summaryrefslogtreecommitdiffstats
diff options
context:
space:
mode:
Diffstat
-rw-r--r--llvm/lib/Target/AMDGPU/AMDGPUAsmPrinter.cpp201
-rw-r--r--llvm/lib/Target/AMDGPU/AMDGPURuntimeMetadata.h39
-rw-r--r--llvm/test/CodeGen/AMDGPU/runtime-metadata.ll1357
3 files changed, 1483 insertions, 114 deletions
diff --git a/llvm/lib/Target/AMDGPU/AMDGPUAsmPrinter.cpp b/llvm/lib/Target/AMDGPU/AMDGPUAsmPrinter.cpp
index 050a22d0cc5..1319f7ed2bd 100644
--- a/llvm/lib/Target/AMDGPU/AMDGPUAsmPrinter.cpp
+++ b/llvm/lib/Target/AMDGPU/AMDGPUAsmPrinter.cpp
@@ -824,6 +824,15 @@ void AMDGPUAsmPrinter::emitStartOfRuntimeMetadata(const Module &M) {
}
}
}
+
+ if (auto MD = M.getNamedMetadata("llvm.printf.fmts")) {
+ for (unsigned I = 0; I < MD->getNumOperands(); ++I) {
+ auto Node = MD->getOperand(I);
+ if (Node->getNumOperands() > 0)
+ emitRuntimeMDStringValue(*OutStreamer, RuntimeMD::KeyPrintfInfo,
+ cast<MDString>(Node->getOperand(0))->getString());
+ }
+ }
}
static std::string getOCLTypeName(Type *Ty, bool Signed) {
@@ -896,6 +905,93 @@ static RuntimeMD::KernelArg::ValueType getRuntimeMDValueType(
}
}
+static RuntimeMD::KernelArg::AddressSpaceQualifer getRuntimeAddrSpace(
+ AMDGPUAS::AddressSpaces A) {
+ switch (A) {
+ case AMDGPUAS::GLOBAL_ADDRESS:
+ return RuntimeMD::KernelArg::Global;
+ case AMDGPUAS::CONSTANT_ADDRESS:
+ return RuntimeMD::KernelArg::Constant;
+ case AMDGPUAS::LOCAL_ADDRESS:
+ return RuntimeMD::KernelArg::Local;
+ case AMDGPUAS::FLAT_ADDRESS:
+ return RuntimeMD::KernelArg::Generic;
+ case AMDGPUAS::REGION_ADDRESS:
+ return RuntimeMD::KernelArg::Region;
+ default:
+ return RuntimeMD::KernelArg::Private;
+ }
+}
+
+static void emitRuntimeMetadataForKernelArg(const DataLayout &DL,
+ MCStreamer &OutStreamer, Type *T,
+ RuntimeMD::KernelArg::Kind Kind,
+ StringRef BaseTypeName = "", StringRef TypeName = "",
+ StringRef ArgName = "", StringRef TypeQual = "", StringRef AccQual = "") {
+ // Emit KeyArgBegin.
+ OutStreamer.EmitIntValue(RuntimeMD::KeyArgBegin, 1);
+
+ // Emit KeyArgSize and KeyArgAlign.
+ emitRuntimeMDIntValue(OutStreamer, RuntimeMD::KeyArgSize,
+ DL.getTypeAllocSize(T), 4);
+ emitRuntimeMDIntValue(OutStreamer, RuntimeMD::KeyArgAlign,
+ DL.getABITypeAlignment(T), 4);
+ if (auto PT = dyn_cast<PointerType>(T)) {
+ auto ET = PT->getElementType();
+ if (PT->getAddressSpace() == AMDGPUAS::LOCAL_ADDRESS && ET->isSized())
+ emitRuntimeMDIntValue(OutStreamer, RuntimeMD::KeyArgPointeeAlign,
+ DL.getABITypeAlignment(ET), 4);
+ }
+
+ // Emit KeyArgTypeName.
+ if (!TypeName.empty())
+ emitRuntimeMDStringValue(OutStreamer, RuntimeMD::KeyArgTypeName, TypeName);
+
+ // Emit KeyArgName.
+ if (!ArgName.empty())
+ emitRuntimeMDStringValue(OutStreamer, RuntimeMD::KeyArgName, ArgName);
+
+ // Emit KeyArgIsVolatile, KeyArgIsRestrict, KeyArgIsConst and KeyArgIsPipe.
+ SmallVector<StringRef, 1> SplitQ;
+ TypeQual.split(SplitQ, " ", -1, false /* Drop empty entry */);
+
+ for (StringRef KeyName : SplitQ) {
+ auto Key = StringSwitch<RuntimeMD::Key>(KeyName)
+ .Case("volatile", RuntimeMD::KeyArgIsVolatile)
+ .Case("restrict", RuntimeMD::KeyArgIsRestrict)
+ .Case("const", RuntimeMD::KeyArgIsConst)
+ .Case("pipe", RuntimeMD::KeyArgIsPipe)
+ .Default(RuntimeMD::KeyNull);
+ OutStreamer.EmitIntValue(Key, 1);
+ }
+
+ // Emit KeyArgKind.
+ emitRuntimeMDIntValue(OutStreamer, RuntimeMD::KeyArgKind, Kind, 1);
+
+ // Emit KeyArgValueType.
+ emitRuntimeMDIntValue(OutStreamer, RuntimeMD::KeyArgValueType,
+ getRuntimeMDValueType(T, BaseTypeName), 2);
+
+ // Emit KeyArgAccQual.
+ if (!AccQual.empty()) {
+ auto AQ = StringSwitch<RuntimeMD::KernelArg::AccessQualifer>(AccQual)
+ .Case("read_only", RuntimeMD::KernelArg::ReadOnly)
+ .Case("write_only", RuntimeMD::KernelArg::WriteOnly)
+ .Case("read_write", RuntimeMD::KernelArg::ReadWrite)
+ .Default(RuntimeMD::KernelArg::None);
+ emitRuntimeMDIntValue(OutStreamer, RuntimeMD::KeyArgAccQual, AQ, 1);
+ }
+
+ // Emit KeyArgAddrQual.
+ if (auto *PT = dyn_cast<PointerType>(T))
+ emitRuntimeMDIntValue(OutStreamer, RuntimeMD::KeyArgAddrQual,
+ getRuntimeAddrSpace(static_cast<AMDGPUAS::AddressSpaces>(
+ PT->getAddressSpace())), 1);
+
+ // Emit KeyArgEnd
+ OutStreamer.EmitIntValue(RuntimeMD::KeyArgEnd, 1);
+}
+
void AMDGPUAsmPrinter::emitRuntimeMetadata(const Function &F) {
if (!F.getMetadata("kernel_arg_type"))
return;
@@ -906,56 +1002,25 @@ void AMDGPUAsmPrinter::emitRuntimeMetadata(const Function &F) {
OutStreamer->EmitIntValue(RuntimeMD::KeyKernelBegin, 1);
emitRuntimeMDStringValue(*OutStreamer, RuntimeMD::KeyKernelName, F.getName());
+ const DataLayout &DL = F.getParent()->getDataLayout();
for (auto &Arg : F.args()) {
- // Emit KeyArgBegin.
unsigned I = Arg.getArgNo();
- OutStreamer->EmitIntValue(RuntimeMD::KeyArgBegin, 1);
-
- // Emit KeyArgSize, KeyArgAlign and KeyArgPointeeAlign.
Type *T = Arg.getType();
- const DataLayout &DL = F.getParent()->getDataLayout();
- emitRuntimeMDIntValue(*OutStreamer, RuntimeMD::KeyArgSize,
- DL.getTypeAllocSize(T), 4);
- emitRuntimeMDIntValue(*OutStreamer, RuntimeMD::KeyArgAlign,
- DL.getABITypeAlignment(T), 4);
- if (auto PT = dyn_cast<PointerType>(T)) {
- auto ET = PT->getElementType();
- if (ET->isSized())
- emitRuntimeMDIntValue(*OutStreamer, RuntimeMD::KeyArgPointeeAlign,
- DL.getABITypeAlignment(ET), 4);
- }
-
- // Emit KeyArgTypeName.
auto TypeName = dyn_cast<MDString>(F.getMetadata(
- "kernel_arg_type")->getOperand(I))->getString();
- emitRuntimeMDStringValue(*OutStreamer, RuntimeMD::KeyArgTypeName, TypeName);
-
- // Emit KeyArgName.
- if (auto ArgNameMD = F.getMetadata("kernel_arg_name")) {
- auto ArgName = cast<MDString>(ArgNameMD->getOperand(I))->getString();
- emitRuntimeMDStringValue(*OutStreamer, RuntimeMD::KeyArgName, ArgName);
- }
-
- // Emit KeyArgIsVolatile, KeyArgIsRestrict, KeyArgIsConst and KeyArgIsPipe.
+ "kernel_arg_type")->getOperand(I))->getString();
+ auto BaseTypeName = cast<MDString>(F.getMetadata(
+ "kernel_arg_base_type")->getOperand(I))->getString();
+ StringRef ArgName;
+ if (auto ArgNameMD = F.getMetadata("kernel_arg_name"))
+ ArgName = cast<MDString>(ArgNameMD->getOperand(I))->getString();
auto TypeQual = cast<MDString>(F.getMetadata(
- "kernel_arg_type_qual")->getOperand(I))->getString();
- SmallVector<StringRef, 1> SplitQ;
- TypeQual.split(SplitQ, " ", -1, false /* Drop empty entry */);
-
- for (StringRef KeyName : SplitQ) {
- auto Key = StringSwitch<RuntimeMD::Key>(KeyName)
- .Case("volatile", RuntimeMD::KeyArgIsVolatile)
- .Case("restrict", RuntimeMD::KeyArgIsRestrict)
- .Case("const", RuntimeMD::KeyArgIsConst)
- .Case("pipe", RuntimeMD::KeyArgIsPipe)
- .Default(RuntimeMD::KeyNull);
- OutStreamer->EmitIntValue(Key, 1);
- }
-
- // Emit KeyArgTypeKind.
- auto BaseTypeName = cast<MDString>(
- F.getMetadata("kernel_arg_base_type")->getOperand(I))->getString();
- auto TypeKind = StringSwitch<RuntimeMD::KernelArg::TypeKind>(BaseTypeName)
+ "kernel_arg_type_qual")->getOperand(I))->getString();
+ auto AccQual = cast<MDString>(F.getMetadata(
+ "kernel_arg_access_qual")->getOperand(I))->getString();
+ RuntimeMD::KernelArg::Kind Kind;
+ if (TypeQual.find("pipe") != StringRef::npos)
+ Kind = RuntimeMD::KernelArg::Pipe;
+ else Kind = StringSwitch<RuntimeMD::KernelArg::Kind>(BaseTypeName)
.Case("sampler_t", RuntimeMD::KernelArg::Sampler)
.Case("queue_t", RuntimeMD::KernelArg::Queue)
.Cases("image1d_t", "image1d_array_t", "image1d_buffer_t",
@@ -965,32 +1030,30 @@ void AMDGPUAsmPrinter::emitRuntimeMetadata(const Function &F) {
"image2d_msaa_depth_t", RuntimeMD::KernelArg::Image)
.Cases("image2d_array_msaa_depth_t", "image3d_t",
RuntimeMD::KernelArg::Image)
- .Default(isa<PointerType>(T) ? RuntimeMD::KernelArg::Pointer :
- RuntimeMD::KernelArg::Value);
- emitRuntimeMDIntValue(*OutStreamer, RuntimeMD::KeyArgTypeKind, TypeKind, 1);
-
- // Emit KeyArgValueType.
- emitRuntimeMDIntValue(*OutStreamer, RuntimeMD::KeyArgValueType,
- getRuntimeMDValueType(T, BaseTypeName), 2);
-
- // Emit KeyArgAccQual.
- auto AccQual = cast<MDString>(F.getMetadata(
- "kernel_arg_access_qual")->getOperand(I))->getString();
- auto AQ = StringSwitch<RuntimeMD::KernelArg::AccessQualifer>(AccQual)
- .Case("read_only", RuntimeMD::KernelArg::ReadOnly)
- .Case("write_only", RuntimeMD::KernelArg::WriteOnly)
- .Case("read_write", RuntimeMD::KernelArg::ReadWrite)
- .Default(RuntimeMD::KernelArg::None);
- emitRuntimeMDIntValue(*OutStreamer, RuntimeMD::KeyArgAccQual, AQ, 1);
+ .Default(isa<PointerType>(T) ?
+ (T->getPointerAddressSpace() == AMDGPUAS::LOCAL_ADDRESS ?
+ RuntimeMD::KernelArg::DynamicSharedPointer :
+ RuntimeMD::KernelArg::GlobalBuffer) :
+ RuntimeMD::KernelArg::ByValue);
+ emitRuntimeMetadataForKernelArg(DL, *OutStreamer, T,
+ Kind, BaseTypeName, TypeName, ArgName, TypeQual, AccQual);
+ }
- // Emit KeyArgAddrQual.
- if (auto *PT = dyn_cast<PointerType>(T)) {
- emitRuntimeMDIntValue(*OutStreamer, RuntimeMD::KeyArgAddrQual,
- PT->getAddressSpace(), 1);
+ // Emit hidden kernel arguments for OpenCL kernels.
+ if (F.getParent()->getNamedMetadata("opencl.ocl.version")) {
+ auto Int64T = Type::getInt64Ty(F.getContext());
+ emitRuntimeMetadataForKernelArg(DL, *OutStreamer, Int64T,
+ RuntimeMD::KernelArg::HiddenGlobalOffsetX);
+ emitRuntimeMetadataForKernelArg(DL, *OutStreamer, Int64T,
+ RuntimeMD::KernelArg::HiddenGlobalOffsetY);
+ emitRuntimeMetadataForKernelArg(DL, *OutStreamer, Int64T,
+ RuntimeMD::KernelArg::HiddenGlobalOffsetZ);
+ if (auto MD = F.getParent()->getNamedMetadata("llvm.printf.fmts")) {
+ auto Int8PtrT = Type::getInt8PtrTy(F.getContext(),
+ RuntimeMD::KernelArg::Global);
+ emitRuntimeMetadataForKernelArg(DL, *OutStreamer, Int8PtrT,
+ RuntimeMD::KernelArg::HiddenPrintfBuffer);
}
-
- // Emit KeyArgEnd
- OutStreamer->EmitIntValue(RuntimeMD::KeyArgEnd, 1);
}
// Emit KeyReqdWorkGroupSize, KeyWorkGroupSizeHint, and KeyVecTypeHint.
diff --git a/llvm/lib/Target/AMDGPU/AMDGPURuntimeMetadata.h b/llvm/lib/Target/AMDGPU/AMDGPURuntimeMetadata.h
index 9050b442b3b..60a86427418 100644
--- a/llvm/lib/Target/AMDGPU/AMDGPURuntimeMetadata.h
+++ b/llvm/lib/Target/AMDGPU/AMDGPURuntimeMetadata.h
@@ -65,7 +65,7 @@ namespace RuntimeMD {
KeyArgAlign = 10, // Kernel arg alignment
KeyArgTypeName = 11, // Kernel type name
KeyArgName = 12, // Kernel name
- KeyArgTypeKind = 13, // Kernel argument type kind
+ KeyArgKind = 13, // Kernel argument kind
KeyArgValueType = 14, // Kernel argument value type
KeyArgAddrQual = 15, // Kernel argument address qualifier
KeyArgAccQual = 16, // Kernel argument access qualifier
@@ -77,13 +77,13 @@ namespace RuntimeMD {
KeyWorkGroupSizeHint = 22, // Work group size hint
KeyVecTypeHint = 23, // Vector type hint
KeyKernelIndex = 24, // Kernel index for device enqueue
- KeySGPRs = 25, // Number of SGPRs
- KeyVGPRs = 26, // Number of VGPRs
- KeyMinWavesPerSIMD = 27, // Minimum number of waves per SIMD
- KeyMaxWavesPerSIMD = 28, // Maximum number of waves per SIMD
- KeyFlatWorkGroupSizeLimits = 29, // Flat work group size limits
- KeyMaxWorkGroupSize = 30, // Maximum work group size
- KeyNoPartialWorkGroups = 31, // No partial work groups
+ KeyMinWavesPerSIMD = 25, // Minimum number of waves per SIMD
+ KeyMaxWavesPerSIMD = 26, // Maximum number of waves per SIMD
+ KeyFlatWorkGroupSizeLimits = 27, // Flat work group size limits
+ KeyMaxWorkGroupSize = 28, // Maximum work group size
+ KeyNoPartialWorkGroups = 29, // No partial work groups
+ KeyPrintfInfo = 30, // Prinf function call information
+ KeyArgActualAcc = 31, // The actual kernel argument access qualifier
KeyArgPointeeAlign = 32, // Alignment of pointee type
};
@@ -103,12 +103,21 @@ namespace RuntimeMD {
};
namespace KernelArg {
- enum TypeKind : uint8_t {
- Value = 0,
- Pointer = 1,
- Image = 2,
- Sampler = 3,
- Queue = 4,
+ enum Kind : uint8_t {
+ ByValue = 0,
+ GlobalBuffer = 1,
+ DynamicSharedPointer = 2,
+ Sampler = 3,
+ Image = 4,
+ Pipe = 5,
+ Queue = 6,
+ HiddenGlobalOffsetX = 7,
+ HiddenGlobalOffsetY = 8,
+ HiddenGlobalOffsetZ = 9,
+ HiddenNone = 10,
+ HiddenPrintfBuffer = 11,
+ HiddenDefaultQueue = 12,
+ HiddenCompletionAction = 13,
};
enum ValueType : uint16_t {
@@ -138,6 +147,8 @@ namespace RuntimeMD {
Global = 1,
Constant = 2,
Local = 3,
+ Generic = 4,
+ Region = 5,
};
} // namespace KernelArg
} // namespace RuntimeMD
diff --git a/llvm/test/CodeGen/AMDGPU/runtime-metadata.ll b/llvm/test/CodeGen/AMDGPU/runtime-metadata.ll
index 7299a67755e..4fcbf701ecf 100644
--- a/llvm/test/CodeGen/AMDGPU/runtime-metadata.ll
+++ b/llvm/test/CodeGen/AMDGPU/runtime-metadata.ll
@@ -16,6 +16,12 @@
; CHECK-NEXT: .byte 0
; CHECK-NEXT: .byte 3
; CHECK-NEXT: .short 200
+; CHECK-NEXT: .byte 30
+; CHECK-NEXT: .long 10
+; CHECK-NEXT: .ascii "1:1:4:%d\\n"
+; CHECK-NEXT: .byte 30
+; CHECK-NEXT: .long 10
+; CHECK-NEXT: .ascii "2:1:8:%g\\n"
; CHECK-LABEL:{{^}}test_char:
; CHECK: .section .AMDGPU.runtime_metadata
@@ -38,6 +44,48 @@
; CHECK-NEXT: .byte 16
; CHECK-NEXT: .byte 0
; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 9
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 9
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 9
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 11
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 1
+; CHECK-NEXT: .byte 15
+; CHECK-NEXT: .byte 1
+; CHECK-NEXT: .byte 8
; CHECK-NEXT: .byte 5
define amdgpu_kernel void @test_char(i8 %a) !kernel_arg_addr_space !1 !kernel_arg_access_qual !2 !kernel_arg_type !9 !kernel_arg_base_type !9 !kernel_arg_type_qual !4 {
@@ -65,6 +113,48 @@ define amdgpu_kernel void @test_char(i8 %a) !kernel_arg_addr_space !1 !kernel_ar
; CHECK-NEXT: .byte 16
; CHECK-NEXT: .byte 0
; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 9
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 9
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 9
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 11
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 1
+; CHECK-NEXT: .byte 15
+; CHECK-NEXT: .byte 1
+; CHECK-NEXT: .byte 8
; CHECK-NEXT: .byte 5
define amdgpu_kernel void @test_ushort2(<2 x i16> %a) !kernel_arg_addr_space !1 !kernel_arg_access_qual !2 !kernel_arg_type !10 !kernel_arg_base_type !10 !kernel_arg_type_qual !4 {
@@ -92,6 +182,48 @@ define amdgpu_kernel void @test_ushort2(<2 x i16> %a) !kernel_arg_addr_space !1
; CHECK-NEXT: .byte 16
; CHECK-NEXT: .byte 0
; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 9
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 9
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 9
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 11
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 1
+; CHECK-NEXT: .byte 15
+; CHECK-NEXT: .byte 1
+; CHECK-NEXT: .byte 8
; CHECK-NEXT: .byte 5
define amdgpu_kernel void @test_int3(<3 x i32> %a) !kernel_arg_addr_space !1 !kernel_arg_access_qual !2 !kernel_arg_type !11 !kernel_arg_base_type !11 !kernel_arg_type_qual !4 {
@@ -119,6 +251,48 @@ define amdgpu_kernel void @test_int3(<3 x i32> %a) !kernel_arg_addr_space !1 !ke
; CHECK-NEXT: .byte 16
; CHECK-NEXT: .byte 0
; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 9
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 9
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 9
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 11
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 1
+; CHECK-NEXT: .byte 15
+; CHECK-NEXT: .byte 1
+; CHECK-NEXT: .byte 8
; CHECK-NEXT: .byte 5
define amdgpu_kernel void @test_ulong4(<4 x i64> %a) !kernel_arg_addr_space !1 !kernel_arg_access_qual !2 !kernel_arg_type !12 !kernel_arg_base_type !12 !kernel_arg_type_qual !4 {
@@ -146,6 +320,48 @@ define amdgpu_kernel void @test_ulong4(<4 x i64> %a) !kernel_arg_addr_space !1 !
; CHECK-NEXT: .byte 16
; CHECK-NEXT: .byte 0
; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 9
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 9
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 9
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 11
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 1
+; CHECK-NEXT: .byte 15
+; CHECK-NEXT: .byte 1
+; CHECK-NEXT: .byte 8
; CHECK-NEXT: .byte 5
define amdgpu_kernel void @test_half8(<8 x half> %a) !kernel_arg_addr_space !1 !kernel_arg_access_qual !2 !kernel_arg_type !13 !kernel_arg_base_type !13 !kernel_arg_type_qual !4 {
@@ -173,6 +389,48 @@ define amdgpu_kernel void @test_half8(<8 x half> %a) !kernel_arg_addr_space !1 !
; CHECK-NEXT: .byte 16
; CHECK-NEXT: .byte 0
; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 9
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 9
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 9
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 11
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 1
+; CHECK-NEXT: .byte 15
+; CHECK-NEXT: .byte 1
+; CHECK-NEXT: .byte 8
; CHECK-NEXT: .byte 5
define amdgpu_kernel void @test_float16(<16 x float> %a) !kernel_arg_addr_space !1 !kernel_arg_access_qual !2 !kernel_arg_type !14 !kernel_arg_base_type !14 !kernel_arg_type_qual !4 {
@@ -200,6 +458,48 @@ define amdgpu_kernel void @test_float16(<16 x float> %a) !kernel_arg_addr_space
; CHECK-NEXT: .byte 16
; CHECK-NEXT: .byte 0
; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 9
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 9
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 9
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 11
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 1
+; CHECK-NEXT: .byte 15
+; CHECK-NEXT: .byte 1
+; CHECK-NEXT: .byte 8
; CHECK-NEXT: .byte 5
define amdgpu_kernel void @test_double16(<16 x double> %a) !kernel_arg_addr_space !1 !kernel_arg_access_qual !2 !kernel_arg_type !15 !kernel_arg_base_type !15 !kernel_arg_type_qual !4 {
@@ -217,8 +517,6 @@ define amdgpu_kernel void @test_double16(<16 x double> %a) !kernel_arg_addr_spac
; CHECK-NEXT: .long 8
; CHECK-NEXT: .byte 10
; CHECK-NEXT: .long 8
-; CHECK-NEXT: .byte 32
-; CHECK-NEXT: .long 4
; CHECK-NEXT: .byte 11
; CHECK-NEXT: .long 5
; CHECK-NEXT: .ascii "int *"
@@ -231,6 +529,48 @@ define amdgpu_kernel void @test_double16(<16 x double> %a) !kernel_arg_addr_spac
; CHECK-NEXT: .byte 15
; CHECK-NEXT: .byte 1
; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 9
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 9
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 9
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 11
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 1
+; CHECK-NEXT: .byte 15
+; CHECK-NEXT: .byte 1
+; CHECK-NEXT: .byte 8
; CHECK-NEXT: .byte 5
define amdgpu_kernel void @test_pointer(i32 addrspace(1)* %a) !kernel_arg_addr_space !1 !kernel_arg_access_qual !2 !kernel_arg_type !16 !kernel_arg_base_type !16 !kernel_arg_type_qual !4 {
@@ -252,7 +592,7 @@ define amdgpu_kernel void @test_pointer(i32 addrspace(1)* %a) !kernel_arg_addr_s
; CHECK-NEXT: .long 9
; CHECK-NEXT: .ascii "image2d_t"
; CHECK-NEXT: .byte 13
-; CHECK-NEXT: .byte 2
+; CHECK-NEXT: .byte 4
; CHECK-NEXT: .byte 14
; CHECK-NEXT: .short 0
; CHECK-NEXT: .byte 16
@@ -260,6 +600,48 @@ define amdgpu_kernel void @test_pointer(i32 addrspace(1)* %a) !kernel_arg_addr_s
; CHECK-NEXT: .byte 15
; CHECK-NEXT: .byte 1
; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 9
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 9
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 9
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 11
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 1
+; CHECK-NEXT: .byte 15
+; CHECK-NEXT: .byte 1
+; CHECK-NEXT: .byte 8
; CHECK-NEXT: .byte 5
define amdgpu_kernel void @test_image(%opencl.image2d_t addrspace(1)* %a) !kernel_arg_addr_space !1 !kernel_arg_access_qual !2 !kernel_arg_type !17 !kernel_arg_base_type !17 !kernel_arg_type_qual !4 {
@@ -287,6 +669,48 @@ define amdgpu_kernel void @test_image(%opencl.image2d_t addrspace(1)* %a) !kerne
; CHECK-NEXT: .byte 16
; CHECK-NEXT: .byte 0
; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 9
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 9
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 9
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 11
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 1
+; CHECK-NEXT: .byte 15
+; CHECK-NEXT: .byte 1
+; CHECK-NEXT: .byte 8
; CHECK-NEXT: .byte 5
define amdgpu_kernel void @test_sampler(i32 %a) !kernel_arg_addr_space !1 !kernel_arg_access_qual !2 !kernel_arg_type !18 !kernel_arg_base_type !18 !kernel_arg_type_qual !4 {
@@ -308,7 +732,7 @@ define amdgpu_kernel void @test_sampler(i32 %a) !kernel_arg_addr_space !1 !kerne
; CHECK-NEXT: .long 7
; CHECK-NEXT: .ascii "queue_t"
; CHECK-NEXT: .byte 13
-; CHECK-NEXT: .byte 4
+; CHECK-NEXT: .byte 6
; CHECK-NEXT: .byte 14
; CHECK-NEXT: .short 0
; CHECK-NEXT: .byte 16
@@ -316,6 +740,48 @@ define amdgpu_kernel void @test_sampler(i32 %a) !kernel_arg_addr_space !1 !kerne
; CHECK-NEXT: .byte 15
; CHECK-NEXT: .byte 1
; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 9
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 9
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 9
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 11
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 1
+; CHECK-NEXT: .byte 15
+; CHECK-NEXT: .byte 1
+; CHECK-NEXT: .byte 8
; CHECK-NEXT: .byte 5
define amdgpu_kernel void @test_queue(%opencl.queue_t addrspace(1)* %a) !kernel_arg_addr_space !1 !kernel_arg_access_qual !2 !kernel_arg_type !19 !kernel_arg_base_type !19 !kernel_arg_type_qual !4 {
@@ -333,8 +799,6 @@ define amdgpu_kernel void @test_queue(%opencl.queue_t addrspace(1)* %a) !kernel_
; CHECK-NEXT: .long 4
; CHECK-NEXT: .byte 10
; CHECK-NEXT: .long 4
-; CHECK-NEXT: .byte 32
-; CHECK-NEXT: .long 4
; CHECK-NEXT: .byte 11
; CHECK-NEXT: .long 8
; CHECK-NEXT: .ascii "struct A"
@@ -347,6 +811,48 @@ define amdgpu_kernel void @test_queue(%opencl.queue_t addrspace(1)* %a) !kernel_
; CHECK-NEXT: .byte 15
; CHECK-NEXT: .byte 0
; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 9
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 9
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 9
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 11
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 1
+; CHECK-NEXT: .byte 15
+; CHECK-NEXT: .byte 1
+; CHECK-NEXT: .byte 8
; CHECK-NEXT: .byte 5
define amdgpu_kernel void @test_struct(%struct.A* byval %a) !kernel_arg_addr_space !1 !kernel_arg_access_qual !2 !kernel_arg_type !20 !kernel_arg_base_type !20 !kernel_arg_type_qual !4 {
@@ -374,6 +880,48 @@ define amdgpu_kernel void @test_struct(%struct.A* byval %a) !kernel_arg_addr_spa
; CHECK-NEXT: .byte 16
; CHECK-NEXT: .byte 0
; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 9
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 9
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 9
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 11
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 1
+; CHECK-NEXT: .byte 15
+; CHECK-NEXT: .byte 1
+; CHECK-NEXT: .byte 8
; CHECK-NEXT: .byte 5
define amdgpu_kernel void @test_i128(i128 %a) !kernel_arg_addr_space !1 !kernel_arg_access_qual !2 !kernel_arg_type !21 !kernel_arg_base_type !21 !kernel_arg_type_qual !4 {
@@ -431,6 +979,48 @@ define amdgpu_kernel void @test_i128(i128 %a) !kernel_arg_addr_space !1 !kernel_
; CHECK-NEXT: .byte 16
; CHECK-NEXT: .byte 0
; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 9
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 9
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 9
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 11
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 1
+; CHECK-NEXT: .byte 15
+; CHECK-NEXT: .byte 1
+; CHECK-NEXT: .byte 8
; CHECK-NEXT: .byte 5
define amdgpu_kernel void @test_multi_arg(i32 %a, <2 x i16> %b, <3 x i8> %c) !kernel_arg_addr_space !22 !kernel_arg_access_qual !23 !kernel_arg_type !24 !kernel_arg_base_type !24 !kernel_arg_type_qual !25 {
@@ -448,8 +1038,6 @@ define amdgpu_kernel void @test_multi_arg(i32 %a, <2 x i16> %b, <3 x i8> %c) !ke
; CHECK-NEXT: .long 8
; CHECK-NEXT: .byte 10
; CHECK-NEXT: .long 8
-; CHECK-NEXT: .byte 32
-; CHECK-NEXT: .long 4
; CHECK-NEXT: .byte 11
; CHECK-NEXT: .long 5
; CHECK-NEXT: .ascii "int *"
@@ -467,8 +1055,6 @@ define amdgpu_kernel void @test_multi_arg(i32 %a, <2 x i16> %b, <3 x i8> %c) !ke
; CHECK-NEXT: .long 8
; CHECK-NEXT: .byte 10
; CHECK-NEXT: .long 8
-; CHECK-NEXT: .byte 32
-; CHECK-NEXT: .long 4
; CHECK-NEXT: .byte 11
; CHECK-NEXT: .long 5
; CHECK-NEXT: .ascii "int *"
@@ -492,7 +1078,7 @@ define amdgpu_kernel void @test_multi_arg(i32 %a, <2 x i16> %b, <3 x i8> %c) !ke
; CHECK-NEXT: .long 5
; CHECK-NEXT: .ascii "int *"
; CHECK-NEXT: .byte 13
-; CHECK-NEXT: .byte 1
+; CHECK-NEXT: .byte 2
; CHECK-NEXT: .byte 14
; CHECK-NEXT: .short 6
; CHECK-NEXT: .byte 16
@@ -500,6 +1086,48 @@ define amdgpu_kernel void @test_multi_arg(i32 %a, <2 x i16> %b, <3 x i8> %c) !ke
; CHECK-NEXT: .byte 15
; CHECK-NEXT: .byte 3
; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 9
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 9
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 9
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 11
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 1
+; CHECK-NEXT: .byte 15
+; CHECK-NEXT: .byte 1
+; CHECK-NEXT: .byte 8
; CHECK-NEXT: .byte 5
define amdgpu_kernel void @test_addr_space(i32 addrspace(1)* %g, i32 addrspace(2)* %c, i32 addrspace(3)* %l) !kernel_arg_addr_space !50 !kernel_arg_access_qual !23 !kernel_arg_type !51 !kernel_arg_base_type !51 !kernel_arg_type_qual !25 {
@@ -517,8 +1145,6 @@ define amdgpu_kernel void @test_addr_space(i32 addrspace(1)* %g, i32 addrspace(2
; CHECK-NEXT: .long 8
; CHECK-NEXT: .byte 10
; CHECK-NEXT: .long 8
-; CHECK-NEXT: .byte 32
-; CHECK-NEXT: .long 4
; CHECK-NEXT: .byte 11
; CHECK-NEXT: .long 5
; CHECK-NEXT: .ascii "int *"
@@ -537,8 +1163,6 @@ define amdgpu_kernel void @test_addr_space(i32 addrspace(1)* %g, i32 addrspace(2
; CHECK-NEXT: .long 8
; CHECK-NEXT: .byte 10
; CHECK-NEXT: .long 8
-; CHECK-NEXT: .byte 32
-; CHECK-NEXT: .long 4
; CHECK-NEXT: .byte 11
; CHECK-NEXT: .long 5
; CHECK-NEXT: .ascii "int *"
@@ -563,7 +1187,7 @@ define amdgpu_kernel void @test_addr_space(i32 addrspace(1)* %g, i32 addrspace(2
; CHECK-NEXT: .ascii "int *"
; CHECK-NEXT: .byte 20
; CHECK-NEXT: .byte 13
-; CHECK-NEXT: .byte 1
+; CHECK-NEXT: .byte 5
; CHECK-NEXT: .byte 14
; CHECK-NEXT: .short 0
; CHECK-NEXT: .byte 16
@@ -571,6 +1195,48 @@ define amdgpu_kernel void @test_addr_space(i32 addrspace(1)* %g, i32 addrspace(2
; CHECK-NEXT: .byte 15
; CHECK-NEXT: .byte 1
; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 9
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 9
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 9
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 11
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 1
+; CHECK-NEXT: .byte 15
+; CHECK-NEXT: .byte 1
+; CHECK-NEXT: .byte 8
; CHECK-NEXT: .byte 5
define amdgpu_kernel void @test_type_qual(i32 addrspace(1)* %a, i32 addrspace(1)* %b, %opencl.pipe_t addrspace(1)* %c) !kernel_arg_addr_space !22 !kernel_arg_access_qual !23 !kernel_arg_type !51 !kernel_arg_base_type !51 !kernel_arg_type_qual !70 {
@@ -592,7 +1258,7 @@ define amdgpu_kernel void @test_type_qual(i32 addrspace(1)* %a, i32 addrspace(1)
; CHECK-NEXT: .long 9
; CHECK-NEXT: .ascii "image1d_t"
; CHECK-NEXT: .byte 13
-; CHECK-NEXT: .byte 2
+; CHECK-NEXT: .byte 4
; CHECK-NEXT: .byte 14
; CHECK-NEXT: .short 0
; CHECK-NEXT: .byte 16
@@ -609,7 +1275,7 @@ define amdgpu_kernel void @test_type_qual(i32 addrspace(1)* %a, i32 addrspace(1)
; CHECK-NEXT: .long 9
; CHECK-NEXT: .ascii "image2d_t"
; CHECK-NEXT: .byte 13
-; CHECK-NEXT: .byte 2
+; CHECK-NEXT: .byte 4
; CHECK-NEXT: .byte 14
; CHECK-NEXT: .short 0
; CHECK-NEXT: .byte 16
@@ -626,7 +1292,7 @@ define amdgpu_kernel void @test_type_qual(i32 addrspace(1)* %a, i32 addrspace(1)
; CHECK-NEXT: .long 9
; CHECK-NEXT: .ascii "image3d_t"
; CHECK-NEXT: .byte 13
-; CHECK-NEXT: .byte 2
+; CHECK-NEXT: .byte 4
; CHECK-NEXT: .byte 14
; CHECK-NEXT: .short 0
; CHECK-NEXT: .byte 16
@@ -634,6 +1300,48 @@ define amdgpu_kernel void @test_type_qual(i32 addrspace(1)* %a, i32 addrspace(1)
; CHECK-NEXT: .byte 15
; CHECK-NEXT: .byte 1
; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 9
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 9
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 9
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 11
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 1
+; CHECK-NEXT: .byte 15
+; CHECK-NEXT: .byte 1
+; CHECK-NEXT: .byte 8
; CHECK-NEXT: .byte 5
define amdgpu_kernel void @test_access_qual(%opencl.image1d_t addrspace(1)* %ro, %opencl.image2d_t addrspace(1)* %wo, %opencl.image3d_t addrspace(1)* %rw) !kernel_arg_addr_space !60 !kernel_arg_access_qual !61 !kernel_arg_type !62 !kernel_arg_base_type !62 !kernel_arg_type_qual !25 {
@@ -661,6 +1369,48 @@ define amdgpu_kernel void @test_access_qual(%opencl.image1d_t addrspace(1)* %ro,
; CHECK-NEXT: .byte 16
; CHECK-NEXT: .byte 0
; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 9
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 9
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 9
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 11
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 1
+; CHECK-NEXT: .byte 15
+; CHECK-NEXT: .byte 1
+; CHECK-NEXT: .byte 8
; CHECK-NEXT: .byte 23
; CHECK-NEXT: .long 4
; CHECK-NEXT: .ascii "half"
@@ -691,6 +1441,48 @@ define amdgpu_kernel void @test_vec_type_hint_half(i32 %a) !kernel_arg_addr_spac
; CHECK-NEXT: .byte 16
; CHECK-NEXT: .byte 0
; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 9
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 9
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 9
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 11
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 1
+; CHECK-NEXT: .byte 15
+; CHECK-NEXT: .byte 1
+; CHECK-NEXT: .byte 8
; CHECK-NEXT: .byte 23
; CHECK-NEXT: .long 5
; CHECK-NEXT: .ascii "float"
@@ -721,6 +1513,48 @@ define amdgpu_kernel void @test_vec_type_hint_float(i32 %a) !kernel_arg_addr_spa
; CHECK-NEXT: .byte 16
; CHECK-NEXT: .byte 0
; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 9
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 9
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 9
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 11
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 1
+; CHECK-NEXT: .byte 15
+; CHECK-NEXT: .byte 1
+; CHECK-NEXT: .byte 8
; CHECK-NEXT: .byte 23
; CHECK-NEXT: .long 6
; CHECK-NEXT: .ascii "double"
@@ -751,6 +1585,48 @@ define amdgpu_kernel void @test_vec_type_hint_double(i32 %a) !kernel_arg_addr_sp
; CHECK-NEXT: .byte 16
; CHECK-NEXT: .byte 0
; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 9
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 9
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 9
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 11
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 1
+; CHECK-NEXT: .byte 15
+; CHECK-NEXT: .byte 1
+; CHECK-NEXT: .byte 8
; CHECK-NEXT: .byte 23
; CHECK-NEXT: .long 4
; CHECK-NEXT: .ascii "char"
@@ -781,6 +1657,48 @@ define amdgpu_kernel void @test_vec_type_hint_char(i32 %a) !kernel_arg_addr_spac
; CHECK-NEXT: .byte 16
; CHECK-NEXT: .byte 0
; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 9
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 9
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 9
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 11
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 1
+; CHECK-NEXT: .byte 15
+; CHECK-NEXT: .byte 1
+; CHECK-NEXT: .byte 8
; CHECK-NEXT: .byte 23
; CHECK-NEXT: .long 5
; CHECK-NEXT: .ascii "short"
@@ -811,6 +1729,48 @@ define amdgpu_kernel void @test_vec_type_hint_short(i32 %a) !kernel_arg_addr_spa
; CHECK-NEXT: .byte 16
; CHECK-NEXT: .byte 0
; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 9
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 9
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 9
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 11
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 1
+; CHECK-NEXT: .byte 15
+; CHECK-NEXT: .byte 1
+; CHECK-NEXT: .byte 8
; CHECK-NEXT: .byte 23
; CHECK-NEXT: .long 4
; CHECK-NEXT: .ascii "long"
@@ -841,6 +1801,48 @@ define amdgpu_kernel void @test_vec_type_hint_long(i32 %a) !kernel_arg_addr_spac
; CHECK-NEXT: .byte 16
; CHECK-NEXT: .byte 0
; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 9
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 9
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 9
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 11
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 1
+; CHECK-NEXT: .byte 15
+; CHECK-NEXT: .byte 1
+; CHECK-NEXT: .byte 8
; CHECK-NEXT: .byte 23
; CHECK-NEXT: .long 7
; CHECK-NEXT: .ascii "unknown"
@@ -871,6 +1873,48 @@ define amdgpu_kernel void @test_vec_type_hint_unknown(i32 %a) !kernel_arg_addr_s
; CHECK-NEXT: .byte 16
; CHECK-NEXT: .byte 0
; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 9
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 9
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 9
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 11
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 1
+; CHECK-NEXT: .byte 15
+; CHECK-NEXT: .byte 1
+; CHECK-NEXT: .byte 8
; CHECK-NEXT: .byte 21
; CHECK-NEXT: .long 1
; CHECK-NEXT: .long 2
@@ -905,6 +1949,48 @@ define amdgpu_kernel void @test_reqd_wgs_vec_type_hint(i32 %a) !kernel_arg_addr_
; CHECK-NEXT: .byte 16
; CHECK-NEXT: .byte 0
; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 9
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 9
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 9
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 11
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 1
+; CHECK-NEXT: .byte 15
+; CHECK-NEXT: .byte 1
+; CHECK-NEXT: .byte 8
; CHECK-NEXT: .byte 22
; CHECK-NEXT: .long 8
; CHECK-NEXT: .long 16
@@ -929,8 +2015,6 @@ define amdgpu_kernel void @test_wgs_hint_vec_type_hint(i32 %a) !kernel_arg_addr_
; CHECK-NEXT: .long 8
; CHECK-NEXT: .byte 10
; CHECK-NEXT: .long 8
-; CHECK-NEXT: .byte 32
-; CHECK-NEXT: .long 4
; CHECK-NEXT: .byte 11
; CHECK-NEXT: .long 6
; CHECK-NEXT: .ascii "int **"
@@ -943,6 +2027,48 @@ define amdgpu_kernel void @test_wgs_hint_vec_type_hint(i32 %a) !kernel_arg_addr_
; CHECK-NEXT: .byte 15
; CHECK-NEXT: .byte 1
; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 9
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 9
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 9
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 11
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 1
+; CHECK-NEXT: .byte 15
+; CHECK-NEXT: .byte 1
+; CHECK-NEXT: .byte 8
; CHECK-NEXT: .byte 5
define amdgpu_kernel void @test_arg_ptr_to_ptr(i32 * addrspace(1)* %a) !kernel_arg_addr_space !81 !kernel_arg_access_qual !2 !kernel_arg_type !80 !kernel_arg_base_type !80 !kernel_arg_type_qual !4 {
@@ -960,8 +2086,6 @@ define amdgpu_kernel void @test_arg_ptr_to_ptr(i32 * addrspace(1)* %a) !kernel_a
; CHECK-NEXT: .long 4
; CHECK-NEXT: .byte 10
; CHECK-NEXT: .long 4
-; CHECK-NEXT: .byte 32
-; CHECK-NEXT: .long 8
; CHECK-NEXT: .byte 11
; CHECK-NEXT: .long 8
; CHECK-NEXT: .ascii "struct B"
@@ -974,6 +2098,48 @@ define amdgpu_kernel void @test_arg_ptr_to_ptr(i32 * addrspace(1)* %a) !kernel_a
; CHECK-NEXT: .byte 15
; CHECK-NEXT: .byte 0
; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 9
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 9
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 9
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 11
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 1
+; CHECK-NEXT: .byte 15
+; CHECK-NEXT: .byte 1
+; CHECK-NEXT: .byte 8
; CHECK-NEXT: .byte 5
define amdgpu_kernel void @test_arg_struct_contains_ptr(%struct.B * byval %a) !kernel_arg_addr_space !1 !kernel_arg_access_qual !2 !kernel_arg_type !82 !kernel_arg_base_type !82 !kernel_arg_type_qual !4 {
@@ -1001,6 +2167,48 @@ define amdgpu_kernel void @test_arg_struct_contains_ptr(%struct.B * byval %a) !k
; CHECK-NEXT: .byte 16
; CHECK-NEXT: .byte 0
; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 9
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 9
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 9
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 11
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 1
+; CHECK-NEXT: .byte 15
+; CHECK-NEXT: .byte 1
+; CHECK-NEXT: .byte 8
; CHECK-NEXT: .byte 5
define amdgpu_kernel void @test_arg_vector_of_ptr(<2 x i32 addrspace(1)*> %a) !kernel_arg_addr_space !1 !kernel_arg_access_qual !2 !kernel_arg_type !83 !kernel_arg_base_type !83 !kernel_arg_type_qual !4 {
@@ -1030,8 +2238,51 @@ define amdgpu_kernel void @test_arg_vector_of_ptr(<2 x i32 addrspace(1)*> %a) !k
; CHECK-NEXT: .byte 15
; CHECK-NEXT: .byte 1
; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 9
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 9
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 9
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 11
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 1
+; CHECK-NEXT: .byte 15
+; CHECK-NEXT: .byte 1
+; CHECK-NEXT: .byte 8
; CHECK-NEXT: .byte 5
+
define amdgpu_kernel void @test_arg_unknown_builtin_type(%opencl.clk_event_t addrspace(1)* %a) !kernel_arg_addr_space !81 !kernel_arg_access_qual !2 !kernel_arg_type !84 !kernel_arg_base_type !84 !kernel_arg_type_qual !4 {
ret void
}
@@ -1047,8 +2298,6 @@ define amdgpu_kernel void @test_arg_unknown_builtin_type(%opencl.clk_event_t add
; CHECK-NEXT: .long 8
; CHECK-NEXT: .byte 10
; CHECK-NEXT: .long 8
-; CHECK-NEXT: .byte 32
-; CHECK-NEXT: .long 8
; CHECK-NEXT: .byte 11
; CHECK-NEXT: .long 6
; CHECK-NEXT: .ascii "long *"
@@ -1072,7 +2321,7 @@ define amdgpu_kernel void @test_arg_unknown_builtin_type(%opencl.clk_event_t add
; CHECK-NEXT: .long 6
; CHECK-NEXT: .ascii "char *"
; CHECK-NEXT: .byte 13
-; CHECK-NEXT: .byte 1
+; CHECK-NEXT: .byte 2
; CHECK-NEXT: .byte 14
; CHECK-NEXT: .short 1
; CHECK-NEXT: .byte 16
@@ -1091,7 +2340,7 @@ define amdgpu_kernel void @test_arg_unknown_builtin_type(%opencl.clk_event_t add
; CHECK-NEXT: .long 7
; CHECK-NEXT: .ascii "char2 *"
; CHECK-NEXT: .byte 13
-; CHECK-NEXT: .byte 1
+; CHECK-NEXT: .byte 2
; CHECK-NEXT: .byte 14
; CHECK-NEXT: .short 1
; CHECK-NEXT: .byte 16
@@ -1110,7 +2359,7 @@ define amdgpu_kernel void @test_arg_unknown_builtin_type(%opencl.clk_event_t add
; CHECK-NEXT: .long 7
; CHECK-NEXT: .ascii "char3 *"
; CHECK-NEXT: .byte 13
-; CHECK-NEXT: .byte 1
+; CHECK-NEXT: .byte 2
; CHECK-NEXT: .byte 14
; CHECK-NEXT: .short 1
; CHECK-NEXT: .byte 16
@@ -1129,7 +2378,7 @@ define amdgpu_kernel void @test_arg_unknown_builtin_type(%opencl.clk_event_t add
; CHECK-NEXT: .long 7
; CHECK-NEXT: .ascii "char4 *"
; CHECK-NEXT: .byte 13
-; CHECK-NEXT: .byte 1
+; CHECK-NEXT: .byte 2
; CHECK-NEXT: .byte 14
; CHECK-NEXT: .short 1
; CHECK-NEXT: .byte 16
@@ -1148,7 +2397,7 @@ define amdgpu_kernel void @test_arg_unknown_builtin_type(%opencl.clk_event_t add
; CHECK-NEXT: .long 7
; CHECK-NEXT: .ascii "char8 *"
; CHECK-NEXT: .byte 13
-; CHECK-NEXT: .byte 1
+; CHECK-NEXT: .byte 2
; CHECK-NEXT: .byte 14
; CHECK-NEXT: .short 1
; CHECK-NEXT: .byte 16
@@ -1167,7 +2416,7 @@ define amdgpu_kernel void @test_arg_unknown_builtin_type(%opencl.clk_event_t add
; CHECK-NEXT: .long 8
; CHECK-NEXT: .ascii "char16 *"
; CHECK-NEXT: .byte 13
-; CHECK-NEXT: .byte 1
+; CHECK-NEXT: .byte 2
; CHECK-NEXT: .byte 14
; CHECK-NEXT: .short 1
; CHECK-NEXT: .byte 16
@@ -1175,12 +2424,56 @@ define amdgpu_kernel void @test_arg_unknown_builtin_type(%opencl.clk_event_t add
; CHECK-NEXT: .byte 15
; CHECK-NEXT: .byte 3
; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 9
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 9
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 9
+; CHECK-NEXT: .byte 8
+; CHECK-NEXT: .byte 7
+; CHECK-NEXT: .byte 9
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 10
+; CHECK-NEXT: .long 8
+; CHECK-NEXT: .byte 13
+; CHECK-NEXT: .byte 11
+; CHECK-NEXT: .byte 14
+; CHECK-NEXT: .short 1
+; CHECK-NEXT: .byte 15
+; CHECK-NEXT: .byte 1
+; CHECK-NEXT: .byte 8
; CHECK-NEXT: .byte 5
define amdgpu_kernel void @test_pointee_align(i64 addrspace(1)* %a, i8 addrspace(3)* %b, <2 x i8> addrspace(3)* %c, <3 x i8> addrspace(3)* %d, <4 x i8> addrspace(3)* %e, <8 x i8> addrspace(3)* %f, <16 x i8> addrspace(3)* %g) !kernel_arg_addr_space !91 !kernel_arg_access_qual !92 !kernel_arg_type !93 !kernel_arg_base_type !93 !kernel_arg_type_qual !94 {
ret void
}
+!llvm.printf.fmts = !{!100, !101}
+
!1 = !{i32 0}
!2 = !{!"none"}
!3 = !{!"int"}
@@ -1230,3 +2523,5 @@ define amdgpu_kernel void @test_pointee_align(i64 addrspace(1)* %a, i8 addrspace
!92 = !{!"none", !"none", !"none", !"none", !"none", !"none", !"none"}
!93 = !{!"long *", !"char *", !"char2 *", !"char3 *", !"char4 *", !"char8 *", !"char16 *"}
!94 = !{!"", !"", !"", !"", !"", !"", !""}
+!100 = !{!"1:1:4:%d\5Cn"}
+!101 = !{!"2:1:8:%g\5Cn"}
OpenPOWER on IntegriCloud