R600/SI: Expand fract to floor, then only select V_FRACT on CI

V_FRACT is buggy on SI.

R600-specific code is left intact.

v2: drop the multiclass, use complex VOP3 patterns

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@233075 91177308-0d34-0410-b5e6-96231b3b80d8

diff --git a/lib/Target/R600/AMDGPUISelLowering.cpp b/lib/Target/R600/AMDGPUISelLowering.cpp
index 4707279ce87087837104dbcb20a90131ae3f6751..62a33fad64bcecd7e433a9479740cb4f6a6cf903 100644 (file)
--- a/lib/Target/R600/AMDGPUISelLowering.cpp
+++ b/lib/Target/R600/AMDGPUISelLowering.cpp
@@ -885,9 +885,6 @@ SDValue AMDGPUTargetLowering::LowerINTRINSIC_WO_CHAIN(SDValue Op,
       return LowerIntrinsicIABS(Op, DAG);
     case AMDGPUIntrinsic::AMDGPU_lrp:
       return LowerIntrinsicLRP(Op, DAG);
-    case AMDGPUIntrinsic::AMDGPU_fract:
-    case AMDGPUIntrinsic::AMDIL_fraction: // Legacy name.
-      return DAG.getNode(AMDGPUISD::FRACT, DL, VT, Op.getOperand(1));
 
     case AMDGPUIntrinsic::AMDGPU_clamp:
     case AMDGPUIntrinsic::AMDIL_clamp: // Legacy name.

diff --git a/lib/Target/R600/R600ISelLowering.cpp b/lib/Target/R600/R600ISelLowering.cpp
index d4f3145e6aef517dd4385b38e1b94e6e773a8a91..06b721796763f3bb89ff950f714718cd9b6807e1 100644 (file)
--- a/lib/Target/R600/R600ISelLowering.cpp
+++ b/lib/Target/R600/R600ISelLowering.cpp
@@ -837,6 +837,10 @@ SDValue R600TargetLowering::LowerOperation(SDValue Op, SelectionDAG &DAG) const
     case Intrinsic::AMDGPU_rsq:
       // XXX - I'm assuming SI's RSQ_LEGACY matches R600's behavior.
       return DAG.getNode(AMDGPUISD::RSQ_LEGACY, DL, VT, Op.getOperand(1));
+
+    case AMDGPUIntrinsic::AMDGPU_fract:
+    case AMDGPUIntrinsic::AMDIL_fraction: // Legacy name.
+      return DAG.getNode(AMDGPUISD::FRACT, DL, VT, Op.getOperand(1));
     }
     // break out of case ISD::INTRINSIC_WO_CHAIN in switch(Op.getOpcode())
     break;

diff --git a/lib/Target/R600/SIISelLowering.cpp b/lib/Target/R600/SIISelLowering.cpp
index 16d67022aa88928d86ef35f62a7492132cdb5f07..01fece2694ff8d0a1a53e9f14e985d049d32fde0 100644 (file)
--- a/lib/Target/R600/SIISelLowering.cpp
+++ b/lib/Target/R600/SIISelLowering.cpp
@@ -928,6 +928,12 @@ SDValue SITargetLowering::LowerINTRINSIC_WO_CHAIN(SDValue Op,
                        Op.getOperand(1),
                        Op.getOperand(2),
                        Op.getOperand(3));
+
+  case AMDGPUIntrinsic::AMDGPU_fract:
+  case AMDGPUIntrinsic::AMDIL_fraction: // Legacy name.
+    return DAG.getNode(ISD::FSUB, DL, VT, Op.getOperand(1),
+                       DAG.getNode(ISD::FFLOOR, DL, VT, Op.getOperand(1)));
+
   default:
     return AMDGPUTargetLowering::LowerOperation(Op, DAG);
   }

diff --git a/lib/Target/R600/SIInstructions.td b/lib/Target/R600/SIInstructions.td
index 10a7aa30eccb07cd95439c35fb98019722a25d69..b06955c83318e6fc68d96f8f011c6b8112fa6753 100644 (file)
--- a/lib/Target/R600/SIInstructions.td
+++ b/lib/Target/R600/SIInstructions.td
@@ -3316,6 +3316,28 @@ def : Pat <
   (V_CNDMASK_B32_e64 $src0, $src1, $src2)
 >;
 
+//===----------------------------------------------------------------------===//
+// Fract Patterns
+//===----------------------------------------------------------------------===//
+
+let Predicates = [isCI] in {
+
+// Convert (x - floor(x)) to fract(x)
+def : Pat <
+  (f32 (fsub (f32 (VOP3Mods f32:$x, i32:$mods)),
+             (f32 (ffloor (f32 (VOP3Mods f32:$x, i32:$mods)))))),
+  (V_FRACT_F32_e64 $mods, $x, DSTCLAMP.NONE, DSTOMOD.NONE)
+>;
+
+// Convert (x + (-floor(x))) to fract(x)
+def : Pat <
+  (f64 (fadd (f64 (VOP3Mods f64:$x, i32:$mods)),
+             (f64 (fneg (f64 (ffloor (f64 (VOP3Mods f64:$x, i32:$mods)))))))),
+  (V_FRACT_F64_e64 $mods, $x, DSTCLAMP.NONE, DSTOMOD.NONE)
+>;
+
+} // End Predicates = [isCI]
+
 //============================================================================//
 // Miscellaneous Optimization Patterns
 //============================================================================//

diff --git a/test/CodeGen/R600/llvm.AMDGPU.fract.ll b/test/CodeGen/R600/llvm.AMDGPU.fract.ll
index f4cf7fc293d3e9f860f0ba741664240126c3fb3a..7501b4b754657ed963d7221d54ac1603aa37673d 100644 (file)
--- a/test/CodeGen/R600/llvm.AMDGPU.fract.ll
+++ b/test/CodeGen/R600/llvm.AMDGPU.fract.ll
@@ -1,14 +1,19 @@
-; RUN: llc -march=amdgcn -mcpu=SI -verify-machineinstrs < %s | FileCheck -check-prefix=SI -check-prefix=FUNC %s
-; RUN: llc -march=amdgcn -mcpu=tonga -verify-machineinstrs < %s | FileCheck -check-prefix=SI -check-prefix=FUNC %s
-; RUN: llc -march=r600 -mcpu=cypress -verify-machineinstrs < %s | FileCheck -check-prefix=EG -check-prefix=FUNC %s
+; RUN: llc -march=amdgcn -mcpu=SI -verify-machineinstrs -enable-unsafe-fp-math < %s | FileCheck -check-prefix=GCN -check-prefix=SI -check-prefix=FUNC %s
+; RUN: llc -march=amdgcn -mcpu=bonaire -verify-machineinstrs -enable-unsafe-fp-math < %s | FileCheck -check-prefix=GCN -check-prefix=CI -check-prefix=FUNC %s
+; RUN: llc -march=amdgcn -mcpu=tonga -verify-machineinstrs -enable-unsafe-fp-math < %s | FileCheck -check-prefix=GCN -check-prefix=CI -check-prefix=FUNC %s
+; RUN: llc -march=r600 -mcpu=cypress -verify-machineinstrs -enable-unsafe-fp-math < %s | FileCheck -check-prefix=EG -check-prefix=FUNC %s
 
+declare float @llvm.fabs.f32(float  %Val)
 declare float @llvm.AMDGPU.fract.f32(float) nounwind readnone
 
 ; Legacy name
 declare float @llvm.AMDIL.fraction.f32(float) nounwind readnone
 
 ; FUNC-LABEL: {{^}}fract_f32:
-; SI: v_fract_f32
+; CI: v_fract_f32_e32 [[RESULT:v[0-9]+]], [[INPUT:v[0-9]+]]
+; SI: v_floor_f32_e32 [[FLR:v[0-9]+]], [[INPUT:v[0-9]+]]
+; SI: v_subrev_f32_e32 [[RESULT:v[0-9]+]], [[FLR]], [[INPUT]]
+; GCN: buffer_store_dword [[RESULT]]
 ; EG: FRACT
 define void @fract_f32(float addrspace(1)* %out, float addrspace(1)* %src) nounwind {
   %val = load float, float addrspace(1)* %src, align 4
@@ -18,7 +23,10 @@ define void @fract_f32(float addrspace(1)* %out, float addrspace(1)* %src) nounw
 }
 
 ; FUNC-LABEL: {{^}}fract_f32_legacy_amdil:
-; SI: v_fract_f32
+; CI: v_fract_f32_e32 [[RESULT:v[0-9]+]], [[INPUT:v[0-9]+]]
+; SI: v_floor_f32_e32 [[FLR:v[0-9]+]], [[INPUT:v[0-9]+]]
+; SI: v_subrev_f32_e32 [[RESULT:v[0-9]+]], [[FLR]], [[INPUT]]
+; GCN: buffer_store_dword [[RESULT]]
 ; EG: FRACT
 define void @fract_f32_legacy_amdil(float addrspace(1)* %out, float addrspace(1)* %src) nounwind {
   %val = load float, float addrspace(1)* %src, align 4
@@ -26,3 +34,32 @@ define void @fract_f32_legacy_amdil(float addrspace(1)* %out, float addrspace(1)
   store float %fract, float addrspace(1)* %out, align 4
   ret void
 }
+
+; FUNC-LABEL: {{^}}fract_f32_neg:
+; CI: v_fract_f32_e64 [[RESULT:v[0-9]+]], -[[INPUT:v[0-9]+]]
+; SI: v_floor_f32_e64 [[FLR:v[0-9]+]], -[[INPUT:v[0-9]+]]
+; SI: v_sub_f32_e64 [[RESULT:v[0-9]+]], -[[INPUT]], [[FLR]]
+; GCN: buffer_store_dword [[RESULT]]
+; EG: FRACT
+define void @fract_f32_neg(float addrspace(1)* %out, float addrspace(1)* %src) nounwind {
+  %val = load float, float addrspace(1)* %src, align 4
+  %neg = fsub float 0.0, %val
+  %fract = call float @llvm.AMDGPU.fract.f32(float %neg) nounwind readnone
+  store float %fract, float addrspace(1)* %out, align 4
+  ret void
+}
+
+; FUNC-LABEL: {{^}}fract_f32_neg_abs:
+; CI: v_fract_f32_e64 [[RESULT:v[0-9]+]], -|[[INPUT:v[0-9]+]]|
+; SI: v_floor_f32_e64 [[FLR:v[0-9]+]], -|[[INPUT:v[0-9]+]]|
+; SI: v_sub_f32_e64 [[RESULT:v[0-9]+]], -|[[INPUT]]|, [[FLR]]
+; GCN: buffer_store_dword [[RESULT]]
+; EG: FRACT
+define void @fract_f32_neg_abs(float addrspace(1)* %out, float addrspace(1)* %src) nounwind {
+  %val = load float, float addrspace(1)* %src, align 4
+  %abs = call float @llvm.fabs.f32(float %val)
+  %neg = fsub float 0.0, %abs
+  %fract = call float @llvm.AMDGPU.fract.f32(float %neg) nounwind readnone
+  store float %fract, float addrspace(1)* %out, align 4
+  ret void
+}