RPCS3/llvm-mirror
1
0
Fork 0
mirror of https://github.com/RPCS3/llvm-mirror.git synced 2024-11-26 12:43:36 +01:00
Code Issues Projects Releases Wiki Activity
master3
llvm-mirror/test/MC/AMDGPU/flat-gfx9.s

151 lines
7.2 KiB
ArmAsm
Raw Permalink Normal View History

AMDGPU: Start adding offset fields to flat instructions llvm-svn: 305194
2017-06-12 17:55:58 +02:00
// RUN: not llvm-mc -arch=amdgcn -mcpu=gfx900 -show-encoding %s | FileCheck -check-prefix=GFX9 -check-prefix=GCN %s
// RUN: not llvm-mc -arch=amdgcn -mcpu=tonga -show-encoding %s | FileCheck -check-prefix=VI -check-prefix=GCN %s
// RUN: not llvm-mc -arch=amdgcn -mcpu=gfx900 -show-encoding 2>&1 %s | FileCheck -check-prefix=GFX9-ERR -check-prefix=GCNERR %s
// RUN: not llvm-mc -arch=amdgcn -mcpu=tonga -show-encoding 2>&1 %s | FileCheck -check-prefix=VI-ERR -check-prefix=GCNERR %s
flat_load_dword v1, v[3:4] offset:0
// GCN: flat_load_dword v1, v[3:4] ; encoding: [0x00,0x00,0x50,0xdc,0x03,0x00,0x00,0x01]
flat_load_dword v1, v[3:4] offset:-1
[AMDGPU][MC] Corrected parsing of FLAT offset modifier Summary of changes: - simplified handling of FLAT offset: offset_s13 and offset_u12 have been replaced with flat_offset; - provided information about error position for pre-gfx9 targets; - improved errors handling. Reviewers: artem.tamazov, arsenm, rampitec Differential Revision: https://reviews.llvm.org/D64244 llvm-svn: 365321
2019-07-08 16:27:37 +02:00
// VI-ERR: :28: error: flat offset modifier is not supported on this GPU
// GFX9-ERR: :28: error: expected a 12-bit unsigned offset
AMDGPU: Start adding offset fields to flat instructions llvm-svn: 305194
2017-06-12 17:55:58 +02:00
// FIXME: Error on VI in wrong column
flat_load_dword v1, v[3:4] offset:4095
// GFX9: flat_load_dword v1, v[3:4] offset:4095 ; encoding: [0xff,0x0f,0x50,0xdc,0x03,0x00,0x00,0x01]
[AMDGPU][MC] Corrected parsing of FLAT offset modifier Summary of changes: - simplified handling of FLAT offset: offset_s13 and offset_u12 have been replaced with flat_offset; - provided information about error position for pre-gfx9 targets; - improved errors handling. Reviewers: artem.tamazov, arsenm, rampitec Differential Revision: https://reviews.llvm.org/D64244 llvm-svn: 365321
2019-07-08 16:27:37 +02:00
// VI-ERR: :28: error: flat offset modifier is not supported on this GPU
AMDGPU: Start adding offset fields to flat instructions llvm-svn: 305194
2017-06-12 17:55:58 +02:00
flat_load_dword v1, v[3:4] offset:4096
[AMDGPU][MC] Corrected parsing of FLAT offset modifier Summary of changes: - simplified handling of FLAT offset: offset_s13 and offset_u12 have been replaced with flat_offset; - provided information about error position for pre-gfx9 targets; - improved errors handling. Reviewers: artem.tamazov, arsenm, rampitec Differential Revision: https://reviews.llvm.org/D64244 llvm-svn: 365321
2019-07-08 16:27:37 +02:00
// VI-ERR: :28: error: flat offset modifier is not supported on this GPU
// GFX9-ERR: :28: error: expected a 12-bit unsigned offset
AMDGPU: Start adding offset fields to flat instructions llvm-svn: 305194
2017-06-12 17:55:58 +02:00
flat_load_dword v1, v[3:4] offset:4 glc
// GFX9: flat_load_dword v1, v[3:4] offset:4 glc ; encoding: [0x04,0x00,0x51,0xdc,0x03,0x00,0x00,0x01]
[AMDGPU][MC] Corrected parsing of FLAT offset modifier Summary of changes: - simplified handling of FLAT offset: offset_s13 and offset_u12 have been replaced with flat_offset; - provided information about error position for pre-gfx9 targets; - improved errors handling. Reviewers: artem.tamazov, arsenm, rampitec Differential Revision: https://reviews.llvm.org/D64244 llvm-svn: 365321
2019-07-08 16:27:37 +02:00
// VI-ERR: :28: error: flat offset modifier is not supported on this GPU
AMDGPU: Start adding offset fields to flat instructions llvm-svn: 305194
2017-06-12 17:55:58 +02:00
flat_load_dword v1, v[3:4] offset:4 glc slc
// GFX9: flat_load_dword v1, v[3:4] offset:4 glc slc ; encoding: [0x04,0x00,0x53,0xdc,0x03,0x00,0x00,0x01]
[AMDGPU][MC] Corrected parsing of FLAT offset modifier Summary of changes: - simplified handling of FLAT offset: offset_s13 and offset_u12 have been replaced with flat_offset; - provided information about error position for pre-gfx9 targets; - improved errors handling. Reviewers: artem.tamazov, arsenm, rampitec Differential Revision: https://reviews.llvm.org/D64244 llvm-svn: 365321
2019-07-08 16:27:37 +02:00
// VI-ERR: :28: error: flat offset modifier is not supported on this GPU
AMDGPU: Start adding offset fields to flat instructions llvm-svn: 305194
2017-06-12 17:55:58 +02:00
flat_atomic_add v[3:4], v5 offset:8 slc
// GFX9: flat_atomic_add v[3:4], v5 offset:8 slc ; encoding: [0x08,0x00,0x0a,0xdd,0x03,0x05,0x00,0x00]
[AMDGPU][MC] Corrected parsing of FLAT offset modifier Summary of changes: - simplified handling of FLAT offset: offset_s13 and offset_u12 have been replaced with flat_offset; - provided information about error position for pre-gfx9 targets; - improved errors handling. Reviewers: artem.tamazov, arsenm, rampitec Differential Revision: https://reviews.llvm.org/D64244 llvm-svn: 365321
2019-07-08 16:27:37 +02:00
// VI-ERR: :28: error: flat offset modifier is not supported on this GPU
AMDGPU: Start adding offset fields to flat instructions llvm-svn: 305194
2017-06-12 17:55:58 +02:00
[AMDGPU][MC][GFX9] Added support of 'inst_offset' modifier for compatibility with SP3 See bug 35329: https://bugs.llvm.org//show_bug.cgi?id=35329 Reviewers: arsenm, vpykhtin, artem.tamazov Differential Revision: https://reviews.llvm.org/D40350 llvm-svn: 318947
2017-11-24 14:22:38 +01:00
flat_atomic_add v[3:4], v5 inst_offset:8 slc
// GFX9: flat_atomic_add v[3:4], v5 offset:8 slc ; encoding: [0x08,0x00,0x0a,0xdd,0x03,0x05,0x00,0x00]
[AMDGPU][MC] Corrected parsing of FLAT offset modifier Summary of changes: - simplified handling of FLAT offset: offset_s13 and offset_u12 have been replaced with flat_offset; - provided information about error position for pre-gfx9 targets; - improved errors handling. Reviewers: artem.tamazov, arsenm, rampitec Differential Revision: https://reviews.llvm.org/D64244 llvm-svn: 365321
2019-07-08 16:27:37 +02:00
// VI-ERR: :28: error: flat offset modifier is not supported on this GPU
[AMDGPU][MC][GFX9] Added support of 'inst_offset' modifier for compatibility with SP3 See bug 35329: https://bugs.llvm.org//show_bug.cgi?id=35329 Reviewers: arsenm, vpykhtin, artem.tamazov Differential Revision: https://reviews.llvm.org/D40350 llvm-svn: 318947
2017-11-24 14:22:38 +01:00
[AMDGPU][MC] Incorrect parsing of flat/global atomic modifiers See bug 35730: https://bugs.llvm.org/show_bug.cgi?id=35730 Differential Revision: https://reviews.llvm.org/D41598 Reviewers: vpykhtin, artem.tamazov, arsenm llvm-svn: 321552
2017-12-29 14:55:11 +01:00
flat_atomic_cmpswap v[1:2], v[3:4] offset:4095
// GFX9: flat_atomic_cmpswap v[1:2], v[3:4] offset:4095 ; encoding: [0xff,0x0f,0x04,0xdd,0x01,0x03,0x00,0x00]
[AMDGPU][MC] Corrected parsing of FLAT offset modifier Summary of changes: - simplified handling of FLAT offset: offset_s13 and offset_u12 have been replaced with flat_offset; - provided information about error position for pre-gfx9 targets; - improved errors handling. Reviewers: artem.tamazov, arsenm, rampitec Differential Revision: https://reviews.llvm.org/D64244 llvm-svn: 365321
2019-07-08 16:27:37 +02:00
// VI-ERR: :36: error: flat offset modifier is not supported on this GPU
[AMDGPU][MC] Incorrect parsing of flat/global atomic modifiers See bug 35730: https://bugs.llvm.org/show_bug.cgi?id=35730 Differential Revision: https://reviews.llvm.org/D41598 Reviewers: vpykhtin, artem.tamazov, arsenm llvm-svn: 321552
2017-12-29 14:55:11 +01:00
flat_atomic_cmpswap v[1:2], v[3:4] offset:4095 slc
// GFX9: flat_atomic_cmpswap v[1:2], v[3:4] offset:4095 slc ; encoding: [0xff,0x0f,0x06,0xdd,0x01,0x03,0x00,0x00]
[AMDGPU][MC] Corrected parsing of FLAT offset modifier Summary of changes: - simplified handling of FLAT offset: offset_s13 and offset_u12 have been replaced with flat_offset; - provided information about error position for pre-gfx9 targets; - improved errors handling. Reviewers: artem.tamazov, arsenm, rampitec Differential Revision: https://reviews.llvm.org/D64244 llvm-svn: 365321
2019-07-08 16:27:37 +02:00
// VI-ERR: :36: error: flat offset modifier is not supported on this GPU
[AMDGPU][MC] Incorrect parsing of flat/global atomic modifiers See bug 35730: https://bugs.llvm.org/show_bug.cgi?id=35730 Differential Revision: https://reviews.llvm.org/D41598 Reviewers: vpykhtin, artem.tamazov, arsenm llvm-svn: 321552
2017-12-29 14:55:11 +01:00
flat_atomic_cmpswap v[1:2], v[3:4]
// GFX9: flat_atomic_cmpswap v[1:2], v[3:4] ; encoding: [0x00,0x00,0x04,0xdd,0x01,0x03,0x00,0x00]
// VI: flat_atomic_cmpswap v[1:2], v[3:4] ; encoding: [0x00,0x00,0x04,0xdd,0x01,0x03,0x00,0x00]
flat_atomic_cmpswap v[1:2], v[3:4] slc
// GFX9: flat_atomic_cmpswap v[1:2], v[3:4] slc ; encoding: [0x00,0x00,0x06,0xdd,0x01,0x03,0x00,0x00]
// VI: flat_atomic_cmpswap v[1:2], v[3:4] slc ; encoding: [0x00,0x00,0x06,0xdd,0x01,0x03,0x00,0x00]
flat_atomic_cmpswap v[1:2], v[3:4] offset:4095 glc
// GCNERR: error: invalid operand for instruction
flat_atomic_cmpswap v[1:2], v[3:4] glc
// GCNERR: error: invalid operand for instruction
flat_atomic_cmpswap v0, v[1:2], v[3:4] offset:4095 glc
// GFX9: flat_atomic_cmpswap v0, v[1:2], v[3:4] offset:4095 glc ; encoding: [0xff,0x0f,0x05,0xdd,0x01,0x03,0x00,0x00]
[AMDGPU][MC] Corrected parsing of FLAT offset modifier Summary of changes: - simplified handling of FLAT offset: offset_s13 and offset_u12 have been replaced with flat_offset; - provided information about error position for pre-gfx9 targets; - improved errors handling. Reviewers: artem.tamazov, arsenm, rampitec Differential Revision: https://reviews.llvm.org/D64244 llvm-svn: 365321
2019-07-08 16:27:37 +02:00
// VI-ERR: :40: error: flat offset modifier is not supported on this GPU
[AMDGPU][MC] Incorrect parsing of flat/global atomic modifiers See bug 35730: https://bugs.llvm.org/show_bug.cgi?id=35730 Differential Revision: https://reviews.llvm.org/D41598 Reviewers: vpykhtin, artem.tamazov, arsenm llvm-svn: 321552
2017-12-29 14:55:11 +01:00
flat_atomic_cmpswap v0, v[1:2], v[3:4] offset:4095 glc slc
// GFX9: flat_atomic_cmpswap v0, v[1:2], v[3:4] offset:4095 glc slc ; encoding: [0xff,0x0f,0x07,0xdd,0x01,0x03,0x00,0x00]
[AMDGPU][MC] Corrected parsing of FLAT offset modifier Summary of changes: - simplified handling of FLAT offset: offset_s13 and offset_u12 have been replaced with flat_offset; - provided information about error position for pre-gfx9 targets; - improved errors handling. Reviewers: artem.tamazov, arsenm, rampitec Differential Revision: https://reviews.llvm.org/D64244 llvm-svn: 365321
2019-07-08 16:27:37 +02:00
// VI-ERR: :40: error: flat offset modifier is not supported on this GPU
[AMDGPU][MC] Incorrect parsing of flat/global atomic modifiers See bug 35730: https://bugs.llvm.org/show_bug.cgi?id=35730 Differential Revision: https://reviews.llvm.org/D41598 Reviewers: vpykhtin, artem.tamazov, arsenm llvm-svn: 321552
2017-12-29 14:55:11 +01:00
flat_atomic_cmpswap v0, v[1:2], v[3:4] glc
// GFX9: flat_atomic_cmpswap v0, v[1:2], v[3:4] glc ; encoding: [0x00,0x00,0x05,0xdd,0x01,0x03,0x00,0x00]
// VI: flat_atomic_cmpswap v0, v[1:2], v[3:4] glc ; encoding: [0x00,0x00,0x05,0xdd,0x01,0x03,0x00,0x00]
flat_atomic_cmpswap v0, v[1:2], v[3:4] glc slc
// GFX9: flat_atomic_cmpswap v0, v[1:2], v[3:4] glc slc ; encoding: [0x00,0x00,0x07,0xdd,0x01,0x03,0x00,0x00]
// VI: flat_atomic_cmpswap v0, v[1:2], v[3:4] glc slc ; encoding: [0x00,0x00,0x07,0xdd,0x01,0x03,0x00,0x00]
flat_atomic_cmpswap v0, v[1:2], v[3:4]
// GFX9: flat_atomic_cmpswap v0, v[1:2], v[3:4] glc ; encoding: [0x00,0x00,0x05,0xdd,0x01,0x03,0x00,0x00]
// VI: flat_atomic_cmpswap v0, v[1:2], v[3:4] glc ; encoding: [0x00,0x00,0x05,0xdd,0x01,0x03,0x00,0x00]
flat_atomic_cmpswap v0, v[1:2], v[3:4] offset:4095
// GCNERR: error: too few operands for instruction
flat_atomic_cmpswap v0, v[1:2], v[3:4] slc
// GCNERR: error: invalid operand for instruction
AMDGPU: Start adding offset fields to flat instructions llvm-svn: 305194
2017-06-12 17:55:58 +02:00
flat_atomic_swap v[3:4], v5 offset:16
// GFX9: flat_atomic_swap v[3:4], v5 offset:16 ; encoding: [0x10,0x00,0x00,0xdd,0x03,0x05,0x00,0x00]
[AMDGPU][MC] Corrected parsing of FLAT offset modifier Summary of changes: - simplified handling of FLAT offset: offset_s13 and offset_u12 have been replaced with flat_offset; - provided information about error position for pre-gfx9 targets; - improved errors handling. Reviewers: artem.tamazov, arsenm, rampitec Differential Revision: https://reviews.llvm.org/D64244 llvm-svn: 365321
2019-07-08 16:27:37 +02:00
// VI-ERR: :29: error: flat offset modifier is not supported on this GPU
AMDGPU: Start adding offset fields to flat instructions llvm-svn: 305194
2017-06-12 17:55:58 +02:00
flat_store_dword v[3:4], v1 offset:16
// GFX9: flat_store_dword v[3:4], v1 offset:16 ; encoding: [0x10,0x00,0x70,0xdc,0x03,0x01,0x00,0x00]
[AMDGPU][MC] Corrected parsing of FLAT offset modifier Summary of changes: - simplified handling of FLAT offset: offset_s13 and offset_u12 have been replaced with flat_offset; - provided information about error position for pre-gfx9 targets; - improved errors handling. Reviewers: artem.tamazov, arsenm, rampitec Differential Revision: https://reviews.llvm.org/D64244 llvm-svn: 365321
2019-07-08 16:27:37 +02:00
// VI-ERR: :29: error: flat offset modifier is not supported on this GPU
AMDGPU: Correct encoding for global instructions The soffset field needs to be be set to 0x7f to disable it, not 0. 0 is interpreted as an SGPR offset. This should be enough to get basic usage of the global instructions working. Technically it is possible to use an SGPR_32 offset, but I'm not sure if it's correct with 64-bit pointers, but that is not handled now. This should also be cleaned up to be more similar to how different MUBUF modes are handled, and to have InstrMappings between the different types. llvm-svn: 308583
2017-07-20 07:17:54 +02:00
flat_store_dword v[3:4], v1, off
// GCNERR: :30: error: invalid operand for instruction
flat_store_dword v[3:4], v1, s[0:1]
// GCNERR: :30: error: invalid operand for instruction
flat_store_dword v[3:4], v1, s0
// GCNERR: :30: error: invalid operand for instruction
flat_load_dword v1, v[3:4], off
// GCNERR: :29: error: invalid operand for instruction
flat_load_dword v1, v[3:4], s[0:1]
// GCNERR: :29: error: invalid operand for instruction
flat_load_dword v1, v[3:4], s0
// GCNERR: :29: error: invalid operand for instruction
flat_load_dword v1, v[3:4], exec_hi
// GCNERR: :29: error: invalid operand for instruction
flat_store_dword v[3:4], v1, exec_hi
// GCNERR: :30: error: invalid operand for instruction
AMDGPU: Add most d16 load/store instruction definitions Doesn't include the tied operand necessary for the loads, but is enough for the assembler to work. llvm-svn: 312347
2017-09-01 20:36:06 +02:00
flat_load_ubyte_d16 v1, v[3:4]
// GFX9: flat_load_ubyte_d16 v1, v[3:4] ; encoding: [0x00,0x00,0x80,0xdc,0x03,0x00,0x00,0x01]
// VI-ERR: error: instruction not supported on this GPU
flat_load_ubyte_d16_hi v1, v[3:4]
// GFX9: flat_load_ubyte_d16_hi v1, v[3:4] ; encoding: [0x00,0x00,0x84,0xdc,0x03,0x00,0x00,0x01]
// VI-ERR: error: instruction not supported on this GPU
flat_load_sbyte_d16 v1, v[3:4]
// GFX9: flat_load_sbyte_d16 v1, v[3:4] ; encoding: [0x00,0x00,0x88,0xdc,0x03,0x00,0x00,0x01]
// VI-ERR: error: instruction not supported on this GPU
flat_load_sbyte_d16_hi v1, v[3:4]
// GFX9: flat_load_sbyte_d16_hi v1, v[3:4] ; encoding: [0x00,0x00,0x8c,0xdc,0x03,0x00,0x00,0x01]
// VI-ERR: error: instruction not supported on this GPU
flat_load_short_d16 v1, v[3:4]
// GFX9: flat_load_short_d16 v1, v[3:4] ; encoding: [0x00,0x00,0x90,0xdc,0x03,0x00,0x00,0x01]
// VI-ERR: error: instruction not supported on this GPU
flat_load_short_d16_hi v1, v[3:4]
// GFX9: flat_load_short_d16_hi v1, v[3:4] ; encoding: [0x00,0x00,0x94,0xdc,0x03,0x00,0x00,0x01]
// VI-ERR: error: instruction not supported on this GPU
flat_store_byte_d16_hi v[3:4], v1
// GFX9: flat_store_byte_d16_hi v[3:4], v1 ; encoding: [0x00,0x00,0x64,0xdc,0x03,0x01,0x00,0x00]
// VI-ERR: error: instruction not supported on this GPU
flat_store_short_d16_hi v[3:4], v1
// GFX9: flat_store_short_d16_hi v[3:4], v1 ; encoding: [0x00,0x00,0x6c,0xdc,0x03,0x01,0x00,0x00
// VI-ERR: error: instruction not supported on this GPU
Reference in New Issue Copy Permalink