summaryrefslogtreecommitdiff
path: root/llvm/lib/Target/AMDGPU/AMDGPUInstrInfo.td
diff options
context:
space:
mode:
Diffstat (limited to 'llvm/lib/Target/AMDGPU/AMDGPUInstrInfo.td')
-rw-r--r--llvm/lib/Target/AMDGPU/AMDGPUInstrInfo.td9
1 files changed, 7 insertions, 2 deletions
diff --git a/llvm/lib/Target/AMDGPU/AMDGPUInstrInfo.td b/llvm/lib/Target/AMDGPU/AMDGPUInstrInfo.td
index 894677ec68b6..0f9cb712f820 100644
--- a/llvm/lib/Target/AMDGPU/AMDGPUInstrInfo.td
+++ b/llvm/lib/Target/AMDGPU/AMDGPUInstrInfo.td
@@ -132,7 +132,6 @@ def AMDGPUpknorm_u16_f32_impl : SDNode<"AMDGPUISD::CVT_PKNORM_U16_F32", AMDGPUFP
def AMDGPUpk_i16_i32_impl : SDNode<"AMDGPUISD::CVT_PK_I16_I32", AMDGPUIntPackOp>;
def AMDGPUpk_u16_u32_impl : SDNode<"AMDGPUISD::CVT_PK_U16_U32", AMDGPUIntPackOp>;
def AMDGPUfp_to_f16 : SDNode<"AMDGPUISD::FP_TO_FP16" , SDTFPToIntOp>;
-def AMDGPUfp16_zext : SDNode<"AMDGPUISD::FP16_ZEXT" , SDTFPToIntOp>;
def AMDGPUfp_class_impl : SDNode<"AMDGPUISD::FP_CLASS", AMDGPUFPClassOp>;
@@ -213,6 +212,8 @@ def AMDGPUcvt_f32_ubyte2 : SDNode<"AMDGPUISD::CVT_F32_UBYTE2",
def AMDGPUcvt_f32_ubyte3 : SDNode<"AMDGPUISD::CVT_F32_UBYTE3",
SDTIntToFPOp, []>;
+def AMDGPUcvt_pk_i16_i32 : SDNode<"AMDGPUISD::CVT_PK_I16_I32",
+ AMDGPUIntPackOp, []>;
// urecip - This operation is a helper for integer division, it returns the
// result of 1 / a as a fractional unsigned integer.
@@ -311,7 +312,7 @@ def AMDGPUfdot2_impl : SDNode<"AMDGPUISD::FDOT2",
SDTCisInt<4>]>,
[]>;
-def AMDGPUperm : SDNode<"AMDGPUISD::PERM", AMDGPUDTIntTernaryOp, []>;
+def AMDGPUperm_impl : SDNode<"AMDGPUISD::PERM", AMDGPUDTIntTernaryOp, []>;
// SI+ export
def AMDGPUExportOp : SDTypeProfile<0, 8, [
@@ -461,3 +462,7 @@ def AMDGPUfdot2 : PatFrags<(ops node:$src0, node:$src1, node:$src2, node:$clamp)
def AMDGPUdiv_fmas : PatFrags<(ops node:$src0, node:$src1, node:$src2, node:$vcc),
[(int_amdgcn_div_fmas node:$src0, node:$src1, node:$src2, node:$vcc),
(AMDGPUdiv_fmas_impl node:$src0, node:$src1, node:$src2, node:$vcc)]>;
+
+def AMDGPUperm : PatFrags<(ops node:$src0, node:$src1, node:$src2),
+ [(int_amdgcn_perm node:$src0, node:$src1, node:$src2),
+ (AMDGPUperm_impl node:$src0, node:$src1, node:$src2)]>;