Skip to content

[VPlan] Handle VPWidenCastRecipe without underlying value in EVL transform #120194

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.

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
Show file tree
Hide file tree
Changes from all commits
Commits
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
15 changes: 7 additions & 8 deletions llvm/lib/Transforms/Vectorize/VPlanTransforms.cpp
Original file line number Diff line number Diff line change
Expand Up @@ -1505,24 +1505,23 @@ static void transformRecipestoEVLRecipes(VPlan &Plan, VPValue &EVL) {
CInst->getDebugLoc());
})
.Case<VPWidenCastRecipe>(
[&](VPWidenCastRecipe *CInst) -> VPRecipeBase * {
auto *CI = dyn_cast<CastInst>(CInst->getUnderlyingInstr());
[&](VPWidenCastRecipe *CastR) -> VPRecipeBase * {
Intrinsic::ID VPID =
VPIntrinsic::getForOpcode(CI->getOpcode());
VPIntrinsic::getForOpcode(CastR->getOpcode());
assert(VPID != Intrinsic::not_intrinsic &&
"Expected vp.casts Instrinsic");

SmallVector<VPValue *> Ops(CInst->operands());
SmallVector<VPValue *> Ops(CastR->operands());
assert(VPIntrinsic::getMaskParamPos(VPID) &&
VPIntrinsic::getVectorLengthParamPos(VPID) &&
"Expected VP intrinsic");
VPValue *Mask = Plan.getOrAddLiveIn(ConstantInt::getTrue(
IntegerType::getInt1Ty(CI->getContext())));
VPValue *Mask =
Plan.getOrAddLiveIn(ConstantInt::getTrue(Ctx));
Ops.push_back(Mask);
Ops.push_back(&EVL);
return new VPWidenIntrinsicRecipe(
VPID, Ops, TypeInfo.inferScalarType(CInst),
CInst->getDebugLoc());
VPID, Ops, TypeInfo.inferScalarType(CastR),
CastR->getDebugLoc());
})
.Case<VPWidenSelectRecipe>([&](VPWidenSelectRecipe *Sel) {
SmallVector<VPValue *> Ops(Sel->operands());
Expand Down
Original file line number Diff line number Diff line change
@@ -0,0 +1,84 @@
; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --version 5
; RUN: opt -passes=loop-vectorize -force-tail-folding-style=data-with-evl -prefer-predicate-over-epilogue=predicate-dont-vectorize -mtriple=riscv64 -mattr=+v -S %s | FileCheck %s

; Make sure we don't crash when transforming a VPWidenCastRecipe created without
; an underlying value to an EVL recipe. This occurs in this test via
; VPlanTransforms::truncateToMinimalBitwidths

define void @truncate_to_minimal_bitwidths_widen_cast_recipe(ptr %src) {
; CHECK-LABEL: define void @truncate_to_minimal_bitwidths_widen_cast_recipe(
; CHECK-SAME: ptr [[SRC:%.*]]) #[[ATTR0:[0-9]+]] {
; CHECK-NEXT: [[ENTRY:.*]]:
; CHECK-NEXT: br i1 false, label %[[SCALAR_PH:.*]], label %[[VECTOR_PH:.*]]
; CHECK: [[VECTOR_PH]]:
; CHECK-NEXT: [[TMP0:%.*]] = call i64 @llvm.vscale.i64()
; CHECK-NEXT: [[TMP1:%.*]] = sub i64 [[TMP0]], 1
; CHECK-NEXT: [[N_RND_UP:%.*]] = add i64 2, [[TMP1]]
; CHECK-NEXT: [[N_MOD_VF:%.*]] = urem i64 [[N_RND_UP]], [[TMP0]]
; CHECK-NEXT: [[N_VEC:%.*]] = sub i64 [[N_RND_UP]], [[N_MOD_VF]]
; CHECK-NEXT: [[TMP2:%.*]] = call i64 @llvm.vscale.i64()
; CHECK-NEXT: br label %[[VECTOR_BODY:.*]]
; CHECK: [[VECTOR_BODY]]:
; CHECK-NEXT: [[INDEX:%.*]] = phi i64 [ 0, %[[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], %[[VECTOR_BODY]] ]
; CHECK-NEXT: [[EVL_BASED_IV:%.*]] = phi i64 [ 0, %[[VECTOR_PH]] ], [ [[INDEX_EVL_NEXT:%.*]], %[[VECTOR_BODY]] ]
; CHECK-NEXT: [[AVL:%.*]] = sub i64 2, [[EVL_BASED_IV]]
; CHECK-NEXT: [[TMP3:%.*]] = call i32 @llvm.experimental.get.vector.length.i64(i64 [[AVL]], i32 1, i1 true)
; CHECK-NEXT: [[TMP4:%.*]] = add i64 [[EVL_BASED_IV]], 0
; CHECK-NEXT: [[TMP5:%.*]] = getelementptr i8, ptr [[SRC]], i64 [[TMP4]]
; CHECK-NEXT: [[TMP6:%.*]] = getelementptr i8, ptr [[TMP5]], i32 0
; CHECK-NEXT: [[VP_OP_LOAD:%.*]] = call <vscale x 1 x i8> @llvm.vp.load.nxv1i8.p0(ptr align 1 [[TMP6]], <vscale x 1 x i1> splat (i1 true), i32 [[TMP3]])
; CHECK-NEXT: [[TMP7:%.*]] = call <vscale x 1 x i16> @llvm.vp.zext.nxv1i16.nxv1i8(<vscale x 1 x i8> [[VP_OP_LOAD]], <vscale x 1 x i1> splat (i1 true), i32 [[TMP3]])
; CHECK-NEXT: [[VP_OP:%.*]] = call <vscale x 1 x i16> @llvm.vp.mul.nxv1i16(<vscale x 1 x i16> zeroinitializer, <vscale x 1 x i16> [[TMP7]], <vscale x 1 x i1> splat (i1 true), i32 [[TMP3]])
; CHECK-NEXT: [[VP_OP1:%.*]] = call <vscale x 1 x i16> @llvm.vp.lshr.nxv1i16(<vscale x 1 x i16> [[VP_OP]], <vscale x 1 x i16> trunc (<vscale x 1 x i32> splat (i32 1) to <vscale x 1 x i16>), <vscale x 1 x i1> splat (i1 true), i32 [[TMP3]])
; CHECK-NEXT: [[TMP8:%.*]] = call <vscale x 1 x i8> @llvm.vp.trunc.nxv1i8.nxv1i16(<vscale x 1 x i16> [[VP_OP1]], <vscale x 1 x i1> splat (i1 true), i32 [[TMP3]])
; CHECK-NEXT: call void @llvm.vp.scatter.nxv1i8.nxv1p0(<vscale x 1 x i8> [[TMP8]], <vscale x 1 x ptr> align 1 zeroinitializer, <vscale x 1 x i1> splat (i1 true), i32 [[TMP3]])
; CHECK-NEXT: [[TMP9:%.*]] = zext i32 [[TMP3]] to i64
; CHECK-NEXT: [[INDEX_EVL_NEXT]] = add nuw i64 [[TMP9]], [[EVL_BASED_IV]]
; CHECK-NEXT: [[INDEX_NEXT]] = add nuw i64 [[INDEX]], [[TMP2]]
; CHECK-NEXT: [[TMP10:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
; CHECK-NEXT: br i1 [[TMP10]], label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP0:![0-9]+]]
; CHECK: [[MIDDLE_BLOCK]]:
; CHECK-NEXT: br i1 true, label %[[EXIT:.*]], label %[[SCALAR_PH]]
; CHECK: [[SCALAR_PH]]:
; CHECK-NEXT: [[BC_RESUME_VAL:%.*]] = phi i64 [ [[N_VEC]], %[[MIDDLE_BLOCK]] ], [ 0, %[[ENTRY]] ]
; CHECK-NEXT: br label %[[LOOP:.*]]
; CHECK: [[LOOP]]:
; CHECK-NEXT: [[IV:%.*]] = phi i64 [ [[BC_RESUME_VAL]], %[[SCALAR_PH]] ], [ [[IV_NEXT:%.*]], %[[LOOP]] ]
; CHECK-NEXT: [[GEP_SRC:%.*]] = getelementptr i8, ptr [[SRC]], i64 [[IV]]
; CHECK-NEXT: [[TMP11:%.*]] = load i8, ptr [[GEP_SRC]], align 1
; CHECK-NEXT: [[CONV:%.*]] = zext i8 [[TMP11]] to i32
; CHECK-NEXT: [[MUL16:%.*]] = mul i32 0, [[CONV]]
; CHECK-NEXT: [[SHR35:%.*]] = lshr i32 [[MUL16]], 1
; CHECK-NEXT: [[CONV36:%.*]] = trunc i32 [[SHR35]] to i8
; CHECK-NEXT: store i8 [[CONV36]], ptr null, align 1
; CHECK-NEXT: [[IV_NEXT]] = add i64 [[IV]], 1
; CHECK-NEXT: [[EC:%.*]] = icmp eq i64 [[IV]], 1
; CHECK-NEXT: br i1 [[EC]], label %[[EXIT]], label %[[LOOP]], !llvm.loop [[LOOP3:![0-9]+]]
; CHECK: [[EXIT]]:
; CHECK-NEXT: ret void
;
entry:
br label %loop

loop: ; preds = %loop, %entry
%iv = phi i64 [ 0, %entry ], [ %iv.next, %loop ]
%gep.src = getelementptr i8, ptr %src, i64 %iv
%0 = load i8, ptr %gep.src, align 1
%conv = zext i8 %0 to i32
%mul16 = mul i32 0, %conv
Copy link
Contributor Author

Choose a reason for hiding this comment

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

This is strange, but running instcombine on this seems to simplify it too much and then we don't get the assertion

%shr35 = lshr i32 %mul16, 1
%conv36 = trunc i32 %shr35 to i8
store i8 %conv36, ptr null, align 1
%iv.next = add i64 %iv, 1
%ec = icmp eq i64 %iv, 1
br i1 %ec, label %exit, label %loop

exit: ; preds = %loop
ret void
}
;.
; CHECK: [[LOOP0]] = distinct !{[[LOOP0]], [[META1:![0-9]+]], [[META2:![0-9]+]]}
; CHECK: [[META1]] = !{!"llvm.loop.isvectorized", i32 1}
; CHECK: [[META2]] = !{!"llvm.loop.unroll.runtime.disable"}
; CHECK: [[LOOP3]] = distinct !{[[LOOP3]], [[META2]], [[META1]]}
;.
Loading