paschalis-mpeis wrote:
The benefits of having`getFRemInstrCost` in my view are the below:
1. frem is a special case anyway:
It's an IR instruction that is not supported by all hw and targets have to
specialize.
Handling it in a dedicated switch case with a dedicated TTI function call,
clearly e
alexey-bataev wrote:
> Changing `getArithmeticInstrCost` is just too dangerous. What if one opcode
> needs TLI for a different reason?
That should be fine, what's the dangerous in it?
https://github.com/llvm/llvm-project/pull/82488
___
llvm-branch-co
paulwalker-arm wrote:
Changing `getArithmeticInstrCost` is just too dangerous. What if one opcode
needs TLI for a different reason? all of a sudden all existing callers are
entered into the contract (FREM is guaranteed to be transformed into a math
call) without ensuring that's actually the c
alexey-bataev wrote:
> Personally I'm happy with keeping this nuance outside of TTI but if we really
> want this captured within TTI then I think it's time to break FREM into its
> own cost function (i.e. implement getFRemInstrCost. That way
> getArithmeticInstrCost can work as it does today a
@@ -8362,9 +8362,12 @@ BoUpSLP::getEntryCost(const TreeEntry *E, ArrayRef VectorizedVals,
unsigned OpIdx = isa(VL0) ? 0 : 1;
TTI::OperandValueInfo Op1Info = getOperandInfo(E->getOperand(0));
TTI::OperandValueInfo Op2Info = getOperandInfo(E->getOperand(OpIdx));
@@ -1031,6 +1032,22 @@ APInt llvm::possiblyDemandedEltsInMask(Value *Mask) {
return DemandedElts;
}
+InstructionCost
+llvm::getVecLibCallCost(const Instruction *I, const TargetTransformInfo *TTI,
+const TargetLibraryInfo *TLI, VectorType *VecTy,
+
@@ -869,6 +870,18 @@ TargetTransformInfo::getOperandInfo(const Value *V) {
return {OpInfo, OpProps};
}
+InstructionCost TargetTransformInfo::getVecLibCallCost(
+const int OpCode, const TargetLibraryInfo *TLI, VectorType *VecTy,
+TTI::TargetCostKind CostKind) {
+ Typ
https://github.com/paulwalker-arm commented:
I've review the patch from both side so most of the comment will be void if you
opt for the new TTI hook. That advantage of the TTI hook is that because it is
specific to FREM you can hardware things like numbers of operands, which should
streamlin
@@ -410,6 +410,14 @@ bool maskIsAllOneOrUndef(Value *Mask);
/// for each lane which may be active.
APInt possiblyDemandedEltsInMask(Value *Mask);
+/// Returns the cost of a call when a target has a vector library function for
+/// the given \p VecTy, otherwise an invalid cost.
@@ -1031,6 +1032,22 @@ APInt llvm::possiblyDemandedEltsInMask(Value *Mask) {
return DemandedElts;
}
+InstructionCost
+llvm::getVecLibCallCost(const Instruction *I, const TargetTransformInfo *TTI,
+const TargetLibraryInfo *TLI, VectorType *VecTy,
+
https://github.com/paulwalker-arm edited
https://github.com/llvm/llvm-project/pull/82488
___
llvm-branch-commits mailing list
llvm-branch-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
@@ -869,6 +870,18 @@ TargetTransformInfo::getOperandInfo(const Value *V) {
return {OpInfo, OpProps};
}
+InstructionCost TargetTransformInfo::getVecLibCallCost(
+const int OpCode, const TargetLibraryInfo *TLI, VectorType *VecTy,
+TTI::TargetCostKind CostKind) {
+ Typ
@@ -869,6 +870,18 @@ TargetTransformInfo::getOperandInfo(const Value *V) {
return {OpInfo, OpProps};
}
+InstructionCost TargetTransformInfo::getVecLibCallCost(
+const int OpCode, const TargetLibraryInfo *TLI, VectorType *VecTy,
+TTI::TargetCostKind CostKind) {
+ Typ
@@ -869,6 +870,18 @@ TargetTransformInfo::getOperandInfo(const Value *V) {
return {OpInfo, OpProps};
}
+InstructionCost TargetTransformInfo::getVecLibCallCost(
+const int OpCode, const TargetLibraryInfo *TLI, VectorType *VecTy,
+TTI::TargetCostKind CostKind) {
+ Typ
@@ -869,6 +870,18 @@ TargetTransformInfo::getOperandInfo(const Value *V) {
return {OpInfo, OpProps};
}
+InstructionCost TargetTransformInfo::getVecLibCallCost(
+const int OpCode, const TargetLibraryInfo *TLI, VectorType *VecTy,
+TTI::TargetCostKind CostKind) {
+ Typ
@@ -869,6 +870,18 @@ TargetTransformInfo::getOperandInfo(const Value *V) {
return {OpInfo, OpProps};
}
+InstructionCost TargetTransformInfo::getVecLibCallCost(
+const int OpCode, const TargetLibraryInfo *TLI, VectorType *VecTy,
+TTI::TargetCostKind CostKind) {
+ Typ
@@ -8362,9 +8362,20 @@ BoUpSLP::getEntryCost(const TreeEntry *E, ArrayRef VectorizedVals,
unsigned OpIdx = isa(VL0) ? 0 : 1;
TTI::OperandValueInfo Op1Info = getOperandInfo(E->getOperand(0));
TTI::OperandValueInfo Op2Info = getOperandInfo(E->getOperand(OpIdx));
@@ -869,6 +870,18 @@ TargetTransformInfo::getOperandInfo(const Value *V) {
return {OpInfo, OpProps};
}
+InstructionCost TargetTransformInfo::getVecLibCallCost(
+const int OpCode, const TargetLibraryInfo *TLI, VectorType *VecTy,
+TTI::TargetCostKind CostKind) {
+ Typ
https://github.com/paschalis-mpeis updated
https://github.com/llvm/llvm-project/pull/82488
>From 641aaf7c13d520bef52b092726f8346bfecb1c8d Mon Sep 17 00:00:00 2001
From: Paschalis Mpeis
Date: Wed, 21 Feb 2024 11:53:00 +
Subject: [PATCH 1/4] SLP cannot vectorize frem calls in AArch64.
It nee
@@ -8362,9 +8362,20 @@ BoUpSLP::getEntryCost(const TreeEntry *E, ArrayRef VectorizedVals,
unsigned OpIdx = isa(VL0) ? 0 : 1;
TTI::OperandValueInfo Op1Info = getOperandInfo(E->getOperand(0));
TTI::OperandValueInfo Op2Info = getOperandInfo(E->getOperand(OpIdx));
@@ -869,6 +870,18 @@ TargetTransformInfo::getOperandInfo(const Value *V) {
return {OpInfo, OpProps};
}
+InstructionCost TargetTransformInfo::getVecLibCallCost(
+const int OpCode, const TargetLibraryInfo *TLI, VectorType *VecTy,
+TTI::TargetCostKind CostKind) {
+ Typ
@@ -869,6 +870,18 @@ TargetTransformInfo::getOperandInfo(const Value *V) {
return {OpInfo, OpProps};
}
+InstructionCost TargetTransformInfo::getVecLibCallCost(
+const int OpCode, const TargetLibraryInfo *TLI, VectorType *VecTy,
+TTI::TargetCostKind CostKind) {
+ Typ
paschalis-mpeis wrote:
Addressed reviewers and rebased to parent pr:
- #80423
Github is now rendering **only** the changes of this patch.
https://github.com/llvm/llvm-project/pull/82488
___
llvm-branch-commits mailing list
llvm-branch-commits@lists.l
https://github.com/paschalis-mpeis edited
https://github.com/llvm/llvm-project/pull/82488
___
llvm-branch-commits mailing list
llvm-branch-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
https://github.com/paschalis-mpeis edited
https://github.com/llvm/llvm-project/pull/82488
___
llvm-branch-commits mailing list
llvm-branch-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
@@ -8362,9 +8362,20 @@ BoUpSLP::getEntryCost(const TreeEntry *E, ArrayRef VectorizedVals,
unsigned OpIdx = isa(VL0) ? 0 : 1;
TTI::OperandValueInfo Op1Info = getOperandInfo(E->getOperand(0));
TTI::OperandValueInfo Op2Info = getOperandInfo(E->getOperand(OpIdx));
https://github.com/paschalis-mpeis edited
https://github.com/llvm/llvm-project/pull/82488
___
llvm-branch-commits mailing list
llvm-branch-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
https://github.com/paschalis-mpeis updated
https://github.com/llvm/llvm-project/pull/82488
>From 641aaf7c13d520bef52b092726f8346bfecb1c8d Mon Sep 17 00:00:00 2001
From: Paschalis Mpeis
Date: Wed, 21 Feb 2024 11:53:00 +
Subject: [PATCH 1/3] SLP cannot vectorize frem calls in AArch64.
It nee
https://github.com/paschalis-mpeis edited
https://github.com/llvm/llvm-project/pull/82488
___
llvm-branch-commits mailing list
llvm-branch-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
29 matches
Mail list logo
