AMDGPU: Add clamp bit to dot builtins

[rampitec]

LGTM

cfe/trunk/include/clang/Basic/BuiltinsAMDGPU.def

 //===----------------------------------------------------------------------===//
 
 TARGET_BUILTIN(__builtin_amdgcn_fmed3h, "hhhh", "nc", "gfx9-insts")
 
 //===----------------------------------------------------------------------===//
 // Deep learning builtins.
 //===----------------------------------------------------------------------===//
 
-TARGET_BUILTIN(__builtin_amdgcn_fdot2, "fV2hV2hf", "nc", "dl-insts")
-TARGET_BUILTIN(__builtin_amdgcn_sdot2, "SiV2SsV2SsSi", "nc", "dl-insts")
-TARGET_BUILTIN(__builtin_amdgcn_udot2, "UiV2UsV2UsUi", "nc", "dl-insts")
-TARGET_BUILTIN(__builtin_amdgcn_sdot4, "SiSiSiSi", "nc", "dl-insts")
-TARGET_BUILTIN(__builtin_amdgcn_udot4, "UiUiUiUi", "nc", "dl-insts")
-TARGET_BUILTIN(__builtin_amdgcn_sdot8, "SiSiSiSi", "nc", "dl-insts")
-TARGET_BUILTIN(__builtin_amdgcn_udot8, "UiUiUiUi", "nc", "dl-insts")
+TARGET_BUILTIN(__builtin_amdgcn_fdot2, "fV2hV2hfIb", "nc", "dl-insts")
+TARGET_BUILTIN(__builtin_amdgcn_sdot2, "SiV2SsV2SsSiIb", "nc", "dl-insts")
+TARGET_BUILTIN(__builtin_amdgcn_udot2, "UiV2UsV2UsUiIb", "nc", "dl-insts")
+TARGET_BUILTIN(__builtin_amdgcn_sdot4, "SiSiSiSiIb", "nc", "dl-insts")
+TARGET_BUILTIN(__builtin_amdgcn_udot4, "UiUiUiUiIb", "nc", "dl-insts")
+TARGET_BUILTIN(__builtin_amdgcn_sdot8, "SiSiSiSiIb", "nc", "dl-insts")
+TARGET_BUILTIN(__builtin_amdgcn_udot8, "UiUiUiUiIb", "nc", "dl-insts")
 
 //===----------------------------------------------------------------------===//
 // Special builtins.
 //===----------------------------------------------------------------------===//
 BUILTIN(__builtin_amdgcn_read_exec, "LUi", "nc")
 BUILTIN(__builtin_amdgcn_read_exec_lo, "Ui", "nc")
 BUILTIN(__builtin_amdgcn_read_exec_hi, "Ui", "nc")
 

cfe/trunk/test/CodeGenOpenCL/builtins-amdgcn-dl-insts-err-clamp.cl

+// REQUIRES: amdgpu-registered-target
+
+// RUN: %clang_cc1 -triple amdgcn-unknown-unknown -target-cpu gfx906 -verify -S -emit-llvm -o - %s
+
+typedef unsigned int uint;
+typedef half __attribute__((ext_vector_type(2))) half2;
+typedef short __attribute__((ext_vector_type(2))) short2;
+typedef unsigned short __attribute__((ext_vector_type(2))) ushort2;
+
+kernel void builtins_amdgcn_dl_insts_err(
+    global float *fOut, global int *siOut, global uint *uiOut,
+    half2 v2hA, half2 v2hB, float fC,
+    short2 v2ssA, short2 v2ssB, int siA, int siB, int siC,
+    ushort2 v2usA, ushort2 v2usB, uint uiA, uint uiB, uint uiC, uint isClamp) {
+  fOut[0] = __builtin_amdgcn_fdot2(v2hA, v2hB, fC, isClamp == 0 ? false : true);     // expected-error {{'__builtin_amdgcn_fdot2' must be a constant integer}}
+
+  siOut[0] = __builtin_amdgcn_sdot2(v2ssA, v2ssB, siC, isClamp == 0 ? false : true); // expected-error {{'__builtin_amdgcn_sdot2' must be a constant integer}}
+  uiOut[0] = __builtin_amdgcn_udot2(v2usA, v2usB, uiC, isClamp == 0 ? false : true); // expected-error {{'__builtin_amdgcn_udot2' must be a constant integer}}
+
+  siOut[1] = __builtin_amdgcn_sdot4(siA, siB, siC, isClamp == 0 ? false : true);     // expected-error {{'__builtin_amdgcn_sdot4' must be a constant integer}}
+  uiOut[1] = __builtin_amdgcn_udot4(uiA, uiB, uiC, isClamp == 0 ? false : true);     // expected-error {{'__builtin_amdgcn_udot4' must be a constant integer}}
+
+  siOut[2] = __builtin_amdgcn_sdot8(siA, siB, siC, isClamp == 0 ? false : true);     // expected-error {{'__builtin_amdgcn_sdot8' must be a constant integer}}
+  uiOut[2] = __builtin_amdgcn_udot8(uiA, uiB, uiC, isClamp == 0 ? false : true);     // expected-error {{'__builtin_amdgcn_udot8' must be a constant integer}}
+}

cfe/trunk/test/CodeGenOpenCL/builtins-amdgcn-dl-insts-err.cl

 // REQUIRES: amdgpu-registered-target
 
 // RUN: %clang_cc1 -triple amdgcn-unknown-unknown -target-cpu gfx900 -verify -S -emit-llvm -o - %s
 
 typedef unsigned int uint;
 typedef half __attribute__((ext_vector_type(2))) half2;
 typedef short __attribute__((ext_vector_type(2))) short2;
 typedef unsigned short __attribute__((ext_vector_type(2))) ushort2;
 
 kernel void builtins_amdgcn_dl_insts_err(
     global float *fOut, global int *siOut, global uint *uiOut,
     half2 v2hA, half2 v2hB, float fC,
     short2 v2ssA, short2 v2ssB, int siA, int siB, int siC,
     ushort2 v2usA, ushort2 v2usB, uint uiA, uint uiB, uint uiC) {
-  fOut[0] = __builtin_amdgcn_fdot2(v2hA, v2hB, fC); // expected-error {{'__builtin_amdgcn_fdot2' needs target feature dl-insts}}
+  fOut[0] = __builtin_amdgcn_fdot2(v2hA, v2hB, fC, false);     // expected-error {{'__builtin_amdgcn_fdot2' needs target feature dl-insts}}
+  fOut[1] = __builtin_amdgcn_fdot2(v2hA, v2hB, fC, true);      // expected-error {{'__builtin_amdgcn_fdot2' needs target feature dl-insts}}
 
-  siOut[0] = __builtin_amdgcn_sdot2(v2ssA, v2ssB, siC); // expected-error {{'__builtin_amdgcn_sdot2' needs target feature dl-insts}}
-  uiOut[0] = __builtin_amdgcn_udot2(v2usA, v2usB, uiC); // expected-error {{'__builtin_amdgcn_udot2' needs target feature dl-insts}}
+  siOut[0] = __builtin_amdgcn_sdot2(v2ssA, v2ssB, siC, false); // expected-error {{'__builtin_amdgcn_sdot2' needs target feature dl-insts}}
+  siOut[1] = __builtin_amdgcn_sdot2(v2ssA, v2ssB, siC, true);  // expected-error {{'__builtin_amdgcn_sdot2' needs target feature dl-insts}}
 
-  siOut[1] = __builtin_amdgcn_sdot4(siA, siB, siC); // expected-error {{'__builtin_amdgcn_sdot4' needs target feature dl-insts}}
-  uiOut[1] = __builtin_amdgcn_udot4(uiA, uiB, uiC); // expected-error {{'__builtin_amdgcn_udot4' needs target feature dl-insts}}
+  uiOut[0] = __builtin_amdgcn_udot2(v2usA, v2usB, uiC, false); // expected-error {{'__builtin_amdgcn_udot2' needs target feature dl-insts}}
+  uiOut[1] = __builtin_amdgcn_udot2(v2usA, v2usB, uiC, true);  // expected-error {{'__builtin_amdgcn_udot2' needs target feature dl-insts}}
 
-  siOut[2] = __builtin_amdgcn_sdot8(siA, siB, siC); // expected-error {{'__builtin_amdgcn_sdot8' needs target feature dl-insts}}
-  uiOut[2] = __builtin_amdgcn_udot8(uiA, uiB, uiC); // expected-error {{'__builtin_amdgcn_udot8' needs target feature dl-insts}}
+  siOut[2] = __builtin_amdgcn_sdot4(siA, siB, siC, false);     // expected-error {{'__builtin_amdgcn_sdot4' needs target feature dl-insts}}
+  siOut[3] = __builtin_amdgcn_sdot4(siA, siB, siC, true);      // expected-error {{'__builtin_amdgcn_sdot4' needs target feature dl-insts}}
+
+  uiOut[2] = __builtin_amdgcn_udot4(uiA, uiB, uiC, false);     // expected-error {{'__builtin_amdgcn_udot4' needs target feature dl-insts}}
+  uiOut[3] = __builtin_amdgcn_udot4(uiA, uiB, uiC, true);      // expected-error {{'__builtin_amdgcn_udot4' needs target feature dl-insts}}
+
+  siOut[4] = __builtin_amdgcn_sdot8(siA, siB, siC, false);     // expected-error {{'__builtin_amdgcn_sdot8' needs target feature dl-insts}}
+  siOut[5] = __builtin_amdgcn_sdot8(siA, siB, siC, true);      // expected-error {{'__builtin_amdgcn_sdot8' needs target feature dl-insts}}
+
+  uiOut[4] = __builtin_amdgcn_udot8(uiA, uiB, uiC, false);     // expected-error {{'__builtin_amdgcn_udot8' needs target feature dl-insts}}
+  uiOut[5] = __builtin_amdgcn_udot8(uiA, uiB, uiC, true);      // expected-error {{'__builtin_amdgcn_udot8' needs target feature dl-insts}}
 }

cfe/trunk/test/CodeGenOpenCL/builtins-amdgcn-dl-insts.cl

 // REQUIRES: amdgpu-registered-target
 
 // RUN: %clang_cc1 -triple amdgcn-unknown-unknown -target-cpu gfx906 -S -emit-llvm -o - %s | FileCheck %s
 
 typedef unsigned int uint;
 typedef half __attribute__((ext_vector_type(2))) half2;
 typedef short __attribute__((ext_vector_type(2))) short2;
 typedef unsigned short __attribute__((ext_vector_type(2))) ushort2;
 
 // CHECK-LABEL: @builtins_amdgcn_dl_insts
-// CHECK: call float @llvm.amdgcn.fdot2
+// CHECK: call float @llvm.amdgcn.fdot2(<2 x half> %v2hA, <2 x half> %v2hB, float %fC, i1 false)
+// CHECK: call float @llvm.amdgcn.fdot2(<2 x half> %v2hA, <2 x half> %v2hB, float %fC, i1 true)
 
-// CHECK: call i32 @llvm.amdgcn.sdot2
-// CHECK: call i32 @llvm.amdgcn.udot2
+// CHECK: call i32 @llvm.amdgcn.sdot2(<2 x i16> %v2ssA, <2 x i16> %v2ssB, i32 %siC, i1 false)
+// CHECK: call i32 @llvm.amdgcn.sdot2(<2 x i16> %v2ssA, <2 x i16> %v2ssB, i32 %siC, i1 true)
 
-// CHECK: call i32 @llvm.amdgcn.sdot4
-// CHECK: call i32 @llvm.amdgcn.udot4
+// CHECK: call i32 @llvm.amdgcn.udot2(<2 x i16> %v2usA, <2 x i16> %v2usB, i32 %uiC, i1 false)
+// CHECK: call i32 @llvm.amdgcn.udot2(<2 x i16> %v2usA, <2 x i16> %v2usB, i32 %uiC, i1 true)
 
-// CHECK: call i32 @llvm.amdgcn.sdot8
-// CHECK: call i32 @llvm.amdgcn.udot8
+// CHECK: call i32 @llvm.amdgcn.sdot4(i32 %siA, i32 %siB, i32 %siC, i1 false)
+// CHECK: call i32 @llvm.amdgcn.sdot4(i32 %siA, i32 %siB, i32 %siC, i1 true)
+
+// CHECK: call i32 @llvm.amdgcn.udot4(i32 %uiA, i32 %uiB, i32 %uiC, i1 false)
+// CHECK: call i32 @llvm.amdgcn.udot4(i32 %uiA, i32 %uiB, i32 %uiC, i1 true)
+
+// CHECK: call i32 @llvm.amdgcn.sdot8(i32 %siA, i32 %siB, i32 %siC, i1 false)
+// CHECK: call i32 @llvm.amdgcn.sdot8(i32 %siA, i32 %siB, i32 %siC, i1 true)
+
+// CHECK: call i32 @llvm.amdgcn.udot8(i32 %uiA, i32 %uiB, i32 %uiC, i1 false)
+// CHECK: call i32 @llvm.amdgcn.udot8(i32 %uiA, i32 %uiB, i32 %uiC, i1 true)
 kernel void builtins_amdgcn_dl_insts(
     global float *fOut, global int *siOut, global uint *uiOut,
     half2 v2hA, half2 v2hB, float fC,
     short2 v2ssA, short2 v2ssB, int siA, int siB, int siC,
     ushort2 v2usA, ushort2 v2usB, uint uiA, uint uiB, uint uiC) {
-  fOut[0] = __builtin_amdgcn_fdot2(v2hA, v2hB, fC);
+  fOut[0] = __builtin_amdgcn_fdot2(v2hA, v2hB, fC, false);
+  fOut[1] = __builtin_amdgcn_fdot2(v2hA, v2hB, fC, true);
+
+  siOut[0] = __builtin_amdgcn_sdot2(v2ssA, v2ssB, siC, false);
+  siOut[1] = __builtin_amdgcn_sdot2(v2ssA, v2ssB, siC, true);
+
+  uiOut[0] = __builtin_amdgcn_udot2(v2usA, v2usB, uiC, false);
+  uiOut[1] = __builtin_amdgcn_udot2(v2usA, v2usB, uiC, true);
+
+  siOut[2] = __builtin_amdgcn_sdot4(siA, siB, siC, false);
+  siOut[3] = __builtin_amdgcn_sdot4(siA, siB, siC, true);
 
-  siOut[0] = __builtin_amdgcn_sdot2(v2ssA, v2ssB, siC);
-  uiOut[0] = __builtin_amdgcn_udot2(v2usA, v2usB, uiC);
+  uiOut[2] = __builtin_amdgcn_udot4(uiA, uiB, uiC, false);
+  uiOut[3] = __builtin_amdgcn_udot4(uiA, uiB, uiC, true);
 
-  siOut[1] = __builtin_amdgcn_sdot4(siA, siB, siC);
-  uiOut[1] = __builtin_amdgcn_udot4(uiA, uiB, uiC);
+  siOut[4] = __builtin_amdgcn_sdot8(siA, siB, siC, false);
+  siOut[5] = __builtin_amdgcn_sdot8(siA, siB, siC, true);
 
-  siOut[2] = __builtin_amdgcn_sdot8(siA, siB, siC);
-  uiOut[2] = __builtin_amdgcn_udot8(uiA, uiB, uiC);
+  uiOut[4] = __builtin_amdgcn_udot8(uiA, uiB, uiC, false);
+  uiOut[5] = __builtin_amdgcn_udot8(uiA, uiB, uiC, true);
 }