85 lines
3.9 KiB
TableGen
85 lines
3.9 KiB
TableGen
//===-- RISCVInstrInfoZfbfmin.td - 'Zfbfmin' instructions --*- tablegen -*-===//
|
|
//
|
|
// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
|
|
// See https://llvm.org/LICENSE.txt for license information.
|
|
// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
|
|
//
|
|
//===----------------------------------------------------------------------===//
|
|
//
|
|
// This file describes the RISC-V instructions from the standard 'Zfbfmin'
|
|
// extension, providing scalar conversion instructions for BFloat16.
|
|
// This version is still experimental as the 'Zfbfmin' extension hasn't been
|
|
// ratified yet.
|
|
//
|
|
//===----------------------------------------------------------------------===//
|
|
|
|
//===----------------------------------------------------------------------===//
|
|
// RISC-V specific DAG Nodes.
|
|
//===----------------------------------------------------------------------===//
|
|
|
|
def SDT_RISCVFP_ROUND_BF16
|
|
: SDTypeProfile<1, 1, [SDTCisVT<0, bf16>, SDTCisVT<1, f32>]>;
|
|
def SDT_RISCVFP_EXTEND_BF16
|
|
: SDTypeProfile<1, 1, [SDTCisVT<0, f32>, SDTCisVT<1, bf16>]>;
|
|
|
|
def riscv_fpround_bf16
|
|
: SDNode<"RISCVISD::FP_ROUND_BF16", SDT_RISCVFP_ROUND_BF16>;
|
|
def riscv_fpextend_bf16
|
|
: SDNode<"RISCVISD::FP_EXTEND_BF16", SDT_RISCVFP_EXTEND_BF16>;
|
|
|
|
//===----------------------------------------------------------------------===//
|
|
// Instructions
|
|
//===----------------------------------------------------------------------===//
|
|
|
|
let Predicates = [HasStdExtZfbfmin] in {
|
|
def FCVT_BF16_S : FPUnaryOp_r_frm<0b0100010, 0b01000, FPR16, FPR32, "fcvt.bf16.s">,
|
|
Sched<[WriteFCvtF32ToF16, ReadFCvtF32ToF16]>;
|
|
def FCVT_S_BF16 : FPUnaryOp_r_frm<0b0100000, 0b00110, FPR32, FPR16, "fcvt.s.bf16">,
|
|
Sched<[WriteFCvtF32ToF16, ReadFCvtF32ToF16]>;
|
|
} // Predicates = [HasStdExtZfbfmin]
|
|
|
|
//===----------------------------------------------------------------------===//
|
|
// Pseudo-instructions and codegen patterns
|
|
//===----------------------------------------------------------------------===//
|
|
|
|
let Predicates = [HasStdExtZfbfmin] in {
|
|
/// Loads
|
|
def : LdPat<load, FLH, bf16>;
|
|
|
|
/// Stores
|
|
def : StPat<store, FSH, FPR16, bf16>;
|
|
|
|
/// Float conversion operations
|
|
// f32 -> bf16, bf16 -> f32
|
|
def : Pat<(bf16 (riscv_fpround_bf16 FPR32:$rs1)),
|
|
(FCVT_BF16_S FPR32:$rs1, FRM_DYN)>;
|
|
def : Pat<(riscv_fpextend_bf16 (bf16 FPR16:$rs1)),
|
|
(FCVT_S_BF16 FPR16:$rs1, FRM_DYN)>;
|
|
|
|
// Moves (no conversion)
|
|
def : Pat<(bf16 (riscv_fmv_h_x GPR:$src)), (FMV_H_X GPR:$src)>;
|
|
def : Pat<(riscv_fmv_x_anyexth (bf16 FPR16:$src)), (FMV_X_H FPR16:$src)>;
|
|
def : Pat<(riscv_fmv_x_signexth (bf16 FPR16:$src)), (FMV_X_H FPR16:$src)>;
|
|
} // Predicates = [HasStdExtZfbfmin]
|
|
|
|
let Predicates = [HasStdExtZfbfmin] in {
|
|
// bf16->[u]int. Round-to-zero must be used for the f32->int step, the
|
|
// rounding mode has no effect for bf16->f32.
|
|
def : Pat<(i32 (any_fp_to_sint (bf16 FPR16:$rs1))), (FCVT_W_S (FCVT_S_BF16 $rs1, FRM_RNE), FRM_RTZ)>;
|
|
def : Pat<(i32 (any_fp_to_uint (bf16 FPR16:$rs1))), (FCVT_WU_S (FCVT_S_BF16 $rs1, FRM_RNE), FRM_RTZ)>;
|
|
|
|
// [u]int->bf16. Match GCC and default to using dynamic rounding mode.
|
|
def : Pat<(bf16 (any_sint_to_fp (i32 GPR:$rs1))), (FCVT_BF16_S (FCVT_S_W $rs1, FRM_DYN), FRM_DYN)>;
|
|
def : Pat<(bf16 (any_uint_to_fp (i32 GPR:$rs1))), (FCVT_BF16_S (FCVT_S_WU $rs1, FRM_DYN), FRM_DYN)>;
|
|
}
|
|
|
|
let Predicates = [HasStdExtZfbfmin, IsRV64] in {
|
|
// bf16->[u]int64. Round-to-zero must be used for the f32->int step, the
|
|
// rounding mode has no effect for bf16->f32.
|
|
def : Pat<(i64 (any_fp_to_sint (bf16 FPR16:$rs1))), (FCVT_L_S (FCVT_S_BF16 $rs1, FRM_RNE), FRM_RTZ)>;
|
|
def : Pat<(i64 (any_fp_to_uint (bf16 FPR16:$rs1))), (FCVT_LU_S (FCVT_S_BF16 $rs1, FRM_RNE), FRM_RTZ)>;
|
|
|
|
// [u]int->bf16. Match GCC and default to using dynamic rounding mode.
|
|
def : Pat<(bf16 (any_sint_to_fp (i64 GPR:$rs1))), (FCVT_BF16_S (FCVT_S_L $rs1, FRM_DYN), FRM_DYN)>;
|
|
def : Pat<(bf16 (any_uint_to_fp (i64 GPR:$rs1))), (FCVT_BF16_S (FCVT_S_LU $rs1, FRM_DYN), FRM_DYN)>;
|
|
}
|