annotate llvm/docs/AMDGPU/AMDGPUAsmGFX908.rst @ 235:edfff9242030 cbc-llvm13

...
author Shinji KONO <kono@ie.u-ryukyu.ac.jp>
date Wed, 21 Jul 2021 11:30:30 +0900
parents 2e18cbf3894f
children c4bab56944e8
Ignore whitespace changes - Everywhere: Within whitespace: At end of lines:
rev   line source
150
anatofuz
parents:
diff changeset
1 ..
anatofuz
parents:
diff changeset
2 **************************************************
anatofuz
parents:
diff changeset
3 * *
anatofuz
parents:
diff changeset
4 * Automatically generated file, do not edit! *
anatofuz
parents:
diff changeset
5 * *
anatofuz
parents:
diff changeset
6 **************************************************
anatofuz
parents:
diff changeset
7
anatofuz
parents:
diff changeset
8 ====================================================================================
anatofuz
parents:
diff changeset
9 Syntax of gfx908 Instructions
anatofuz
parents:
diff changeset
10 ====================================================================================
anatofuz
parents:
diff changeset
11
anatofuz
parents:
diff changeset
12 .. contents::
anatofuz
parents:
diff changeset
13 :local:
anatofuz
parents:
diff changeset
14
anatofuz
parents:
diff changeset
15 Introduction
anatofuz
parents:
diff changeset
16 ============
anatofuz
parents:
diff changeset
17
anatofuz
parents:
diff changeset
18 This document describes the syntax of *instructions specific to gfx908*.
anatofuz
parents:
diff changeset
19
anatofuz
parents:
diff changeset
20 For a description of other gfx908 instructions see :doc:`Syntax of Core GFX9 Instructions<AMDGPUAsmGFX9>`.
anatofuz
parents:
diff changeset
21
anatofuz
parents:
diff changeset
22 Notation
anatofuz
parents:
diff changeset
23 ========
anatofuz
parents:
diff changeset
24
anatofuz
parents:
diff changeset
25 Notation used in this document is explained :ref:`here<amdgpu_syn_instruction_notation>`.
anatofuz
parents:
diff changeset
26
anatofuz
parents:
diff changeset
27 Overview
anatofuz
parents:
diff changeset
28 ========
anatofuz
parents:
diff changeset
29
anatofuz
parents:
diff changeset
30 An overview of generic syntax and other features of AMDGPU instructions may be found :ref:`in this document<amdgpu_syn_instructions>`.
anatofuz
parents:
diff changeset
31
anatofuz
parents:
diff changeset
32 Instructions
anatofuz
parents:
diff changeset
33 ============
anatofuz
parents:
diff changeset
34
anatofuz
parents:
diff changeset
35
anatofuz
parents:
diff changeset
36 FLAT
anatofuz
parents:
diff changeset
37 -----------------------
anatofuz
parents:
diff changeset
38
anatofuz
parents:
diff changeset
39 .. parsed-literal::
anatofuz
parents:
diff changeset
40
anatofuz
parents:
diff changeset
41 **INSTRUCTION** **DST** **SRC0** **SRC1** **SRC2** **MODIFIERS**
anatofuz
parents:
diff changeset
42 \ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|
anatofuz
parents:
diff changeset
43 global_atomic_add_f32 :ref:`vdst<amdgpu_synid908_dst_flat_atomic32>`::ref:`opt<amdgpu_synid908_opt>`, :ref:`vaddr<amdgpu_synid908_vaddr_flat_global>`, :ref:`vdata<amdgpu_synid908_vdata32_0>`, :ref:`saddr<amdgpu_synid908_saddr_flat_global>` :ref:`offset13s<amdgpu_synid_flat_offset13s>` :ref:`slc<amdgpu_synid_slc>`
anatofuz
parents:
diff changeset
44 global_atomic_pk_add_f16 :ref:`vdst<amdgpu_synid908_dst_flat_atomic32>`::ref:`opt<amdgpu_synid908_opt>`::ref:`f16x2<amdgpu_synid908_type_dev>`, :ref:`vaddr<amdgpu_synid908_vaddr_flat_global>`, :ref:`vdata<amdgpu_synid908_vdata32_0>`::ref:`f16x2<amdgpu_synid908_type_dev>`, :ref:`saddr<amdgpu_synid908_saddr_flat_global>` :ref:`offset13s<amdgpu_synid_flat_offset13s>` :ref:`slc<amdgpu_synid_slc>`
anatofuz
parents:
diff changeset
45
anatofuz
parents:
diff changeset
46 MUBUF
anatofuz
parents:
diff changeset
47 -----------------------
anatofuz
parents:
diff changeset
48
anatofuz
parents:
diff changeset
49 .. parsed-literal::
anatofuz
parents:
diff changeset
50
anatofuz
parents:
diff changeset
51 **INSTRUCTION** **SRC0** **SRC1** **SRC2** **SRC3** **MODIFIERS**
anatofuz
parents:
diff changeset
52 \ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|
anatofuz
parents:
diff changeset
53 buffer_atomic_add_f32 :ref:`vdata<amdgpu_synid908_data_buf_atomic32>`::ref:`dst<amdgpu_synid908_ret>`, :ref:`vaddr<amdgpu_synid908_addr_buf>`, :ref:`srsrc<amdgpu_synid908_rsrc_buf>`, :ref:`soffset<amdgpu_synid908_offset_buf>` :ref:`idxen<amdgpu_synid_idxen>` :ref:`offen<amdgpu_synid_offen>` :ref:`offset12<amdgpu_synid_buf_offset12>` :ref:`slc<amdgpu_synid_slc>`
anatofuz
parents:
diff changeset
54 buffer_atomic_pk_add_f16 :ref:`vdata<amdgpu_synid908_data_buf_atomic32>`::ref:`dst<amdgpu_synid908_ret>`::ref:`f16x2<amdgpu_synid908_type_dev>`, :ref:`vaddr<amdgpu_synid908_addr_buf>`, :ref:`srsrc<amdgpu_synid908_rsrc_buf>`, :ref:`soffset<amdgpu_synid908_offset_buf>` :ref:`idxen<amdgpu_synid_idxen>` :ref:`offen<amdgpu_synid_offen>` :ref:`offset12<amdgpu_synid_buf_offset12>` :ref:`slc<amdgpu_synid_slc>`
anatofuz
parents:
diff changeset
55
anatofuz
parents:
diff changeset
56 VOP2
anatofuz
parents:
diff changeset
57 -----------------------
anatofuz
parents:
diff changeset
58
anatofuz
parents:
diff changeset
59 .. parsed-literal::
anatofuz
parents:
diff changeset
60
anatofuz
parents:
diff changeset
61 **INSTRUCTION** **DST** **SRC0** **SRC1** **MODIFIERS**
anatofuz
parents:
diff changeset
62 \ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|
anatofuz
parents:
diff changeset
63 v_dot2c_f32_f16 :ref:`vdst<amdgpu_synid908_vdst32_0>`, :ref:`src0<amdgpu_synid908_src32_0>`::ref:`f16x2<amdgpu_synid908_type_dev>`, :ref:`vsrc1<amdgpu_synid908_vsrc32_0>`::ref:`f16x2<amdgpu_synid908_type_dev>`
anatofuz
parents:
diff changeset
64 v_dot2c_f32_f16_dpp :ref:`vdst<amdgpu_synid908_vdst32_0>`, :ref:`vsrc0<amdgpu_synid908_vsrc32_0>`::ref:`f16x2<amdgpu_synid908_type_dev>`, :ref:`vsrc1<amdgpu_synid908_vsrc32_0>`::ref:`f16x2<amdgpu_synid908_type_dev>` :ref:`dpp_ctrl<amdgpu_synid_dpp_ctrl>` :ref:`row_mask<amdgpu_synid_row_mask>` :ref:`bank_mask<amdgpu_synid_bank_mask>` :ref:`bound_ctrl<amdgpu_synid_bound_ctrl>`
anatofuz
parents:
diff changeset
65 v_dot2c_i32_i16 :ref:`vdst<amdgpu_synid908_vdst32_0>`, :ref:`src0<amdgpu_synid908_src32_0>`::ref:`i16x2<amdgpu_synid908_type_dev>`, :ref:`vsrc1<amdgpu_synid908_vsrc32_0>`::ref:`i16x2<amdgpu_synid908_type_dev>`
anatofuz
parents:
diff changeset
66 v_dot2c_i32_i16_dpp :ref:`vdst<amdgpu_synid908_vdst32_0>`, :ref:`vsrc0<amdgpu_synid908_vsrc32_0>`::ref:`i16x2<amdgpu_synid908_type_dev>`, :ref:`vsrc1<amdgpu_synid908_vsrc32_0>`::ref:`i16x2<amdgpu_synid908_type_dev>` :ref:`dpp_ctrl<amdgpu_synid_dpp_ctrl>` :ref:`row_mask<amdgpu_synid_row_mask>` :ref:`bank_mask<amdgpu_synid_bank_mask>` :ref:`bound_ctrl<amdgpu_synid_bound_ctrl>`
anatofuz
parents:
diff changeset
67 v_dot4c_i32_i8 :ref:`vdst<amdgpu_synid908_vdst32_0>`, :ref:`src0<amdgpu_synid908_src32_0>`::ref:`i8x4<amdgpu_synid908_type_dev>`, :ref:`vsrc1<amdgpu_synid908_vsrc32_0>`::ref:`i8x4<amdgpu_synid908_type_dev>`
anatofuz
parents:
diff changeset
68 v_dot4c_i32_i8_dpp :ref:`vdst<amdgpu_synid908_vdst32_0>`, :ref:`vsrc0<amdgpu_synid908_vsrc32_0>`::ref:`i8x4<amdgpu_synid908_type_dev>`, :ref:`vsrc1<amdgpu_synid908_vsrc32_0>`::ref:`i8x4<amdgpu_synid908_type_dev>` :ref:`dpp_ctrl<amdgpu_synid_dpp_ctrl>` :ref:`row_mask<amdgpu_synid_row_mask>` :ref:`bank_mask<amdgpu_synid_bank_mask>` :ref:`bound_ctrl<amdgpu_synid_bound_ctrl>`
anatofuz
parents:
diff changeset
69 v_dot8c_i32_i4 :ref:`vdst<amdgpu_synid908_vdst32_0>`, :ref:`src0<amdgpu_synid908_src32_0>`::ref:`i4x8<amdgpu_synid908_type_dev>`, :ref:`vsrc1<amdgpu_synid908_vsrc32_0>`::ref:`i4x8<amdgpu_synid908_type_dev>`
anatofuz
parents:
diff changeset
70 v_dot8c_i32_i4_dpp :ref:`vdst<amdgpu_synid908_vdst32_0>`, :ref:`vsrc0<amdgpu_synid908_vsrc32_0>`::ref:`i4x8<amdgpu_synid908_type_dev>`, :ref:`vsrc1<amdgpu_synid908_vsrc32_0>`::ref:`i4x8<amdgpu_synid908_type_dev>` :ref:`dpp_ctrl<amdgpu_synid_dpp_ctrl>` :ref:`row_mask<amdgpu_synid_row_mask>` :ref:`bank_mask<amdgpu_synid_bank_mask>` :ref:`bound_ctrl<amdgpu_synid_bound_ctrl>`
anatofuz
parents:
diff changeset
71 v_fmac_f32 :ref:`vdst<amdgpu_synid908_vdst32_0>`, :ref:`src0<amdgpu_synid908_src32_0>`, :ref:`vsrc1<amdgpu_synid908_vsrc32_0>`
anatofuz
parents:
diff changeset
72 v_fmac_f32_dpp :ref:`vdst<amdgpu_synid908_vdst32_0>`, :ref:`vsrc0<amdgpu_synid908_vsrc32_0>`::ref:`m<amdgpu_synid908_mod_dpp_sdwa_abs_neg>`, :ref:`vsrc1<amdgpu_synid908_vsrc32_0>`::ref:`m<amdgpu_synid908_mod_dpp_sdwa_abs_neg>` :ref:`dpp_ctrl<amdgpu_synid_dpp_ctrl>` :ref:`row_mask<amdgpu_synid_row_mask>` :ref:`bank_mask<amdgpu_synid_bank_mask>` :ref:`bound_ctrl<amdgpu_synid_bound_ctrl>`
anatofuz
parents:
diff changeset
73 v_pk_fmac_f16 :ref:`vdst<amdgpu_synid908_vdst32_0>`::ref:`f16x2<amdgpu_synid908_type_dev>`, :ref:`src0<amdgpu_synid908_src32_0>`::ref:`f16x2<amdgpu_synid908_type_dev>`, :ref:`vsrc1<amdgpu_synid908_vsrc32_0>`::ref:`f16x2<amdgpu_synid908_type_dev>`
anatofuz
parents:
diff changeset
74 v_xnor_b32 :ref:`vdst<amdgpu_synid908_vdst32_0>`, :ref:`src0<amdgpu_synid908_src32_0>`, :ref:`vsrc1<amdgpu_synid908_vsrc32_0>`
anatofuz
parents:
diff changeset
75 v_xnor_b32_dpp :ref:`vdst<amdgpu_synid908_vdst32_0>`, :ref:`vsrc0<amdgpu_synid908_vsrc32_0>`, :ref:`vsrc1<amdgpu_synid908_vsrc32_0>` :ref:`dpp_ctrl<amdgpu_synid_dpp_ctrl>` :ref:`row_mask<amdgpu_synid_row_mask>` :ref:`bank_mask<amdgpu_synid_bank_mask>` :ref:`bound_ctrl<amdgpu_synid_bound_ctrl>`
anatofuz
parents:
diff changeset
76 v_xnor_b32_sdwa :ref:`vdst<amdgpu_synid908_vdst32_0>`, :ref:`src0<amdgpu_synid908_src32_0>`::ref:`m<amdgpu_synid908_mod_sdwa_sext>`, :ref:`vsrc1<amdgpu_synid908_vsrc32_0>`::ref:`m<amdgpu_synid908_mod_sdwa_sext>` :ref:`dst_sel<amdgpu_synid_dst_sel>` :ref:`dst_unused<amdgpu_synid_dst_unused>` :ref:`src0_sel<amdgpu_synid_src0_sel>` :ref:`src1_sel<amdgpu_synid_src1_sel>`
anatofuz
parents:
diff changeset
77
anatofuz
parents:
diff changeset
78 VOP3
anatofuz
parents:
diff changeset
79 -----------------------
anatofuz
parents:
diff changeset
80
anatofuz
parents:
diff changeset
81 .. parsed-literal::
anatofuz
parents:
diff changeset
82
anatofuz
parents:
diff changeset
83 **INSTRUCTION** **DST** **SRC0** **SRC1** **MODIFIERS**
anatofuz
parents:
diff changeset
84 \ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|
anatofuz
parents:
diff changeset
85 v_fmac_f32_e64 :ref:`vdst<amdgpu_synid908_vdst32_0>`, :ref:`src0<amdgpu_synid908_src32_1>`::ref:`m<amdgpu_synid908_mod_vop3_abs_neg>`, :ref:`src1<amdgpu_synid908_src32_2>`::ref:`m<amdgpu_synid908_mod_vop3_abs_neg>` :ref:`clamp<amdgpu_synid_clamp>` :ref:`omod<amdgpu_synid_omod>`
anatofuz
parents:
diff changeset
86 v_xnor_b32_e64 :ref:`vdst<amdgpu_synid908_vdst32_0>`, :ref:`src0<amdgpu_synid908_src32_1>`, :ref:`src1<amdgpu_synid908_src32_2>`
anatofuz
parents:
diff changeset
87
anatofuz
parents:
diff changeset
88 VOP3P
anatofuz
parents:
diff changeset
89 -----------------------
anatofuz
parents:
diff changeset
90
anatofuz
parents:
diff changeset
91 .. parsed-literal::
anatofuz
parents:
diff changeset
92
anatofuz
parents:
diff changeset
93 **INSTRUCTION** **DST** **SRC0** **SRC1** **SRC2** **MODIFIERS**
anatofuz
parents:
diff changeset
94 \ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|\ |---|
anatofuz
parents:
diff changeset
95 v_accvgpr_read_b32 :ref:`vdst<amdgpu_synid908_vdst32_0>`, :ref:`asrc<amdgpu_synid908_asrc32_0>`
anatofuz
parents:
diff changeset
96 v_accvgpr_write_b32 :ref:`adst<amdgpu_synid908_adst32_0>`, :ref:`src<amdgpu_synid908_src32_3>`
anatofuz
parents:
diff changeset
97 v_dot2_f32_f16 :ref:`vdst<amdgpu_synid908_vdst32_0>`, :ref:`src0<amdgpu_synid908_src32_1>`::ref:`f16x2<amdgpu_synid908_type_dev>`, :ref:`src1<amdgpu_synid908_src32_2>`::ref:`f16x2<amdgpu_synid908_type_dev>`, :ref:`src2<amdgpu_synid908_src32_2>`::ref:`f32<amdgpu_synid908_type_dev>` :ref:`neg_lo<amdgpu_synid_neg_lo>` :ref:`neg_hi<amdgpu_synid_neg_hi>` :ref:`clamp<amdgpu_synid_clamp>`
207
Shinji KONO <kono@ie.u-ryukyu.ac.jp>
parents: 150
diff changeset
98 v_dot2_i32_i16 :ref:`vdst<amdgpu_synid908_vdst32_0>`, :ref:`src0<amdgpu_synid908_src32_4>`::ref:`i16x2<amdgpu_synid908_type_dev>`, :ref:`src1<amdgpu_synid908_src32_5>`::ref:`i16x2<amdgpu_synid908_type_dev>`, :ref:`src2<amdgpu_synid908_src32_2>`::ref:`i32<amdgpu_synid908_type_dev>` :ref:`clamp<amdgpu_synid_clamp>`
Shinji KONO <kono@ie.u-ryukyu.ac.jp>
parents: 150
diff changeset
99 v_dot2_u32_u16 :ref:`vdst<amdgpu_synid908_vdst32_0>`, :ref:`src0<amdgpu_synid908_src32_4>`::ref:`u16x2<amdgpu_synid908_type_dev>`, :ref:`src1<amdgpu_synid908_src32_5>`::ref:`u16x2<amdgpu_synid908_type_dev>`, :ref:`src2<amdgpu_synid908_src32_2>`::ref:`u32<amdgpu_synid908_type_dev>` :ref:`clamp<amdgpu_synid_clamp>`
150
anatofuz
parents:
diff changeset
100 v_dot4_i32_i8 :ref:`vdst<amdgpu_synid908_vdst32_0>`, :ref:`src0<amdgpu_synid908_src32_1>`::ref:`i8x4<amdgpu_synid908_type_dev>`, :ref:`src1<amdgpu_synid908_src32_2>`::ref:`i8x4<amdgpu_synid908_type_dev>`, :ref:`src2<amdgpu_synid908_src32_2>`::ref:`i32<amdgpu_synid908_type_dev>` :ref:`clamp<amdgpu_synid_clamp>`
anatofuz
parents:
diff changeset
101 v_dot4_u32_u8 :ref:`vdst<amdgpu_synid908_vdst32_0>`, :ref:`src0<amdgpu_synid908_src32_1>`::ref:`u8x4<amdgpu_synid908_type_dev>`, :ref:`src1<amdgpu_synid908_src32_2>`::ref:`u8x4<amdgpu_synid908_type_dev>`, :ref:`src2<amdgpu_synid908_src32_2>`::ref:`u32<amdgpu_synid908_type_dev>` :ref:`clamp<amdgpu_synid_clamp>`
anatofuz
parents:
diff changeset
102 v_dot8_i32_i4 :ref:`vdst<amdgpu_synid908_vdst32_0>`, :ref:`src0<amdgpu_synid908_src32_1>`::ref:`i4x8<amdgpu_synid908_type_dev>`, :ref:`src1<amdgpu_synid908_src32_2>`::ref:`i4x8<amdgpu_synid908_type_dev>`, :ref:`src2<amdgpu_synid908_src32_2>`::ref:`i32<amdgpu_synid908_type_dev>` :ref:`clamp<amdgpu_synid_clamp>`
anatofuz
parents:
diff changeset
103 v_dot8_u32_u4 :ref:`vdst<amdgpu_synid908_vdst32_0>`, :ref:`src0<amdgpu_synid908_src32_1>`::ref:`u4x8<amdgpu_synid908_type_dev>`, :ref:`src1<amdgpu_synid908_src32_2>`::ref:`u4x8<amdgpu_synid908_type_dev>`, :ref:`src2<amdgpu_synid908_src32_2>`::ref:`u32<amdgpu_synid908_type_dev>` :ref:`clamp<amdgpu_synid_clamp>`
anatofuz
parents:
diff changeset
104 v_fma_mix_f32 :ref:`vdst<amdgpu_synid908_vdst32_0>`, :ref:`src0<amdgpu_synid908_src32_1>`::ref:`m<amdgpu_synid908_mod_vop3_abs_neg>`::ref:`fx<amdgpu_synid908_mad_type_dev>`, :ref:`src1<amdgpu_synid908_src32_2>`::ref:`m<amdgpu_synid908_mod_vop3_abs_neg>`::ref:`fx<amdgpu_synid908_mad_type_dev>`, :ref:`src2<amdgpu_synid908_src32_2>`::ref:`m<amdgpu_synid908_mod_vop3_abs_neg>`::ref:`fx<amdgpu_synid908_mad_type_dev>` :ref:`m_op_sel<amdgpu_synid_mad_mix_op_sel>` :ref:`m_op_sel_hi<amdgpu_synid_mad_mix_op_sel_hi>` :ref:`clamp<amdgpu_synid_clamp>`
anatofuz
parents:
diff changeset
105 v_fma_mixhi_f16 :ref:`vdst<amdgpu_synid908_vdst32_0>`, :ref:`src0<amdgpu_synid908_src32_1>`::ref:`m<amdgpu_synid908_mod_vop3_abs_neg>`::ref:`fx<amdgpu_synid908_mad_type_dev>`, :ref:`src1<amdgpu_synid908_src32_2>`::ref:`m<amdgpu_synid908_mod_vop3_abs_neg>`::ref:`fx<amdgpu_synid908_mad_type_dev>`, :ref:`src2<amdgpu_synid908_src32_2>`::ref:`m<amdgpu_synid908_mod_vop3_abs_neg>`::ref:`fx<amdgpu_synid908_mad_type_dev>` :ref:`m_op_sel<amdgpu_synid_mad_mix_op_sel>` :ref:`m_op_sel_hi<amdgpu_synid_mad_mix_op_sel_hi>` :ref:`clamp<amdgpu_synid_clamp>`
anatofuz
parents:
diff changeset
106 v_fma_mixlo_f16 :ref:`vdst<amdgpu_synid908_vdst32_0>`, :ref:`src0<amdgpu_synid908_src32_1>`::ref:`m<amdgpu_synid908_mod_vop3_abs_neg>`::ref:`fx<amdgpu_synid908_mad_type_dev>`, :ref:`src1<amdgpu_synid908_src32_2>`::ref:`m<amdgpu_synid908_mod_vop3_abs_neg>`::ref:`fx<amdgpu_synid908_mad_type_dev>`, :ref:`src2<amdgpu_synid908_src32_2>`::ref:`m<amdgpu_synid908_mod_vop3_abs_neg>`::ref:`fx<amdgpu_synid908_mad_type_dev>` :ref:`m_op_sel<amdgpu_synid_mad_mix_op_sel>` :ref:`m_op_sel_hi<amdgpu_synid_mad_mix_op_sel_hi>` :ref:`clamp<amdgpu_synid_clamp>`
anatofuz
parents:
diff changeset
107 v_mfma_f32_16x16x16f16 :ref:`adst<amdgpu_synid908_adst128_0>`::ref:`f32x4<amdgpu_synid908_type_dev>`, :ref:`vasrc0<amdgpu_synid908_vasrc64_0>`::ref:`f16x4<amdgpu_synid908_type_dev>`, :ref:`vasrc1<amdgpu_synid908_vasrc64_0>`::ref:`f16x4<amdgpu_synid908_type_dev>`, :ref:`asrc2<amdgpu_synid908_asrc128_0>`::ref:`f32x4<amdgpu_synid908_type_dev>` :ref:`cbsz<amdgpu_synid_cbsz>` :ref:`abid<amdgpu_synid_abid>` :ref:`blgp<amdgpu_synid_blgp>`
anatofuz
parents:
diff changeset
108 v_mfma_f32_16x16x1f32 :ref:`adst<amdgpu_synid908_adst512_0>`::ref:`f32x16<amdgpu_synid908_type_dev>`, :ref:`vasrc0<amdgpu_synid908_vasrc32_0>`::ref:`f32<amdgpu_synid908_type_dev>`, :ref:`vasrc1<amdgpu_synid908_vasrc32_0>`::ref:`f32<amdgpu_synid908_type_dev>`, :ref:`asrc2<amdgpu_synid908_asrc512_0>`::ref:`f32x16<amdgpu_synid908_type_dev>` :ref:`cbsz<amdgpu_synid_cbsz>` :ref:`abid<amdgpu_synid_abid>` :ref:`blgp<amdgpu_synid_blgp>`
anatofuz
parents:
diff changeset
109 v_mfma_f32_16x16x2bf16 :ref:`adst<amdgpu_synid908_adst512_0>`::ref:`f32x16<amdgpu_synid908_type_dev>`, :ref:`vasrc0<amdgpu_synid908_vasrc32_0>`::ref:`bf16x2<amdgpu_synid908_type_dev>`, :ref:`vasrc1<amdgpu_synid908_vasrc32_0>`::ref:`bf16x2<amdgpu_synid908_type_dev>`, :ref:`asrc2<amdgpu_synid908_asrc512_0>`::ref:`f32x16<amdgpu_synid908_type_dev>` :ref:`cbsz<amdgpu_synid_cbsz>` :ref:`abid<amdgpu_synid_abid>` :ref:`blgp<amdgpu_synid_blgp>`
anatofuz
parents:
diff changeset
110 v_mfma_f32_16x16x4f16 :ref:`adst<amdgpu_synid908_adst512_0>`::ref:`f32x16<amdgpu_synid908_type_dev>`, :ref:`vasrc0<amdgpu_synid908_vasrc64_0>`::ref:`f16x4<amdgpu_synid908_type_dev>`, :ref:`vasrc1<amdgpu_synid908_vasrc64_0>`::ref:`f16x4<amdgpu_synid908_type_dev>`, :ref:`asrc2<amdgpu_synid908_asrc512_0>`::ref:`f32x16<amdgpu_synid908_type_dev>` :ref:`cbsz<amdgpu_synid_cbsz>` :ref:`abid<amdgpu_synid_abid>` :ref:`blgp<amdgpu_synid_blgp>`
anatofuz
parents:
diff changeset
111 v_mfma_f32_16x16x4f32 :ref:`adst<amdgpu_synid908_adst128_0>`::ref:`f32x4<amdgpu_synid908_type_dev>`, :ref:`vasrc0<amdgpu_synid908_vasrc32_0>`::ref:`f32<amdgpu_synid908_type_dev>`, :ref:`vasrc1<amdgpu_synid908_vasrc32_0>`::ref:`f32<amdgpu_synid908_type_dev>`, :ref:`asrc2<amdgpu_synid908_asrc128_0>`::ref:`f32x4<amdgpu_synid908_type_dev>` :ref:`cbsz<amdgpu_synid_cbsz>` :ref:`abid<amdgpu_synid_abid>` :ref:`blgp<amdgpu_synid_blgp>`
anatofuz
parents:
diff changeset
112 v_mfma_f32_16x16x8bf16 :ref:`adst<amdgpu_synid908_adst128_0>`::ref:`f32x4<amdgpu_synid908_type_dev>`, :ref:`vasrc0<amdgpu_synid908_vasrc32_0>`::ref:`bf16x2<amdgpu_synid908_type_dev>`, :ref:`vasrc1<amdgpu_synid908_vasrc32_0>`::ref:`bf16x2<amdgpu_synid908_type_dev>`, :ref:`asrc2<amdgpu_synid908_asrc128_0>`::ref:`f32x4<amdgpu_synid908_type_dev>` :ref:`cbsz<amdgpu_synid_cbsz>` :ref:`abid<amdgpu_synid_abid>` :ref:`blgp<amdgpu_synid_blgp>`
anatofuz
parents:
diff changeset
113 v_mfma_f32_32x32x1f32 :ref:`adst<amdgpu_synid908_adst1024_0>`::ref:`f32x32<amdgpu_synid908_type_dev>`, :ref:`vasrc0<amdgpu_synid908_vasrc32_0>`::ref:`f32<amdgpu_synid908_type_dev>`, :ref:`vasrc1<amdgpu_synid908_vasrc32_0>`::ref:`f32<amdgpu_synid908_type_dev>`, :ref:`asrc2<amdgpu_synid908_asrc1024_0>`::ref:`f32x32<amdgpu_synid908_type_dev>` :ref:`cbsz<amdgpu_synid_cbsz>` :ref:`abid<amdgpu_synid_abid>` :ref:`blgp<amdgpu_synid_blgp>`
anatofuz
parents:
diff changeset
114 v_mfma_f32_32x32x2bf16 :ref:`adst<amdgpu_synid908_adst1024_0>`::ref:`f32x32<amdgpu_synid908_type_dev>`, :ref:`vasrc0<amdgpu_synid908_vasrc32_0>`::ref:`bf16x2<amdgpu_synid908_type_dev>`, :ref:`vasrc1<amdgpu_synid908_vasrc32_0>`::ref:`bf16x2<amdgpu_synid908_type_dev>`, :ref:`asrc2<amdgpu_synid908_asrc1024_0>`::ref:`f32x32<amdgpu_synid908_type_dev>` :ref:`cbsz<amdgpu_synid_cbsz>` :ref:`abid<amdgpu_synid_abid>` :ref:`blgp<amdgpu_synid_blgp>`
anatofuz
parents:
diff changeset
115 v_mfma_f32_32x32x2f32 :ref:`adst<amdgpu_synid908_adst512_0>`::ref:`f32x16<amdgpu_synid908_type_dev>`, :ref:`vasrc0<amdgpu_synid908_vasrc32_0>`::ref:`f32<amdgpu_synid908_type_dev>`, :ref:`vasrc1<amdgpu_synid908_vasrc32_0>`::ref:`f32<amdgpu_synid908_type_dev>`, :ref:`asrc2<amdgpu_synid908_asrc512_0>`::ref:`f32x16<amdgpu_synid908_type_dev>` :ref:`cbsz<amdgpu_synid_cbsz>` :ref:`abid<amdgpu_synid_abid>` :ref:`blgp<amdgpu_synid_blgp>`
anatofuz
parents:
diff changeset
116 v_mfma_f32_32x32x4bf16 :ref:`adst<amdgpu_synid908_adst512_0>`::ref:`f32x16<amdgpu_synid908_type_dev>`, :ref:`vasrc0<amdgpu_synid908_vasrc32_0>`::ref:`bf16x2<amdgpu_synid908_type_dev>`, :ref:`vasrc1<amdgpu_synid908_vasrc32_0>`::ref:`bf16x2<amdgpu_synid908_type_dev>`, :ref:`asrc2<amdgpu_synid908_asrc512_0>`::ref:`f32x16<amdgpu_synid908_type_dev>` :ref:`cbsz<amdgpu_synid_cbsz>` :ref:`abid<amdgpu_synid_abid>` :ref:`blgp<amdgpu_synid_blgp>`
anatofuz
parents:
diff changeset
117 v_mfma_f32_32x32x4f16 :ref:`adst<amdgpu_synid908_adst1024_0>`::ref:`f32x32<amdgpu_synid908_type_dev>`, :ref:`vasrc0<amdgpu_synid908_vasrc64_0>`::ref:`f16x4<amdgpu_synid908_type_dev>`, :ref:`vasrc1<amdgpu_synid908_vasrc64_0>`::ref:`f16x4<amdgpu_synid908_type_dev>`, :ref:`asrc2<amdgpu_synid908_asrc1024_0>`::ref:`f32x32<amdgpu_synid908_type_dev>` :ref:`cbsz<amdgpu_synid_cbsz>` :ref:`abid<amdgpu_synid_abid>` :ref:`blgp<amdgpu_synid_blgp>`
anatofuz
parents:
diff changeset
118 v_mfma_f32_32x32x8f16 :ref:`adst<amdgpu_synid908_adst512_0>`::ref:`f32x16<amdgpu_synid908_type_dev>`, :ref:`vasrc0<amdgpu_synid908_vasrc64_0>`::ref:`f16x4<amdgpu_synid908_type_dev>`, :ref:`vasrc1<amdgpu_synid908_vasrc64_0>`::ref:`f16x4<amdgpu_synid908_type_dev>`, :ref:`asrc2<amdgpu_synid908_asrc512_0>`::ref:`f32x16<amdgpu_synid908_type_dev>` :ref:`cbsz<amdgpu_synid_cbsz>` :ref:`abid<amdgpu_synid_abid>` :ref:`blgp<amdgpu_synid_blgp>`
anatofuz
parents:
diff changeset
119 v_mfma_f32_4x4x1f32 :ref:`adst<amdgpu_synid908_adst128_0>`::ref:`f32x4<amdgpu_synid908_type_dev>`, :ref:`vasrc0<amdgpu_synid908_vasrc32_0>`::ref:`f32<amdgpu_synid908_type_dev>`, :ref:`vasrc1<amdgpu_synid908_vasrc32_0>`::ref:`f32<amdgpu_synid908_type_dev>`, :ref:`asrc2<amdgpu_synid908_asrc128_0>`::ref:`f32x4<amdgpu_synid908_type_dev>` :ref:`cbsz<amdgpu_synid_cbsz>` :ref:`abid<amdgpu_synid_abid>` :ref:`blgp<amdgpu_synid_blgp>`
anatofuz
parents:
diff changeset
120 v_mfma_f32_4x4x2bf16 :ref:`adst<amdgpu_synid908_adst128_0>`::ref:`f32x4<amdgpu_synid908_type_dev>`, :ref:`vasrc0<amdgpu_synid908_vasrc32_0>`::ref:`bf16x2<amdgpu_synid908_type_dev>`, :ref:`vasrc1<amdgpu_synid908_vasrc32_0>`::ref:`bf16x2<amdgpu_synid908_type_dev>`, :ref:`asrc2<amdgpu_synid908_asrc128_0>`::ref:`f32x4<amdgpu_synid908_type_dev>` :ref:`cbsz<amdgpu_synid_cbsz>` :ref:`abid<amdgpu_synid_abid>` :ref:`blgp<amdgpu_synid_blgp>`
anatofuz
parents:
diff changeset
121 v_mfma_f32_4x4x4f16 :ref:`adst<amdgpu_synid908_adst128_0>`::ref:`f32x4<amdgpu_synid908_type_dev>`, :ref:`vasrc0<amdgpu_synid908_vasrc64_0>`::ref:`f16x4<amdgpu_synid908_type_dev>`, :ref:`vasrc1<amdgpu_synid908_vasrc64_0>`::ref:`f16x4<amdgpu_synid908_type_dev>`, :ref:`asrc2<amdgpu_synid908_asrc128_0>`::ref:`f32x4<amdgpu_synid908_type_dev>` :ref:`cbsz<amdgpu_synid_cbsz>` :ref:`abid<amdgpu_synid_abid>` :ref:`blgp<amdgpu_synid_blgp>`
anatofuz
parents:
diff changeset
122 v_mfma_i32_16x16x16i8 :ref:`adst<amdgpu_synid908_adst128_0>`::ref:`i32x4<amdgpu_synid908_type_dev>`, :ref:`vasrc0<amdgpu_synid908_vasrc32_0>`::ref:`i8x4<amdgpu_synid908_type_dev>`, :ref:`vasrc1<amdgpu_synid908_vasrc32_0>`::ref:`i8x4<amdgpu_synid908_type_dev>`, :ref:`asrc2<amdgpu_synid908_asrc128_0>`::ref:`i32x4<amdgpu_synid908_type_dev>` :ref:`cbsz<amdgpu_synid_cbsz>` :ref:`abid<amdgpu_synid_abid>` :ref:`blgp<amdgpu_synid_blgp>`
anatofuz
parents:
diff changeset
123 v_mfma_i32_16x16x4i8 :ref:`adst<amdgpu_synid908_adst512_0>`::ref:`i32x16<amdgpu_synid908_type_dev>`, :ref:`vasrc0<amdgpu_synid908_vasrc32_0>`::ref:`i8x4<amdgpu_synid908_type_dev>`, :ref:`vasrc1<amdgpu_synid908_vasrc32_0>`::ref:`i8x4<amdgpu_synid908_type_dev>`, :ref:`asrc2<amdgpu_synid908_asrc512_0>`::ref:`i32x16<amdgpu_synid908_type_dev>` :ref:`cbsz<amdgpu_synid_cbsz>` :ref:`abid<amdgpu_synid_abid>` :ref:`blgp<amdgpu_synid_blgp>`
anatofuz
parents:
diff changeset
124 v_mfma_i32_32x32x4i8 :ref:`adst<amdgpu_synid908_adst1024_0>`::ref:`i32x32<amdgpu_synid908_type_dev>`, :ref:`vasrc0<amdgpu_synid908_vasrc32_0>`::ref:`i8x4<amdgpu_synid908_type_dev>`, :ref:`vasrc1<amdgpu_synid908_vasrc32_0>`::ref:`i8x4<amdgpu_synid908_type_dev>`, :ref:`asrc2<amdgpu_synid908_asrc1024_0>`::ref:`i32x32<amdgpu_synid908_type_dev>` :ref:`cbsz<amdgpu_synid_cbsz>` :ref:`abid<amdgpu_synid_abid>` :ref:`blgp<amdgpu_synid_blgp>`
anatofuz
parents:
diff changeset
125 v_mfma_i32_32x32x8i8 :ref:`adst<amdgpu_synid908_adst512_0>`::ref:`i32x16<amdgpu_synid908_type_dev>`, :ref:`vasrc0<amdgpu_synid908_vasrc32_0>`::ref:`i8x4<amdgpu_synid908_type_dev>`, :ref:`vasrc1<amdgpu_synid908_vasrc32_0>`::ref:`i8x4<amdgpu_synid908_type_dev>`, :ref:`asrc2<amdgpu_synid908_asrc512_0>`::ref:`i32x16<amdgpu_synid908_type_dev>` :ref:`cbsz<amdgpu_synid_cbsz>` :ref:`abid<amdgpu_synid_abid>` :ref:`blgp<amdgpu_synid_blgp>`
anatofuz
parents:
diff changeset
126 v_mfma_i32_4x4x4i8 :ref:`adst<amdgpu_synid908_adst128_0>`::ref:`i32x4<amdgpu_synid908_type_dev>`, :ref:`vasrc0<amdgpu_synid908_vasrc32_0>`::ref:`i8x4<amdgpu_synid908_type_dev>`, :ref:`vasrc1<amdgpu_synid908_vasrc32_0>`::ref:`i8x4<amdgpu_synid908_type_dev>`, :ref:`asrc2<amdgpu_synid908_asrc128_0>`::ref:`i32x4<amdgpu_synid908_type_dev>` :ref:`cbsz<amdgpu_synid_cbsz>` :ref:`abid<amdgpu_synid_abid>` :ref:`blgp<amdgpu_synid_blgp>`
anatofuz
parents:
diff changeset
127
anatofuz
parents:
diff changeset
128 .. |---| unicode:: U+02014 .. em dash
anatofuz
parents:
diff changeset
129
anatofuz
parents:
diff changeset
130
anatofuz
parents:
diff changeset
131 .. toctree::
anatofuz
parents:
diff changeset
132 :hidden:
anatofuz
parents:
diff changeset
133
anatofuz
parents:
diff changeset
134 AMDGPUAsmGFX9
anatofuz
parents:
diff changeset
135 gfx908_addr_buf
anatofuz
parents:
diff changeset
136 gfx908_adst1024_0
anatofuz
parents:
diff changeset
137 gfx908_adst128_0
anatofuz
parents:
diff changeset
138 gfx908_adst32_0
anatofuz
parents:
diff changeset
139 gfx908_adst512_0
anatofuz
parents:
diff changeset
140 gfx908_asrc1024_0
anatofuz
parents:
diff changeset
141 gfx908_asrc128_0
anatofuz
parents:
diff changeset
142 gfx908_asrc32_0
anatofuz
parents:
diff changeset
143 gfx908_asrc512_0
anatofuz
parents:
diff changeset
144 gfx908_data_buf_atomic32
anatofuz
parents:
diff changeset
145 gfx908_dst_flat_atomic32
anatofuz
parents:
diff changeset
146 gfx908_offset_buf
anatofuz
parents:
diff changeset
147 gfx908_rsrc_buf
anatofuz
parents:
diff changeset
148 gfx908_saddr_flat_global
anatofuz
parents:
diff changeset
149 gfx908_src32_0
anatofuz
parents:
diff changeset
150 gfx908_src32_1
anatofuz
parents:
diff changeset
151 gfx908_src32_2
anatofuz
parents:
diff changeset
152 gfx908_src32_3
207
Shinji KONO <kono@ie.u-ryukyu.ac.jp>
parents: 150
diff changeset
153 gfx908_src32_4
Shinji KONO <kono@ie.u-ryukyu.ac.jp>
parents: 150
diff changeset
154 gfx908_src32_5
150
anatofuz
parents:
diff changeset
155 gfx908_vaddr_flat_global
anatofuz
parents:
diff changeset
156 gfx908_vasrc32_0
anatofuz
parents:
diff changeset
157 gfx908_vasrc64_0
anatofuz
parents:
diff changeset
158 gfx908_vdata32_0
anatofuz
parents:
diff changeset
159 gfx908_vdst32_0
anatofuz
parents:
diff changeset
160 gfx908_vsrc32_0
anatofuz
parents:
diff changeset
161 gfx908_mad_type_dev
anatofuz
parents:
diff changeset
162 gfx908_mod_dpp_sdwa_abs_neg
anatofuz
parents:
diff changeset
163 gfx908_mod_sdwa_sext
anatofuz
parents:
diff changeset
164 gfx908_mod_vop3_abs_neg
anatofuz
parents:
diff changeset
165 gfx908_opt
anatofuz
parents:
diff changeset
166 gfx908_ret
anatofuz
parents:
diff changeset
167 gfx908_type_dev