llvm-6502/lib/Target/R600/SIIntrinsics.td
Marek Olsak f286d63757 R600/SI: add gather4 and getlod intrinsics (v3)
This contains all the previous patches + getlod support on top of it.
It doesn't use SDNodes anymore, so it's quite small.
It also adds v16i8 to SReg_128, which is used for the sampler descriptor.

Reviewed-by: Tom Stellard

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@211228 91177308-0d34-0410-b5e6-96231b3b80d8
2014-06-18 22:00:29 +00:00

133 lines
5.2 KiB
TableGen

//===-- SIIntrinsics.td - SI Intrinsic defs ----------------*- tablegen -*-===//
//
// The LLVM Compiler Infrastructure
//
// This file is distributed under the University of Illinois Open Source
// License. See LICENSE.TXT for details.
//
//===----------------------------------------------------------------------===//
//
// SI Intrinsic Definitions
//
//===----------------------------------------------------------------------===//
let TargetPrefix = "SI", isTarget = 1 in {
def int_SI_tid : Intrinsic <[llvm_i32_ty], [], [IntrNoMem]>;
def int_SI_packf16 : Intrinsic <[llvm_i32_ty], [llvm_float_ty, llvm_float_ty], [IntrNoMem]>;
def int_SI_export : Intrinsic <[], [llvm_i32_ty, llvm_i32_ty, llvm_i32_ty, llvm_i32_ty, llvm_i32_ty, llvm_float_ty, llvm_float_ty, llvm_float_ty, llvm_float_ty], []>;
def int_SI_load_const : Intrinsic <[llvm_float_ty], [llvm_anyint_ty, llvm_i32_ty], [IntrNoMem]>;
def int_SI_vs_load_input : Intrinsic <[llvm_v4f32_ty], [llvm_anyint_ty, llvm_i16_ty, llvm_i32_ty], [IntrNoMem]> ;
// Fully-flexible TBUFFER_STORE_FORMAT_* except for the ADDR64 bit, which is not exposed
def int_SI_tbuffer_store : Intrinsic <
[],
[llvm_anyint_ty, // rsrc(SGPR)
llvm_anyint_ty, // vdata(VGPR), overloaded for types i32, v2i32, v4i32
llvm_i32_ty, // num_channels(imm), selects opcode suffix: 1=X, 2=XY, 3=XYZ, 4=XYZW
llvm_i32_ty, // vaddr(VGPR)
llvm_i32_ty, // soffset(SGPR)
llvm_i32_ty, // inst_offset(imm)
llvm_i32_ty, // dfmt(imm)
llvm_i32_ty, // nfmt(imm)
llvm_i32_ty, // offen(imm)
llvm_i32_ty, // idxen(imm)
llvm_i32_ty, // glc(imm)
llvm_i32_ty, // slc(imm)
llvm_i32_ty], // tfe(imm)
[]>;
// Fully-flexible BUFFER_LOAD_DWORD_* except for the ADDR64 bit, which is not exposed
def int_SI_buffer_load_dword : Intrinsic <
[llvm_anyint_ty], // vdata(VGPR), overloaded for types i32, v2i32, v4i32
[llvm_anyint_ty, // rsrc(SGPR)
llvm_anyint_ty, // vaddr(VGPR)
llvm_i32_ty, // soffset(SGPR)
llvm_i32_ty, // inst_offset(imm)
llvm_i32_ty, // offen(imm)
llvm_i32_ty, // idxen(imm)
llvm_i32_ty, // glc(imm)
llvm_i32_ty, // slc(imm)
llvm_i32_ty], // tfe(imm)
[IntrReadArgMem]>;
def int_SI_sendmsg : Intrinsic <[], [llvm_i32_ty, llvm_i32_ty], [IntrNoMem]>;
class Sample : Intrinsic <[llvm_v4f32_ty], [llvm_anyvector_ty, llvm_v32i8_ty, llvm_anyint_ty, llvm_i32_ty], [IntrNoMem]>;
// Fully-flexible SAMPLE instruction.
class SampleRaw : Intrinsic <
[llvm_v4f32_ty], // vdata(VGPR)
[llvm_anyint_ty, // vaddr(VGPR)
llvm_v32i8_ty, // rsrc(SGPR)
llvm_v16i8_ty, // sampler(SGPR)
llvm_i32_ty, // dmask(imm)
llvm_i32_ty, // unorm(imm)
llvm_i32_ty, // r128(imm)
llvm_i32_ty, // da(imm)
llvm_i32_ty, // glc(imm)
llvm_i32_ty, // slc(imm)
llvm_i32_ty, // tfe(imm)
llvm_i32_ty], // lwe(imm)
[IntrNoMem]>;
def int_SI_sample : Sample;
def int_SI_sampleb : Sample;
def int_SI_sampled : Sample;
def int_SI_samplel : Sample;
// Basic gather4
def int_SI_gather4 : SampleRaw;
def int_SI_gather4_cl : SampleRaw;
def int_SI_gather4_l : SampleRaw;
def int_SI_gather4_b : SampleRaw;
def int_SI_gather4_b_cl : SampleRaw;
def int_SI_gather4_lz : SampleRaw;
// Gather4 with comparison
def int_SI_gather4_c : SampleRaw;
def int_SI_gather4_c_cl : SampleRaw;
def int_SI_gather4_c_l : SampleRaw;
def int_SI_gather4_c_b : SampleRaw;
def int_SI_gather4_c_b_cl : SampleRaw;
def int_SI_gather4_c_lz : SampleRaw;
// Gather4 with offsets
def int_SI_gather4_o : SampleRaw;
def int_SI_gather4_cl_o : SampleRaw;
def int_SI_gather4_l_o : SampleRaw;
def int_SI_gather4_b_o : SampleRaw;
def int_SI_gather4_b_cl_o : SampleRaw;
def int_SI_gather4_lz_o : SampleRaw;
// Gather4 with comparison and offsets
def int_SI_gather4_c_o : SampleRaw;
def int_SI_gather4_c_cl_o : SampleRaw;
def int_SI_gather4_c_l_o : SampleRaw;
def int_SI_gather4_c_b_o : SampleRaw;
def int_SI_gather4_c_b_cl_o : SampleRaw;
def int_SI_gather4_c_lz_o : SampleRaw;
def int_SI_getlod : SampleRaw;
def int_SI_imageload : Intrinsic <[llvm_v4i32_ty], [llvm_anyvector_ty, llvm_v32i8_ty, llvm_i32_ty], [IntrNoMem]>;
def int_SI_resinfo : Intrinsic <[llvm_v4i32_ty], [llvm_i32_ty, llvm_v32i8_ty, llvm_i32_ty], [IntrNoMem]>;
/* Interpolation Intrinsics */
def int_SI_fs_constant : Intrinsic <[llvm_float_ty], [llvm_i32_ty, llvm_i32_ty, llvm_i32_ty], [IntrNoMem]>;
def int_SI_fs_interp : Intrinsic <[llvm_float_ty], [llvm_i32_ty, llvm_i32_ty, llvm_i32_ty, llvm_v2i32_ty], [IntrNoMem]>;
/* Control flow Intrinsics */
def int_SI_if : Intrinsic<[llvm_i64_ty], [llvm_i1_ty, llvm_empty_ty], []>;
def int_SI_else : Intrinsic<[llvm_i64_ty], [llvm_i64_ty, llvm_empty_ty], []>;
def int_SI_break : Intrinsic<[llvm_i64_ty], [llvm_i64_ty], []>;
def int_SI_if_break : Intrinsic<[llvm_i64_ty], [llvm_i1_ty, llvm_i64_ty], []>;
def int_SI_else_break : Intrinsic<[llvm_i64_ty], [llvm_i64_ty, llvm_i64_ty], []>;
def int_SI_loop : Intrinsic<[], [llvm_i64_ty, llvm_empty_ty], []>;
def int_SI_end_cf : Intrinsic<[], [llvm_i64_ty], []>;
}