Skip to content

Commit 3453f3d

Browse files
aakanksha555aakanksha555
committed
[AMDGPU] Add gfx1035 target
Differential Revision: https://reviews.llvm.org/D104804
1 parent d064182 commit 3453f3d

File tree

29 files changed

+66
-4
lines changed

29 files changed

+66
-4
lines changed

‎clang/include/clang/Basic/Cuda.h

+1
Original file line numberDiff line numberDiff line change
@@ -89,6 +89,7 @@ enum class CudaArch {
8989
GFX1032,
9090
GFX1033,
9191
GFX1034,
92+
GFX1035,
9293
LAST,
9394
};
9495

‎clang/lib/Basic/Cuda.cpp

+1
Original file line numberDiff line numberDiff line change
@@ -111,6 +111,7 @@ static const CudaArchToStringMap arch_names[] = {
111111
GFX(1032), // gfx1032
112112
GFX(1033), // gfx1033
113113
GFX(1034), // gfx1034
114+
GFX(1035), // gfx1035
114115
// clang-format on
115116
};
116117
#undef SM

‎clang/lib/Basic/Targets/AMDGPU.cpp

+1
Original file line numberDiff line numberDiff line change
@@ -184,6 +184,7 @@ bool AMDGPUTargetInfo::initFeatureMap(
184184
// XXX - What does the member GPU mean if device name string passed here?
185185
if (isAMDGCN(getTriple())) {
186186
switch (llvm::AMDGPU::parseArchAMDGCN(CPU)) {
187+
case GK_GFX1035:
187188
case GK_GFX1034:
188189
case GK_GFX1033:
189190
case GK_GFX1032:

‎clang/lib/Basic/Targets/NVPTX.cpp

+1
Original file line numberDiff line numberDiff line change
@@ -213,6 +213,7 @@ void NVPTXTargetInfo::getTargetDefines(const LangOptions &Opts,
213213
case CudaArch::GFX1032:
214214
case CudaArch::GFX1033:
215215
case CudaArch::GFX1034:
216+
case CudaArch::GFX1035:
216217
case CudaArch::LAST:
217218
break;
218219
case CudaArch::UNUSED:

‎clang/lib/CodeGen/CGOpenMPRuntimeGPU.cpp

+1
Original file line numberDiff line numberDiff line change
@@ -4194,6 +4194,7 @@ void CGOpenMPRuntimeGPU::processRequiresDirective(
41944194
case CudaArch::GFX1032:
41954195
case CudaArch::GFX1033:
41964196
case CudaArch::GFX1034:
4197+
case CudaArch::GFX1035:
41974198
case CudaArch::UNUSED:
41984199
case CudaArch::UNKNOWN:
41994200
break;

‎clang/test/CodeGenOpenCL/amdgpu-features.cl

+2
Original file line numberDiff line numberDiff line change
@@ -34,6 +34,7 @@
3434
// RUN: %clang_cc1 -triple amdgcn -target-cpu gfx1032 -S -emit-llvm -o - %s | FileCheck --check-prefix=GFX1032 %s
3535
// RUN: %clang_cc1 -triple amdgcn -target-cpu gfx1033 -S -emit-llvm -o - %s | FileCheck --check-prefix=GFX1033 %s
3636
// RUN: %clang_cc1 -triple amdgcn -target-cpu gfx1034 -S -emit-llvm -o - %s | FileCheck --check-prefix=GFX1034 %s
37+
// RUN: %clang_cc1 -triple amdgcn -target-cpu gfx1035 -S -emit-llvm -o - %s | FileCheck --check-prefix=GFX1035 %s
3738

3839
// GFX600: "target-features"="+s-memtime-inst"
3940
// GFX601: "target-features"="+s-memtime-inst"
@@ -66,5 +67,6 @@
6667
// GFX1032: "target-features"="+16-bit-insts,+ci-insts,+dl-insts,+dot1-insts,+dot2-insts,+dot5-insts,+dot6-insts,+dot7-insts,+dpp,+flat-address-space,+gfx10-3-insts,+gfx10-insts,+gfx8-insts,+gfx9-insts,+s-memrealtime,+s-memtime-inst"
6768
// GFX1033: "target-features"="+16-bit-insts,+ci-insts,+dl-insts,+dot1-insts,+dot2-insts,+dot5-insts,+dot6-insts,+dot7-insts,+dpp,+flat-address-space,+gfx10-3-insts,+gfx10-insts,+gfx8-insts,+gfx9-insts,+s-memrealtime,+s-memtime-inst"
6869
// GFX1034: "target-features"="+16-bit-insts,+ci-insts,+dl-insts,+dot1-insts,+dot2-insts,+dot5-insts,+dot6-insts,+dot7-insts,+dpp,+flat-address-space,+gfx10-3-insts,+gfx10-insts,+gfx8-insts,+gfx9-insts,+s-memrealtime,+s-memtime-inst"
70+
// GFX1035: "target-features"="+16-bit-insts,+ci-insts,+dl-insts,+dot1-insts,+dot2-insts,+dot5-insts,+dot6-insts,+dot7-insts,+dpp,+flat-address-space,+gfx10-3-insts,+gfx10-insts,+gfx8-insts,+gfx9-insts,+s-memrealtime,+s-memtime-inst"
6971

7072
kernel void test() {}

‎clang/test/Driver/amdgpu-macros.cl

+1
Original file line numberDiff line numberDiff line change
@@ -116,6 +116,7 @@
116116
// RUN: %clang -E -dM -target amdgcn -mcpu=gfx1032 %s 2>&1 | FileCheck --check-prefixes=ARCH-GCN,FAST_FMAF %s -DWAVEFRONT_SIZE=32 -DCPU=gfx1032
117117
// RUN: %clang -E -dM -target amdgcn -mcpu=gfx1033 %s 2>&1 | FileCheck --check-prefixes=ARCH-GCN,FAST_FMAF %s -DWAVEFRONT_SIZE=32 -DCPU=gfx1033
118118
// RUN: %clang -E -dM -target amdgcn -mcpu=gfx1034 %s 2>&1 | FileCheck --check-prefixes=ARCH-GCN,FAST_FMAF %s -DWAVEFRONT_SIZE=32 -DCPU=gfx1034
119+
// RUN: %clang -E -dM -target amdgcn -mcpu=gfx1035 %s 2>&1 | FileCheck --check-prefixes=ARCH-GCN,FAST_FMAF %s -DWAVEFRONT_SIZE=32 -DCPU=gfx1035
119120

120121
// ARCH-GCN-DAG: #define FP_FAST_FMA 1
121122

‎clang/test/Driver/amdgpu-mcpu.cl

+2
Original file line numberDiff line numberDiff line change
@@ -101,6 +101,7 @@
101101
// RUN: %clang -### -target amdgcn -mcpu=gfx1032 %s 2>&1 | FileCheck --check-prefix=GFX1032 %s
102102
// RUN: %clang -### -target amdgcn -mcpu=gfx1033 %s 2>&1 | FileCheck --check-prefix=GFX1033 %s
103103
// RUN: %clang -### -target amdgcn -mcpu=gfx1034 %s 2>&1 | FileCheck --check-prefix=GFX1034 %s
104+
// RUN: %clang -### -target amdgcn -mcpu=gfx1035 %s 2>&1 | FileCheck --check-prefix=GFX1035 %s
104105

105106
// GCNDEFAULT-NOT: -target-cpu
106107
// GFX600: "-target-cpu" "gfx600"
@@ -134,3 +135,4 @@
134135
// GFX1032: "-target-cpu" "gfx1032"
135136
// GFX1033: "-target-cpu" "gfx1033"
136137
// GFX1034: "-target-cpu" "gfx1034"
138+
// GFX1035: "-target-cpu" "gfx1035"

‎clang/test/Misc/target-invalid-cpu-note.c

+1-1
Original file line numberDiff line numberDiff line change
@@ -87,7 +87,7 @@
8787
// AMDGCN-SAME: gfx802, iceland, tonga, gfx803, fiji, polaris10, polaris11,
8888
// AMDGCN-SAME: gfx805, tongapro, gfx810, stoney, gfx900, gfx902, gfx904, gfx906,
8989
// AMDGCN-SAME: gfx908, gfx909, gfx90a, gfx90c, gfx1010, gfx1011, gfx1012, gfx1013, gfx1030, gfx1031,
90-
// AMDGCN-SAME: gfx1032, gfx1033, gfx1034
90+
// AMDGCN-SAME: gfx1032, gfx1033, gfx1034, gfx1035
9191

9292
// RUN: not %clang_cc1 -triple wasm64--- -target-cpu not-a-cpu -fsyntax-only %s 2>&1 | FileCheck %s --check-prefix WEBASM
9393
// WEBASM: error: unknown target CPU 'not-a-cpu'

‎llvm/docs/AMDGPUUsage.rst

+7-1
Original file line numberDiff line numberDiff line change
@@ -421,6 +421,12 @@ Every processor supports every OS ABI (see :ref:`amdgpu-os`) with the following
421421
Add product
422422
names.
423423

424+
``gfx1035`` ``amdgcn`` APU - cumode - Absolute - *pal-amdpal* *TBA*
425+
- wavefrontsize64 flat
426+
scratch .. TODO::
427+
Add product
428+
names.
429+
424430
=========== =============== ============ ===== ================= =============== =============== ======================
425431

426432
.. _amdgpu-target-features:
@@ -1151,7 +1157,7 @@ The AMDGPU backend uses the following ELF header:
11511157
``EF_AMDGPU_MACH_AMDGCN_GFX602`` 0x03a ``gfx602``
11521158
``EF_AMDGPU_MACH_AMDGCN_GFX705`` 0x03b ``gfx705``
11531159
``EF_AMDGPU_MACH_AMDGCN_GFX805`` 0x03c ``gfx805``
1154-
*reserved* 0x03d Reserved.
1160+
``EF_AMDGPU_MACH_AMDGCN_GFX1035`` 0x03d ``gfx1035``
11551161
``EF_AMDGPU_MACH_AMDGCN_GFX1034`` 0x03e ``gfx1034``
11561162
``EF_AMDGPU_MACH_AMDGCN_GFX90A`` 0x03f ``gfx90a``
11571163
*reserved* 0x040 Reserved.

‎llvm/include/llvm/BinaryFormat/ELF.h

+1-1
Original file line numberDiff line numberDiff line change
@@ -737,7 +737,7 @@ enum : unsigned {
737737
EF_AMDGPU_MACH_AMDGCN_GFX602 = 0x03a,
738738
EF_AMDGPU_MACH_AMDGCN_GFX705 = 0x03b,
739739
EF_AMDGPU_MACH_AMDGCN_GFX805 = 0x03c,
740-
EF_AMDGPU_MACH_AMDGCN_RESERVED_0X3D = 0x03d,
740+
EF_AMDGPU_MACH_AMDGCN_GFX1035 = 0x03d,
741741
EF_AMDGPU_MACH_AMDGCN_GFX1034 = 0x03e,
742742
EF_AMDGPU_MACH_AMDGCN_GFX90A = 0x03f,
743743
EF_AMDGPU_MACH_AMDGCN_RESERVED_0X40 = 0x040,

‎llvm/include/llvm/Support/TargetParser.h

+2-1
Original file line numberDiff line numberDiff line change
@@ -95,9 +95,10 @@ enum GPUKind : uint32_t {
9595
GK_GFX1032 = 77,
9696
GK_GFX1033 = 78,
9797
GK_GFX1034 = 79,
98+
GK_GFX1035 = 80,
9899

99100
GK_AMDGCN_FIRST = GK_GFX600,
100-
GK_AMDGCN_LAST = GK_GFX1034,
101+
GK_AMDGCN_LAST = GK_GFX1035,
101102
};
102103

103104
/// Instruction set architecture version.

‎llvm/lib/Object/ELFObjectFile.cpp

+2
Original file line numberDiff line numberDiff line change
@@ -481,6 +481,8 @@ StringRef ELFObjectFileBase::getAMDGPUCPUName() const {
481481
return "gfx1033";
482482
case ELF::EF_AMDGPU_MACH_AMDGCN_GFX1034:
483483
return "gfx1034";
484+
case ELF::EF_AMDGPU_MACH_AMDGCN_GFX1035:
485+
return "gfx1035";
484486
default:
485487
llvm_unreachable("Unknown EF_AMDGPU_MACH value");
486488
}

‎llvm/lib/ObjectYAML/ELFYAML.cpp

+1
Original file line numberDiff line numberDiff line change
@@ -555,6 +555,7 @@ void ScalarBitSetTraits<ELFYAML::ELF_EF>::bitset(IO &IO,
555555
BCaseMask(EF_AMDGPU_MACH_AMDGCN_GFX1032, EF_AMDGPU_MACH);
556556
BCaseMask(EF_AMDGPU_MACH_AMDGCN_GFX1033, EF_AMDGPU_MACH);
557557
BCaseMask(EF_AMDGPU_MACH_AMDGCN_GFX1034, EF_AMDGPU_MACH);
558+
BCaseMask(EF_AMDGPU_MACH_AMDGCN_GFX1035, EF_AMDGPU_MACH);
558559
switch (Object->Header.ABIVersion) {
559560
default:
560561
// ELFOSABI_AMDGPU_PAL, ELFOSABI_AMDGPU_MESA3D support *_V3 flags.

‎llvm/lib/Support/TargetParser.cpp

+2
Original file line numberDiff line numberDiff line change
@@ -115,6 +115,7 @@ constexpr GPUInfo AMDGCNGPUs[] = {
115115
{{"gfx1032"}, {"gfx1032"}, GK_GFX1032, FEATURE_FAST_FMA_F32|FEATURE_FAST_DENORMAL_F32|FEATURE_WAVE32},
116116
{{"gfx1033"}, {"gfx1033"}, GK_GFX1033, FEATURE_FAST_FMA_F32|FEATURE_FAST_DENORMAL_F32|FEATURE_WAVE32},
117117
{{"gfx1034"}, {"gfx1034"}, GK_GFX1034, FEATURE_FAST_FMA_F32|FEATURE_FAST_DENORMAL_F32|FEATURE_WAVE32},
118+
{{"gfx1035"}, {"gfx1035"}, GK_GFX1035, FEATURE_FAST_FMA_F32|FEATURE_FAST_DENORMAL_F32|FEATURE_WAVE32},
118119
};
119120

120121
const GPUInfo *getArchEntry(AMDGPU::GPUKind AK, ArrayRef<GPUInfo> Table) {
@@ -227,6 +228,7 @@ AMDGPU::IsaVersion AMDGPU::getIsaVersion(StringRef GPU) {
227228
case GK_GFX1032: return {10, 3, 2};
228229
case GK_GFX1033: return {10, 3, 3};
229230
case GK_GFX1034: return {10, 3, 4};
231+
case GK_GFX1035: return {10, 3, 5};
230232
default: return {0, 0, 0};
231233
}
232234
}

‎llvm/lib/Target/AMDGPU/GCNProcessors.td

+4
Original file line numberDiff line numberDiff line change
@@ -231,3 +231,7 @@ def : ProcessorModel<"gfx1033", GFX10SpeedModel,
231231
def : ProcessorModel<"gfx1034", GFX10SpeedModel,
232232
FeatureISAVersion10_3_0.Features
233233
>;
234+
235+
def : ProcessorModel<"gfx1035", GFX10SpeedModel,
236+
FeatureISAVersion10_3_0.Features
237+
>;

‎llvm/lib/Target/AMDGPU/MCTargetDesc/AMDGPUTargetStreamer.cpp

+2
Original file line numberDiff line numberDiff line change
@@ -111,6 +111,7 @@ StringRef AMDGPUTargetStreamer::getArchNameFromElfMach(unsigned ElfMach) {
111111
case ELF::EF_AMDGPU_MACH_AMDGCN_GFX1032: AK = GK_GFX1032; break;
112112
case ELF::EF_AMDGPU_MACH_AMDGCN_GFX1033: AK = GK_GFX1033; break;
113113
case ELF::EF_AMDGPU_MACH_AMDGCN_GFX1034: AK = GK_GFX1034; break;
114+
case ELF::EF_AMDGPU_MACH_AMDGCN_GFX1035: AK = GK_GFX1035; break;
114115
case ELF::EF_AMDGPU_MACH_NONE: AK = GK_NONE; break;
115116
}
116117

@@ -173,6 +174,7 @@ unsigned AMDGPUTargetStreamer::getElfMach(StringRef GPU) {
173174
case GK_GFX1032: return ELF::EF_AMDGPU_MACH_AMDGCN_GFX1032;
174175
case GK_GFX1033: return ELF::EF_AMDGPU_MACH_AMDGCN_GFX1033;
175176
case GK_GFX1034: return ELF::EF_AMDGPU_MACH_AMDGCN_GFX1034;
177+
case GK_GFX1035: return ELF::EF_AMDGPU_MACH_AMDGCN_GFX1035;
176178
case GK_NONE: return ELF::EF_AMDGPU_MACH_NONE;
177179
}
178180

‎llvm/test/CodeGen/AMDGPU/directive-amdgcn-target.ll

+4
Original file line numberDiff line numberDiff line change
@@ -88,6 +88,7 @@
8888
; RUN: llc -mtriple=amdgcn-amd-amdhsa --amdhsa-code-object-version=3 -mcpu=gfx1032 < %s | FileCheck --check-prefixes=V3-GFX1032 %s
8989
; RUN: llc -mtriple=amdgcn-amd-amdhsa --amdhsa-code-object-version=3 -mcpu=gfx1033 < %s | FileCheck --check-prefixes=V3-GFX1033 %s
9090
; RUN: llc -mtriple=amdgcn-amd-amdhsa --amdhsa-code-object-version=3 -mcpu=gfx1034 < %s | FileCheck --check-prefixes=V3-GFX1034 %s
91+
; RUN: llc -mtriple=amdgcn-amd-amdhsa --amdhsa-code-object-version=3 -mcpu=gfx1035 < %s | FileCheck --check-prefixes=V3-GFX1035 %s
9192

9293
; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx600 < %s | FileCheck --check-prefixes=GFX600 %s
9394
; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=tahiti < %s | FileCheck --check-prefixes=GFX600 %s
@@ -179,6 +180,7 @@
179180
; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx1032 < %s | FileCheck --check-prefixes=GFX1032 %s
180181
; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx1033 < %s | FileCheck --check-prefixes=GFX1033 %s
181182
; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx1034 < %s | FileCheck --check-prefixes=GFX1034 %s
183+
; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx1035 < %s | FileCheck --check-prefixes=GFX1035 %s
182184

183185
; V3-GFX600: .amdgcn_target "amdgcn-amd-amdhsa--gfx600"
184186
; V3-GFX601: .amdgcn_target "amdgcn-amd-amdhsa--gfx601"
@@ -227,6 +229,7 @@
227229
; V3-GFX1032: .amdgcn_target "amdgcn-amd-amdhsa--gfx1032"
228230
; V3-GFX1033: .amdgcn_target "amdgcn-amd-amdhsa--gfx1033"
229231
; V3-GFX1034: .amdgcn_target "amdgcn-amd-amdhsa--gfx1034"
232+
; V3-GFX1035: .amdgcn_target "amdgcn-amd-amdhsa--gfx1035"
230233

231234
; GFX600: .amdgcn_target "amdgcn-amd-amdhsa--gfx600"
232235
; GFX601: .amdgcn_target "amdgcn-amd-amdhsa--gfx601"
@@ -296,6 +299,7 @@
296299
; GFX1032: .amdgcn_target "amdgcn-amd-amdhsa--gfx1032"
297300
; GFX1033: .amdgcn_target "amdgcn-amd-amdhsa--gfx1033"
298301
; GFX1034: .amdgcn_target "amdgcn-amd-amdhsa--gfx1034"
302+
; GFX1035: .amdgcn_target "amdgcn-amd-amdhsa--gfx1035"
299303

300304
define amdgpu_kernel void @directive_amdgcn_target() {
301305
ret void

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

+2
Original file line numberDiff line numberDiff line change
@@ -63,6 +63,7 @@
6363
; RUN: llc -filetype=obj -march=amdgcn -mcpu=gfx1032 < %s | llvm-readobj -file-headers - | FileCheck --check-prefixes=ALL,ARCH-GCN,GFX1032 %s
6464
; RUN: llc -filetype=obj -march=amdgcn -mcpu=gfx1033 < %s | llvm-readobj -file-headers - | FileCheck --check-prefixes=ALL,ARCH-GCN,GFX1033 %s
6565
; RUN: llc -filetype=obj -march=amdgcn -mcpu=gfx1034 < %s | llvm-readobj -file-headers - | FileCheck --check-prefixes=ALL,ARCH-GCN,GFX1034 %s
66+
; RUN: llc -filetype=obj -march=amdgcn -mcpu=gfx1035 < %s | llvm-readobj -file-headers - | FileCheck --check-prefixes=ALL,ARCH-GCN,GFX1035 %s
6667

6768
; FIXME: With the default attributes the eflags are not accurate for
6869
; xnack and sramecc. Subsequent Target-ID patches will address this.
@@ -123,6 +124,7 @@
123124
; GFX1032: EF_AMDGPU_MACH_AMDGCN_GFX1032 (0x38)
124125
; GFX1033: EF_AMDGPU_MACH_AMDGCN_GFX1033 (0x39)
125126
; GFX1034: EF_AMDGPU_MACH_AMDGCN_GFX1034 (0x3E)
127+
; GFX1035: EF_AMDGPU_MACH_AMDGCN_GFX1035 (0x3D)
126128
; ALL: ]
127129

128130
define amdgpu_kernel void @elf_header() {

‎llvm/test/MC/AMDGPU/gfx1011_dlops.s

+1
Original file line numberDiff line numberDiff line change
@@ -5,6 +5,7 @@
55
// RUN: llvm-mc -arch=amdgcn -mcpu=gfx1032 -show-encoding %s | FileCheck --check-prefix=GFX10 %s
66
// RUN: llvm-mc -arch=amdgcn -mcpu=gfx1033 -show-encoding %s | FileCheck --check-prefix=GFX10 %s
77
// RUN: llvm-mc -arch=amdgcn -mcpu=gfx1034 -show-encoding %s | FileCheck --check-prefix=GFX10 %s
8+
// RUN: llvm-mc -arch=amdgcn -mcpu=gfx1035 -show-encoding %s | FileCheck --check-prefix=GFX10 %s
89

910
v_dot2_f32_f16 v0, v1, v2, v3
1011
// GFX10: encoding: [0x00,0x40,0x13,0xcc,0x01,0x05,0x0e,0x1c]

‎llvm/test/MC/AMDGPU/gfx1030_err.s

+1
Original file line numberDiff line numberDiff line change
@@ -3,6 +3,7 @@
33
// RUN: not llvm-mc -arch=amdgcn -mcpu=gfx1032 %s 2>&1 | FileCheck --check-prefix=GFX10 --implicit-check-not=error: %s
44
// RUN: not llvm-mc -arch=amdgcn -mcpu=gfx1033 %s 2>&1 | FileCheck --check-prefix=GFX10 --implicit-check-not=error: %s
55
// RUN: not llvm-mc -arch=amdgcn -mcpu=gfx1034 %s 2>&1 | FileCheck --check-prefix=GFX10 --implicit-check-not=error: %s
6+
// RUN: not llvm-mc -arch=amdgcn -mcpu=gfx1035 %s 2>&1 | FileCheck --check-prefix=GFX10 --implicit-check-not=error: %s
67

78
v_dot8c_i32_i4 v5, v1, v2
89
// GFX10: :[[@LINE-1]]:{{[0-9]+}}: error: instruction not supported on this GPU

‎llvm/test/MC/AMDGPU/gfx1030_new.s

+1
Original file line numberDiff line numberDiff line change
@@ -3,6 +3,7 @@
33
// RUN: llvm-mc -arch=amdgcn -mcpu=gfx1032 -show-encoding %s | FileCheck --check-prefix=GFX10 %s
44
// RUN: llvm-mc -arch=amdgcn -mcpu=gfx1033 -show-encoding %s | FileCheck --check-prefix=GFX10 %s
55
// RUN: llvm-mc -arch=amdgcn -mcpu=gfx1034 -show-encoding %s | FileCheck --check-prefix=GFX10 %s
6+
// RUN: llvm-mc -arch=amdgcn -mcpu=gfx1035 -show-encoding %s | FileCheck --check-prefix=GFX10 %s
67

78
global_load_dword_addtid v1, s[2:3] offset:16
89
// GFX10: encoding: [0x10,0x80,0x58,0xdc,0x00,0x00,0x02,0x01]

‎llvm/test/MC/Disassembler/AMDGPU/gfx1011_dasm_dlops.txt

+1
Original file line numberDiff line numberDiff line change
@@ -5,6 +5,7 @@
55
# RUN: llvm-mc -arch=amdgcn -mcpu=gfx1032 -disassemble -show-encoding < %s | FileCheck -check-prefix=GFX10 %s
66
# RUN: llvm-mc -arch=amdgcn -mcpu=gfx1033 -disassemble -show-encoding < %s | FileCheck -check-prefix=GFX10 %s
77
# RUN: llvm-mc -arch=amdgcn -mcpu=gfx1034 -disassemble -show-encoding < %s | FileCheck -check-prefix=GFX10 %s
8+
# RUN: llvm-mc -arch=amdgcn -mcpu=gfx1035 -disassemble -show-encoding < %s | FileCheck -check-prefix=GFX10 %s
89

910
# GFX10: v_dot2_f32_f16 v0, v1, v2, v3 ; encoding: [0x00,0x40,0x13,0xcc,0x01,0x05,0x0e,0x1c]
1011
0x00,0x40,0x13,0xcc,0x01,0x05,0x0e,0x1c

‎llvm/test/MC/Disassembler/AMDGPU/gfx1030_dasm_new.txt

+1
Original file line numberDiff line numberDiff line change
@@ -3,6 +3,7 @@
33
# RUN: llvm-mc -arch=amdgcn -mcpu=gfx1032 -disassemble -show-encoding < %s | FileCheck -check-prefix=GFX10 %s
44
# RUN: llvm-mc -arch=amdgcn -mcpu=gfx1033 -disassemble -show-encoding < %s | FileCheck -check-prefix=GFX10 %s
55
# RUN: llvm-mc -arch=amdgcn -mcpu=gfx1034 -disassemble -show-encoding < %s | FileCheck -check-prefix=GFX10 %s
6+
# RUN: llvm-mc -arch=amdgcn -mcpu=gfx1035 -disassemble -show-encoding < %s | FileCheck -check-prefix=GFX10 %s
67

78
# GFX10: global_load_dword_addtid v1, s[2:3] offset:16
89
0x10,0x80,0x58,0xdc,0x00,0x00,0x02,0x01

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

+7
Original file line numberDiff line numberDiff line change
@@ -186,6 +186,10 @@
186186
# RUN: llvm-readobj -S --file-headers %t.o.AMDGCN_GFX1034 | FileCheck --check-prefixes=ELF-AMDGCN-ALL,ELF-AMDGCN-GFX1034 %s
187187
# RUN: obj2yaml %t.o.AMDGCN_GFX1034 | FileCheck --check-prefixes=YAML-AMDGCN-ALL,YAML-AMDGCN-GFX1034 %s
188188

189+
# RUN: sed -e 's/<BITS>/64/' -e 's/<MACH>/AMDGCN_GFX1035/' %s | yaml2obj -o %t.o.AMDGCN_GFX1035
190+
# RUN: llvm-readobj -S --file-headers %t.o.AMDGCN_GFX1035 | FileCheck --check-prefixes=ELF-AMDGCN-ALL,ELF-AMDGCN-GFX1035 %s
191+
# RUN: obj2yaml %t.o.AMDGCN_GFX1035 | FileCheck --check-prefixes=YAML-AMDGCN-ALL,YAML-AMDGCN-GFX1035 %s
192+
189193
# ELF-R600-ALL: Format: elf32-amdgpu
190194
# ELF-R600-ALL: Arch: r600
191195
# ELF-R600-ALL: AddressSize: 32bit
@@ -344,6 +348,9 @@
344348
# ELF-AMDGCN-GFX1034: EF_AMDGPU_MACH_AMDGCN_GFX1034 (0x3E)
345349
# YAML-AMDGCN-GFX1034: Flags: [ EF_AMDGPU_MACH_AMDGCN_GFX1034 ]
346350

351+
# ELF-AMDGCN-GFX1035: EF_AMDGPU_MACH_AMDGCN_GFX1035 (0x3D)
352+
# YAML-AMDGCN-GFX1035: Flags: [ EF_AMDGPU_MACH_AMDGCN_GFX1035 ]
353+
#
347354
# ELF-AMDGCN-ALL: ]
348355

349356

‎llvm/test/tools/llvm-objdump/ELF/AMDGPU/subtarget.ll

+4
Original file line numberDiff line numberDiff line change
@@ -5,6 +5,10 @@ define amdgpu_kernel void @test_kernel() {
55
; Test subtarget detection. Disassembly is only supported for GFX8 and beyond.
66
;
77
; ----------------------------------GFX10--------------------------------------
8+
; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx1035 -filetype=obj -O0 -o %t.o %s
9+
; RUN: llvm-objdump -D --arch-name=amdgcn --mcpu=gfx1035 %t.o > %t-specify.txt
10+
; RUN: llvm-objdump -D %t.o > %t-detect.txt
11+
; RUN: diff %t-specify.txt %t-detect.txt
812
;
913
; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx1034 -filetype=obj -O0 -o %t.o %s
1014
; RUN: llvm-objdump -D --arch-name=amdgcn --mcpu=gfx1034 %t.o > %t-specify.txt

‎llvm/test/tools/llvm-readobj/ELF/amdgpu-elf-headers.test

+8
Original file line numberDiff line numberDiff line change
@@ -277,6 +277,14 @@
277277
# RUN: yaml2obj %s -o %t -DABI_VERSION=2 -DFLAG_NAME=EF_AMDGPU_MACH_AMDGCN_GFX1034
278278
# 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_GFX1034 -DFLAG_VALUE=0x3E
279279

280+
# RUN: yaml2obj %s -o %t -DABI_VERSION=0 -DFLAG_NAME=EF_AMDGPU_MACH_AMDGCN_GFX1035
281+
# 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_GFX1035 -DFLAG_VALUE=0x3D
282+
#
283+
# RUN: yaml2obj %s -o %t -DABI_VERSION=1 -DFLAG_NAME=EF_AMDGPU_MACH_AMDGCN_GFX1035
284+
# 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_GFX1035 -DFLAG_VALUE=0x3D
285+
#
286+
# RUN: yaml2obj %s -o %t -DABI_VERSION=2 -DFLAG_NAME=EF_AMDGPU_MACH_AMDGCN_GFX1035
287+
# 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_GFX1035 -DFLAG_VALUE=0x3D
280288

281289
# RUN: yaml2obj %s -o %t -DABI_VERSION=0 -DFLAG_NAME="EF_AMDGPU_MACH_AMDGCN_GFX90A, EF_AMDGPU_FEATURE_XNACK_V3"
282290
# RUN: llvm-readobj -h %t | FileCheck %s --check-prefixes=ALL,KNOWN-ABI-VERSION,DOUBLE-FLAG --match-full-lines -DABI_VERSION=0 -DFILE=%t -DFLAG_0="EF_AMDGPU_FEATURE_XNACK_V3 (0x100)" -DFLAG_1="EF_AMDGPU_MACH_AMDGCN_GFX90A (0x3F)" -DFLAG_VALUE=0x13F

‎llvm/tools/llvm-readobj/ELFDumper.cpp

+2
Original file line numberDiff line numberDiff line change
@@ -1493,6 +1493,7 @@ static const EnumEntry<unsigned> ElfHeaderAMDGPUFlagsABIVersion3[] = {
14931493
LLVM_READOBJ_ENUM_ENT(ELF, EF_AMDGPU_MACH_AMDGCN_GFX1032),
14941494
LLVM_READOBJ_ENUM_ENT(ELF, EF_AMDGPU_MACH_AMDGCN_GFX1033),
14951495
LLVM_READOBJ_ENUM_ENT(ELF, EF_AMDGPU_MACH_AMDGCN_GFX1034),
1496+
LLVM_READOBJ_ENUM_ENT(ELF, EF_AMDGPU_MACH_AMDGCN_GFX1035),
14961497
LLVM_READOBJ_ENUM_ENT(ELF, EF_AMDGPU_FEATURE_XNACK_V3),
14971498
LLVM_READOBJ_ENUM_ENT(ELF, EF_AMDGPU_FEATURE_SRAMECC_V3)
14981499
};
@@ -1546,6 +1547,7 @@ static const EnumEntry<unsigned> ElfHeaderAMDGPUFlagsABIVersion4[] = {
15461547
LLVM_READOBJ_ENUM_ENT(ELF, EF_AMDGPU_MACH_AMDGCN_GFX1032),
15471548
LLVM_READOBJ_ENUM_ENT(ELF, EF_AMDGPU_MACH_AMDGCN_GFX1033),
15481549
LLVM_READOBJ_ENUM_ENT(ELF, EF_AMDGPU_MACH_AMDGCN_GFX1034),
1550+
LLVM_READOBJ_ENUM_ENT(ELF, EF_AMDGPU_MACH_AMDGCN_GFX1035),
15491551
LLVM_READOBJ_ENUM_ENT(ELF, EF_AMDGPU_FEATURE_XNACK_ANY_V4),
15501552
LLVM_READOBJ_ENUM_ENT(ELF, EF_AMDGPU_FEATURE_XNACK_OFF_V4),
15511553
LLVM_READOBJ_ENUM_ENT(ELF, EF_AMDGPU_FEATURE_XNACK_ON_V4),

‎openmp/libomptarget/plugins/amdgpu/impl/get_elf_mach_gfx_name.cpp

+2
Original file line numberDiff line numberDiff line change
@@ -51,6 +51,8 @@ const char *get_elf_mach_gfx_name(uint32_t EFlags) {
5151
return "gfx1033";
5252
case EF_AMDGPU_MACH_AMDGCN_GFX1034:
5353
return "gfx1034";
54+
case EF_AMDGPU_MACH_AMDGCN_GFX1035:
55+
return "gfx1035";
5456
default:
5557
return "--unknown gfx";
5658
}

0 commit comments

Comments
 (0)