[AMDGPU] Add gfx1013 target

[foad]

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

[foad]

llvm/docs/AMDGPUUsage.rst

389

Is it dGPU or APU?

Every other entry with *TBA* also has a TODO:: message

llvm/lib/Target/AMDGPU/AMDGPU.td

468

What is this new encoding? It doesn't seem to be used for anything.

llvm/lib/Target/AMDGPU/GCNSubtarget.h

879

Stray whitespace on this line.

llvm/lib/Target/AMDGPU/Utils/AMDGPUBaseInfo.cpp

1453

Stray whitespace.

llvm/test/CodeGen/AMDGPU/GlobalISel/llvm.amdgcn.intersect_ray.ll

4

This test surely should not pass for gfx1012, since it does not have these instructions. And with your patch as written it should fail for gfx1013 too, since they are predicated on HasGFX10_BEncoding.

@rampitec any idea what is wrong here? Apparently the backend will happily generate image_bvh_intersect_ray instructions even for gfx900!

llvm/test/CodeGen/AMDGPU/llvm.amdgcn.intersect_ray.ll

3

Likewise.

[rampitec]

You need to replace HasGFX10_BEncoding with HasGFX10_AEncoding in the BVH and IMAGE_MSAA_LOAD_X. You also need to update llvm.amdgcn.image.msaa.load.x.ll test to include gfx1013.

llvm/lib/Target/AMDGPU/AMDGPU.td

1106

gfx1030 should now include FeatureGFX10_AEncoding as well. 10_B is an extension above 10_A.

llvm/test/CodeGen/AMDGPU/GlobalISel/llvm.amdgcn.intersect_ray.ll

4

Indeed. MIMG_IntersectRay has this:

let SubtargetPredicate = HasGFX10_BEncoding,
    AssemblerPredicate = HasGFX10_BEncoding,

but apparently SubtargetPredicate did not work. It needs to be fixed.

gfx1012 does not have it, gfx1013 does though. That is what GFX10_A encoding is about, 10_B it has to be replaced with 10_A in BVH and MSAA load.

[rampitec]

llvm/test/CodeGen/AMDGPU/GlobalISel/llvm.amdgcn.intersect_ray.ll

4

Image lowering and selection is not really done like everything else. For BVH it just lowers intrinsic to opcode. I think the easiest fix is to add to SIISelLowering.cpp where we lower Intrinsic::amdgcn_image_bvh_intersect_ray something like this:

if (!Subtarget->hasGFX10_AEncoding())
  report_fatal_error(
      "requested image instruction is not supported on this GPU");

[bcahoon]

Addressed review comments. Updated the patch to use the new AEncoding target feature
correctly. Added code to report an error for the image intersect intrinsics for
unsupported targets.

[bcahoon]

llvm/docs/AMDGPUUsage.rst

389

It is APU.

llvm/lib/Target/AMDGPU/AMDGPU.td

468

Fixed this. The BVH raytracing instructions use the encoding.

1106

I had added FeatureGFX10_AEncoding as an Implies feature for FeatureGFX10_BEncoding in the previous patch. But, I've changed the patch so that it's no longer an Implies feature.

llvm/test/CodeGen/AMDGPU/GlobalISel/llvm.amdgcn.intersect_ray.ll

4

I ended up using emitRemovedIntrinsicError, which uses DiagnosticInfoUnsupported. This way the failure isn't a crash dump.

[rampitec]

llvm/lib/Target/AMDGPU/AMDGPULegalizerInfo.cpp

4690

80 chars per line.

4693

Just return false like in other places.

llvm/lib/Target/AMDGPU/SIISelLowering.cpp

7345

return emitRemovedIntrinsicError();

llvm/test/CodeGen/AMDGPU/GlobalISel/llvm.amdgcn.intersect_ray.ll

4

not --crash llc ...

4

I ended up using emitRemovedIntrinsicError, which uses DiagnosticInfoUnsupported. This way the failure isn't a crash dump.

Diagnostics is a good thing, but we still have to fail the compilation.

llvm/test/CodeGen/AMDGPU/llvm.amdgcn.intersect_ray.ll

3

not --crash llc

[bcahoon]

llvm/test/CodeGen/AMDGPU/GlobalISel/llvm.amdgcn.intersect_ray.ll

4

The diagnostic is marked as an error, so the compilation fails in that llc returns a non-zero return code. This mechanism is used in other places in the back-end to report similar types of errors. The alternative, if I understand correctly, is that a crash occurs with an error message that indicates that the bug is in LLVM (rather the the input source file).

[rampitec]

llvm/test/CodeGen/AMDGPU/GlobalISel/llvm.amdgcn.intersect_ray.ll

4

We do not seem to be consistent here and return either undef or SDValue(), but as far as I can see we never continue selecting code though, like here in SIISelLowering and always return false from the AMDGPUInstructionSelector.

[bcahoon]

Addressed review comments

[bcahoon]

llvm/lib/Target/AMDGPU/SIISelLowering.cpp

7345

I've changed this to return. Thanks for catching that. But, it returns a UNDEF value instead of SDValue() so that it doesn't crash. I can change the behavior if that's preferred.

llvm/test/CodeGen/AMDGPU/GlobalISel/llvm.amdgcn.intersect_ray.ll

4

I've left the patch so that it doesn't crash. But, let me know if you think we should return false and crash, and I'll make that change.

[rampitec]

llvm/lib/Target/AMDGPU/AMDGPULegalizerInfo.cpp

4693

Just return false. I see that is like this in the whole file.

[bcahoon]

Changed legalizer to return false for raytracing intrinsics that are not supported by the subtarget. I changed both GlobalISel and regular ISel to work similarly. A crash occurs with a message that the intrinsic cannot be legalized, and only the first instance is reported.

[bcahoon]

llvm/lib/Target/AMDGPU/AMDGPULegalizerInfo.cpp

4693

Changed this to false, and also changed SIISelLlowering to return SDValue so that both fail in a similar way.

[rampitec]

llvm/lib/Target/AMDGPU/AMDGPULegalizerInfo.cpp

4694

You can just omit undef and erase.

[bcahoon]

Addressed review comment.

[foad]

llvm/lib/Target/AMDGPU/AMDGPU.td

471

I realise you're just following the precedent set by GFX10_B, but is this terminology actually used in any documentation anywhere? And if not could we describe it a little better here?

llvm/lib/Target/AMDGPU/SIISelLowering.cpp

7345

Personally I would follow all the existing precedents and "return emitRemovedIntrinsicError(...)". I don't see any value in deliberately trying to make the compiler crash harder.

[foad]

LGTM anyway, with or without any action on my last couple of comments.

[bcahoon]

llvm/lib/Target/AMDGPU/AMDGPU.td

471

I changed the description to be specific w.r.t what the target feature enables.

llvm/lib/Target/AMDGPU/SIISelLowering.cpp

7345

The handling of diagnostic errors is a little inconsistent in ISelLowering. Sometimes SDValue() is returned and other times it's Undef. I have it return SDValue() so that the failure mode is consistent with how GlobaISel handles these intrinsics. It's probably worth a discussion to decide how best to handle diagnostic errors. I'm happy to submit a follow-on patch as needed.

As an aside, return emitRemovedIntrinsicError() isn't enough because the intrinsic has both a return value and a chain edge. So, something like return DAG.getMergeValues({emitRemovedIntrinsicError(), Op.getValue(0)}, DL) is needed.

[foad]

llvm/lib/Target/AMDGPU/AMDGPU.td

471

Thank you. I think that is much more useful.

clang/include/clang/Basic/Cuda.h

enum class CudaArch {
   GFX906,
   GFX908,
   GFX909,
   GFX90a,
   GFX90c,
   GFX1010,
   GFX1011,
   GFX1012,
+  GFX1013,
   GFX1030,
   GFX1031,
   GFX1032,
   GFX1033,
   GFX1034,
   LAST,
 };
 

clang/lib/Basic/Cuda.cpp

static const CudaArchToStringMap arch_names[] = {
     GFX(906),  // gfx906
     GFX(908),  // gfx908
     GFX(909),  // gfx909
     GFX(90a),  // gfx90a
     GFX(90c),  // gfx90c
     GFX(1010), // gfx1010
     GFX(1011), // gfx1011
     GFX(1012), // gfx1012
+    GFX(1013), // gfx1013
     GFX(1030), // gfx1030
     GFX(1031), // gfx1031
     GFX(1032), // gfx1032
     GFX(1033), // gfx1033
     GFX(1034), // gfx1034
     // clang-format on
 };
 #undef SM

clang/lib/Basic/Targets/AMDGPU.cpp

if (isAMDGCN(getTriple())) {
     case GK_GFX1012:
     case GK_GFX1011:
       Features["dot1-insts"] = true;
       Features["dot2-insts"] = true;
       Features["dot5-insts"] = true;
       Features["dot6-insts"] = true;
       Features["dot7-insts"] = true;
       LLVM_FALLTHROUGH;
+    case GK_GFX1013:
     case GK_GFX1010:
       Features["dl-insts"] = true;
       Features["ci-insts"] = true;
       Features["flat-address-space"] = true;
       Features["16-bit-insts"] = true;
       Features["dpp"] = true;
       Features["gfx8-insts"] = true;
       Features["gfx9-insts"] = true;

clang/lib/Basic/Targets/NVPTX.cpp

std::string CUDAArchCode = [this] {
       case CudaArch::GFX906:
       case CudaArch::GFX908:
       case CudaArch::GFX909:
       case CudaArch::GFX90a:
       case CudaArch::GFX90c:
       case CudaArch::GFX1010:
       case CudaArch::GFX1011:
       case CudaArch::GFX1012:
+      case CudaArch::GFX1013:
       case CudaArch::GFX1030:
       case CudaArch::GFX1031:
       case CudaArch::GFX1032:
       case CudaArch::GFX1033:
       case CudaArch::GFX1034:
       case CudaArch::LAST:
         break;
       case CudaArch::UNUSED:

clang/lib/CodeGen/CGOpenMPRuntimeGPU.cpp

if (Clause->getClauseKind() == OMPC_unified_shared_memory) {
       case CudaArch::GFX906:
       case CudaArch::GFX908:
       case CudaArch::GFX909:
       case CudaArch::GFX90a:
       case CudaArch::GFX90c:
       case CudaArch::GFX1010:
       case CudaArch::GFX1011:
       case CudaArch::GFX1012:
+      case CudaArch::GFX1013:
       case CudaArch::GFX1030:
       case CudaArch::GFX1031:
       case CudaArch::GFX1032:
       case CudaArch::GFX1033:
       case CudaArch::GFX1034:
       case CudaArch::UNUSED:
       case CudaArch::UNKNOWN:
         break;
static std::pair<unsigned, unsigned> getSMsBlocksPerSM(CodeGenModule &CGM) {
   case CudaArch::GFX906:
   case CudaArch::GFX908:
   case CudaArch::GFX909:
   case CudaArch::GFX90a:
   case CudaArch::GFX90c:
   case CudaArch::GFX1010:
   case CudaArch::GFX1011:
   case CudaArch::GFX1012:
+  case CudaArch::GFX1013:
   case CudaArch::GFX1030:
   case CudaArch::GFX1031:
   case CudaArch::GFX1032:
   case CudaArch::GFX1033:
   case CudaArch::GFX1034:
   case CudaArch::UNUSED:
   case CudaArch::UNKNOWN:
     break;

clang/test/CodeGenOpenCL/amdgpu-features.cl

 // RUN: %clang_cc1 -triple amdgcn -target-cpu gfx906 -S -emit-llvm -o - %s | FileCheck --check-prefix=GFX906 %s
 // RUN: %clang_cc1 -triple amdgcn -target-cpu gfx908 -S -emit-llvm -o - %s | FileCheck --check-prefix=GFX908 %s
 // RUN: %clang_cc1 -triple amdgcn -target-cpu gfx909 -S -emit-llvm -o - %s | FileCheck --check-prefix=GFX909 %s
 // RUN: %clang_cc1 -triple amdgcn -target-cpu gfx90a -S -emit-llvm -o - %s | FileCheck --check-prefix=GFX90A %s
 // RUN: %clang_cc1 -triple amdgcn -target-cpu gfx90c -S -emit-llvm -o - %s | FileCheck --check-prefix=GFX90C %s
 // RUN: %clang_cc1 -triple amdgcn -target-cpu gfx1010 -S -emit-llvm -o - %s | FileCheck --check-prefix=GFX1010 %s
 // RUN: %clang_cc1 -triple amdgcn -target-cpu gfx1011 -S -emit-llvm -o - %s | FileCheck --check-prefix=GFX1011 %s
 // RUN: %clang_cc1 -triple amdgcn -target-cpu gfx1012 -S -emit-llvm -o - %s | FileCheck --check-prefix=GFX1012 %s
+// RUN: %clang_cc1 -triple amdgcn -target-cpu gfx1013 -S -emit-llvm -o - %s | FileCheck --check-prefix=GFX1013 %s
 // RUN: %clang_cc1 -triple amdgcn -target-cpu gfx1030 -S -emit-llvm -o - %s | FileCheck --check-prefix=GFX1030 %s
 // RUN: %clang_cc1 -triple amdgcn -target-cpu gfx1031 -S -emit-llvm -o - %s | FileCheck --check-prefix=GFX1031 %s
 // RUN: %clang_cc1 -triple amdgcn -target-cpu gfx1032 -S -emit-llvm -o - %s | FileCheck --check-prefix=GFX1032 %s
 // RUN: %clang_cc1 -triple amdgcn -target-cpu gfx1033 -S -emit-llvm -o - %s | FileCheck --check-prefix=GFX1033 %s
 // RUN: %clang_cc1 -triple amdgcn -target-cpu gfx1034 -S -emit-llvm -o - %s | FileCheck --check-prefix=GFX1034 %s
 
 // GFX600: "target-features"="+s-memtime-inst"
 // GFX601: "target-features"="+s-memtime-inst"

 // GFX906: "target-features"="+16-bit-insts,+ci-insts,+dl-insts,+dot1-insts,+dot2-insts,+dot7-insts,+dpp,+flat-address-space,+gfx8-insts,+gfx9-insts,+s-memrealtime,+s-memtime-inst"
 // GFX908: "target-features"="+16-bit-insts,+ci-insts,+dl-insts,+dot1-insts,+dot2-insts,+dot3-insts,+dot4-insts,+dot5-insts,+dot6-insts,+dot7-insts,+dpp,+flat-address-space,+gfx8-insts,+gfx9-insts,+mai-insts,+s-memrealtime,+s-memtime-inst"
 // GFX909: "target-features"="+16-bit-insts,+ci-insts,+dpp,+flat-address-space,+gfx8-insts,+gfx9-insts,+s-memrealtime,+s-memtime-inst"
 // GFX90A: "target-features"="+16-bit-insts,+ci-insts,+dl-insts,+dot1-insts,+dot2-insts,+dot3-insts,+dot4-insts,+dot5-insts,+dot6-insts,+dot7-insts,+dpp,+flat-address-space,+gfx8-insts,+gfx9-insts,+gfx90a-insts,+mai-insts,+s-memrealtime,+s-memtime-inst"
 // GFX90C: "target-features"="+16-bit-insts,+ci-insts,+dpp,+flat-address-space,+gfx8-insts,+gfx9-insts,+s-memrealtime,+s-memtime-inst"
 // GFX1010: "target-features"="+16-bit-insts,+ci-insts,+dl-insts,+dpp,+flat-address-space,+gfx10-insts,+gfx8-insts,+gfx9-insts,+s-memrealtime,+s-memtime-inst"
 // GFX1011: "target-features"="+16-bit-insts,+ci-insts,+dl-insts,+dot1-insts,+dot2-insts,+dot5-insts,+dot6-insts,+dot7-insts,+dpp,+flat-address-space,+gfx10-insts,+gfx8-insts,+gfx9-insts,+s-memrealtime,+s-memtime-inst"
 // GFX1012: "target-features"="+16-bit-insts,+ci-insts,+dl-insts,+dot1-insts,+dot2-insts,+dot5-insts,+dot6-insts,+dot7-insts,+dpp,+flat-address-space,+gfx10-insts,+gfx8-insts,+gfx9-insts,+s-memrealtime,+s-memtime-inst"
+// GFX1013: "target-features"="+16-bit-insts,+ci-insts,+dl-insts,+dpp,+flat-address-space,+gfx10-insts,+gfx8-insts,+gfx9-insts,+s-memrealtime,+s-memtime-inst"
 // GFX1030: "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"
 // GFX1031: "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"
 // 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"
 // 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"
 // 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"
 
 kernel void test() {}

clang/test/Driver/amdgpu-macros.cl

 // RUN: %clang -E -dM -target amdgcn -mcpu=gfx906 %s 2>&1 | FileCheck --check-prefixes=ARCH-GCN,FAST_FMAF %s -DWAVEFRONT_SIZE=64 -DCPU=gfx906
 // RUN: %clang -E -dM -target amdgcn -mcpu=gfx908 %s 2>&1 | FileCheck --check-prefixes=ARCH-GCN,FAST_FMAF %s -DWAVEFRONT_SIZE=64 -DCPU=gfx908
 // RUN: %clang -E -dM -target amdgcn -mcpu=gfx909 %s 2>&1 | FileCheck --check-prefixes=ARCH-GCN,FAST_FMAF %s -DWAVEFRONT_SIZE=64 -DCPU=gfx909
 // RUN: %clang -E -dM -target amdgcn -mcpu=gfx90a %s 2>&1 | FileCheck --check-prefixes=ARCH-GCN,FAST_FMAF %s -DWAVEFRONT_SIZE=64 -DCPU=gfx90a
 // RUN: %clang -E -dM -target amdgcn -mcpu=gfx90c %s 2>&1 | FileCheck --check-prefixes=ARCH-GCN,FAST_FMAF %s -DWAVEFRONT_SIZE=64 -DCPU=gfx90c
 // RUN: %clang -E -dM -target amdgcn -mcpu=gfx1010 %s 2>&1 | FileCheck --check-prefixes=ARCH-GCN,FAST_FMAF %s -DWAVEFRONT_SIZE=32 -DCPU=gfx1010
 // RUN: %clang -E -dM -target amdgcn -mcpu=gfx1011 %s 2>&1 | FileCheck --check-prefixes=ARCH-GCN,FAST_FMAF %s -DWAVEFRONT_SIZE=32 -DCPU=gfx1011
 // RUN: %clang -E -dM -target amdgcn -mcpu=gfx1012 %s 2>&1 | FileCheck --check-prefixes=ARCH-GCN,FAST_FMAF %s -DWAVEFRONT_SIZE=32 -DCPU=gfx1012
+// RUN: %clang -E -dM -target amdgcn -mcpu=gfx1013 %s 2>&1 | FileCheck --check-prefixes=ARCH-GCN,FAST_FMAF %s -DWAVEFRONT_SIZE=32 -DCPU=gfx1013
 // RUN: %clang -E -dM -target amdgcn -mcpu=gfx1030 %s 2>&1 | FileCheck --check-prefixes=ARCH-GCN,FAST_FMAF %s -DWAVEFRONT_SIZE=32 -DCPU=gfx1030
 // RUN: %clang -E -dM -target amdgcn -mcpu=gfx1031 %s 2>&1 | FileCheck --check-prefixes=ARCH-GCN,FAST_FMAF %s -DWAVEFRONT_SIZE=32 -DCPU=gfx1031
 // RUN: %clang -E -dM -target amdgcn -mcpu=gfx1032 %s 2>&1 | FileCheck --check-prefixes=ARCH-GCN,FAST_FMAF %s -DWAVEFRONT_SIZE=32 -DCPU=gfx1032
 // RUN: %clang -E -dM -target amdgcn -mcpu=gfx1033 %s 2>&1 | FileCheck --check-prefixes=ARCH-GCN,FAST_FMAF %s -DWAVEFRONT_SIZE=32 -DCPU=gfx1033
 // RUN: %clang -E -dM -target amdgcn -mcpu=gfx1034 %s 2>&1 | FileCheck --check-prefixes=ARCH-GCN,FAST_FMAF %s -DWAVEFRONT_SIZE=32 -DCPU=gfx1034
 
 // ARCH-GCN-DAG: #define FP_FAST_FMA 1
 

clang/test/Driver/amdgpu-mcpu.cl

 // RUN: %clang -### -target amdgcn -mcpu=gfx906 %s 2>&1 | FileCheck --check-prefix=GFX906 %s
 // RUN: %clang -### -target amdgcn -mcpu=gfx908 %s 2>&1 | FileCheck --check-prefix=GFX908 %s
 // RUN: %clang -### -target amdgcn -mcpu=gfx909 %s 2>&1 | FileCheck --check-prefix=GFX909 %s
 // RUN: %clang -### -target amdgcn -mcpu=gfx90a %s 2>&1 | FileCheck --check-prefix=GFX90A %s
 // RUN: %clang -### -target amdgcn -mcpu=gfx90c %s 2>&1 | FileCheck --check-prefix=GFX90C %s
 // RUN: %clang -### -target amdgcn -mcpu=gfx1010 %s 2>&1 | FileCheck --check-prefix=GFX1010 %s
 // RUN: %clang -### -target amdgcn -mcpu=gfx1011 %s 2>&1 | FileCheck --check-prefix=GFX1011 %s
 // RUN: %clang -### -target amdgcn -mcpu=gfx1012 %s 2>&1 | FileCheck --check-prefix=GFX1012 %s
+// RUN: %clang -### -target amdgcn -mcpu=gfx1013 %s 2>&1 | FileCheck --check-prefix=GFX1013 %s
 // RUN: %clang -### -target amdgcn -mcpu=gfx1030 %s 2>&1 | FileCheck --check-prefix=GFX1030 %s
 // RUN: %clang -### -target amdgcn -mcpu=gfx1031 %s 2>&1 | FileCheck --check-prefix=GFX1031 %s
 // RUN: %clang -### -target amdgcn -mcpu=gfx1032 %s 2>&1 | FileCheck --check-prefix=GFX1032 %s
 // RUN: %clang -### -target amdgcn -mcpu=gfx1033 %s 2>&1 | FileCheck --check-prefix=GFX1033 %s
 // RUN: %clang -### -target amdgcn -mcpu=gfx1034 %s 2>&1 | FileCheck --check-prefix=GFX1034 %s
 
 // GCNDEFAULT-NOT: -target-cpu
 // GFX600:    "-target-cpu" "gfx600"

 // GFX906:    "-target-cpu" "gfx906"
 // GFX908:    "-target-cpu" "gfx908"
 // GFX909:    "-target-cpu" "gfx909"
 // GFX90A:    "-target-cpu" "gfx90a"
 // GFX90C:    "-target-cpu" "gfx90c"
 // GFX1010:   "-target-cpu" "gfx1010"
 // GFX1011:   "-target-cpu" "gfx1011"
 // GFX1012:   "-target-cpu" "gfx1012"
+// GFX1013:   "-target-cpu" "gfx1013"
 // GFX1030:   "-target-cpu" "gfx1030"
 // GFX1031:   "-target-cpu" "gfx1031"
 // GFX1032:   "-target-cpu" "gfx1032"
 // GFX1033:   "-target-cpu" "gfx1033"
 // GFX1034:   "-target-cpu" "gfx1034"

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

 
 // RUN: not %clang_cc1 -triple amdgcn--- -target-cpu not-a-cpu -fsyntax-only %s 2>&1 | FileCheck %s --check-prefix AMDGCN
 // AMDGCN: error: unknown target CPU 'not-a-cpu'
 // AMDGCN: note: valid target CPU values are: gfx600, tahiti, gfx601, pitcairn, verde,
 // AMDGCN-SAME: gfx602, hainan, oland, gfx700, kaveri, gfx701, hawaii, gfx702,
 // AMDGCN-SAME: gfx703, kabini, mullins, gfx704, bonaire, gfx705, gfx801, carrizo,
 // AMDGCN-SAME: gfx802, iceland, tonga, gfx803, fiji, polaris10, polaris11,
 // AMDGCN-SAME: gfx805, tongapro, gfx810, stoney, gfx900, gfx902, gfx904, gfx906,
-// AMDGCN-SAME: gfx908, gfx909, gfx90a, gfx90c, gfx1010, gfx1011, gfx1012, gfx1030, gfx1031,
+// AMDGCN-SAME: gfx908, gfx909, gfx90a, gfx90c, gfx1010, gfx1011, gfx1012, gfx1013, gfx1030, gfx1031,
 // AMDGCN-SAME: gfx1032, gfx1033, gfx1034
 
 // RUN: not %clang_cc1 -triple wasm64--- -target-cpu not-a-cpu -fsyntax-only %s 2>&1 | FileCheck %s --check-prefix WEBASM
 // WEBASM: error: unknown target CPU 'not-a-cpu'
 // WEBASM: note: valid target CPU values are: mvp, bleeding-edge, generic
 
 // RUN: not %clang_cc1 -triple systemz--- -target-cpu not-a-cpu -fsyntax-only %s 2>&1 | FileCheck %s --check-prefix SYSTEMZ
 // SYSTEMZ: error: unknown target CPU 'not-a-cpu'

llvm/docs/AMDGPUUsage.rst

``gfx1010``                 ``amdgcn``   dGPU  - cumode          - Absolute      - *rocm-amdhsa* - Radeon RX 5700
                                                                                                       - Radeon Pro 5600M
      ``gfx1011``                 ``amdgcn``   dGPU  - cumode                          - *rocm-amdhsa* - Radeon Pro V520
                                                     - wavefrontsize64 - Absolute      - *pal-amdhsa*
                                                     - xnack             flat          - *pal-amdpal*
                                                                         scratch
      ``gfx1012``                 ``amdgcn``   dGPU  - cumode          - Absolute      - *rocm-amdhsa* - Radeon RX 5500
                                                     - wavefrontsize64   flat          - *pal-amdhsa*  - Radeon RX 5500 XT
                                                     - xnack             scratch       - *pal-amdpal*
+     ``gfx1013``                 ``amdgcn``   APU   - cumode          - Absolute      - *rocm-amdhsa* *TBA*
+                                                    - wavefrontsize64   flat          - *pal-amdhsa*
+                                                    - xnack             scratch       - *pal-amdpal*  .. TODO::
+
+                                                                                                        Add product
+                                                                                                        names.
+
      **GCN GFX10 (RDNA 2)** [AMD-GCN-GFX10-RDNA2]_
      -----------------------------------------------------------------------------------------------------------------------
      ``gfx1030``                 ``amdgcn``   dGPU  - cumode          - Absolute      - *rocm-amdhsa* - Radeon RX 6800
                                                     - wavefrontsize64   flat          - *pal-amdhsa*  - Radeon RX 6800 XT
                                                                         scratch       - *pal-amdpal*  - Radeon RX 6900 XT
      ``gfx1031``                 ``amdgcn``   dGPU  - cumode          - Absolute      - *rocm-amdhsa* - Radeon RX 6700 XT
                                                     - wavefrontsize64   flat          - *pal-amdhsa*
                                                                         scratch       - *pal-amdpal*
.. table:: AMDGPU ``EF_AMDGPU_MACH`` Values
      ``EF_AMDGPU_MACH_AMDGCN_GFX602``     0x03a      ``gfx602``
      ``EF_AMDGPU_MACH_AMDGCN_GFX705``     0x03b      ``gfx705``
      ``EF_AMDGPU_MACH_AMDGCN_GFX805``     0x03c      ``gfx805``
      *reserved*                           0x03d      Reserved.
      ``EF_AMDGPU_MACH_AMDGCN_GFX1034``    0x03e      ``gfx1034``
      ``EF_AMDGPU_MACH_AMDGCN_GFX90A``     0x03f      ``gfx90a``
      *reserved*                           0x040      Reserved.
      *reserved*                           0x041      Reserved.
+     ``EF_AMDGPU_MACH_AMDGCN_GFX1013``    0x042      ``gfx1013``
      ==================================== ========== =============================
 
 Sections
 --------
 
 An AMDGPU target ELF code object has the standard ELF sections which include:
 
   .. table:: AMDGPU ELF Sections

llvm/include/llvm/BinaryFormat/ELF.h

enum : unsigned {
   EF_AMDGPU_MACH_AMDGCN_GFX602        = 0x03a,
   EF_AMDGPU_MACH_AMDGCN_GFX705        = 0x03b,
   EF_AMDGPU_MACH_AMDGCN_GFX805        = 0x03c,
   EF_AMDGPU_MACH_AMDGCN_RESERVED_0X3D = 0x03d,
   EF_AMDGPU_MACH_AMDGCN_GFX1034       = 0x03e,
   EF_AMDGPU_MACH_AMDGCN_GFX90A        = 0x03f,
   EF_AMDGPU_MACH_AMDGCN_RESERVED_0X40 = 0x040,
   EF_AMDGPU_MACH_AMDGCN_RESERVED_0X41 = 0x041,
+  EF_AMDGPU_MACH_AMDGCN_GFX1013       = 0x042,
 
   // First/last AMDGCN-based processors.
   EF_AMDGPU_MACH_AMDGCN_FIRST = EF_AMDGPU_MACH_AMDGCN_GFX600,
-  EF_AMDGPU_MACH_AMDGCN_LAST = EF_AMDGPU_MACH_AMDGCN_GFX90A,
+  EF_AMDGPU_MACH_AMDGCN_LAST = EF_AMDGPU_MACH_AMDGCN_GFX1013,
 
   // Indicates if the "xnack" target feature is enabled for all code contained
   // in the object.
   //
   // Only valid for ELFOSABI_AMDGPU_HSA and ELFABIVERSION_AMDGPU_HSA_V2.
   EF_AMDGPU_FEATURE_XNACK_V2 = 0x01,
   // Indicates if the trap handler is enabled for all code contained
   // in the object.

llvm/include/llvm/Support/TargetParser.h

enum GPUKind : uint32_t {
   GK_GFX908 = 64,
   GK_GFX909 = 65,
   GK_GFX90A = 66,
   GK_GFX90C = 67,
 
   GK_GFX1010 = 71,
   GK_GFX1011 = 72,
   GK_GFX1012 = 73,
+  GK_GFX1013 = 74,
   GK_GFX1030 = 75,
   GK_GFX1031 = 76,
   GK_GFX1032 = 77,
   GK_GFX1033 = 78,
   GK_GFX1034 = 79,
 
   GK_AMDGCN_FIRST = GK_GFX600,
   GK_AMDGCN_LAST = GK_GFX1034,

llvm/lib/Object/ELFObjectFile.cpp

StringRef ELFObjectFileBase::getAMDGPUCPUName() const {
 
   // AMDGCN GFX10.
   case ELF::EF_AMDGPU_MACH_AMDGCN_GFX1010:
     return "gfx1010";
   case ELF::EF_AMDGPU_MACH_AMDGCN_GFX1011:
     return "gfx1011";
   case ELF::EF_AMDGPU_MACH_AMDGCN_GFX1012:
     return "gfx1012";
+  case ELF::EF_AMDGPU_MACH_AMDGCN_GFX1013:
+    return "gfx1013";
   case ELF::EF_AMDGPU_MACH_AMDGCN_GFX1030:
     return "gfx1030";
   case ELF::EF_AMDGPU_MACH_AMDGCN_GFX1031:
     return "gfx1031";
   case ELF::EF_AMDGPU_MACH_AMDGCN_GFX1032:
     return "gfx1032";
   case ELF::EF_AMDGPU_MACH_AMDGCN_GFX1033:
     return "gfx1033";

llvm/lib/ObjectYAML/ELFYAML.cpp

case ELF::EM_AMDGPU:
     BCaseMask(EF_AMDGPU_MACH_AMDGCN_GFX906, EF_AMDGPU_MACH);
     BCaseMask(EF_AMDGPU_MACH_AMDGCN_GFX908, EF_AMDGPU_MACH);
     BCaseMask(EF_AMDGPU_MACH_AMDGCN_GFX909, EF_AMDGPU_MACH);
     BCaseMask(EF_AMDGPU_MACH_AMDGCN_GFX90A, EF_AMDGPU_MACH);
     BCaseMask(EF_AMDGPU_MACH_AMDGCN_GFX90C, EF_AMDGPU_MACH);
     BCaseMask(EF_AMDGPU_MACH_AMDGCN_GFX1010, EF_AMDGPU_MACH);
     BCaseMask(EF_AMDGPU_MACH_AMDGCN_GFX1011, EF_AMDGPU_MACH);
     BCaseMask(EF_AMDGPU_MACH_AMDGCN_GFX1012, EF_AMDGPU_MACH);
+    BCaseMask(EF_AMDGPU_MACH_AMDGCN_GFX1013, EF_AMDGPU_MACH);
     BCaseMask(EF_AMDGPU_MACH_AMDGCN_GFX1030, EF_AMDGPU_MACH);
     BCaseMask(EF_AMDGPU_MACH_AMDGCN_GFX1031, EF_AMDGPU_MACH);
     BCaseMask(EF_AMDGPU_MACH_AMDGCN_GFX1032, EF_AMDGPU_MACH);
     BCaseMask(EF_AMDGPU_MACH_AMDGCN_GFX1033, EF_AMDGPU_MACH);
     BCaseMask(EF_AMDGPU_MACH_AMDGCN_GFX1034, EF_AMDGPU_MACH);
     switch (Object->Header.ABIVersion) {
     default:
       // ELFOSABI_AMDGPU_PAL, ELFOSABI_AMDGPU_MESA3D support *_V3 flags.

llvm/lib/Support/TargetParser.cpp

constexpr GPUInfo AMDGCNGPUs[] = {
   {{"gfx906"},    {"gfx906"},  GK_GFX906,  FEATURE_FAST_FMA_F32|FEATURE_FAST_DENORMAL_F32|FEATURE_XNACK|FEATURE_SRAMECC},
   {{"gfx908"},    {"gfx908"},  GK_GFX908,  FEATURE_FAST_FMA_F32|FEATURE_FAST_DENORMAL_F32|FEATURE_XNACK|FEATURE_SRAMECC},
   {{"gfx909"},    {"gfx909"},  GK_GFX909,  FEATURE_FAST_FMA_F32|FEATURE_FAST_DENORMAL_F32|FEATURE_XNACK},
   {{"gfx90a"},    {"gfx90a"},  GK_GFX90A,  FEATURE_FAST_FMA_F32|FEATURE_FAST_DENORMAL_F32|FEATURE_XNACK|FEATURE_SRAMECC},
   {{"gfx90c"},    {"gfx90c"},  GK_GFX90C,  FEATURE_FAST_FMA_F32|FEATURE_FAST_DENORMAL_F32|FEATURE_XNACK},
   {{"gfx1010"},   {"gfx1010"}, GK_GFX1010, FEATURE_FAST_FMA_F32|FEATURE_FAST_DENORMAL_F32|FEATURE_WAVE32|FEATURE_XNACK},
   {{"gfx1011"},   {"gfx1011"}, GK_GFX1011, FEATURE_FAST_FMA_F32|FEATURE_FAST_DENORMAL_F32|FEATURE_WAVE32|FEATURE_XNACK},
   {{"gfx1012"},   {"gfx1012"}, GK_GFX1012, FEATURE_FAST_FMA_F32|FEATURE_FAST_DENORMAL_F32|FEATURE_WAVE32|FEATURE_XNACK},
+  {{"gfx1013"},   {"gfx1013"}, GK_GFX1013, FEATURE_FAST_FMA_F32|FEATURE_FAST_DENORMAL_F32|FEATURE_WAVE32|FEATURE_XNACK},
   {{"gfx1030"},   {"gfx1030"}, GK_GFX1030, FEATURE_FAST_FMA_F32|FEATURE_FAST_DENORMAL_F32|FEATURE_WAVE32},
   {{"gfx1031"},   {"gfx1031"}, GK_GFX1031, FEATURE_FAST_FMA_F32|FEATURE_FAST_DENORMAL_F32|FEATURE_WAVE32},
   {{"gfx1032"},   {"gfx1032"}, GK_GFX1032, FEATURE_FAST_FMA_F32|FEATURE_FAST_DENORMAL_F32|FEATURE_WAVE32},
   {{"gfx1033"},   {"gfx1033"}, GK_GFX1033, FEATURE_FAST_FMA_F32|FEATURE_FAST_DENORMAL_F32|FEATURE_WAVE32},
   {{"gfx1034"},   {"gfx1034"}, GK_GFX1034, FEATURE_FAST_FMA_F32|FEATURE_FAST_DENORMAL_F32|FEATURE_WAVE32},
 };
 
 const GPUInfo *getArchEntry(AMDGPU::GPUKind AK, ArrayRef<GPUInfo> Table) {
AMDGPU::IsaVersion AMDGPU::getIsaVersion(StringRef GPU) {
   case GK_GFX906:  return {9, 0, 6};
   case GK_GFX908:  return {9, 0, 8};
   case GK_GFX909:  return {9, 0, 9};
   case GK_GFX90A:  return {9, 0, 10};
   case GK_GFX90C:  return {9, 0, 12};
   case GK_GFX1010: return {10, 1, 0};
   case GK_GFX1011: return {10, 1, 1};
   case GK_GFX1012: return {10, 1, 2};
+  case GK_GFX1013: return {10, 1, 3};
   case GK_GFX1030: return {10, 3, 0};
   case GK_GFX1031: return {10, 3, 1};
   case GK_GFX1032: return {10, 3, 2};
   case GK_GFX1033: return {10, 3, 3};
   case GK_GFX1034: return {10, 3, 4};
   default:         return {0, 0, 0};
   }
 }

llvm/lib/Target/AMDGPU/AMDGPU.td

 >;
 
 def FeatureExtendedImageInsts : SubtargetFeature<"extended-image-insts",
   "HasExtendedImageInsts",
   "true",
   "Support mips != 0, lod != 0, gather4, and get_lod"
 >;
 
+def FeatureGFX10_AEncoding : SubtargetFeature<"gfx10_a-encoding",
+  "GFX10_AEncoding",
+  "true",
+  "Has BVH ray tracing instructions"
+>;
+
 def FeatureGFX10_BEncoding : SubtargetFeature<"gfx10_b-encoding",
   "GFX10_BEncoding",
   "true",
   "Encoding format GFX10_B"
 >;
 
 def FeatureIntClamp : SubtargetFeature<"int-clamp-insts",
   "HasIntClamp",
!listconcat(FeatureGroup.GFX10_1_Bugs,
      FeatureScalarAtomics,
      FeatureScalarFlatScratchInsts,
      FeatureGetWaveIdInst,
      FeatureMadMacF32Insts,
      FeatureDsSrc2Insts,
      FeatureLdsMisalignedBug,
      FeatureSupportsXNACK])>;
 
+def FeatureISAVersion10_1_3 : FeatureSet<
+  !listconcat(FeatureGroup.GFX10_1_Bugs,
+    [FeatureGFX10,
+     FeatureGFX10_AEncoding,
+     FeatureLDSBankCount32,
+     FeatureDLInsts,
+     FeatureNSAEncoding,
+     FeatureWavefrontSize32,
+     FeatureScalarStores,
+     FeatureScalarAtomics,
+     FeatureScalarFlatScratchInsts,
+     FeatureGetWaveIdInst,
+     FeatureMadMacF32Insts,
+     FeatureDsSrc2Insts,
+     FeatureLdsMisalignedBug,
+     FeatureSupportsXNACK])>;
+
 def FeatureISAVersion10_3_0 : FeatureSet<
   [FeatureGFX10,
+   FeatureGFX10_AEncoding,
    FeatureGFX10_BEncoding,
    FeatureGFX10_3Insts,
    FeatureLDSBankCount32,
    FeatureDLInsts,
    FeatureDot1Insts,
    FeatureDot2Insts,
    FeatureDot5Insts,
    FeatureDot6Insts,
    FeatureDot7Insts,

 def HasScalarFlatScratchInsts : Predicate<"Subtarget->hasScalarFlatScratchInsts()">,
   AssemblerPredicate<(all_of FeatureScalarFlatScratchInsts)>;
 def HasD16LoadStore : Predicate<"Subtarget->hasD16LoadStore()">,
   AssemblerPredicate<(all_of FeatureGFX9Insts)>;
 
 def HasFlatScratchSTMode : Predicate<"Subtarget->hasFlatScratchSTMode()">,
   AssemblerPredicate<(any_of FeatureGFX10_3Insts)>;
 
+def HasGFX10_AEncoding : Predicate<"Subtarget->hasGFX10_AEncoding()">,
+  AssemblerPredicate<(all_of FeatureGFX10_AEncoding)>;
+
 def HasGFX10_BEncoding : Predicate<"Subtarget->hasGFX10_BEncoding()">,
   AssemblerPredicate<(all_of FeatureGFX10_BEncoding)>;
 
 def HasUnpackedD16VMem : Predicate<"Subtarget->hasUnpackedD16VMem()">,
   AssemblerPredicate<(all_of FeatureUnpackedD16VMem)>;
 def HasPackedD16VMem : Predicate<"!Subtarget->hasUnpackedD16VMem()">,
   AssemblerPredicate<(all_of (not FeatureUnpackedD16VMem))>;
 

llvm/lib/Target/AMDGPU/AMDGPULegalizerInfo.cpp

bool AMDGPULegalizerInfo::legalizeBVHIntrinsic(MachineInstr &MI,
   Register DstReg = MI.getOperand(0).getReg();
   Register NodePtr = MI.getOperand(2).getReg();
   Register RayExtent = MI.getOperand(3).getReg();
   Register RayOrigin = MI.getOperand(4).getReg();
   Register RayDir = MI.getOperand(5).getReg();
   Register RayInvDir = MI.getOperand(6).getReg();
   Register TDescr = MI.getOperand(7).getReg();
 
+  if (!ST.hasGFX10_AEncoding()) {
+    DiagnosticInfoUnsupported BadIntrin(B.getMF().getFunction(),
+                                        "intrinsic not supported on subtarget",
+                                        MI.getDebugLoc());
+    B.getMF().getFunction().getContext().diagnose(BadIntrin);
+    MI.eraseFromParent();
+    return false;
+  }
+
   bool IsA16 = MRI.getType(RayDir).getElementType().getSizeInBits() == 16;
   bool Is64 =  MRI.getType(NodePtr).getSizeInBits() == 64;
   unsigned Opcode = IsA16 ? Is64 ? AMDGPU::IMAGE_BVH64_INTERSECT_RAY_a16_nsa
                                  : AMDGPU::IMAGE_BVH_INTERSECT_RAY_a16_nsa
                           : Is64 ? AMDGPU::IMAGE_BVH64_INTERSECT_RAY_nsa
                                  : AMDGPU::IMAGE_BVH_INTERSECT_RAY_nsa;
 
   SmallVector<Register, 12> Ops;

llvm/lib/Target/AMDGPU/AMDGPUSubtarget.cpp

GCNSubtarget::GCNSubtarget(const Triple &TT, StringRef GPU, StringRef FS,
     HasDPP8(false),
     Has64BitDPP(false),
     HasPackedFP32Ops(false),
     HasExtendedImageInsts(false),
     HasR128A16(false),
     HasGFX10A16(false),
     HasG16(false),
     HasNSAEncoding(false),
+    GFX10_AEncoding(false),
     GFX10_BEncoding(false),
     HasDLInsts(false),
     HasDot1Insts(false),
     HasDot2Insts(false),
     HasDot3Insts(false),
     HasDot4Insts(false),
     HasDot5Insts(false),
     HasDot6Insts(false),

llvm/lib/Target/AMDGPU/GCNProcessors.td

 def : ProcessorModel<"gfx1011", GFX10SpeedModel,
   FeatureISAVersion10_1_1.Features
 >;
 
 def : ProcessorModel<"gfx1012", GFX10SpeedModel,
   FeatureISAVersion10_1_2.Features
 >;
 
+def : ProcessorModel<"gfx1013", GFX10SpeedModel,
+  FeatureISAVersion10_1_3.Features
+>;
+
 def : ProcessorModel<"gfx1030", GFX10SpeedModel,
   FeatureISAVersion10_3_0.Features
 >;
 
 def : ProcessorModel<"gfx1031", GFX10SpeedModel,
   FeatureISAVersion10_3_0.Features
 >;
 

llvm/lib/Target/AMDGPU/GCNSubtarget.h

protected:
   bool HasDPP8;
   bool Has64BitDPP;
   bool HasPackedFP32Ops;
   bool HasExtendedImageInsts;
   bool HasR128A16;
   bool HasGFX10A16;
   bool HasG16;
   bool HasNSAEncoding;
+  bool GFX10_AEncoding;
   bool GFX10_BEncoding;
   bool HasDLInsts;
   bool HasDot1Insts;
   bool HasDot2Insts;
   bool HasDot3Insts;
   bool HasDot4Insts;
   bool HasDot5Insts;
   bool HasDot6Insts;
public:
   }
 
   bool hasImageStoreD16Bug() const { return HasImageStoreD16Bug; }
 
   bool hasImageGather4D16Bug() const { return HasImageGather4D16Bug; }
 
   bool hasNSAEncoding() const { return HasNSAEncoding; }
 
+  bool hasGFX10_AEncoding() const {
+    return GFX10_AEncoding;
+  }
+
   bool hasGFX10_BEncoding() const {
     return GFX10_BEncoding;
   }
 
   bool hasGFX10_3Insts() const {
     return GFX10_3Insts;
   }
 

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

StringRef AMDGPUTargetStreamer::getArchNameFromElfMach(unsigned ElfMach) {
   case ELF::EF_AMDGPU_MACH_AMDGCN_GFX906:  AK = GK_GFX906;  break;
   case ELF::EF_AMDGPU_MACH_AMDGCN_GFX908:  AK = GK_GFX908;  break;
   case ELF::EF_AMDGPU_MACH_AMDGCN_GFX909:  AK = GK_GFX909;  break;
   case ELF::EF_AMDGPU_MACH_AMDGCN_GFX90A:  AK = GK_GFX90A;  break;
   case ELF::EF_AMDGPU_MACH_AMDGCN_GFX90C:  AK = GK_GFX90C;  break;
   case ELF::EF_AMDGPU_MACH_AMDGCN_GFX1010: AK = GK_GFX1010; break;
   case ELF::EF_AMDGPU_MACH_AMDGCN_GFX1011: AK = GK_GFX1011; break;
   case ELF::EF_AMDGPU_MACH_AMDGCN_GFX1012: AK = GK_GFX1012; break;
+  case ELF::EF_AMDGPU_MACH_AMDGCN_GFX1013: AK = GK_GFX1013; break;
   case ELF::EF_AMDGPU_MACH_AMDGCN_GFX1030: AK = GK_GFX1030; break;
   case ELF::EF_AMDGPU_MACH_AMDGCN_GFX1031: AK = GK_GFX1031; break;
   case ELF::EF_AMDGPU_MACH_AMDGCN_GFX1032: AK = GK_GFX1032; break;
   case ELF::EF_AMDGPU_MACH_AMDGCN_GFX1033: AK = GK_GFX1033; break;
   case ELF::EF_AMDGPU_MACH_AMDGCN_GFX1034: AK = GK_GFX1034; break;
   case ELF::EF_AMDGPU_MACH_NONE:           AK = GK_NONE;    break;
   }
 
unsigned AMDGPUTargetStreamer::getElfMach(StringRef GPU) {
   case GK_GFX906:  return ELF::EF_AMDGPU_MACH_AMDGCN_GFX906;
   case GK_GFX908:  return ELF::EF_AMDGPU_MACH_AMDGCN_GFX908;
   case GK_GFX909:  return ELF::EF_AMDGPU_MACH_AMDGCN_GFX909;
   case GK_GFX90A:  return ELF::EF_AMDGPU_MACH_AMDGCN_GFX90A;
   case GK_GFX90C:  return ELF::EF_AMDGPU_MACH_AMDGCN_GFX90C;
   case GK_GFX1010: return ELF::EF_AMDGPU_MACH_AMDGCN_GFX1010;
   case GK_GFX1011: return ELF::EF_AMDGPU_MACH_AMDGCN_GFX1011;
   case GK_GFX1012: return ELF::EF_AMDGPU_MACH_AMDGCN_GFX1012;
+  case GK_GFX1013: return ELF::EF_AMDGPU_MACH_AMDGCN_GFX1013;
   case GK_GFX1030: return ELF::EF_AMDGPU_MACH_AMDGCN_GFX1030;
   case GK_GFX1031: return ELF::EF_AMDGPU_MACH_AMDGCN_GFX1031;
   case GK_GFX1032: return ELF::EF_AMDGPU_MACH_AMDGCN_GFX1032;
   case GK_GFX1033: return ELF::EF_AMDGPU_MACH_AMDGCN_GFX1033;
   case GK_GFX1034: return ELF::EF_AMDGPU_MACH_AMDGCN_GFX1034;
   case GK_NONE:    return ELF::EF_AMDGPU_MACH_NONE;
   }
 

llvm/lib/Target/AMDGPU/MIMGInstructions.td

class MIMG_IntersectRay_nsa_gfx10<mimgopc op, string opcode, int num_addrs, bit A16>
   let InOperandList = !con(nsah.AddrIns,
                            (ins SReg_128:$srsrc),
                            !if(A16, (ins GFX10A16:$a16), (ins)));
   let AsmString = opcode#" $vdata, "#nsah.AddrAsm#", $srsrc"#!if(A16, "$a16", "");
 }
 
 multiclass MIMG_IntersectRay<mimgopc op, string opcode, int num_addrs, bit A16> {
   def "" : MIMGBaseOpcode;
-  let SubtargetPredicate = HasGFX10_BEncoding,
-      AssemblerPredicate = HasGFX10_BEncoding,
+  let SubtargetPredicate = HasGFX10_AEncoding,
+      AssemblerPredicate = HasGFX10_AEncoding,
       AsmMatchConverter = !if(A16, "cvtIntersectRay", ""),
       dmask = 0xf,
       unorm = 1,
       d16 = 0,
       cpol = 0,
       tfe = 0,
       lwe = 0,
       r128 = 1,

 defm IMAGE_SAMPLE_CD_O_G16      : MIMG_Sampler <mimgopc<0xec>, AMDGPUSample_cd_o, 0, 1>;
 defm IMAGE_SAMPLE_CD_CL_O_G16   : MIMG_Sampler <mimgopc<0xed>, AMDGPUSample_cd_cl_o, 0, 1>;
 defm IMAGE_SAMPLE_C_CD_O_G16    : MIMG_Sampler <mimgopc<0xee>, AMDGPUSample_c_cd_o, 0, 1>;
 defm IMAGE_SAMPLE_C_CD_CL_O_G16 : MIMG_Sampler <mimgopc<0xef>, AMDGPUSample_c_cd_cl_o, 0, 1>;
 } // End OtherPredicates = [HasExtendedImageInsts]
 //def IMAGE_RSRC256 : MIMG_NoPattern_RSRC256 <"image_rsrc256", 0x0000007e>;
 //def IMAGE_SAMPLER : MIMG_NoPattern_ <"image_sampler", 0x0000007f>;
 
-let SubtargetPredicate = HasGFX10_BEncoding in
+let SubtargetPredicate = HasGFX10_AEncoding in
 defm IMAGE_MSAA_LOAD_X : MIMG_NoSampler <mimgopc<0x80>, "image_msaa_load", 1, 0, 0, 1>;
 
 defm IMAGE_BVH_INTERSECT_RAY       : MIMG_IntersectRay<mimgopc<0xe6>, "image_bvh_intersect_ray", 11, 0>;
 defm IMAGE_BVH_INTERSECT_RAY_a16   : MIMG_IntersectRay<mimgopc<0xe6>, "image_bvh_intersect_ray", 8, 1>;
 defm IMAGE_BVH64_INTERSECT_RAY     : MIMG_IntersectRay<mimgopc<0xe7>, "image_bvh64_intersect_ray", 12, 0>;
 defm IMAGE_BVH64_INTERSECT_RAY_a16 : MIMG_IntersectRay<mimgopc<0xe7>, "image_bvh64_intersect_ray", 9, 1>;
 
 /********** ========================================= **********/

llvm/lib/Target/AMDGPU/SIISelLowering.cpp

case Intrinsic::amdgcn_image_bvh_intersect_ray: {
     SDValue RayInvDir = M->getOperand(6);
     SDValue TDescr = M->getOperand(7);
 
     assert(NodePtr.getValueType() == MVT::i32 ||
            NodePtr.getValueType() == MVT::i64);
     assert(RayDir.getValueType() == MVT::v4f16 ||
            RayDir.getValueType() == MVT::v4f32);
 
+    if (!Subtarget->hasGFX10_AEncoding()) {
+      emitRemovedIntrinsicError(DAG, DL, Op.getValueType());
+      return SDValue();
+    }
+
     bool IsA16 = RayDir.getValueType().getVectorElementType() == MVT::f16;
     bool Is64 = NodePtr.getValueType() == MVT::i64;
     unsigned Opcode = IsA16 ? Is64 ? AMDGPU::IMAGE_BVH64_INTERSECT_RAY_a16_nsa
                                    : AMDGPU::IMAGE_BVH_INTERSECT_RAY_a16_nsa
                             : Is64 ? AMDGPU::IMAGE_BVH64_INTERSECT_RAY_nsa
                                    : AMDGPU::IMAGE_BVH_INTERSECT_RAY_nsa;
 
     SmallVector<SDValue, 16> Ops;

llvm/lib/Target/AMDGPU/Utils/AMDGPUBaseInfo.h

 bool isSI(const MCSubtargetInfo &STI);
 bool isCI(const MCSubtargetInfo &STI);
 bool isVI(const MCSubtargetInfo &STI);
 bool isGFX9(const MCSubtargetInfo &STI);
 bool isGFX9Plus(const MCSubtargetInfo &STI);
 bool isGFX10(const MCSubtargetInfo &STI);
 bool isGFX10Plus(const MCSubtargetInfo &STI);
 bool isGCN3Encoding(const MCSubtargetInfo &STI);
+bool isGFX10_AEncoding(const MCSubtargetInfo &STI);
 bool isGFX10_BEncoding(const MCSubtargetInfo &STI);
 bool hasGFX10_3Insts(const MCSubtargetInfo &STI);
 bool isGFX90A(const MCSubtargetInfo &STI);
 bool hasArchitectedFlatScratch(const MCSubtargetInfo &STI);
 
 /// Is Reg - scalar register
 bool isSGPR(unsigned Reg, const MCRegisterInfo* TRI);
 

llvm/lib/Target/AMDGPU/Utils/AMDGPUBaseInfo.cpp

 }
 
 bool isGFX10Plus(const MCSubtargetInfo &STI) { return isGFX10(STI); }
 
 bool isGCN3Encoding(const MCSubtargetInfo &STI) {
   return STI.getFeatureBits()[AMDGPU::FeatureGCN3Encoding];
 }
 
+bool isGFX10_AEncoding(const MCSubtargetInfo &STI) {
+  return STI.getFeatureBits()[AMDGPU::FeatureGFX10_AEncoding];
+}
+
 bool isGFX10_BEncoding(const MCSubtargetInfo &STI) {
   return STI.getFeatureBits()[AMDGPU::FeatureGFX10_BEncoding];
 }
 
 bool hasGFX10_3Insts(const MCSubtargetInfo &STI) {
   return STI.getFeatureBits()[AMDGPU::FeatureGFX10_3Insts];
 }
 

llvm/test/CodeGen/AMDGPU/GlobalISel/llvm.amdgcn.intersect_ray.ll

 ; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
 ; RUN: llc -global-isel -march=amdgcn -mcpu=gfx1030 -verify-machineinstrs < %s | FileCheck -check-prefix=GCN %s
+; RUN: llc -global-isel -march=amdgcn -mcpu=gfx1013 -verify-machineinstrs < %s | FileCheck -check-prefix=GCN %s
+; RUN: not --crash llc -global-isel -march=amdgcn -mcpu=gfx1012 -verify-machineinstrs < %s -o /dev/null 2>&1 | FileCheck -check-prefix=ERR %s
 
 ; uint4 llvm.amdgcn.image.bvh.intersect.ray.i32.v4f32(uint node_ptr, float ray_extent, float4 ray_origin, float4 ray_dir, float4 ray_inv_dir, uint4 texture_descr)
 ; uint4 llvm.amdgcn.image.bvh.intersect.ray.i32.v4f16(uint node_ptr, float ray_extent, float4 ray_origin, half4 ray_dir, half4 ray_inv_dir, uint4 texture_descr)
 ; uint4 llvm.amdgcn.image.bvh.intersect.ray.i64.v4f32(ulong node_ptr, float ray_extent, float4 ray_origin, float4 ray_dir, float4 ray_inv_dir, uint4 texture_descr)
 ; uint4 llvm.amdgcn.image.bvh.intersect.ray.i64.v4f16(ulong node_ptr, float ray_extent, float4 ray_origin, half4 ray_dir, half4 ray_inv_dir, uint4 texture_descr)
 
 declare <4 x i32> @llvm.amdgcn.image.bvh.intersect.ray.i32.v4f32(i32, float, <4 x float>, <4 x float>, <4 x float>, <4 x i32>)
 declare <4 x i32> @llvm.amdgcn.image.bvh.intersect.ray.i32.v4f16(i32, float, <4 x float>, <4 x half>, <4 x half>, <4 x i32>)
 declare <4 x i32> @llvm.amdgcn.image.bvh.intersect.ray.i64.v4f32(i64, float, <4 x float>, <4 x float>, <4 x float>, <4 x i32>)
 declare <4 x i32> @llvm.amdgcn.image.bvh.intersect.ray.i64.v4f16(i64, float, <4 x float>, <4 x half>, <4 x half>, <4 x i32>)
 
 define amdgpu_ps <4 x float> @image_bvh_intersect_ray(i32 %node_ptr, float %ray_extent, <4 x float> %ray_origin, <4 x float> %ray_dir, <4 x float> %ray_inv_dir, <4 x i32> inreg %tdescr) {
 ; GCN-LABEL: image_bvh_intersect_ray:
 ; GCN:       ; %bb.0:
 ; GCN-NEXT:    image_bvh_intersect_ray v[0:3], [v0, v1, v2, v3, v4, v6, v7, v8, v10, v11, v12], s[0:3]
 ; GCN-NEXT:    s_waitcnt vmcnt(0)
 ; GCN-NEXT:    ; return to shader part epilog
+; ERR: in function image_bvh_intersect_ray{{.*}}intrinsic not supported on subtarget
   %v = call <4 x i32> @llvm.amdgcn.image.bvh.intersect.ray.i32.v4f32(i32 %node_ptr, float %ray_extent, <4 x float> %ray_origin, <4 x float> %ray_dir, <4 x float> %ray_inv_dir, <4 x i32> %tdescr)
   %r = bitcast <4 x i32> %v to <4 x float>
   ret <4 x float> %r
 }
 
 define amdgpu_ps <4 x float> @image_bvh_intersect_ray_a16(i32 %node_ptr, float %ray_extent, <4 x float> %ray_origin, <4 x half> %ray_dir, <4 x half> %ray_inv_dir, <4 x i32> inreg %tdescr) {
 ; GCN-LABEL: image_bvh_intersect_ray_a16:
 ; GCN:       ; %bb.0:

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

 ; RUN: llc -mtriple=amdgcn-amd-amdhsa --amdhsa-code-object-version=3 -mcpu=gfx1010 -mattr=-xnack < %s | FileCheck --check-prefixes=V3-GFX1010-NOXNACK %s
 ; RUN: llc -mtriple=amdgcn-amd-amdhsa --amdhsa-code-object-version=3 -mcpu=gfx1010 -mattr=+xnack < %s | FileCheck --check-prefixes=V3-GFX1010-XNACK %s
 ; RUN: llc -mtriple=amdgcn-amd-amdhsa --amdhsa-code-object-version=3 -mcpu=gfx1011 < %s | FileCheck --check-prefixes=V3-GFX1011-XNACK %s
 ; RUN: llc -mtriple=amdgcn-amd-amdhsa --amdhsa-code-object-version=3 -mcpu=gfx1011 -mattr=-xnack < %s | FileCheck --check-prefixes=V3-GFX1011-NOXNACK %s
 ; RUN: llc -mtriple=amdgcn-amd-amdhsa --amdhsa-code-object-version=3 -mcpu=gfx1011 -mattr=+xnack < %s | FileCheck --check-prefixes=V3-GFX1011-XNACK %s
 ; RUN: llc -mtriple=amdgcn-amd-amdhsa --amdhsa-code-object-version=3 -mcpu=gfx1012 < %s | FileCheck --check-prefixes=V3-GFX1012-XNACK %s
 ; RUN: llc -mtriple=amdgcn-amd-amdhsa --amdhsa-code-object-version=3 -mcpu=gfx1012 -mattr=-xnack < %s | FileCheck --check-prefixes=V3-GFX1012-NOXNACK %s
 ; RUN: llc -mtriple=amdgcn-amd-amdhsa --amdhsa-code-object-version=3 -mcpu=gfx1012 -mattr=+xnack < %s | FileCheck --check-prefixes=V3-GFX1012-XNACK %s
+; RUN: llc -mtriple=amdgcn-amd-amdhsa --amdhsa-code-object-version=3 -mcpu=gfx1013 < %s | FileCheck --check-prefixes=V3-GFX1013-XNACK %s
+; RUN: llc -mtriple=amdgcn-amd-amdhsa --amdhsa-code-object-version=3 -mcpu=gfx1013 -mattr=-xnack < %s | FileCheck --check-prefixes=V3-GFX1013-NOXNACK %s
+; RUN: llc -mtriple=amdgcn-amd-amdhsa --amdhsa-code-object-version=3 -mcpu=gfx1013 -mattr=+xnack < %s | FileCheck --check-prefixes=V3-GFX1013-XNACK %s
 ; RUN: llc -mtriple=amdgcn-amd-amdhsa --amdhsa-code-object-version=3 -mcpu=gfx1030 < %s | FileCheck --check-prefixes=V3-GFX1030 %s
 ; RUN: llc -mtriple=amdgcn-amd-amdhsa --amdhsa-code-object-version=3 -mcpu=gfx1031 < %s | FileCheck --check-prefixes=V3-GFX1031 %s
 ; RUN: llc -mtriple=amdgcn-amd-amdhsa --amdhsa-code-object-version=3 -mcpu=gfx1032 < %s | FileCheck --check-prefixes=V3-GFX1032 %s
 ; RUN: llc -mtriple=amdgcn-amd-amdhsa --amdhsa-code-object-version=3 -mcpu=gfx1033 < %s | FileCheck --check-prefixes=V3-GFX1033 %s
 ; RUN: llc -mtriple=amdgcn-amd-amdhsa --amdhsa-code-object-version=3 -mcpu=gfx1034 < %s | FileCheck --check-prefixes=V3-GFX1034 %s
 
 ; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx600 < %s | FileCheck --check-prefixes=GFX600 %s
 ; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=tahiti < %s | FileCheck --check-prefixes=GFX600 %s

 ; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx1010 -mattr=-xnack < %s | FileCheck --check-prefixes=GFX1010-NOXNACK %s
 ; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx1010 -mattr=+xnack < %s | FileCheck --check-prefixes=GFX1010-XNACK %s
 ; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx1011 < %s | FileCheck --check-prefixes=GFX1011 %s
 ; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx1011 -mattr=-xnack < %s | FileCheck --check-prefixes=GFX1011-NOXNACK %s
 ; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx1011 -mattr=+xnack < %s | FileCheck --check-prefixes=GFX1011-XNACK %s
 ; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx1012 < %s | FileCheck --check-prefixes=GFX1012 %s
 ; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx1012 -mattr=-xnack < %s | FileCheck --check-prefixes=GFX1012-NOXNACK %s
 ; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx1012 -mattr=+xnack < %s | FileCheck --check-prefixes=GFX1012-XNACK %s
+; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx1013 < %s | FileCheck --check-prefixes=GFX1013 %s
+; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx1013 -mattr=-xnack < %s | FileCheck --check-prefixes=GFX1013-NOXNACK %s
+; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx1013 -mattr=+xnack < %s | FileCheck --check-prefixes=GFX1013-XNACK %s
 ; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx1030 < %s | FileCheck --check-prefixes=GFX1030 %s
 ; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx1031 < %s | FileCheck --check-prefixes=GFX1031 %s
 ; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx1032 < %s | FileCheck --check-prefixes=GFX1032 %s
 ; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx1033 < %s | FileCheck --check-prefixes=GFX1033 %s
 ; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx1034 < %s | FileCheck --check-prefixes=GFX1034 %s
 
 ; V3-GFX600: .amdgcn_target "amdgcn-amd-amdhsa--gfx600"
 ; V3-GFX601: .amdgcn_target "amdgcn-amd-amdhsa--gfx601"

 ; V3-GFX90C-NOXNACK: .amdgcn_target "amdgcn-amd-amdhsa--gfx90c"
 ; V3-GFX90C-XNACK: .amdgcn_target "amdgcn-amd-amdhsa--gfx90c+xnack"
 ; V3-GFX1010-NOXNACK: .amdgcn_target "amdgcn-amd-amdhsa--gfx1010"
 ; V3-GFX1010-XNACK: .amdgcn_target "amdgcn-amd-amdhsa--gfx1010+xnack"
 ; V3-GFX1011-NOXNACK: .amdgcn_target "amdgcn-amd-amdhsa--gfx1011"
 ; V3-GFX1011-XNACK: .amdgcn_target "amdgcn-amd-amdhsa--gfx1011+xnack"
 ; V3-GFX1012-NOXNACK: .amdgcn_target "amdgcn-amd-amdhsa--gfx1012"
 ; V3-GFX1012-XNACK: .amdgcn_target "amdgcn-amd-amdhsa--gfx1012+xnack"
+; V3-GFX1013-NOXNACK: .amdgcn_target "amdgcn-amd-amdhsa--gfx1013"
+; V3-GFX1013-XNACK: .amdgcn_target "amdgcn-amd-amdhsa--gfx1013+xnack"
 ; V3-GFX1030: .amdgcn_target "amdgcn-amd-amdhsa--gfx1030"
 ; V3-GFX1031: .amdgcn_target "amdgcn-amd-amdhsa--gfx1031"
 ; V3-GFX1032: .amdgcn_target "amdgcn-amd-amdhsa--gfx1032"
 ; V3-GFX1033: .amdgcn_target "amdgcn-amd-amdhsa--gfx1033"
 ; V3-GFX1034: .amdgcn_target "amdgcn-amd-amdhsa--gfx1034"
 
 ; GFX600: .amdgcn_target "amdgcn-amd-amdhsa--gfx600"
 ; GFX601: .amdgcn_target "amdgcn-amd-amdhsa--gfx601"

 ; GFX1010-NOXNACK: .amdgcn_target "amdgcn-amd-amdhsa--gfx1010:xnack-"
 ; GFX1010-XNACK: .amdgcn_target "amdgcn-amd-amdhsa--gfx1010:xnack+"
 ; GFX1011: .amdgcn_target "amdgcn-amd-amdhsa--gfx1011"
 ; GFX1011-NOXNACK: .amdgcn_target "amdgcn-amd-amdhsa--gfx1011:xnack-"
 ; GFX1011-XNACK: .amdgcn_target "amdgcn-amd-amdhsa--gfx1011:xnack+"
 ; GFX1012: .amdgcn_target "amdgcn-amd-amdhsa--gfx1012"
 ; GFX1012-NOXNACK: .amdgcn_target "amdgcn-amd-amdhsa--gfx1012:xnack-"
 ; GFX1012-XNACK: .amdgcn_target "amdgcn-amd-amdhsa--gfx1012:xnack+"
+; GFX1013: .amdgcn_target "amdgcn-amd-amdhsa--gfx1013"
+; GFX1013-NOXNACK: .amdgcn_target "amdgcn-amd-amdhsa--gfx1013:xnack-"
+; GFX1013-XNACK: .amdgcn_target "amdgcn-amd-amdhsa--gfx1013:xnack+"
 ; GFX1030: .amdgcn_target "amdgcn-amd-amdhsa--gfx1030"
 ; GFX1031: .amdgcn_target "amdgcn-amd-amdhsa--gfx1031"
 ; GFX1032: .amdgcn_target "amdgcn-amd-amdhsa--gfx1032"
 ; GFX1033: .amdgcn_target "amdgcn-amd-amdhsa--gfx1033"
 ; GFX1034: .amdgcn_target "amdgcn-amd-amdhsa--gfx1034"
 
 define amdgpu_kernel void @directive_amdgcn_target() {
   ret void
 }

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

 ; RUN: llc -filetype=obj -march=amdgcn -mcpu=gfx906 < %s | llvm-readobj -file-headers - | FileCheck --check-prefixes=ALL,ARCH-GCN,GFX906 %s
 ; RUN: llc -filetype=obj -march=amdgcn -mcpu=gfx908 < %s | llvm-readobj -file-headers - | FileCheck --check-prefixes=ALL,ARCH-GCN,GFX908 %s
 ; RUN: llc -filetype=obj -march=amdgcn -mcpu=gfx909 < %s | llvm-readobj -file-headers - | FileCheck --check-prefixes=ALL,ARCH-GCN,GFX909 %s
 ; RUN: llc -filetype=obj -march=amdgcn -mcpu=gfx90a < %s | llvm-readobj -file-headers - | FileCheck --check-prefixes=ALL,ARCH-GCN,GFX90A %s
 ; RUN: llc -filetype=obj -march=amdgcn -mcpu=gfx90c < %s | llvm-readobj -file-headers - | FileCheck --check-prefixes=ALL,ARCH-GCN,GFX90C %s
 ; RUN: llc -filetype=obj -march=amdgcn -mcpu=gfx1010 < %s | llvm-readobj -file-headers - | FileCheck --check-prefixes=ALL,ARCH-GCN,GFX1010 %s
 ; RUN: llc -filetype=obj -march=amdgcn -mcpu=gfx1011 < %s | llvm-readobj -file-headers - | FileCheck --check-prefixes=ALL,ARCH-GCN,GFX1011 %s
 ; RUN: llc -filetype=obj -march=amdgcn -mcpu=gfx1012 < %s | llvm-readobj -file-headers - | FileCheck --check-prefixes=ALL,ARCH-GCN,GFX1012 %s
+; RUN: llc -filetype=obj -march=amdgcn -mcpu=gfx1013 < %s | llvm-readobj -file-headers - | FileCheck --check-prefixes=ALL,ARCH-GCN,GFX1013 %s
 ; RUN: llc -filetype=obj -march=amdgcn -mcpu=gfx1030 < %s | llvm-readobj -file-headers - | FileCheck --check-prefixes=ALL,ARCH-GCN,GFX1030 %s
 ; RUN: llc -filetype=obj -march=amdgcn -mcpu=gfx1031 < %s | llvm-readobj -file-headers - | FileCheck --check-prefixes=ALL,ARCH-GCN,GFX1031 %s
 ; RUN: llc -filetype=obj -march=amdgcn -mcpu=gfx1032 < %s | llvm-readobj -file-headers - | FileCheck --check-prefixes=ALL,ARCH-GCN,GFX1032 %s
 ; RUN: llc -filetype=obj -march=amdgcn -mcpu=gfx1033 < %s | llvm-readobj -file-headers - | FileCheck --check-prefixes=ALL,ARCH-GCN,GFX1033 %s
 ; RUN: llc -filetype=obj -march=amdgcn -mcpu=gfx1034 < %s | llvm-readobj -file-headers - | FileCheck --check-prefixes=ALL,ARCH-GCN,GFX1034 %s
 
 ; FIXME: With the default attributes the eflags are not accurate for
 ; xnack and sramecc. Subsequent Target-ID patches will address this.

 ; GFX906:        EF_AMDGPU_MACH_AMDGCN_GFX906 (0x2F)
 ; GFX908:        EF_AMDGPU_MACH_AMDGCN_GFX908 (0x30)
 ; GFX909:        EF_AMDGPU_MACH_AMDGCN_GFX909 (0x31)
 ; GFX90A:        EF_AMDGPU_MACH_AMDGCN_GFX90A (0x3F)
 ; GFX90C:        EF_AMDGPU_MACH_AMDGCN_GFX90C (0x32)
 ; GFX1010:       EF_AMDGPU_MACH_AMDGCN_GFX1010 (0x33)
 ; GFX1011:       EF_AMDGPU_MACH_AMDGCN_GFX1011 (0x34)
 ; GFX1012:       EF_AMDGPU_MACH_AMDGCN_GFX1012 (0x35)
+; GFX1013:       EF_AMDGPU_MACH_AMDGCN_GFX1013 (0x42)
 ; GFX1030:       EF_AMDGPU_MACH_AMDGCN_GFX1030 (0x36)
 ; GFX1031:       EF_AMDGPU_MACH_AMDGCN_GFX1031 (0x37)
 ; GFX1032:       EF_AMDGPU_MACH_AMDGCN_GFX1032 (0x38)
 ; GFX1033:       EF_AMDGPU_MACH_AMDGCN_GFX1033 (0x39)
 ; GFX1034:       EF_AMDGPU_MACH_AMDGCN_GFX1034 (0x3E)
 ; ALL:         ]
 
 define amdgpu_kernel void @elf_header() {
   ret void
 }

llvm/test/CodeGen/AMDGPU/llvm.amdgcn.intersect_ray.ll

 ; RUN: llc -march=amdgcn -mcpu=gfx1030 -verify-machineinstrs < %s | FileCheck -check-prefix=GCN %s
+; RUN: llc -march=amdgcn -mcpu=gfx1013 -verify-machineinstrs < %s | FileCheck -check-prefix=GCN %s
+; RUN: not --crash llc -march=amdgcn -mcpu=gfx1012 -verify-machineinstrs < %s 2>&1 | FileCheck -check-prefix=ERR %s
 
 ; uint4 llvm.amdgcn.image.bvh.intersect.ray.i32.v4f32(uint node_ptr, float ray_extent, float4 ray_origin, float4 ray_dir, float4 ray_inv_dir, uint4 texture_descr)
 ; uint4 llvm.amdgcn.image.bvh.intersect.ray.i32.v4f16(uint node_ptr, float ray_extent, float4 ray_origin, half4 ray_dir, half4 ray_inv_dir, uint4 texture_descr)
 ; uint4 llvm.amdgcn.image.bvh.intersect.ray.i64.v4f32(ulong node_ptr, float ray_extent, float4 ray_origin, float4 ray_dir, float4 ray_inv_dir, uint4 texture_descr)
 ; uint4 llvm.amdgcn.image.bvh.intersect.ray.i64.v4f16(ulong node_ptr, float ray_extent, float4 ray_origin, half4 ray_dir, half4 ray_inv_dir, uint4 texture_descr)
 
 declare <4 x i32> @llvm.amdgcn.image.bvh.intersect.ray.i32.v4f32(i32, float, <4 x float>, <4 x float>, <4 x float>, <4 x i32>)
 declare <4 x i32> @llvm.amdgcn.image.bvh.intersect.ray.i32.v4f16(i32, float, <4 x float>, <4 x half>, <4 x half>, <4 x i32>)
 declare <4 x i32> @llvm.amdgcn.image.bvh.intersect.ray.i64.v4f32(i64, float, <4 x float>, <4 x float>, <4 x float>, <4 x i32>)
 declare <4 x i32> @llvm.amdgcn.image.bvh.intersect.ray.i64.v4f16(i64, float, <4 x float>, <4 x half>, <4 x half>, <4 x i32>)
 
 ; GCN-LABEL: {{^}}image_bvh_intersect_ray:
 ; GCN: image_bvh_intersect_ray v[0:3], v[0:15], s[0:3]{{$}}
+; ERR: in function image_bvh_intersect_ray{{.*}}intrinsic not supported on subtarget
 ; Arguments are flattened to represent the actual VGPR_A layout, so we have no
 ; extra moves in the generated kernel.
 define amdgpu_ps <4 x float> @image_bvh_intersect_ray(i32 %node_ptr, float %ray_extent, float %ray_origin_x, float %ray_origin_y, float %ray_origin_z, float %ray_dir_x, float %ray_dir_y, float %ray_dir_z, float %ray_inv_dir_x, float %ray_inv_dir_y, float %ray_inv_dir_z, <4 x i32> inreg %tdescr) {
 main_body:
   %ray_origin0 = insertelement <4 x float> undef, float %ray_origin_x, i32 0
   %ray_origin1 = insertelement <4 x float> %ray_origin0, float %ray_origin_y, i32 1
   %ray_origin = insertelement <4 x float> %ray_origin1, float %ray_origin_z, i32 2
   %ray_dir0 = insertelement <4 x float> undef, float %ray_dir_x, i32 0

llvm/test/MC/AMDGPU/dl-insts-err.s

 // RUN: not llvm-mc -arch=amdgcn -mcpu=gfx800 %s 2>&1 | FileCheck %s
 // RUN: not llvm-mc -arch=amdgcn -mcpu=gfx900 %s 2>&1 | FileCheck %s
 // RUN: not llvm-mc -arch=amdgcn -mcpu=gfx906 %s 2>&1 | FileCheck %s --check-prefix=GFX906-GFX908
 // RUN: not llvm-mc -arch=amdgcn -mcpu=gfx908 %s 2>&1 | FileCheck %s --check-prefix=GFX906-GFX908
+// RUN: not llvm-mc -arch=amdgcn -mcpu=gfx1013 %s 2>&1 | FileCheck %s --check-prefix=GFX1013
 
 //
 // Test unsupported GPUs.
 //
 
 // CHECK: error: instruction not supported on this GPU
 v_fmac_f32 v0, v1, v2
 // CHECK: error: instruction not supported on this GPU
 v_xnor_b32 v0, v1, v2
 // CHECK: error: instruction not supported on this GPU
+// GFX1013: error: instruction not supported on this GPU
 v_dot2_f32_f16 v0, v1, v2, v3
 // CHECK: error: instruction not supported on this GPU
+// GFX1013: error: instruction not supported on this GPU
 v_dot2_i32_i16 v0, v1, v2, v3
 // CHECK: error: instruction not supported on this GPU
+// GFX1013: error: instruction not supported on this GPU
 v_dot2_u32_u16 v0, v1, v2, v3
 // CHECK: error: instruction not supported on this GPU
+// GFX1013: error: instruction not supported on this GPU
 v_dot4_i32_i8 v0, v1, v2, v3
 // CHECK: error: instruction not supported on this GPU
+// GFX1013: error: instruction not supported on this GPU
 v_dot4_u32_u8 v0, v1, v2, v3
 // CHECK: error: instruction not supported on this GPU
+// GFX1013: error: instruction not supported on this GPU
 v_dot8_i32_i4 v0, v1, v2, v3
 // CHECK: error: instruction not supported on this GPU
+// GFX1013: error: instruction not supported on this GPU
 v_dot8_u32_u4 v0, v1, v2, v3
 
 //
 // Test invalid operands.
 //
 
 // GFX906-GFX908: error: invalid operand for instruction
 v_dot2_f32_f16 v0, v1, v2, v3 op_sel

llvm/test/MC/AMDGPU/gfx10_unsupported.s

 // RUN: not llvm-mc -arch=amdgcn -mcpu=gfx1010 -mattr=+wavefrontsize32,-wavefrontsize64 %s 2>&1 | FileCheck --implicit-check-not=error: %s
 // RUN: not llvm-mc -arch=amdgcn -mcpu=gfx1010 -mattr=-wavefrontsize32,+wavefrontsize64 %s 2>&1 | FileCheck --implicit-check-not=error: %s
+// RUN: not llvm-mc -arch=amdgcn -mcpu=gfx1013 -mattr=-wavefrontsize32,+wavefrontsize64 %s 2>&1 | FileCheck --implicit-check-not=error: %s
 
 //===----------------------------------------------------------------------===//
 // Unsupported instructions.
 //===----------------------------------------------------------------------===//
 
 buffer_atomic_add_f32 v255, off, s[8:11], s3 offset:4095
 // CHECK: :[[@LINE-1]]:{{[0-9]+}}: error: instruction not supported on this GPU
 

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

 # RUN: sed -e 's/<BITS>/64/' -e 's/<MACH>/AMDGCN_GFX1011/' %s | yaml2obj -o %t.o.AMDGCN_GFX1011
 # RUN: llvm-readobj -S --file-headers %t.o.AMDGCN_GFX1011 | FileCheck --check-prefixes=ELF-AMDGCN-ALL,ELF-AMDGCN-GFX1011 %s
 # RUN: obj2yaml %t.o.AMDGCN_GFX1011 | FileCheck --check-prefixes=YAML-AMDGCN-ALL,YAML-AMDGCN-GFX1011 %s
 
 # RUN: sed -e 's/<BITS>/64/' -e 's/<MACH>/AMDGCN_GFX1012/' %s | yaml2obj -o %t.o.AMDGCN_GFX1012
 # RUN: llvm-readobj -S --file-headers %t.o.AMDGCN_GFX1012 | FileCheck --check-prefixes=ELF-AMDGCN-ALL,ELF-AMDGCN-GFX1012 %s
 # RUN: obj2yaml %t.o.AMDGCN_GFX1012 | FileCheck --check-prefixes=YAML-AMDGCN-ALL,YAML-AMDGCN-GFX1012 %s
 
+# RUN: sed -e 's/<BITS>/64/' -e 's/<MACH>/AMDGCN_GFX1013/' %s | yaml2obj -o %t.o.AMDGCN_GFX1013
+# RUN: llvm-readobj -S --file-headers %t.o.AMDGCN_GFX1013 | FileCheck --check-prefixes=ELF-AMDGCN-ALL,ELF-AMDGCN-GFX1013 %s
+# RUN: obj2yaml %t.o.AMDGCN_GFX1013 | FileCheck --check-prefixes=YAML-AMDGCN-ALL,YAML-AMDGCN-GFX1013 %s
+
 # RUN: sed -e 's/<BITS>/64/' -e 's/<MACH>/AMDGCN_GFX1030/' %s | yaml2obj -o %t.o.AMDGCN_GFX1030
 # RUN: llvm-readobj -S --file-headers %t.o.AMDGCN_GFX1030 | FileCheck --check-prefixes=ELF-AMDGCN-ALL,ELF-AMDGCN-GFX1030 %s
 # RUN: obj2yaml %t.o.AMDGCN_GFX1030 | FileCheck --check-prefixes=YAML-AMDGCN-ALL,YAML-AMDGCN-GFX1030 %s
 
 # RUN: sed -e 's/<BITS>/64/' -e 's/<MACH>/AMDGCN_GFX1031/' %s | yaml2obj -o %t.o.AMDGCN_GFX1031
 # RUN: llvm-readobj -S --file-headers %t.o.AMDGCN_GFX1031 | FileCheck --check-prefixes=ELF-AMDGCN-ALL,ELF-AMDGCN-GFX1031 %s
 # RUN: obj2yaml %t.o.AMDGCN_GFX1031 | FileCheck --check-prefixes=YAML-AMDGCN-ALL,YAML-AMDGCN-GFX1031 %s
 

 # YAML-AMDGCN-GFX1010:  Flags: [ EF_AMDGPU_MACH_AMDGCN_GFX1010 ]
 
 # ELF-AMDGCN-GFX1011:   EF_AMDGPU_MACH_AMDGCN_GFX1011 (0x34)
 # YAML-AMDGCN-GFX1011:  Flags: [ EF_AMDGPU_MACH_AMDGCN_GFX1011 ]
 
 # ELF-AMDGCN-GFX1012:   EF_AMDGPU_MACH_AMDGCN_GFX1012 (0x35)
 # YAML-AMDGCN-GFX1012:  Flags: [ EF_AMDGPU_MACH_AMDGCN_GFX1012 ]
 
+# ELF-AMDGCN-GFX1013:   EF_AMDGPU_MACH_AMDGCN_GFX1013 (0x42)
+# YAML-AMDGCN-GFX1013:  Flags: [ EF_AMDGPU_MACH_AMDGCN_GFX1013 ]
+
 # ELF-AMDGCN-GFX1030:   EF_AMDGPU_MACH_AMDGCN_GFX1030 (0x36)
 # YAML-AMDGCN-GFX1030:  Flags: [ EF_AMDGPU_MACH_AMDGCN_GFX1030 ]
 
 # ELF-AMDGCN-GFX1031:   EF_AMDGPU_MACH_AMDGCN_GFX1031 (0x37)
 # YAML-AMDGCN-GFX1031:  Flags: [ EF_AMDGPU_MACH_AMDGCN_GFX1031 ]
 
 # ELF-AMDGCN-GFX1032:   EF_AMDGPU_MACH_AMDGCN_GFX1032 (0x38)
 # YAML-AMDGCN-GFX1032:  Flags: [ EF_AMDGPU_MACH_AMDGCN_GFX1032 ]

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

 ; RUN: llvm-objdump -D %t.o > %t-detect.txt
 ; RUN: diff %t-specify.txt %t-detect.txt
 
 ; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx1030 -filetype=obj -O0 -o %t.o %s
 ; RUN: llvm-objdump -D --arch-name=amdgcn --mcpu=gfx1030 %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=gfx1013 -filetype=obj -O0 -o %t.o %s
+; RUN: llvm-objdump -D --arch-name=amdgcn --mcpu=gfx1013 %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=gfx1012 -filetype=obj -O0 -o %t.o %s
 ; RUN: llvm-objdump -D --arch-name=amdgcn --mcpu=gfx1012 %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=gfx1011 -filetype=obj -O0 -o %t.o %s
 ; RUN: llvm-objdump -D --arch-name=amdgcn --mcpu=gfx1011 %t.o > %t-specify.txt
 ; RUN: llvm-objdump -D %t.o > %t-detect.txt

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

 # 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_GFX1012 -DFLAG_VALUE=0x35
 
 # RUN: yaml2obj %s -o %t -DABI_VERSION=1 -DFLAG_NAME=EF_AMDGPU_MACH_AMDGCN_GFX1012
 # 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_GFX1012 -DFLAG_VALUE=0x35
 
 # RUN: yaml2obj %s -o %t -DABI_VERSION=2 -DFLAG_NAME=EF_AMDGPU_MACH_AMDGCN_GFX1012
 # 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_GFX1012 -DFLAG_VALUE=0x35
 
+# RUN: yaml2obj %s -o %t -DABI_VERSION=0 -DFLAG_NAME=EF_AMDGPU_MACH_AMDGCN_GFX1013
+# 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_GFX1013 -DFLAG_VALUE=0x42
+
+# RUN: yaml2obj %s -o %t -DABI_VERSION=1 -DFLAG_NAME=EF_AMDGPU_MACH_AMDGCN_GFX1013
+# 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_GFX1013 -DFLAG_VALUE=0x42
+
+# RUN: yaml2obj %s -o %t -DABI_VERSION=2 -DFLAG_NAME=EF_AMDGPU_MACH_AMDGCN_GFX1013
+# 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_GFX1013 -DFLAG_VALUE=0x42
+
 # RUN: yaml2obj %s -o %t -DABI_VERSION=0 -DFLAG_NAME=EF_AMDGPU_MACH_AMDGCN_GFX1030
 # 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_GFX1030 -DFLAG_VALUE=0x36
 
 # RUN: yaml2obj %s -o %t -DABI_VERSION=1 -DFLAG_NAME=EF_AMDGPU_MACH_AMDGCN_GFX1030
 # 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_GFX1030 -DFLAG_VALUE=0x36
 
 # RUN: yaml2obj %s -o %t -DABI_VERSION=2 -DFLAG_NAME=EF_AMDGPU_MACH_AMDGCN_GFX1030
 # 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_GFX1030 -DFLAG_VALUE=0x36

llvm/tools/llvm-readobj/ELFDumper.cpp

static const EnumEntry<unsigned> ElfHeaderAMDGPUFlagsABIVersion3[] = {
   LLVM_READOBJ_ENUM_ENT(ELF, EF_AMDGPU_MACH_AMDGCN_GFX906),
   LLVM_READOBJ_ENUM_ENT(ELF, EF_AMDGPU_MACH_AMDGCN_GFX908),
   LLVM_READOBJ_ENUM_ENT(ELF, EF_AMDGPU_MACH_AMDGCN_GFX909),
   LLVM_READOBJ_ENUM_ENT(ELF, EF_AMDGPU_MACH_AMDGCN_GFX90A),
   LLVM_READOBJ_ENUM_ENT(ELF, EF_AMDGPU_MACH_AMDGCN_GFX90C),
   LLVM_READOBJ_ENUM_ENT(ELF, EF_AMDGPU_MACH_AMDGCN_GFX1010),
   LLVM_READOBJ_ENUM_ENT(ELF, EF_AMDGPU_MACH_AMDGCN_GFX1011),
   LLVM_READOBJ_ENUM_ENT(ELF, EF_AMDGPU_MACH_AMDGCN_GFX1012),
+  LLVM_READOBJ_ENUM_ENT(ELF, EF_AMDGPU_MACH_AMDGCN_GFX1013),
   LLVM_READOBJ_ENUM_ENT(ELF, EF_AMDGPU_MACH_AMDGCN_GFX1030),
   LLVM_READOBJ_ENUM_ENT(ELF, EF_AMDGPU_MACH_AMDGCN_GFX1031),
   LLVM_READOBJ_ENUM_ENT(ELF, EF_AMDGPU_MACH_AMDGCN_GFX1032),
   LLVM_READOBJ_ENUM_ENT(ELF, EF_AMDGPU_MACH_AMDGCN_GFX1033),
   LLVM_READOBJ_ENUM_ENT(ELF, EF_AMDGPU_MACH_AMDGCN_GFX1034),
   LLVM_READOBJ_ENUM_ENT(ELF, EF_AMDGPU_FEATURE_XNACK_V3),
   LLVM_READOBJ_ENUM_ENT(ELF, EF_AMDGPU_FEATURE_SRAMECC_V3)
 };
static const EnumEntry<unsigned> ElfHeaderAMDGPUFlagsABIVersion4[] = {
   LLVM_READOBJ_ENUM_ENT(ELF, EF_AMDGPU_MACH_AMDGCN_GFX906),
   LLVM_READOBJ_ENUM_ENT(ELF, EF_AMDGPU_MACH_AMDGCN_GFX908),
   LLVM_READOBJ_ENUM_ENT(ELF, EF_AMDGPU_MACH_AMDGCN_GFX909),
   LLVM_READOBJ_ENUM_ENT(ELF, EF_AMDGPU_MACH_AMDGCN_GFX90A),
   LLVM_READOBJ_ENUM_ENT(ELF, EF_AMDGPU_MACH_AMDGCN_GFX90C),
   LLVM_READOBJ_ENUM_ENT(ELF, EF_AMDGPU_MACH_AMDGCN_GFX1010),
   LLVM_READOBJ_ENUM_ENT(ELF, EF_AMDGPU_MACH_AMDGCN_GFX1011),
   LLVM_READOBJ_ENUM_ENT(ELF, EF_AMDGPU_MACH_AMDGCN_GFX1012),
+  LLVM_READOBJ_ENUM_ENT(ELF, EF_AMDGPU_MACH_AMDGCN_GFX1013),
   LLVM_READOBJ_ENUM_ENT(ELF, EF_AMDGPU_MACH_AMDGCN_GFX1030),
   LLVM_READOBJ_ENUM_ENT(ELF, EF_AMDGPU_MACH_AMDGCN_GFX1031),
   LLVM_READOBJ_ENUM_ENT(ELF, EF_AMDGPU_MACH_AMDGCN_GFX1032),
   LLVM_READOBJ_ENUM_ENT(ELF, EF_AMDGPU_MACH_AMDGCN_GFX1033),
   LLVM_READOBJ_ENUM_ENT(ELF, EF_AMDGPU_MACH_AMDGCN_GFX1034),
   LLVM_READOBJ_ENUM_ENT(ELF, EF_AMDGPU_FEATURE_XNACK_ANY_V4),
   LLVM_READOBJ_ENUM_ENT(ELF, EF_AMDGPU_FEATURE_XNACK_OFF_V4),
   LLVM_READOBJ_ENUM_ENT(ELF, EF_AMDGPU_FEATURE_XNACK_ON_V4),

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

const char *get_elf_mach_gfx_name(uint32_t EFlags) {
   case EF_AMDGPU_MACH_AMDGCN_GFX90C:
     return "gfx90c";
   case EF_AMDGPU_MACH_AMDGCN_GFX1010:
     return "gfx1010";
   case EF_AMDGPU_MACH_AMDGCN_GFX1011:
     return "gfx1011";
   case EF_AMDGPU_MACH_AMDGCN_GFX1012:
     return "gfx1012";
+  case EF_AMDGPU_MACH_AMDGCN_GFX1013:
+    return "gfx1013";
   case EF_AMDGPU_MACH_AMDGCN_GFX1030:
     return "gfx1030";
   case EF_AMDGPU_MACH_AMDGCN_GFX1031:
     return "gfx1031";
   case EF_AMDGPU_MACH_AMDGCN_GFX1032:
     return "gfx1032";
   case EF_AMDGPU_MACH_AMDGCN_GFX1033:
     return "gfx1033";
   case EF_AMDGPU_MACH_AMDGCN_GFX1034:
     return "gfx1034";
   default:
     return "--unknown gfx";
   }
 }