RPCSX/llvm
1
0
Fork 0
mirror of https://github.com/RPCSX/llvm.git synced 2025-02-23 22:52:48 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity

[ARM] Make f16 interleaved accesses expensive.

Browse Source 
There are no vldN/vstN f16 variants, even with +fullfp16.
We could use the i16 variants, but, in practice, even with +fullfp16,
the f16 sequence leading to the i16 shuffle usually gets scalarized.
We'd need to improve our support for f16 codegen before getting there.

Teach the cost model to consider f16 interleaved operations as
expensive.  Otherwise, we are all but guaranteed to end up with
a large block of scalarized vector code.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@294819 91177308-0d34-0410-b5e6-96231b3b80d8
This commit is contained in:
Ahmed Bougacha 2017-02-11 01:53:04 +00:00
parent bd8e11d920
commit 6ea43f266f
2 changed files with 33 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

3
lib/Target/ARM/ARMTargetTransformInfo.cpp
View File

@ -533,7 +533,8 @@ int ARMTTIImpl::getInterleavedMemoryOpCost(unsigned Opcode, Type *VecTy,
unsigned SubVecSize = DL.getTypeSizeInBits(SubVecTy);
// vldN/vstN only support legal vector types of size 64 or 128 in bits.
if (NumElts % Factor == 0 && (SubVecSize == 64 || SubVecSize == 128))
if (NumElts % Factor == 0 && (SubVecSize == 64 || SubVecSize == 128) &&
!VecTy->getScalarType()->isHalfTy())
return Factor;
}

31
test/Transforms/LoopVectorize/ARM/interleaved_cost.ll
View File

@ -99,3 +99,34 @@ for.body:
for.end:
ret void
}
%half.2 = type {half, half}
define void @half_factor_2(%half.2* %data, i64 %n) {
entry:
br label %for.body
; VF_4-LABEL: Checking a loop in "half_factor_2"
; VF_4: Found an estimated cost of 40 for VF 4 For instruction: %tmp2 = load half, half* %tmp0, align 2
; VF_4-NEXT: Found an estimated cost of 0 for VF 4 For instruction: %tmp3 = load half, half* %tmp1, align 2
; VF_4-NEXT: Found an estimated cost of 0 for VF 4 For instruction: store half 0xH0000, half* %tmp0, align 2
; VF_4-NEXT: Found an estimated cost of 40 for VF 4 For instruction: store half 0xH0000, half* %tmp1, align 2
; VF_8-LABEL: Checking a loop in "half_factor_2"
; VF_8: Found an estimated cost of 80 for VF 8 For instruction: %tmp2 = load half, half* %tmp0, align 2
; VF_8-NEXT: Found an estimated cost of 0 for VF 8 For instruction: %tmp3 = load half, half* %tmp1, align 2
; VF_8-NEXT: Found an estimated cost of 0 for VF 8 For instruction: store half 0xH0000, half* %tmp0, align 2
; VF_8-NEXT: Found an estimated cost of 80 for VF 8 For instruction: store half 0xH0000, half* %tmp1, align 2
for.body:
%i = phi i64 [ 0, %entry ], [ %i.next, %for.body ]
%tmp0 = getelementptr inbounds %half.2, %half.2* %data, i64 %i, i32 0
%tmp1 = getelementptr inbounds %half.2, %half.2* %data, i64 %i, i32 1
%tmp2 = load half, half* %tmp0, align 2
%tmp3 = load half, half* %tmp1, align 2
store half 0., half* %tmp0, align 2
store half 0., half* %tmp1, align 2
%i.next = add nuw nsw i64 %i, 1
%cond = icmp slt i64 %i.next, %n
br i1 %cond, label %for.body, label %for.end
for.end:
ret void
}