Skip to content

[LV][EVL] Support cast instruction with EVL-vectorization #108351

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 12 commits into from
Dec 11, 2024
Merged

[LV][EVL] Support cast instruction with EVL-vectorization #108351

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
12 commits
Select commit Hold shift + click to select a range
28e4d5c
[LV][EVL] Support cast instruction with EVL-vectorization
LiqinWeng Sep 12, 2024
4d3be15
Rebase and Revert the implement of castEVLRecipe
LiqinWeng Oct 18, 2024
3f4e0e5
[VPlan] Use VPWidenIntrinsicRecipe to vp.cast
LiqinWeng Oct 18, 2024
cc04719
fix the comments
LiqinWeng Oct 31, 2024
363d625
fix the comments
LiqinWeng Nov 6, 2024
0e45dc4
fix the commets
LiqinWeng Nov 14, 2024
1d8f27c
fix the comments
LiqinWeng Nov 17, 2024
5bd40b9
Rebase && add ir test && address the failed tests
LiqinWeng Nov 28, 2024
118f55a
Merge branch 'main' into widen-vp-with-evl
LiqinWeng Dec 2, 2024
6b8fb80
fix the comments
LiqinWeng Dec 3, 2024
1e54505
address the comments
LiqinWeng Dec 10, 2024
b7bc8a8
address the comments
LiqinWeng Dec 10, 2024
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
3 changes: 3 additions & 0 deletions llvm/lib/Analysis/VectorUtils.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -144,6 +144,9 @@ bool llvm::isVectorIntrinsicWithOverloadTypeAtArg(
if (TTI && Intrinsic::isTargetIntrinsic(ID))
return TTI->isVectorIntrinsicWithOverloadTypeAtArg(ID, OpdIdx);

if (VPCastIntrinsic::isVPCast(ID))
return OpdIdx == -1 || OpdIdx == 0;

switch (ID) {
case Intrinsic::fptosi_sat:
case Intrinsic::fptoui_sat:
Expand Down
3 changes: 2 additions & 1 deletion llvm/lib/Transforms/Vectorize/VPlanRecipes.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -964,7 +964,8 @@ void VPWidenIntrinsicRecipe::execute(VPTransformState &State) {
Module *M = State.Builder.GetInsertBlock()->getModule();
Function *VectorF =
Intrinsic::getOrInsertDeclaration(M, VectorIntrinsicID, TysForDecl);
assert(VectorF && "Can't retrieve vector intrinsic.");
Copy link
Contributor

Choose a reason for hiding this comment

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

Is this change still need?

Copy link
Contributor

Choose a reason for hiding this comment

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

Probably sufficient to keep the existing message

assert(VectorF &&
"Can't retrieve vector intrinsic or vector-predication intrinsics.");

auto *CI = cast_or_null<CallInst>(getUnderlyingValue());
SmallVector<OperandBundleDef, 1> OpBundles;
Expand Down
20 changes: 20 additions & 0 deletions llvm/lib/Transforms/Vectorize/VPlanTransforms.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1500,6 +1500,26 @@ static void transformRecipestoEVLRecipes(VPlan &Plan, VPValue &EVL) {
*CI, VPID, Ops, TypeInfo.inferScalarType(CInst),
CInst->getDebugLoc());
})
.Case<VPWidenCastRecipe>(
[&](VPWidenCastRecipe *CInst) -> VPRecipeBase * {
auto *CI = dyn_cast<CastInst>(CInst->getUnderlyingInstr());
Intrinsic::ID VPID =
VPIntrinsic::getForOpcode(CI->getOpcode());
assert(VPID != Intrinsic::not_intrinsic &&
"Expected vp.casts Instrinsic");

SmallVector<VPValue *> Ops(CInst->operands());
assert(VPIntrinsic::getMaskParamPos(VPID) &&
VPIntrinsic::getVectorLengthParamPos(VPID) &&
"Expected VP intrinsic");
VPValue *Mask = Plan.getOrAddLiveIn(ConstantInt::getTrue(
IntegerType::getInt1Ty(CI->getContext())));
Ops.push_back(Mask);
Ops.push_back(&EVL);
return new VPWidenIntrinsicRecipe(
VPID, Ops, TypeInfo.inferScalarType(CInst),
CInst->getDebugLoc());
})
.Case<VPWidenSelectRecipe>([&](VPWidenSelectRecipe *Sel) {
SmallVector<VPValue *> Ops(Sel->operands());
Ops.push_back(&EVL);
Expand Down
20 changes: 10 additions & 10 deletions llvm/test/Transforms/LoopVectorize/RISCV/inloop-reduction.ll
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -143,7 +143,7 @@ define i32 @add_i16_i32(ptr nocapture readonly %x, i32 %n) {
; IF-EVL-OUTLOOP-NEXT: [[TMP7:%.*]] = getelementptr inbounds i16, ptr [[X:%.*]], i32 [[TMP6]]
; IF-EVL-OUTLOOP-NEXT: [[TMP8:%.*]] = getelementptr inbounds i16, ptr [[TMP7]], i32 0
; IF-EVL-OUTLOOP-NEXT: [[VP_OP_LOAD:%.*]] = call <vscale x 4 x i16> @llvm.vp.load.nxv4i16.p0(ptr align 2 [[TMP8]], <vscale x 4 x i1> splat (i1 true), i32 [[TMP5]])
; IF-EVL-OUTLOOP-NEXT: [[TMP9:%.*]] = sext <vscale x 4 x i16> [[VP_OP_LOAD]] to <vscale x 4 x i32>
; IF-EVL-OUTLOOP-NEXT: [[TMP9:%.*]] = call <vscale x 4 x i32> @llvm.vp.sext.nxv4i32.nxv4i16(<vscale x 4 x i16> [[VP_OP_LOAD]], <vscale x 4 x i1> splat (i1 true), i32 [[TMP5]])
; IF-EVL-OUTLOOP-NEXT: [[VP_OP:%.*]] = call <vscale x 4 x i32> @llvm.vp.add.nxv4i32(<vscale x 4 x i32> [[VEC_PHI]], <vscale x 4 x i32> [[TMP9]], <vscale x 4 x i1> splat (i1 true), i32 [[TMP5]])
; IF-EVL-OUTLOOP-NEXT: [[TMP10]] = call <vscale x 4 x i32> @llvm.vp.merge.nxv4i32(<vscale x 4 x i1> splat (i1 true), <vscale x 4 x i32> [[VP_OP]], <vscale x 4 x i32> [[VEC_PHI]], i32 [[TMP5]])
; IF-EVL-OUTLOOP-NEXT: [[INDEX_EVL_NEXT]] = add nuw i32 [[TMP5]], [[EVL_BASED_IV]]
Expand Down Expand Up @@ -194,16 +194,16 @@ define i32 @add_i16_i32(ptr nocapture readonly %x, i32 %n) {
; IF-EVL-INLOOP-NEXT: [[INDEX:%.*]] = phi i32 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
; IF-EVL-INLOOP-NEXT: [[EVL_BASED_IV:%.*]] = phi i32 [ 0, [[VECTOR_PH]] ], [ [[INDEX_EVL_NEXT:%.*]], [[VECTOR_BODY]] ]
; IF-EVL-INLOOP-NEXT: [[VEC_PHI:%.*]] = phi i32 [ 0, [[VECTOR_PH]] ], [ [[TMP11:%.*]], [[VECTOR_BODY]] ]
; IF-EVL-INLOOP-NEXT: [[AVL:%.*]] = sub i32 [[N]], [[EVL_BASED_IV]]
; IF-EVL-INLOOP-NEXT: [[TMP5:%.*]] = call i32 @llvm.experimental.get.vector.length.i32(i32 [[AVL]], i32 8, i1 true)
; IF-EVL-INLOOP-NEXT: [[TMP6:%.*]] = add i32 [[EVL_BASED_IV]], 0
; IF-EVL-INLOOP-NEXT: [[TMP7:%.*]] = getelementptr inbounds i16, ptr [[X:%.*]], i32 [[TMP6]]
; IF-EVL-INLOOP-NEXT: [[TMP8:%.*]] = getelementptr inbounds i16, ptr [[TMP7]], i32 0
; IF-EVL-INLOOP-NEXT: [[VP_OP_LOAD:%.*]] = call <vscale x 8 x i16> @llvm.vp.load.nxv8i16.p0(ptr align 2 [[TMP8]], <vscale x 8 x i1> splat (i1 true), i32 [[TMP5]])
; IF-EVL-INLOOP-NEXT: [[TMP9:%.*]] = sext <vscale x 8 x i16> [[VP_OP_LOAD]] to <vscale x 8 x i32>
; IF-EVL-INLOOP-NEXT: [[TMP10:%.*]] = call i32 @llvm.vp.reduce.add.nxv8i32(i32 0, <vscale x 8 x i32> [[TMP9]], <vscale x 8 x i1> splat (i1 true), i32 [[TMP5]])
; IF-EVL-INLOOP-NEXT: [[TMP5:%.*]] = sub i32 [[N]], [[EVL_BASED_IV]]
; IF-EVL-INLOOP-NEXT: [[TMP6:%.*]] = call i32 @llvm.experimental.get.vector.length.i32(i32 [[TMP5]], i32 8, i1 true)
; IF-EVL-INLOOP-NEXT: [[TMP7:%.*]] = add i32 [[EVL_BASED_IV]], 0
; IF-EVL-INLOOP-NEXT: [[TMP8:%.*]] = getelementptr inbounds i16, ptr [[X:%.*]], i32 [[TMP7]]
; IF-EVL-INLOOP-NEXT: [[TMP9:%.*]] = getelementptr inbounds i16, ptr [[TMP8]], i32 0
; IF-EVL-INLOOP-NEXT: [[VP_OP_LOAD:%.*]] = call <vscale x 8 x i16> @llvm.vp.load.nxv8i16.p0(ptr align 2 [[TMP9]], <vscale x 8 x i1> splat (i1 true), i32 [[TMP6]])
; IF-EVL-INLOOP-NEXT: [[TMP14:%.*]] = call <vscale x 8 x i32> @llvm.vp.sext.nxv8i32.nxv8i16(<vscale x 8 x i16> [[VP_OP_LOAD]], <vscale x 8 x i1> splat (i1 true), i32 [[TMP6]])
; IF-EVL-INLOOP-NEXT: [[TMP10:%.*]] = call i32 @llvm.vp.reduce.add.nxv8i32(i32 0, <vscale x 8 x i32> [[TMP14]], <vscale x 8 x i1> splat (i1 true), i32 [[TMP6]])
; IF-EVL-INLOOP-NEXT: [[TMP11]] = add i32 [[TMP10]], [[VEC_PHI]]
; IF-EVL-INLOOP-NEXT: [[INDEX_EVL_NEXT]] = add nuw i32 [[TMP5]], [[EVL_BASED_IV]]
; IF-EVL-INLOOP-NEXT: [[INDEX_EVL_NEXT]] = add nuw i32 [[TMP6]], [[EVL_BASED_IV]]
; IF-EVL-INLOOP-NEXT: [[INDEX_NEXT]] = add nuw i32 [[INDEX]], [[TMP4]]
; IF-EVL-INLOOP-NEXT: [[TMP12:%.*]] = icmp eq i32 [[INDEX_NEXT]], [[N_VEC]]
; IF-EVL-INLOOP-NEXT: br i1 [[TMP12]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP0:![0-9]+]]
Expand Down
8 changes: 4 additions & 4 deletions llvm/test/Transforms/LoopVectorize/RISCV/vectorize-force-tail-with-evl-call-intrinsics.ll
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -708,9 +708,9 @@ define void @vp_lrint(ptr %a, ptr %b, i64 %N) {
; IF-EVL-NEXT: [[TMP11:%.*]] = getelementptr inbounds float, ptr [[B]], i64 [[TMP10]]
; IF-EVL-NEXT: [[TMP12:%.*]] = getelementptr inbounds float, ptr [[TMP11]], i32 0
; IF-EVL-NEXT: [[VP_OP_LOAD:%.*]] = call <vscale x 4 x float> @llvm.vp.load.nxv4f32.p0(ptr align 4 [[TMP12]], <vscale x 4 x i1> splat (i1 true), i32 [[TMP9]])
; IF-EVL-NEXT: [[TMP13:%.*]] = fpext <vscale x 4 x float> [[VP_OP_LOAD]] to <vscale x 4 x double>
; IF-EVL-NEXT: [[TMP13:%.*]] = call <vscale x 4 x double> @llvm.vp.fpext.nxv4f64.nxv4f32(<vscale x 4 x float> [[VP_OP_LOAD]], <vscale x 4 x i1> splat (i1 true), i32 [[TMP9]])
; IF-EVL-NEXT: [[TMP14:%.*]] = call <vscale x 4 x i64> @llvm.vp.lrint.nxv4i64.nxv4f64(<vscale x 4 x double> [[TMP13]], <vscale x 4 x i1> splat (i1 true), i32 [[TMP9]])
; IF-EVL-NEXT: [[TMP15:%.*]] = trunc <vscale x 4 x i64> [[TMP14]] to <vscale x 4 x i32>
; IF-EVL-NEXT: [[TMP15:%.*]] = call <vscale x 4 x i32> @llvm.vp.trunc.nxv4i32.nxv4i64(<vscale x 4 x i64> [[TMP14]], <vscale x 4 x i1> splat (i1 true), i32 [[TMP9]])
; IF-EVL-NEXT: [[TMP16:%.*]] = getelementptr inbounds i32, ptr [[A]], i64 [[TMP10]]
; IF-EVL-NEXT: [[TMP17:%.*]] = getelementptr inbounds i32, ptr [[TMP16]], i32 0
; IF-EVL-NEXT: call void @llvm.vp.store.nxv4i32.p0(<vscale x 4 x i32> [[TMP15]], ptr align 4 [[TMP17]], <vscale x 4 x i1> splat (i1 true), i32 [[TMP9]])
Expand Down Expand Up @@ -817,9 +817,9 @@ define void @vp_llrint(ptr %a, ptr %b, i64 %N) {
; IF-EVL-NEXT: [[TMP11:%.*]] = getelementptr inbounds float, ptr [[B]], i64 [[TMP10]]
; IF-EVL-NEXT: [[TMP12:%.*]] = getelementptr inbounds float, ptr [[TMP11]], i32 0
; IF-EVL-NEXT: [[VP_OP_LOAD:%.*]] = call <vscale x 4 x float> @llvm.vp.load.nxv4f32.p0(ptr align 4 [[TMP12]], <vscale x 4 x i1> splat (i1 true), i32 [[TMP9]])
; IF-EVL-NEXT: [[TMP13:%.*]] = fpext <vscale x 4 x float> [[VP_OP_LOAD]] to <vscale x 4 x double>
; IF-EVL-NEXT: [[TMP13:%.*]] = call <vscale x 4 x double> @llvm.vp.fpext.nxv4f64.nxv4f32(<vscale x 4 x float> [[VP_OP_LOAD]], <vscale x 4 x i1> splat (i1 true), i32 [[TMP9]])
; IF-EVL-NEXT: [[TMP14:%.*]] = call <vscale x 4 x i64> @llvm.vp.llrint.nxv4i64.nxv4f64(<vscale x 4 x double> [[TMP13]], <vscale x 4 x i1> splat (i1 true), i32 [[TMP9]])
; IF-EVL-NEXT: [[TMP15:%.*]] = trunc <vscale x 4 x i64> [[TMP14]] to <vscale x 4 x i32>
; IF-EVL-NEXT: [[TMP15:%.*]] = call <vscale x 4 x i32> @llvm.vp.trunc.nxv4i32.nxv4i64(<vscale x 4 x i64> [[TMP14]], <vscale x 4 x i1> splat (i1 true), i32 [[TMP9]])
; IF-EVL-NEXT: [[TMP16:%.*]] = getelementptr inbounds i32, ptr [[A]], i64 [[TMP10]]
; IF-EVL-NEXT: [[TMP17:%.*]] = getelementptr inbounds i32, ptr [[TMP16]], i32 0
; IF-EVL-NEXT: call void @llvm.vp.store.nxv4i32.p0(<vscale x 4 x i32> [[TMP15]], ptr align 4 [[TMP17]], <vscale x 4 x i1> splat (i1 true), i32 [[TMP9]])
Expand Down
Loading
Loading