]] = add i32 [[N]], -9
; CHECK-NEXT: [[TMP3:%.*]] = zext i32 [[TMP2]] to i64
; CHECK-NEXT: [[MIN_ITERS_CHECK:%.*]] = icmp ult i64 [[TMP3]], 8
; CHECK-NEXT: br i1 [[MIN_ITERS_CHECK]], label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
; CHECK: vector.ph:
; CHECK-NEXT: [[N_MOD_VF:%.*]] = urem i64 [[TMP3]], 8
; CHECK-NEXT: [[N_VEC:%.*]] = sub i64 [[TMP3]], [[N_MOD_VF]]
; CHECK-NEXT: [[IND_END:%.*]] = add i64 9, [[N_VEC]]
; CHECK-NEXT: br label [[VECTOR_BODY:%.*]]
; CHECK: vector.body:
; CHECK-NEXT: [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
; CHECK-NEXT: [[VEC_IND:%.*]] = phi <2 x i64> [ <i64 9, i64 10>, [[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], [[VECTOR_BODY]] ]
; CHECK-NEXT: [[STEP_ADD:%.*]] = add <2 x i64> [[VEC_IND]], splat (i64 2)
; CHECK-NEXT: [[STEP_ADD_2:%.*]] = add <2 x i64> [[STEP_ADD]], splat (i64 2)
; CHECK-NEXT: [[STEP_ADD_3:%.*]] = add <2 x i64> [[STEP_ADD_2]], splat (i64 2)
; CHECK-NEXT: [[OFFSET_IDX:%.*]] = add i64 9, [[INDEX]]
; CHECK-NEXT: [[TMP8:%.*]] = add nsw <2 x i64> [[VEC_IND]], splat (i64 3)
; CHECK-NEXT: [[TMP9:%.*]] = add nsw <2 x i64> [[STEP_ADD]], splat (i64 3)
; CHECK-NEXT: [[TMP10:%.*]] = add nsw <2 x i64> [[STEP_ADD_2]], splat (i64 3)
; CHECK-NEXT: [[TMP11:%.*]] = add nsw <2 x i64> [[STEP_ADD_3]], splat (i64 3)
; CHECK-NEXT: [[TMP12:%.*]] = sitofp <2 x i64> [[TMP8]] to <2 x float>
; CHECK-NEXT: [[TMP18:%.*]] = sitofp <2 x i64> [[TMP9]] to <2 x float>
; CHECK-NEXT: [[TMP19:%.*]] = sitofp <2 x i64> [[TMP10]] to <2 x float>
; CHECK-NEXT: [[TMP20:%.*]] = sitofp <2 x i64> [[TMP11]] to <2 x float>
; CHECK-NEXT: [[TMP13:%.*]] = getelementptr inbounds float, ptr [[B:%.*]], i64 [[OFFSET_IDX]]
; CHECK-NEXT: [[TMP15:%.*]] = getelementptr inbounds float, ptr [[TMP13]], i64 2
; CHECK-NEXT: [[TMP16:%.*]] = getelementptr inbounds float, ptr [[TMP13]], i64 4
; CHECK-NEXT: [[TMP17:%.*]] = getelementptr inbounds float, ptr [[TMP13]], i64 6
; CHECK-NEXT: store <2 x float> [[TMP12]], ptr [[TMP13]], align 4
; CHECK-NEXT: store <2 x float> [[TMP18]], ptr [[TMP15]], align 4
; CHECK-NEXT: store <2 x float> [[TMP19]], ptr [[TMP16]], align 4
; CHECK-NEXT: store <2 x float> [[TMP20]], ptr [[TMP17]], align 4
; CHECK-NEXT: [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 8
; CHECK-NEXT: [[VEC_IND_NEXT]] = add <2 x i64> [[STEP_ADD_3]], splat (i64 2)
; CHECK-NEXT: [[TMP24:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
; CHECK-NEXT: br i1 [[TMP24]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP6:![0-9]+]]
; CHECK: middle.block:
; CHECK-NEXT: [[CMP_N:%.*]] = icmp eq i64 [[TMP3]], [[N_VEC]]
; CHECK-NEXT: br i1 [[CMP_N]], label [[DOT_CRIT_EDGE_LOOPEXIT:%.*]], label [[SCALAR_PH]]
; CHECK: scalar.ph:
; CHECK-NEXT: [[BC_RESUME_VAL:%.*]] = phi i64 [ [[IND_END]], [[MIDDLE_BLOCK]] ], [ 9, [[DOTLR_PH_PREHEADER]] ]
; CHECK-NEXT: br label [[DOTLR_PH:%.*]]
; CHECK: .lr.ph:
; CHECK-NEXT: [[INDVARS_IV:%.*]] = phi i64 [ [[INDVARS_IV_NEXT:%.*]], [[DOTLR_PH]] ], [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ]
; CHECK-NEXT: [[ADD:%.*]] = add nsw i64 [[INDVARS_IV]], 3
; CHECK-NEXT: [[TOFP:%.*]] = sitofp i64 [[ADD]] to float
; CHECK-NEXT: [[GEP:%.*]] = getelementptr inbounds float, ptr [[B]], i64 [[INDVARS_IV]]
; CHECK-NEXT: store float [[TOFP]], ptr [[GEP]], align 4
; CHECK-NEXT: [[INDVARS_IV_NEXT]] = add i64 [[INDVARS_IV]], 1
; CHECK-NEXT: [[LFTR_WIDEIV:%.*]] = trunc i64 [[INDVARS_IV_NEXT]] to i32
; CHECK-NEXT: [[EXITCOND:%.*]] = icmp eq i32 [[LFTR_WIDEIV]], [[N]]
; CHECK-NEXT: br i1 [[EXITCOND]], label [[DOT_CRIT_EDGE_LOOPEXIT]], label [[DOTLR_PH]], !llvm.loop [[LOOP7:![0-9]+]]
; CHECK: ._crit_edge.loopexit:
; CHECK-NEXT: br label [[DOT_CRIT_EDGE]]
; CHECK: ._crit_edge:
; CHECK-NEXT: ret void
;
%1 = icmp sgt i32 %n, 9
br i1 %1, label %.lr.ph, label %._crit_edge
.lr.ph: ; preds = %0, %.lr.ph
%indvars.iv = phi i64 [ %indvars.iv.next, %.lr.ph ], [ 9, %0 ]
%add = add nsw i64 %indvars.iv, 3
%tofp = sitofp i64 %add to float
%gep = getelementptr inbounds float, ptr %B, i64 %indvars.iv
store float %tofp, ptr %gep, align 4
%indvars.iv.next = add i64 %indvars.iv, 1
%lftr.wideiv = trunc i64 %indvars.iv.next to i32
%exitcond = icmp eq i32 %lftr.wideiv, %n
br i1 %exitcond, label %._crit_edge, label %.lr.ph
._crit_edge: ; preds = %.lr.ph, %0
ret void
}
Reference in New Issue View Git Blame Copy Permalink