shylie/llvm-project
1
0
Fork 0
Code Issues Pull Requests Actions 6 Packages Projects Releases Wiki Activity

[AMDGPU] Define new targets gfx1171 and gfx1172 (#187735)

Browse Source
This commit is contained in:
Mirko Brkušanin 2026-04-01 18:16:11 +02:00 committed by GitHub
parent b528f6746d
commit 5d9eb0c76a
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
20 changed files with 113 additions and 11 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
clang/include/clang/Basic/OffloadArch.h
View File

@ -104,6 +104,8 @@ enum class OffloadArch {
GFX1152,
GFX1153,
GFX1170,
GFX1171,
GFX1172,
GFX12_GENERIC,
GFX1200,
GFX1201,

2
clang/lib/Basic/OffloadArch.cpp
View File

@ -99,6 +99,8 @@ static const OffloadArchToStringMap ArchNames[] = {
GFX(1152), // gfx1152
GFX(1153), // gfx1153
GFX(1170), // gfx1170
GFX(1171), // gfx1171
GFX(1172), // gfx1172
{OffloadArch::GFX12_GENERIC, "gfx12-generic", "compute_amdgcn"},
GFX(1200), // gfx1200
GFX(1201), // gfx1201

2
clang/lib/CodeGen/CGOpenMPRuntimeGPU.cpp
View File

@ -2363,6 +2363,8 @@ void CGOpenMPRuntimeGPU::processRequiresDirective(const OMPRequiresDecl *D) {
case OffloadArch::GFX1152:
case OffloadArch::GFX1153:
case OffloadArch::GFX1170:
case OffloadArch::GFX1171:
case OffloadArch::GFX1172:
case OffloadArch::GFX12_GENERIC:
case OffloadArch::GFX1200:
case OffloadArch::GFX1201:

2
clang/test/CodeGenOpenCL/builtins-amdgcn-gfx11.cl
View File

@ -8,6 +8,8 @@
// RUN: %clang_cc1 -triple amdgcn-unknown-unknown -target-cpu gfx1152 -emit-llvm -o - %s | FileCheck --check-prefixes=CHECK,GCN %s
// RUN: %clang_cc1 -triple amdgcn-unknown-unknown -target-cpu gfx1153 -emit-llvm -o - %s | FileCheck --check-prefixes=CHECK,GCN %s
// RUN: %clang_cc1 -triple amdgcn-unknown-unknown -target-cpu gfx1170 -emit-llvm -o - %s | FileCheck --check-prefixes=CHECK,GCN %s
// RUN: %clang_cc1 -triple amdgcn-unknown-unknown -target-cpu gfx1171 -emit-llvm -o - %s | FileCheck --check-prefixes=CHECK,GCN %s
// RUN: %clang_cc1 -triple amdgcn-unknown-unknown -target-cpu gfx1172 -emit-llvm -o - %s | FileCheck --check-prefixes=CHECK,GCN %s
// RUN: %clang_cc1 -triple spirv64-amd-amdhsa -emit-llvm -o - %s | FileCheck --check-prefixes=CHECK,AMDGCNSPIRV %s
typedef unsigned int uint;

2
clang/test/Driver/amdgpu-macros.cl
View File

@ -130,6 +130,8 @@
// RUN: %clang -E -dM -target amdgcn -mcpu=gfx1152 %s 2>&1 | FileCheck --check-prefixes=ARCH-GCN,FAST_FMAF %s -DWAVEFRONT_SIZE=32 -DCPU=gfx1152 -DFAMILY=GFX11
// RUN: %clang -E -dM -target amdgcn -mcpu=gfx1153 %s 2>&1 | FileCheck --check-prefixes=ARCH-GCN,FAST_FMAF %s -DWAVEFRONT_SIZE=32 -DCPU=gfx1153 -DFAMILY=GFX11
// RUN: %clang -E -dM -target amdgcn -mcpu=gfx1170 %s 2>&1 | FileCheck --check-prefixes=ARCH-GCN,FAST_FMAF %s -DWAVEFRONT_SIZE=32 -DCPU=gfx1170 -DFAMILY=GFX11
// RUN: %clang -E -dM -target amdgcn -mcpu=gfx1171 %s 2>&1 | FileCheck --check-prefixes=ARCH-GCN,FAST_FMAF %s -DWAVEFRONT_SIZE=32 -DCPU=gfx1171 -DFAMILY=GFX11
// RUN: %clang -E -dM -target amdgcn -mcpu=gfx1172 %s 2>&1 | FileCheck --check-prefixes=ARCH-GCN,FAST_FMAF %s -DWAVEFRONT_SIZE=32 -DCPU=gfx1172 -DFAMILY=GFX11
// RUN: %clang -E -dM -target amdgcn -mcpu=gfx1200 %s 2>&1 | FileCheck --check-prefixes=ARCH-GCN,FAST_FMAF %s -DWAVEFRONT_SIZE=32 -DCPU=gfx1200 -DFAMILY=GFX12
// RUN: %clang -E -dM -target amdgcn -mcpu=gfx1201 %s 2>&1 | FileCheck --check-prefixes=ARCH-GCN,FAST_FMAF %s -DWAVEFRONT_SIZE=32 -DCPU=gfx1201 -DFAMILY=GFX12
// RUN: %clang -E -dM -target amdgcn -mcpu=gfx1250 %s 2>&1 | FileCheck --check-prefixes=ARCH-GCN,FAST_FMAF %s -DWAVEFRONT_SIZE=32 -DCPU=gfx1250 -DFAMILY=GFX12

4
clang/test/Driver/amdgpu-mcpu.cl
View File

@ -114,6 +114,8 @@
// RUN: %clang -### -target amdgcn -mcpu=gfx1152 %s 2>&1 | FileCheck --check-prefix=GFX1152 %s
// RUN: %clang -### -target amdgcn -mcpu=gfx1153 %s 2>&1 | FileCheck --check-prefix=GFX1153 %s
// RUN: %clang -### -target amdgcn -mcpu=gfx1170 %s 2>&1 | FileCheck --check-prefix=GFX1170 %s
// RUN: %clang -### -target amdgcn -mcpu=gfx1171 %s 2>&1 | FileCheck --check-prefix=GFX1171 %s
// RUN: %clang -### -target amdgcn -mcpu=gfx1172 %s 2>&1 | FileCheck --check-prefix=GFX1172 %s
// RUN: %clang -### -target amdgcn -mcpu=gfx1200 %s 2>&1 | FileCheck --check-prefix=GFX1200 %s
// RUN: %clang -### -target amdgcn -mcpu=gfx1201 %s 2>&1 | FileCheck --check-prefix=GFX1201 %s
// RUN: %clang -### -target amdgcn -mcpu=gfx1250 %s 2>&1 | FileCheck --check-prefix=GFX1250 %s
@ -173,6 +175,8 @@
// GFX1152: "-target-cpu" "gfx1152"
// GFX1153: "-target-cpu" "gfx1153"
// GFX1170: "-target-cpu" "gfx1170"
// GFX1171: "-target-cpu" "gfx1171"
// GFX1172: "-target-cpu" "gfx1172"
// GFX1200: "-target-cpu" "gfx1200"
// GFX1201: "-target-cpu" "gfx1201"
// GFX1250: "-target-cpu" "gfx1250"

2
clang/test/Misc/target-invalid-cpu-note/amdgcn.c
View File

@ -67,6 +67,8 @@
// CHECK-SAME: {{^}}, gfx1152
// CHECK-SAME: {{^}}, gfx1153
// CHECK-SAME: {{^}}, gfx1170
// CHECK-SAME: {{^}}, gfx1171
// CHECK-SAME: {{^}}, gfx1172
// CHECK-SAME: {{^}}, gfx1200
// CHECK-SAME: {{^}}, gfx1201
// CHECK-SAME: {{^}}, gfx1250

2
clang/test/Misc/target-invalid-cpu-note/nvptx.c
View File

@ -88,6 +88,8 @@
// CHECK-SAME: {{^}}, gfx1152
// CHECK-SAME: {{^}}, gfx1153
// CHECK-SAME: {{^}}, gfx1170
// CHECK-SAME: {{^}}, gfx1171
// CHECK-SAME: {{^}}, gfx1172
// CHECK-SAME: {{^}}, gfx12-generic
// CHECK-SAME: {{^}}, gfx1200
// CHECK-SAME: {{^}}, gfx1201

32
llvm/docs/AMDGPUUsage.rst
View File

@ -525,6 +525,20 @@ Every processor supports every OS ABI (see :ref:`amdgpu-os`) with the following
work-item Add product
IDs names.
``gfx1171`` ``amdgcn`` APU - cumode - Architected *TBA*
- wavefrontsize64 flat
scratch .. TODO::
- Packed
work-item Add product
IDs names.
``gfx1172`` ``amdgcn`` APU - cumode - Architected *TBA*
- wavefrontsize64 flat
scratch .. TODO::
- Packed
work-item Add product
IDs names.
**GCN GFX12 (RDNA 4)** [AMD-GCN-GFX12-RDNA4]_
-----------------------------------------------------------------------------------------------------------------------
``gfx1200`` ``amdgcn`` dGPU - cumode - Architected - Radeon RX 9060
@ -2000,11 +2014,11 @@ The AMDGPU backend supports the following LLVM IR attributes.
"amdgpu-flat-work-group-size" value, the implied occupancy
bounds by the workgroup size takes precedence.
"amdgpu-ieee" true/false. GFX6-GFX11 (Except GFX1170) Only
"amdgpu-ieee" true/false. GFX6-GFX11 (Except GFX11.7) Only
Specify whether the function expects the IEEE field of the
mode register to be set on entry. Overrides the default for
the calling convention.
"amdgpu-dx10-clamp" true/false. GFX6-GFX11 (Except GFX1170) Only
"amdgpu-dx10-clamp" true/false. GFX6-GFX11 (Except GFX11.7) Only
Specify whether the function expects the DX10_CLAMP field of
the mode register to be set on entry. Overrides the default
for the calling convention.
@ -2760,7 +2774,9 @@ The AMDGPU backend uses the following ELF header:
``EF_AMDGPU_MACH_AMDGCN_GFX12_GENERIC`` 0x059 ``gfx12-generic``
``EF_AMDGPU_MACH_AMDGCN_GFX1251`` 0x05a ``gfx1251``
``EF_AMDGPU_MACH_AMDGCN_GFX12_5_GENERIC`` 0x05b ``gfx12-5-generic``
``EF_AMDGPU_MACH_AMDGCN_GFX1172`` 0x05c ``gfx1172``
``EF_AMDGPU_MACH_AMDGCN_GFX1170`` 0x05d ``gfx1170``
``EF_AMDGPU_MACH_AMDGCN_GFX1171`` 0x05e ``gfx1171``
``EF_AMDGPU_MACH_AMDGCN_GFX9_4_GENERIC`` 0x05f ``gfx9-4-generic``
*reserved* 0x060 Reserved.
*reserved* 0x070 Reserved.
@ -5789,7 +5805,7 @@ The fields used by CP for code objects before V3 also match those specified in
CP is responsible for
filling in
``COMPUTE_PGM_RSRC1.PRIV``.
21 1 bit ENABLE_DX10_CLAMP GFX9-GFX11 (except GFX1170)
21 1 bit ENABLE_DX10_CLAMP GFX9-GFX11 (except GFX11.7)
Wavefront starts execution
with DX10 clamp mode
enabled. Used by the vector
@ -5801,7 +5817,7 @@ The fields used by CP for code objects before V3 also match those specified in
Used by CP to set up
``COMPUTE_PGM_RSRC1.DX10_CLAMP``.
GFX1170
GFX11.7
Reserved. Must be 0.
WG_RR_EN GFX12
If 1, wavefronts are scheduled
@ -5820,7 +5836,7 @@ The fields used by CP for code objects before V3 also match those specified in
CP is responsible for
filling in
``COMPUTE_PGM_RSRC1.DEBUG_MODE``.
23 1 bit ENABLE_IEEE_MODE GFX9-GFX11 (except GFX1170)
23 1 bit ENABLE_IEEE_MODE GFX9-GFX11 (except GFX11.7)
Wavefront starts execution
with IEEE mode
enabled. Floating point
@ -5836,7 +5852,7 @@ The fields used by CP for code objects before V3 also match those specified in
Used by CP to set up
``COMPUTE_PGM_RSRC1.IEEE_MODE``.
GFX1170
GFX11.7
Reserved. Must be 0.
DISABLE_PERF GFX12
Reserved. Must be 0.
@ -21534,10 +21550,10 @@ terminated by an ``.end_amdhsa_kernel`` directive.
:ref:`amdgpu-amdhsa-floating-point-denorm-mode-enumeration-values-table`.
``.amdhsa_dx10_clamp`` 1 GFX6-GFX11 Controls ENABLE_DX10_CLAMP in
(except :ref:`amdgpu-amdhsa-compute_pgm_rsrc1-gfx6-gfx12-table`.
GFX1170)
GFX11.7)
``.amdhsa_ieee_mode`` 1 GFX6-GFX11 Controls ENABLE_IEEE_MODE in
(except :ref:`amdgpu-amdhsa-compute_pgm_rsrc1-gfx6-gfx12-table`.
GFX1170)
GFX11.7)
``.amdhsa_round_robin_scheduling`` 0 GFX12 Controls ENABLE_WG_RR_EN in
:ref:`amdgpu-amdhsa-compute_pgm_rsrc1-gfx6-gfx12-table`.
``.amdhsa_fp16_overflow`` 0 GFX9-GFX12 Controls FP16_OVFL in

2
llvm/include/llvm/BinaryFormat/ELF.h
View File

@ -834,7 +834,9 @@ enum {
X(0x59, EF_AMDGPU_MACH_AMDGCN_GFX12_GENERIC, "gfx12-generic") \
X(0x5a, EF_AMDGPU_MACH_AMDGCN_GFX1251, "gfx1251") \
X(0x5b, EF_AMDGPU_MACH_AMDGCN_GFX12_5_GENERIC, "gfx12-5-generic") \
X(0x5c, EF_AMDGPU_MACH_AMDGCN_GFX1172, "gfx1172") \
X(0x5d, EF_AMDGPU_MACH_AMDGCN_GFX1170, "gfx1170") \
X(0x5e, EF_AMDGPU_MACH_AMDGCN_GFX1171, "gfx1171") \
X(0x5f, EF_AMDGPU_MACH_AMDGCN_GFX9_4_GENERIC, "gfx9-4-generic")
enum : unsigned {

2
llvm/include/llvm/TargetParser/AMDGPUTargetParser.def
View File

@ -119,6 +119,8 @@ AMDGCN_GPU ("gfx1151", GK_GFX1151, (11, 5, 1), FEATURE_FAST_FMA_F32|FEAT
AMDGCN_GPU ("gfx1152", GK_GFX1152, (11, 5, 2), FEATURE_FAST_FMA_F32|FEATURE_FAST_DENORMAL_F32|FEATURE_WAVE32|FEATURE_WGP)
AMDGCN_GPU ("gfx1153", GK_GFX1153, (11, 5, 3), FEATURE_FAST_FMA_F32|FEATURE_FAST_DENORMAL_F32|FEATURE_WAVE32|FEATURE_WGP)
AMDGCN_GPU ("gfx1170", GK_GFX1170, (11, 7, 0), FEATURE_FAST_FMA_F32|FEATURE_FAST_DENORMAL_F32|FEATURE_WAVE32|FEATURE_WGP)
AMDGCN_GPU ("gfx1171", GK_GFX1171, (11, 7, 1), FEATURE_FAST_FMA_F32|FEATURE_FAST_DENORMAL_F32|FEATURE_WAVE32|FEATURE_WGP)
AMDGCN_GPU ("gfx1172", GK_GFX1172, (11, 7, 2), FEATURE_FAST_FMA_F32|FEATURE_FAST_DENORMAL_F32|FEATURE_WAVE32|FEATURE_WGP)
AMDGCN_GPU ("gfx1200", GK_GFX1200, (12, 0, 0), FEATURE_FAST_FMA_F32|FEATURE_FAST_DENORMAL_F32|FEATURE_WAVE32|FEATURE_WGP)
AMDGCN_GPU ("gfx1201", GK_GFX1201, (12, 0, 1), FEATURE_FAST_FMA_F32|FEATURE_FAST_DENORMAL_F32|FEATURE_WAVE32|FEATURE_WGP)
AMDGCN_GPU ("gfx1250", GK_GFX1250, (12, 5, 0), FEATURE_FAST_FMA_F32|FEATURE_FAST_DENORMAL_F32|FEATURE_WAVE32|FEATURE_XNACK_ALWAYS)

4
llvm/lib/Target/AMDGPU/AMDGPU.td
View File

@ -1893,7 +1893,7 @@ def FeatureISAVersion11_Common : FeatureSet<
// There are few workarounds that need to be added to all targets. This
// pessimizes codegen a bit on the generic GFX11 target. This generic target
// does not include GFX1170 due to incompatible changes.
// does not include GFX11_7 due to incompatible changes.
def FeatureISAVersion11_Generic: FeatureSet<
!listconcat(FeatureISAVersion11_Common.Features,
[FeatureMSAALoadDstSelBug,
@ -1960,7 +1960,7 @@ def FeatureISAVersion11_5_3 : FeatureSet<
!listconcat(FeatureISAVersion11_5_Common.Features,
[])>;
def FeatureISAVersion11_7_0 : FeatureSet<
def FeatureISAVersion11_7_Common : FeatureSet<
!listconcat(FeatureISAVersion11_Common.Features,
[FeatureGFX11_7Insts,
FeatureSALUFloatInsts,

10
llvm/lib/Target/AMDGPU/GCNProcessors.td
View File

@ -305,7 +305,15 @@ def : ProcessorModel<"gfx1153", GFX11SpeedModel,
>;
def : ProcessorModel<"gfx1170", GFX11SpeedModel,
FeatureISAVersion11_7_0.Features
FeatureISAVersion11_7_Common.Features
>;
def : ProcessorModel<"gfx1171", GFX11SpeedModel,
FeatureISAVersion11_7_Common.Features
>;
def : ProcessorModel<"gfx1172", GFX11SpeedModel,
FeatureISAVersion11_7_Common.Features
>;
// [gfx1100, gfx1101, gfx1102, gfx1103, gfx1150, gfx1151, gfx1152, gfx1153]

4
llvm/lib/Target/AMDGPU/MCTargetDesc/AMDGPUTargetStreamer.cpp
View File

@ -116,6 +116,8 @@ StringRef AMDGPUTargetStreamer::getArchNameFromElfMach(unsigned ElfMach) {
case ELF::EF_AMDGPU_MACH_AMDGCN_GFX1152: AK = GK_GFX1152; break;
case ELF::EF_AMDGPU_MACH_AMDGCN_GFX1153: AK = GK_GFX1153; break;
case ELF::EF_AMDGPU_MACH_AMDGCN_GFX1170: AK = GK_GFX1170; break;
case ELF::EF_AMDGPU_MACH_AMDGCN_GFX1171: AK = GK_GFX1171; break;
case ELF::EF_AMDGPU_MACH_AMDGCN_GFX1172: AK = GK_GFX1172; break;
case ELF::EF_AMDGPU_MACH_AMDGCN_GFX1200: AK = GK_GFX1200; break;
case ELF::EF_AMDGPU_MACH_AMDGCN_GFX1201: AK = GK_GFX1201; break;
case ELF::EF_AMDGPU_MACH_AMDGCN_GFX1250: AK = GK_GFX1250; break;
@ -206,6 +208,8 @@ unsigned AMDGPUTargetStreamer::getElfMach(StringRef GPU) {
case GK_GFX1152: return ELF::EF_AMDGPU_MACH_AMDGCN_GFX1152;
case GK_GFX1153: return ELF::EF_AMDGPU_MACH_AMDGCN_GFX1153;
case GK_GFX1170: return ELF::EF_AMDGPU_MACH_AMDGCN_GFX1170;
case GK_GFX1171: return ELF::EF_AMDGPU_MACH_AMDGCN_GFX1171;
case GK_GFX1172: return ELF::EF_AMDGPU_MACH_AMDGCN_GFX1172;
case GK_GFX1200: return ELF::EF_AMDGPU_MACH_AMDGCN_GFX1200;
case GK_GFX1201: return ELF::EF_AMDGPU_MACH_AMDGCN_GFX1201;
case GK_GFX1250: return ELF::EF_AMDGPU_MACH_AMDGCN_GFX1250;

2
llvm/lib/TargetParser/TargetParser.cpp
View File

@ -340,6 +340,8 @@ static void fillAMDGCNFeatureMap(StringRef GPU, const Triple &T,
Features["atomic-fmin-fmax-global-f32"] = true;
break;
case GK_GFX1170:
case GK_GFX1171:
case GK_GFX1172:
Features["ci-insts"] = true;
Features["dot7-insts"] = true;
Features["dot8-insts"] = true;

4
llvm/test/CodeGen/AMDGPU/directive-amdgcn-target.ll
View File

@ -105,6 +105,8 @@
; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx1152 < %s | FileCheck --check-prefixes=GFX1152 %s
; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx1153 < %s | FileCheck --check-prefixes=GFX1153 %s
; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx1170 < %s | FileCheck --check-prefixes=GFX1170 %s
; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx1171 < %s | FileCheck --check-prefixes=GFX1171 %s
; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx1172 < %s | FileCheck --check-prefixes=GFX1172 %s
; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx1200 < %s | FileCheck --check-prefixes=GFX1200 %s
; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx1201 < %s | FileCheck --check-prefixes=GFX1201 %s
; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx1250 < %s | FileCheck --check-prefixes=GFX1250 %s
@ -207,6 +209,8 @@
; GFX1152: .amdgcn_target "amdgcn-amd-amdhsa--gfx1152"
; GFX1153: .amdgcn_target "amdgcn-amd-amdhsa--gfx1153"
; GFX1170: .amdgcn_target "amdgcn-amd-amdhsa--gfx1170"
; GFX1171: .amdgcn_target "amdgcn-amd-amdhsa--gfx1171"
; GFX1172: .amdgcn_target "amdgcn-amd-amdhsa--gfx1172"
; GFX1200: .amdgcn_target "amdgcn-amd-amdhsa--gfx1200"
; GFX1201: .amdgcn_target "amdgcn-amd-amdhsa--gfx1201"
; GFX1250: .amdgcn_target "amdgcn-amd-amdhsa--gfx1250"

4
llvm/test/CodeGen/AMDGPU/elf-header-flags-mach.ll
View File

@ -76,6 +76,8 @@
; RUN: llc -filetype=obj -mtriple=amdgcn -mcpu=gfx1152 < %s | llvm-readobj --file-header - | FileCheck --check-prefixes=ALL,ARCH-GCN,GFX1152 %s
; RUN: llc -filetype=obj -mtriple=amdgcn -mcpu=gfx1153 < %s | llvm-readobj --file-header - | FileCheck --check-prefixes=ALL,ARCH-GCN,GFX1153 %s
; RUN: llc -filetype=obj -mtriple=amdgcn -mcpu=gfx1170 < %s | llvm-readobj --file-header - | FileCheck --check-prefixes=ALL,ARCH-GCN,GFX1170 %s
; RUN: llc -filetype=obj -mtriple=amdgcn -mcpu=gfx1171 < %s | llvm-readobj --file-header - | FileCheck --check-prefixes=ALL,ARCH-GCN,GFX1171 %s
; RUN: llc -filetype=obj -mtriple=amdgcn -mcpu=gfx1172 < %s | llvm-readobj --file-header - | FileCheck --check-prefixes=ALL,ARCH-GCN,GFX1172 %s
; RUN: llc -filetype=obj -mtriple=amdgcn -mcpu=gfx1200 < %s | llvm-readobj --file-header - | FileCheck --check-prefixes=ALL,ARCH-GCN,GFX1200 %s
; RUN: llc -filetype=obj -mtriple=amdgcn -mcpu=gfx1201 < %s | llvm-readobj --file-header - | FileCheck --check-prefixes=ALL,ARCH-GCN,GFX1201 %s
; RUN: llc -filetype=obj -mtriple=amdgcn -mcpu=gfx1250 < %s | llvm-readobj --file-header - | FileCheck --check-prefixes=ALL,ARCH-GCN,GFX1250 %s
@ -162,6 +164,8 @@
; GFX1152: EF_AMDGPU_MACH_AMDGCN_GFX1152 (0x55)
; GFX1153: EF_AMDGPU_MACH_AMDGCN_GFX1153 (0x58)
; GFX1170: EF_AMDGPU_MACH_AMDGCN_GFX1170 (0x5D)
; GFX1171: EF_AMDGPU_MACH_AMDGCN_GFX1171 (0x5E)
; GFX1172: EF_AMDGPU_MACH_AMDGCN_GFX1172 (0x5C)
; GFX1200: EF_AMDGPU_MACH_AMDGCN_GFX1200 (0x48)
; GFX1201: EF_AMDGPU_MACH_AMDGCN_GFX1201 (0x4E)
; GFX1250: EF_AMDGPU_MACH_AMDGCN_GFX1250 (0x49)

14
llvm/test/Object/AMDGPU/elf-header-flags-mach.yaml
View File

@ -238,6 +238,14 @@
# RUN: llvm-readobj -S --file-headers %t.o.AMDGCN_GFX1170 | FileCheck --check-prefixes=ELF-AMDGCN-ALL,ELF-AMDGCN-GFX1170 %s
# RUN: obj2yaml %t.o.AMDGCN_GFX1170 | FileCheck --check-prefixes=YAML-AMDGCN-ALL,YAML-AMDGCN-GFX1170 %s
# RUN: sed -e 's/<BITS>/64/' -e 's/<MACH>/AMDGCN_GFX1171/' %s | yaml2obj -o %t.o.AMDGCN_GFX1171
# RUN: llvm-readobj -S --file-headers %t.o.AMDGCN_GFX1171 | FileCheck --check-prefixes=ELF-AMDGCN-ALL,ELF-AMDGCN-GFX1171 %s
# RUN: obj2yaml %t.o.AMDGCN_GFX1171 | FileCheck --check-prefixes=YAML-AMDGCN-ALL,YAML-AMDGCN-GFX1171 %s
# RUN: sed -e 's/<BITS>/64/' -e 's/<MACH>/AMDGCN_GFX1172/' %s | yaml2obj -o %t.o.AMDGCN_GFX1172
# RUN: llvm-readobj -S --file-headers %t.o.AMDGCN_GFX1172 | FileCheck --check-prefixes=ELF-AMDGCN-ALL,ELF-AMDGCN-GFX1172 %s
# RUN: obj2yaml %t.o.AMDGCN_GFX1172 | FileCheck --check-prefixes=YAML-AMDGCN-ALL,YAML-AMDGCN-GFX1172 %s
# RUN: sed -e 's/<BITS>/64/' -e 's/<MACH>/AMDGCN_GFX1200/' %s | yaml2obj -o %t.o.AMDGCN_GFX1200
# RUN: llvm-readobj -S --file-headers %t.o.AMDGCN_GFX1200 | FileCheck --check-prefixes=ELF-AMDGCN-ALL,ELF-AMDGCN-GFX1200 %s
# RUN: obj2yaml %t.o.AMDGCN_GFX1200 | FileCheck --check-prefixes=YAML-AMDGCN-ALL,YAML-AMDGCN-GFX1200 %s
@ -484,6 +492,12 @@
# ELF-AMDGCN-GFX1170: EF_AMDGPU_MACH_AMDGCN_GFX1170 (0x5D)
# YAML-AMDGCN-GFX1170: Flags: [ EF_AMDGPU_MACH_AMDGCN_GFX1170 ]
# ELF-AMDGCN-GFX1171: EF_AMDGPU_MACH_AMDGCN_GFX1171 (0x5E)
# YAML-AMDGCN-GFX1171: Flags: [ EF_AMDGPU_MACH_AMDGCN_GFX1171 ]
# ELF-AMDGCN-GFX1172: EF_AMDGPU_MACH_AMDGCN_GFX1172 (0x5C)
# YAML-AMDGCN-GFX1172: Flags: [ EF_AMDGPU_MACH_AMDGCN_GFX1172 ]
# ELF-AMDGCN-GFX1200: EF_AMDGPU_MACH_AMDGCN_GFX1200 (0x48)
# YAML-AMDGCN-GFX1200: Flags: [ EF_AMDGPU_MACH_AMDGCN_GFX1200 ]

10
llvm/test/tools/llvm-objdump/ELF/AMDGPU/subtarget.ll
View File

@ -50,6 +50,16 @@ define amdgpu_kernel void @test_kernel() {
; RUN: llvm-objdump -D -mllvm --amdhsa-code-object-version=6 %t.o > %t-detect.txt
; RUN: diff %t-specify.txt %t-detect.txt
; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx1172 -filetype=obj -O0 -o %t.o %s
; RUN: llvm-objdump -D --arch-name=amdgcn --mcpu=gfx1172 %t.o > %t-specify.txt
; RUN: llvm-objdump -D %t.o > %t-detect.txt
; RUN: diff %t-specify.txt %t-detect.txt
; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx1171 -filetype=obj -O0 -o %t.o %s
; RUN: llvm-objdump -D --arch-name=amdgcn --mcpu=gfx1171 %t.o > %t-specify.txt
; RUN: llvm-objdump -D %t.o > %t-detect.txt
; RUN: diff %t-specify.txt %t-detect.txt
; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx1170 -filetype=obj -O0 -o %t.o %s
; RUN: llvm-objdump -D --arch-name=amdgcn --mcpu=gfx1170 %t.o > %t-specify.txt
; RUN: llvm-objdump -D %t.o > %t-detect.txt

18
llvm/test/tools/llvm-readobj/ELF/AMDGPU/elf-headers.test
View File

@ -442,6 +442,24 @@
# RUN: yaml2obj %s -o %t -DABI_VERSION=2 -DFLAG_NAME=EF_AMDGPU_MACH_AMDGCN_GFX1170
# RUN: llvm-readobj -h %t | FileCheck %s --check-prefixes=ALL,KNOWN-ABI-VERSION,SINGLE-FLAG --match-full-lines -DABI_VERSION=2 -DFILE=%t -DFLAG_NAME=EF_AMDGPU_MACH_AMDGCN_GFX1170 -DFLAG_VALUE=0x5D
# RUN: yaml2obj %s -o %t -DABI_VERSION=0 -DFLAG_NAME=EF_AMDGPU_MACH_AMDGCN_GFX1171
# RUN: llvm-readobj -h %t | FileCheck %s --check-prefixes=ALL,KNOWN-ABI-VERSION,SINGLE-FLAG --match-full-lines -DABI_VERSION=0 -DFILE=%t -DFLAG_NAME=EF_AMDGPU_MACH_AMDGCN_GFX1171 -DFLAG_VALUE=0x5E
# RUN: yaml2obj %s -o %t -DABI_VERSION=1 -DFLAG_NAME=EF_AMDGPU_MACH_AMDGCN_GFX1171
# RUN: llvm-readobj -h %t | FileCheck %s --check-prefixes=ALL,KNOWN-ABI-VERSION,SINGLE-FLAG --match-full-lines -DABI_VERSION=1 -DFILE=%t -DFLAG_NAME=EF_AMDGPU_MACH_AMDGCN_GFX1171 -DFLAG_VALUE=0x5E
# RUN: yaml2obj %s -o %t -DABI_VERSION=2 -DFLAG_NAME=EF_AMDGPU_MACH_AMDGCN_GFX1171
# RUN: llvm-readobj -h %t | FileCheck %s --check-prefixes=ALL,KNOWN-ABI-VERSION,SINGLE-FLAG --match-full-lines -DABI_VERSION=2 -DFILE=%t -DFLAG_NAME=EF_AMDGPU_MACH_AMDGCN_GFX1171 -DFLAG_VALUE=0x5E
# RUN: yaml2obj %s -o %t -DABI_VERSION=0 -DFLAG_NAME=EF_AMDGPU_MACH_AMDGCN_GFX1172
# RUN: llvm-readobj -h %t | FileCheck %s --check-prefixes=ALL,KNOWN-ABI-VERSION,SINGLE-FLAG --match-full-lines -DABI_VERSION=0 -DFILE=%t -DFLAG_NAME=EF_AMDGPU_MACH_AMDGCN_GFX1172 -DFLAG_VALUE=0x5C
# RUN: yaml2obj %s -o %t -DABI_VERSION=1 -DFLAG_NAME=EF_AMDGPU_MACH_AMDGCN_GFX1172
# RUN: llvm-readobj -h %t | FileCheck %s --check-prefixes=ALL,KNOWN-ABI-VERSION,SINGLE-FLAG --match-full-lines -DABI_VERSION=1 -DFILE=%t -DFLAG_NAME=EF_AMDGPU_MACH_AMDGCN_GFX1172 -DFLAG_VALUE=0x5C
# RUN: yaml2obj %s -o %t -DABI_VERSION=2 -DFLAG_NAME=EF_AMDGPU_MACH_AMDGCN_GFX1172
# RUN: llvm-readobj -h %t | FileCheck %s --check-prefixes=ALL,KNOWN-ABI-VERSION,SINGLE-FLAG --match-full-lines -DABI_VERSION=2 -DFILE=%t -DFLAG_NAME=EF_AMDGPU_MACH_AMDGCN_GFX1172 -DFLAG_VALUE=0x5C
# RUN: yaml2obj %s -o %t -DABI_VERSION=0 -DFLAG_NAME=EF_AMDGPU_MACH_AMDGCN_GFX1200
# RUN: llvm-readobj -h %t | FileCheck %s --check-prefixes=ALL,KNOWN-ABI-VERSION,SINGLE-FLAG --match-full-lines -DABI_VERSION=0 -DFILE=%t -DFLAG_NAME=EF_AMDGPU_MACH_AMDGCN_GFX1200 -DFLAG_VALUE=0x48