maxjhandsome
/
llvm-project
forked from OSchip/llvm-project
1
0
Fork 0
Code Issues Pull Requests Packages Releases Wiki Activity

[DAGCombiner] try repeated fdiv divisor transform before building estimate (2nd try) 

The original patch was committed at rL359398 and reverted at rL359695 because of
infinite looping.

This includes a fix to check for a vector splat of "1.0" to avoid the infinite loop.

Original commit message:

This was originally part of D61028, but it's an independent diff.

If we try the repeated divisor reciprocal transform before producing an estimate sequence,
then we have an opportunity to use scalar fdiv. On x86, the trade-off is 1 divss vs. 5
vector FP ops in the default estimate sequence. On recent chips (Skylake, Ryzen), the
full-precision division is only 3 cycle throughput, so that's probably the better perf
default option and avoids problems from x86's inaccurate estimates.

The last 2 tests show that users still have the option to override the defaults by using
the function attributes for reciprocal estimates, but those patterns are potentially made
faster by converting the vector ops (including ymm ops) to scalar math.

Differential Revision: https://reviews.llvm.org/D61149

llvm-svn: 359793
This commit is contained in:
Sanjay Patel 2019-05-02 15:02:08 +00:00
parent 4ad6dbc5fd
commit 1972826178
2 changed files with 31 additions and 43 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

8
llvm/lib/CodeGen/SelectionDAG/DAGCombiner.cpp
View File

@ -11915,7 +11915,7 @@ SDValue DAGCombiner::combineRepeatedFPDivisors(SDNode *N) {
// Skip if current node is a reciprocal.
SDValue N0 = N->getOperand(0);
ConstantFPSDNode *N0CFP = dyn_cast<ConstantFPSDNode>(N0);
ConstantFPSDNode *N0CFP = isConstOrConstSplatFP(N0, /* AllowUndefs */ true);
if (N0CFP && N0CFP->isExactlyValue(1.0))
return SDValue();
@ -11993,6 +11993,9 @@ SDValue DAGCombiner::visitFDIV(SDNode *N) {
if (SDValue NewSel = foldBinOpIntoSelect(N))
return NewSel;
if (SDValue V = combineRepeatedFPDivisors(N))
return V;
if (Options.UnsafeFPMath || Flags.hasAllowReciprocal()) {
// fold (fdiv X, c2) -> fmul X, 1/c2 if losing precision is acceptable.
if (N1CFP) {
@ -12082,9 +12085,6 @@ SDValue DAGCombiner::visitFDIV(SDNode *N) {
}
}
if (SDValue CombineRepeatedDivisors = combineRepeatedFPDivisors(N))
return CombineRepeatedDivisors;
return SDValue();
}

66
llvm/test/CodeGen/X86/fdiv-combine-vec.ll
View File

@ -51,25 +51,17 @@ define <4 x double> @splat_fdiv_v4f64(<4 x double> %x, double %y) {
define <4 x float> @splat_fdiv_v4f32(<4 x float> %x, float %y) {
; SSE-LABEL: splat_fdiv_v4f32:
; SSE: # %bb.0:
; SSE-NEXT: shufps {{.*#+}} xmm1 = xmm1[0,0,0,0]
; SSE-NEXT: rcpps %xmm1, %xmm2
; SSE-NEXT: mulps %xmm2, %xmm1
; SSE-NEXT: movaps {{.*#+}} xmm3 = [1.0E+0,1.0E+0,1.0E+0,1.0E+0]
; SSE-NEXT: subps %xmm1, %xmm3
; SSE-NEXT: mulps %xmm2, %xmm3
; SSE-NEXT: addps %xmm2, %xmm3
; SSE-NEXT: mulps %xmm3, %xmm0
; SSE-NEXT: movss {{.*#+}} xmm2 = mem[0],zero,zero,zero
; SSE-NEXT: divss %xmm1, %xmm2
; SSE-NEXT: shufps {{.*#+}} xmm2 = xmm2[0,0,0,0]
; SSE-NEXT: mulps %xmm2, %xmm0
; SSE-NEXT: retq
;
; AVX-LABEL: splat_fdiv_v4f32:
; AVX: # %bb.0:
; AVX-NEXT: vmovss {{.*#+}} xmm2 = mem[0],zero,zero,zero
; AVX-NEXT: vdivss %xmm1, %xmm2, %xmm1
; AVX-NEXT: vpermilps {{.*#+}} xmm1 = xmm1[0,0,0,0]
; AVX-NEXT: vrcpps %xmm1, %xmm2
; AVX-NEXT: vmulps %xmm2, %xmm1, %xmm1
; AVX-NEXT: vmovaps {{.*#+}} xmm3 = [1.0E+0,1.0E+0,1.0E+0,1.0E+0]
; AVX-NEXT: vsubps %xmm1, %xmm3, %xmm1
; AVX-NEXT: vmulps %xmm1, %xmm2, %xmm1
; AVX-NEXT: vaddps %xmm1, %xmm2, %xmm1
; AVX-NEXT: vmulps %xmm1, %xmm0, %xmm0
; AVX-NEXT: retq
%vy = insertelement <4 x float> undef, float %y, i32 0
@ -90,14 +82,10 @@ define <8 x float> @splat_fdiv_v8f32(<8 x float> %x, float %y) {
;
; AVX-LABEL: splat_fdiv_v8f32:
; AVX: # %bb.0:
; AVX-NEXT: vmovss {{.*#+}} xmm2 = mem[0],zero,zero,zero
; AVX-NEXT: vdivss %xmm1, %xmm2, %xmm1
; AVX-NEXT: vpermilps {{.*#+}} xmm1 = xmm1[0,0,0,0]
; AVX-NEXT: vinsertf128 $1, %xmm1, %ymm1, %ymm1
; AVX-NEXT: vrcpps %ymm1, %ymm2
; AVX-NEXT: vmulps %ymm2, %ymm1, %ymm1
; AVX-NEXT: vmovaps {{.*#+}} ymm3 = [1.0E+0,1.0E+0,1.0E+0,1.0E+0,1.0E+0,1.0E+0,1.0E+0,1.0E+0]
; AVX-NEXT: vsubps %ymm1, %ymm3, %ymm1
; AVX-NEXT: vmulps %ymm1, %ymm2, %ymm1
; AVX-NEXT: vaddps %ymm1, %ymm2, %ymm1
; AVX-NEXT: vmulps %ymm1, %ymm0, %ymm0
; AVX-NEXT: retq
%vy = insertelement <8 x float> undef, float %y, i32 0
@ -109,25 +97,25 @@ define <8 x float> @splat_fdiv_v8f32(<8 x float> %x, float %y) {
define <4 x float> @splat_fdiv_v4f32_estimate(<4 x float> %x, float %y) #0 {
; SSE-LABEL: splat_fdiv_v4f32_estimate:
; SSE: # %bb.0:
; SSE-NEXT: shufps {{.*#+}} xmm1 = xmm1[0,0,0,0]
; SSE-NEXT: rcpps %xmm1, %xmm2
; SSE-NEXT: mulps %xmm2, %xmm1
; SSE-NEXT: movaps {{.*#+}} xmm3 = [1.0E+0,1.0E+0,1.0E+0,1.0E+0]
; SSE-NEXT: subps %xmm1, %xmm3
; SSE-NEXT: mulps %xmm2, %xmm3
; SSE-NEXT: addps %xmm2, %xmm3
; SSE-NEXT: rcpss %xmm1, %xmm2
; SSE-NEXT: mulss %xmm2, %xmm1
; SSE-NEXT: movss {{.*#+}} xmm3 = mem[0],zero,zero,zero
; SSE-NEXT: subss %xmm1, %xmm3
; SSE-NEXT: mulss %xmm2, %xmm3
; SSE-NEXT: addss %xmm2, %xmm3
; SSE-NEXT: shufps {{.*#+}} xmm3 = xmm3[0,0,0,0]
; SSE-NEXT: mulps %xmm3, %xmm0
; SSE-NEXT: retq
;
; AVX-LABEL: splat_fdiv_v4f32_estimate:
; AVX: # %bb.0:
; AVX-NEXT: vrcpss %xmm1, %xmm1, %xmm2
; AVX-NEXT: vmulss %xmm2, %xmm1, %xmm1
; AVX-NEXT: vmovss {{.*#+}} xmm3 = mem[0],zero,zero,zero
; AVX-NEXT: vsubss %xmm1, %xmm3, %xmm1
; AVX-NEXT: vmulss %xmm1, %xmm2, %xmm1
; AVX-NEXT: vaddss %xmm1, %xmm2, %xmm1
; AVX-NEXT: vpermilps {{.*#+}} xmm1 = xmm1[0,0,0,0]
; AVX-NEXT: vrcpps %xmm1, %xmm2
; AVX-NEXT: vmulps %xmm2, %xmm1, %xmm1
; AVX-NEXT: vmovaps {{.*#+}} xmm3 = [1.0E+0,1.0E+0,1.0E+0,1.0E+0]
; AVX-NEXT: vsubps %xmm1, %xmm3, %xmm1
; AVX-NEXT: vmulps %xmm1, %xmm2, %xmm1
; AVX-NEXT: vaddps %xmm1, %xmm2, %xmm1
; AVX-NEXT: vmulps %xmm1, %xmm0, %xmm0
; AVX-NEXT: retq
%vy = insertelement <4 x float> undef, float %y, i32 0
@ -152,14 +140,14 @@ define <8 x float> @splat_fdiv_v8f32_estimate(<8 x float> %x, float %y) #0 {
;
; AVX-LABEL: splat_fdiv_v8f32_estimate:
; AVX: # %bb.0:
; AVX-NEXT: vrcpss %xmm1, %xmm1, %xmm2
; AVX-NEXT: vmulss %xmm2, %xmm1, %xmm1
; AVX-NEXT: vmovss {{.*#+}} xmm3 = mem[0],zero,zero,zero
; AVX-NEXT: vsubss %xmm1, %xmm3, %xmm1
; AVX-NEXT: vmulss %xmm1, %xmm2, %xmm1
; AVX-NEXT: vaddss %xmm1, %xmm2, %xmm1
; AVX-NEXT: vpermilps {{.*#+}} xmm1 = xmm1[0,0,0,0]
; AVX-NEXT: vinsertf128 $1, %xmm1, %ymm1, %ymm1
; AVX-NEXT: vrcpps %ymm1, %ymm2
; AVX-NEXT: vmulps %ymm2, %ymm1, %ymm1
; AVX-NEXT: vmovaps {{.*#+}} ymm3 = [1.0E+0,1.0E+0,1.0E+0,1.0E+0,1.0E+0,1.0E+0,1.0E+0,1.0E+0]
; AVX-NEXT: vsubps %ymm1, %ymm3, %ymm1
; AVX-NEXT: vmulps %ymm1, %ymm2, %ymm1
; AVX-NEXT: vaddps %ymm1, %ymm2, %ymm1
; AVX-NEXT: vmulps %ymm1, %ymm0, %ymm0
; AVX-NEXT: retq
%vy = insertelement <8 x float> undef, float %y, i32 0