[PATCH] D21761: AMDGPU: Fix fdiv lowering when f32 denormals supported

[Matt Arsenault via llvm-commits]

arsenm created this revision.
arsenm added a reviewer: tstellarAMD.
arsenm added a subscriber: llvm-commits.
Herald added subscribers: kzhuravl, arsenm.

Also fix test not actually using function labels.

http://reviews.llvm.org/D21761

Files:
  lib/Target/AMDGPU/SIISelLowering.cpp
  test/CodeGen/AMDGPU/fdiv.ll

Index: test/CodeGen/AMDGPU/fdiv.ll
===================================================================
--- test/CodeGen/AMDGPU/fdiv.ll
+++ test/CodeGen/AMDGPU/fdiv.ll
@@ -1,20 +1,19 @@
-; RUN: llc -march=amdgcn -verify-machineinstrs < %s | FileCheck -check-prefix=SI %s
-; RUN: llc -march=amdgcn -mcpu=tonga -verify-machineinstrs -amdgpu-fast-fdiv < %s | FileCheck -check-prefix=SI %s
-; RUN: llc -march=amdgcn -mcpu=fiji -verify-machineinstrs < %s | FileCheck -check-prefix=I754 %s
-; RUN: llc -march=amdgcn -mcpu=fiji -verify-machineinstrs -enable-unsafe-fp-math < %s | FileCheck -check-prefix=UNSAFE-FP %s
-; RUN: llc -march=r600 -mcpu=redwood < %s | FileCheck -check-prefix=R600 %s
+; RUN: llc -march=amdgcn -verify-machineinstrs < %s | FileCheck -check-prefix=SI -check-prefix=FUNC %s
+; RUN: llc -march=amdgcn -mcpu=tonga -mattr=+fp32-denormals -verify-machineinstrs < %s | FileCheck -check-prefix=I754 -check-prefix=FUNC %s
+; RUN: llc -march=amdgcn -mcpu=tonga -verify-machineinstrs -amdgpu-fast-fdiv < %s | FileCheck -check-prefix=SI -check-prefix=FUNC %s
+; RUN: llc -march=amdgcn -mcpu=fiji -verify-machineinstrs < %s | FileCheck -check-prefix=I754 -check-prefix=FUNC %s
+; RUN: llc -march=amdgcn -mcpu=fiji -verify-machineinstrs -enable-unsafe-fp-math < %s | FileCheck -check-prefix=UNSAFE-FP -check-prefix=FUNC %s
+; RUN: llc -march=r600 -mcpu=redwood < %s | FileCheck -check-prefix=R600 -check-prefix=FUNC %s
 
 ; These tests check that fdiv is expanded correctly and also test that the
 ; scheduler is scheduling the RECIP_IEEE and MUL_IEEE instructions in separate
 ; instruction groups.
 
 ; These test check that fdiv using unsafe_fp_math, coarse fp div, and IEEE754 fp div.
 
 ; FUNC-LABEL: {{^}}fdiv_f32:
-; R600-DAG: RECIP_IEEE * T{{[0-9]+\.[XYZW]}}, KC0[3].Z
-; R600-DAG: RECIP_IEEE * T{{[0-9]+\.[XYZW]}}, KC0[3].Y
-; R600-DAG: MUL_IEEE {{\** *}}T{{[0-9]+\.[XYZW]}}, KC0[3].X, PS
-; R600-DAG: MUL_IEEE {{\** *}}T{{[0-9]+\.[XYZW]}}, KC0[2].W, PS
+; R600-DAG: RECIP_IEEE * T{{[0-9]+\.[XYZW]}}, KC0[2].W
+; R600-DAG: MUL_IEEE {{\** *}}T{{[0-9]+\.[XYZW]}}, KC0[2].Z, PS
 
 ; UNSAFE-FP: v_rcp_f32
 ; UNSAFE-FP: v_mul_f32_e32
@@ -36,10 +35,8 @@
 }
 
 ; FUNC-LABEL: {{^}}fdiv_f32_fast_math:
-; R600-DAG: RECIP_IEEE * T{{[0-9]+\.[XYZW]}}, KC0[3].Z
-; R600-DAG: RECIP_IEEE * T{{[0-9]+\.[XYZW]}}, KC0[3].Y
-; R600-DAG: MUL_IEEE {{\** *}}T{{[0-9]+\.[XYZW]}}, KC0[3].X, PS
-; R600-DAG: MUL_IEEE {{\** *}}T{{[0-9]+\.[XYZW]}}, KC0[2].W, PS
+; R600-DAG: RECIP_IEEE * T{{[0-9]+\.[XYZW]}}, KC0[2].W
+; R600-DAG: MUL_IEEE {{\** *}}T{{[0-9]+\.[XYZW]}}, KC0[2].Z, PS
 
 ; UNSAFE-FP: v_rcp_f32
 ; UNSAFE-FP: v_mul_f32_e32
@@ -54,10 +51,8 @@
 }
 
 ; FUNC-LABEL: {{^}}fdiv_f32_arcp_math:
-; R600-DAG: RECIP_IEEE * T{{[0-9]+\.[XYZW]}}, KC0[3].Z
-; R600-DAG: RECIP_IEEE * T{{[0-9]+\.[XYZW]}}, KC0[3].Y
-; R600-DAG: MUL_IEEE {{\** *}}T{{[0-9]+\.[XYZW]}}, KC0[3].X, PS
-; R600-DAG: MUL_IEEE {{\** *}}T{{[0-9]+\.[XYZW]}}, KC0[2].W, PS
+; R600-DAG: RECIP_IEEE * T{{[0-9]+\.[XYZW]}}, KC0[2].W
+; R600-DAG: MUL_IEEE {{\** *}}T{{[0-9]+\.[XYZW]}}, KC0[2].Z, PS
 
 ; UNSAFE-FP: v_rcp_f32
 ; UNSAFE-FP: v_mul_f32_e32
Index: lib/Target/AMDGPU/SIISelLowering.cpp
===================================================================
--- lib/Target/AMDGPU/SIISelLowering.cpp
+++ lib/Target/AMDGPU/SIISelLowering.cpp
@@ -2067,17 +2067,13 @@
   if (SDValue FastLowered = LowerFastFDIV(Op, DAG))
     return FastLowered;
 
-  // This uses v_rcp_f32 which does not handle denormals. Let this hit a
-  // selection error for now rather than do something incorrect.
-  if (Subtarget->hasFP32Denormals())
-    return SDValue();
-
   SDLoc SL(Op);
   SDValue LHS = Op.getOperand(0);
   SDValue RHS = Op.getOperand(1);
 
   // faster 2.5 ulp fdiv when using -amdgpu-fast-fdiv flag
   if (EnableAMDGPUFastFDIV) {
+    // This does not support denormals.
     SDValue r1 = DAG.getNode(ISD::FABS, SL, MVT::f32, RHS);
 
     const APFloat K0Val(BitsToFloat(0x6f800000));
@@ -2099,6 +2095,7 @@
 
     r1 = DAG.getNode(ISD::FMUL, SL, MVT::f32, RHS, r3);
 
+    // rcp does not support denormals.
     SDValue r0 = DAG.getNode(AMDGPUISD::RCP, SL, MVT::f32, r1);
 
     SDValue Mul = DAG.getNode(ISD::FMUL, SL, MVT::f32, LHS, r0);
@@ -2114,6 +2111,7 @@
   SDValue DenominatorScaled = DAG.getNode(AMDGPUISD::DIV_SCALE, SL, ScaleVT, RHS, RHS, LHS);
   SDValue NumeratorScaled = DAG.getNode(AMDGPUISD::DIV_SCALE, SL, ScaleVT, LHS, RHS, LHS);
 
+  // Denominator is scaled to not be denormal, so using rcp is ok.
   SDValue ApproxRcp = DAG.getNode(AMDGPUISD::RCP, SL, MVT::f32, DenominatorScaled);
 
   SDValue NegDivScale0 = DAG.getNode(ISD::FNEG, SL, MVT::f32, DenominatorScaled);


-------------- next part --------------
A non-text attachment was scrubbed...
Name: D21761.61993.patch
Type: text/x-patch
Size: 4644 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20160627/9ec837a6/attachment.bin>