[llvm] r336697 - AMDGPU: Make hidden argument metadata consistent with
via llvm-commits
llvm-commits at lists.llvm.org
Tue Jul 10 13:16:49 PDT 2018
This should be fixed in https://reviews.llvm.org/rL336722
Scott
On 2018-07-10 11:41, Konstantin Zhuravlyov wrote:
> +Scott
> -------------------------
>
> FROM: Zaara Syeda <syzaara at ca.ibm.com>
> SENT: Tuesday, July 10, 2018 2:40 PM
> TO: Zaara Syeda
> CC: kzhuravl_dev at outlook.com; llvm-commits at lists.llvm.org
> SUBJECT: Re: [llvm] r336697 - AMDGPU: Make hidden argument metadata
> consistent with
>
> Opps sorry, I meant to reply regarding commit
> https://reviews.llvm.org/rL336707 [11].
>
>> ----- Original message -----
>> From: Zaara Syeda via llvm-commits <llvm-commits at lists.llvm.org>
>> Sent by: "llvm-commits" <llvm-commits-bounces at lists.llvm.org>
>> To: kzhuravl_dev at outlook.com
>> Cc: llvm-commits at lists.llvm.org
>> Subject: Re: [llvm] r336697 - AMDGPU: Make hidden argument metadata
>> consistent with
>> Date: Tue, Jul 10, 2018 2:11 PM
>>
>> I believe this patch is causing failure on:
>>
> http://lab.llvm.org:8011/builders/clang-ppc64le-linux-multistage/builds/7287
>> [9]
>>
>>
> http://lab.llvm.org:8011/builders/clang-ppc64le-linux-multistage/builds/7287/steps/build%20stage%201/logs/stdio
>> [10]
>>
>>> ----- Original message -----
>>> From: Konstantin Zhuravlyov via llvm-commits
>>> <llvm-commits at lists.llvm.org>
>>> Sent by: "llvm-commits" <llvm-commits-bounces at lists.llvm.org>
>>> To: llvm-commits at lists.llvm.org
>>> Cc:
>>> Subject: [llvm] r336697 - AMDGPU: Make hidden argument metadata
>>> consistent with
>>> Date: Tue, Jul 10, 2018 12:17 PM
>>>
>>> Author: kzhuravl
>>> Date: Tue Jul 10 09:12:51 2018
>>> New Revision: 336697
>>>
>>> URL: http://llvm.org/viewvc/llvm-project?rev=336697&view=rev [1]
>>> Log:
>>> AMDGPU: Make hidden argument metadata consistent with
>>> amdgpu-implicitarg-num-bytes attribute
>>>
>>> Differential Revision: https://reviews.llvm.org/D49096 [2]
>>>
>>> Modified:
>>>
>>
> llvm/trunk/lib/Target/AMDGPU/MCTargetDesc/AMDGPUHSAMetadataStreamer.cpp
>>>
>>
> llvm/trunk/lib/Target/AMDGPU/MCTargetDesc/AMDGPUHSAMetadataStreamer.h
>>> llvm/trunk/test/CodeGen/AMDGPU/hsa-metadata-enqueue-kernel.ll
>>> llvm/trunk/test/CodeGen/AMDGPU/hsa-metadata-from-llvm-ir-full.ll
>>> llvm/trunk/test/CodeGen/AMDGPU/hsa-metadata-hidden-args.ll
>>>
>>> Modified:
>>>
>>
> llvm/trunk/lib/Target/AMDGPU/MCTargetDesc/AMDGPUHSAMetadataStreamer.cpp
>>> URL:
>>>
>>
> http://llvm.org/viewvc/llvm-project/llvm/trunk/lib/Target/AMDGPU/MCTargetDesc/AMDGPUHSAMetadataStreamer.cpp?rev=336697&r1=336696&r2=336697&view=diff
>>> [3]
>>>
>>
> ==============================================================================
>>> ---
>>>
>>
> llvm/trunk/lib/Target/AMDGPU/MCTargetDesc/AMDGPUHSAMetadataStreamer.cpp
>>> (original)
>>> +++
>>>
>>
> llvm/trunk/lib/Target/AMDGPU/MCTargetDesc/AMDGPUHSAMetadataStreamer.cpp
>>> Tue Jul 10 09:12:51 2018
>>> @@ -14,7 +14,7 @@
>>>
>>
> //===----------------------------------------------------------------------===//
>>>
>>> #include "AMDGPUHSAMetadataStreamer.h"
>>> -#include "AMDGPU.h"
>>> +#include "Utils/AMDGPUBaseInfo.h"
>>> #include "llvm/ADT/StringSwitch.h"
>>> #include "llvm/IR/Constants.h"
>>> #include "llvm/IR/Module.h"
>>> @@ -255,36 +255,7 @@ void MetadataStreamer::emitKernelArgs(co
>>> for (auto &Arg : Func.args())
>>> emitKernelArg(Arg);
>>>
>>> - // TODO: What about other languages?
>>> - if (!Func.getParent()->getNamedMetadata("opencl.ocl.version"))
>>> - return;
>>> -
>>> - auto &DL = Func.getParent()->getDataLayout();
>>> - auto Int64Ty = Type::getInt64Ty(Func.getContext());
>>> -
>>> - emitKernelArg(DL, Int64Ty, ValueKind::HiddenGlobalOffsetX);
>>> - emitKernelArg(DL, Int64Ty, ValueKind::HiddenGlobalOffsetY);
>>> - emitKernelArg(DL, Int64Ty, ValueKind::HiddenGlobalOffsetZ);
>>> -
>>> - auto Int8PtrTy = Type::getInt8PtrTy(Func.getContext(),
>>> - AMDGPUASI.GLOBAL_ADDRESS);
>>> -
>>> - // Emit "printf buffer" argument if printf is used, otherwise
>>> emit dummy
>>> - // "none" argument.
>>> - if (Func.getParent()->getNamedMetadata("llvm.printf.fmts"))
>>> - emitKernelArg(DL, Int8PtrTy, ValueKind::HiddenPrintfBuffer);
>>> - else
>>> - emitKernelArg(DL, Int8PtrTy, ValueKind::HiddenNone);
>>> -
>>> - // Emit "default queue" and "completion action" arguments if
>>> enqueue kernel is
>>> - // used, otherwise emit dummy "none" arguments.
>>> - if (Func.hasFnAttribute("calls-enqueue-kernel")) {
>>> - emitKernelArg(DL, Int8PtrTy, ValueKind::HiddenDefaultQueue);
>>> - emitKernelArg(DL, Int8PtrTy, ValueKind::HiddenCompletionAction);
>>> - } else {
>>> - emitKernelArg(DL, Int8PtrTy, ValueKind::HiddenNone);
>>> - emitKernelArg(DL, Int8PtrTy, ValueKind::HiddenNone);
>>> - }
>>> + emitHiddenKernelArgs(Func);
>>> }
>>>
>>> void MetadataStreamer::emitKernelArg(const Argument &Arg) {
>>> @@ -378,6 +349,48 @@ void MetadataStreamer::emitKernelArg(con
>>> }
>>> }
>>>
>>> +void MetadataStreamer::emitHiddenKernelArgs(const Function &Func)
>>> {
>>> + int HiddenArgNumBytes =
>>> + getIntegerAttribute(Func, "amdgpu-implicitarg-num-bytes", 0);
>>> +
>>> + if (!HiddenArgNumBytes)
>>> + return;
>>> +
>>> + auto &DL = Func.getParent()->getDataLayout();
>>> + auto Int64Ty = Type::getInt64Ty(Func.getContext());
>>> +
>>> + if (HiddenArgNumBytes >= 8)
>>> + emitKernelArg(DL, Int64Ty, ValueKind::HiddenGlobalOffsetX);
>>> + if (HiddenArgNumBytes >= 16)
>>> + emitKernelArg(DL, Int64Ty, ValueKind::HiddenGlobalOffsetY);
>>> + if (HiddenArgNumBytes >= 24)
>>> + emitKernelArg(DL, Int64Ty, ValueKind::HiddenGlobalOffsetZ);
>>> +
>>> + auto Int8PtrTy = Type::getInt8PtrTy(Func.getContext(),
>>> + AMDGPUASI.GLOBAL_ADDRESS);
>>> +
>>> + // Emit "printf buffer" argument if printf is used, otherwise
>>> emit dummy
>>> + // "none" argument.
>>> + if (HiddenArgNumBytes >= 32) {
>>> + if (Func.getParent()->getNamedMetadata("llvm.printf.fmts"))
>>> + emitKernelArg(DL, Int8PtrTy, ValueKind::HiddenPrintfBuffer);
>>> + else
>>> + emitKernelArg(DL, Int8PtrTy, ValueKind::HiddenNone);
>>> + }
>>> +
>>> + // Emit "default queue" and "completion action" arguments if
>>> enqueue kernel is
>>> + // used, otherwise emit dummy "none" arguments.
>>> + if (HiddenArgNumBytes >= 48) {
>>> + if (Func.hasFnAttribute("calls-enqueue-kernel")) {
>>> + emitKernelArg(DL, Int8PtrTy, ValueKind::HiddenDefaultQueue);
>>> + emitKernelArg(DL, Int8PtrTy, ValueKind::HiddenCompletionAction);
>>> + } else {
>>> + emitKernelArg(DL, Int8PtrTy, ValueKind::HiddenNone);
>>> + emitKernelArg(DL, Int8PtrTy, ValueKind::HiddenNone);
>>> + }
>>> + }
>>> +}
>>> +
>>> void MetadataStreamer::begin(const Module &Mod) {
>>> AMDGPUASI = getAMDGPUAS(Mod);
>>> emitVersion();
>>>
>>> Modified:
>>>
>>
> llvm/trunk/lib/Target/AMDGPU/MCTargetDesc/AMDGPUHSAMetadataStreamer.h
>>> URL:
>>>
>>
> http://llvm.org/viewvc/llvm-project/llvm/trunk/lib/Target/AMDGPU/MCTargetDesc/AMDGPUHSAMetadataStreamer.h?rev=336697&r1=336696&r2=336697&view=diff
>>> [4]
>>>
>>
> ==============================================================================
>>> ---
>>>
>>
> llvm/trunk/lib/Target/AMDGPU/MCTargetDesc/AMDGPUHSAMetadataStreamer.h
>>> (original)
>>> +++
>>>
>>
> llvm/trunk/lib/Target/AMDGPU/MCTargetDesc/AMDGPUHSAMetadataStreamer.h
>>> Tue Jul 10 09:12:51 2018
>>> @@ -73,6 +73,8 @@ private:
>>> StringRef BaseTypeName = "", StringRef AccQual = "",
>>> StringRef TypeQual = "");
>>>
>>> + void emitHiddenKernelArgs(const Function &Func);
>>> +
>>> public:
>>> MetadataStreamer() = default;
>>> ~MetadataStreamer() = default;
>>>
>>> Modified:
>>> llvm/trunk/test/CodeGen/AMDGPU/hsa-metadata-enqueue-kernel.ll
>>> URL:
>>>
>>
> http://llvm.org/viewvc/llvm-project/llvm/trunk/test/CodeGen/AMDGPU/hsa-metadata-enqueue-kernel.ll?rev=336697&r1=336696&r2=336697&view=diff
>>> [5]
>>>
>>
> ==============================================================================
>>> --- llvm/trunk/test/CodeGen/AMDGPU/hsa-metadata-enqueue-kernel.ll
>>> (original)
>>> +++ llvm/trunk/test/CodeGen/AMDGPU/hsa-metadata-enqueue-kernel.ll
>>> Tue Jul 10 09:12:51 2018
>>> @@ -32,7 +32,7 @@
>>> ; CHECK-NEXT: ValueType: I64
>>> ; CHECK-NOT: ValueKind: HiddenDefaultQueue
>>> ; CHECK-NOT: ValueKind: HiddenCompletionAction
>>> -define amdgpu_kernel void @test_non_enqueue_kernel_caller(i8 %a)
>>> +define amdgpu_kernel void @test_non_enqueue_kernel_caller(i8 %a)
>>> #0
>>> !kernel_arg_addr_space !1 !kernel_arg_access_qual !2
>>> !kernel_arg_type !3
>>> !kernel_arg_base_type !3 !kernel_arg_type_qual !4 {
>>> ret void
>>> @@ -77,13 +77,14 @@ define amdgpu_kernel void @test_non_enqu
>>> ; CHECK-NEXT: ValueKind: HiddenCompletionAction
>>> ; CHECK-NEXT: ValueType: I8
>>> ; CHECK-NEXT: AddrSpaceQual: Global
>>> -define amdgpu_kernel void @test_enqueue_kernel_caller(i8 %a) #0
>>> +define amdgpu_kernel void @test_enqueue_kernel_caller(i8 %a) #1
>>> !kernel_arg_addr_space !1 !kernel_arg_access_qual !2
>>> !kernel_arg_type !3
>>> !kernel_arg_base_type !3 !kernel_arg_type_qual !4 {
>>> ret void
>>> }
>>>
>>> -attributes #0 = { "calls-enqueue-kernel" }
>>> +attributes #0 = { "amdgpu-implicitarg-num-bytes"="48" }
>>> +attributes #1 = { "calls-enqueue-kernel"
>>> "amdgpu-implicitarg-num-bytes"="48" }
>>>
>>> !1 = !{i32 0}
>>> !2 = !{!"none"}
>>> @@ -93,5 +94,4 @@ attributes #0 = { "calls-enqueue-kernel"
>>> !opencl.ocl.version = !{!90}
>>> !90 = !{i32 2, i32 0}
>>>
>>> -
>>> ; PARSER: AMDGPU HSA Metadata Parser Test: PASS
>>>
>>> Modified:
>>> llvm/trunk/test/CodeGen/AMDGPU/hsa-metadata-from-llvm-ir-full.ll
>>> URL:
>>>
>>
> http://llvm.org/viewvc/llvm-project/llvm/trunk/test/CodeGen/AMDGPU/hsa-metadata-from-llvm-ir-full.ll?rev=336697&r1=336696&r2=336697&view=diff
>>> [6]
>>>
>>
> ==============================================================================
>>> ---
>>> llvm/trunk/test/CodeGen/AMDGPU/hsa-metadata-from-llvm-ir-full.ll
>>> (original)
>>> +++
>>> llvm/trunk/test/CodeGen/AMDGPU/hsa-metadata-from-llvm-ir-full.ll
>>> Tue Jul 10 09:12:51 2018
>>> @@ -54,7 +54,7 @@
>>> ; CHECK-NEXT: AddrSpaceQual: Global
>>> ; CHECK-NOT: ValueKind: HiddenDefaultQueue
>>> ; CHECK-NOT: ValueKind: HiddenCompletionAction
>>> -define amdgpu_kernel void @test_char(i8 %a)
>>> +define amdgpu_kernel void @test_char(i8 %a) #0
>>> !kernel_arg_addr_space !1 !kernel_arg_access_qual !2
>>> !kernel_arg_type !9
>>> !kernel_arg_base_type !9 !kernel_arg_type_qual !4 {
>>> ret void
>>> @@ -89,7 +89,7 @@ define amdgpu_kernel void @test_char(i8
>>> ; CHECK-NEXT: ValueKind: HiddenPrintfBuffer
>>> ; CHECK-NEXT: ValueType: I8
>>> ; CHECK-NEXT: AddrSpaceQual: Global
>>> -define amdgpu_kernel void @test_ushort2(<2 x i16> %a)
>>> +define amdgpu_kernel void @test_ushort2(<2 x i16> %a) #0
>>> !kernel_arg_addr_space !1 !kernel_arg_access_qual !2
>>> !kernel_arg_type !10
>>> !kernel_arg_base_type !10 !kernel_arg_type_qual !4 {
>>> ret void
>>> @@ -124,7 +124,7 @@ define amdgpu_kernel void @test_ushort2(
>>> ; CHECK-NEXT: ValueKind: HiddenPrintfBuffer
>>> ; CHECK-NEXT: ValueType: I8
>>> ; CHECK-NEXT: AddrSpaceQual: Global
>>> -define amdgpu_kernel void @test_int3(<3 x i32> %a)
>>> +define amdgpu_kernel void @test_int3(<3 x i32> %a) #0
>>> !kernel_arg_addr_space !1 !kernel_arg_access_qual !2
>>> !kernel_arg_type !11
>>> !kernel_arg_base_type !11 !kernel_arg_type_qual !4 {
>>> ret void
>>> @@ -159,7 +159,7 @@ define amdgpu_kernel void @test_int3(<3
>>> ; CHECK-NEXT: ValueKind: HiddenPrintfBuffer
>>> ; CHECK-NEXT: ValueType: I8
>>> ; CHECK-NEXT: AddrSpaceQual: Global
>>> -define amdgpu_kernel void @test_ulong4(<4 x i64> %a)
>>> +define amdgpu_kernel void @test_ulong4(<4 x i64> %a) #0
>>> !kernel_arg_addr_space !1 !kernel_arg_access_qual !2
>>> !kernel_arg_type !12
>>> !kernel_arg_base_type !12 !kernel_arg_type_qual !4 {
>>> ret void
>>> @@ -194,7 +194,7 @@ define amdgpu_kernel void @test_ulong4(<
>>> ; CHECK-NEXT: ValueKind: HiddenPrintfBuffer
>>> ; CHECK-NEXT: ValueType: I8
>>> ; CHECK-NEXT: AddrSpaceQual: Global
>>> -define amdgpu_kernel void @test_half8(<8 x half> %a)
>>> +define amdgpu_kernel void @test_half8(<8 x half> %a) #0
>>> !kernel_arg_addr_space !1 !kernel_arg_access_qual !2
>>> !kernel_arg_type !13
>>> !kernel_arg_base_type !13 !kernel_arg_type_qual !4 {
>>> ret void
>>> @@ -229,7 +229,7 @@ define amdgpu_kernel void @test_half8(<8
>>> ; CHECK-NEXT: ValueKind: HiddenPrintfBuffer
>>> ; CHECK-NEXT: ValueType: I8
>>> ; CHECK-NEXT: AddrSpaceQual: Global
>>> -define amdgpu_kernel void @test_float16(<16 x float> %a)
>>> +define amdgpu_kernel void @test_float16(<16 x float> %a) #0
>>> !kernel_arg_addr_space !1 !kernel_arg_access_qual !2
>>> !kernel_arg_type !14
>>> !kernel_arg_base_type !14 !kernel_arg_type_qual !4 {
>>> ret void
>>> @@ -264,7 +264,7 @@ define amdgpu_kernel void @test_float16(
>>> ; CHECK-NEXT: ValueKind: HiddenPrintfBuffer
>>> ; CHECK-NEXT: ValueType: I8
>>> ; CHECK-NEXT: AddrSpaceQual: Global
>>> -define amdgpu_kernel void @test_double16(<16 x double> %a)
>>> +define amdgpu_kernel void @test_double16(<16 x double> %a) #0
>>> !kernel_arg_addr_space !1 !kernel_arg_access_qual !2
>>> !kernel_arg_type !15
>>> !kernel_arg_base_type !15 !kernel_arg_type_qual !4 {
>>> ret void
>>> @@ -300,7 +300,7 @@ define amdgpu_kernel void @test_double16
>>> ; CHECK-NEXT: ValueKind: HiddenPrintfBuffer
>>> ; CHECK-NEXT: ValueType: I8
>>> ; CHECK-NEXT: AddrSpaceQual: Global
>>> -define amdgpu_kernel void @test_pointer(i32 addrspace(1)* %a)
>>> +define amdgpu_kernel void @test_pointer(i32 addrspace(1)* %a) #0
>>> !kernel_arg_addr_space !1 !kernel_arg_access_qual !2
>>> !kernel_arg_type !16
>>> !kernel_arg_base_type !16 !kernel_arg_type_qual !4 {
>>> ret void
>>> @@ -336,7 +336,7 @@ define amdgpu_kernel void @test_pointer(
>>> ; CHECK-NEXT: ValueKind: HiddenPrintfBuffer
>>> ; CHECK-NEXT: ValueType: I8
>>> ; CHECK-NEXT: AddrSpaceQual: Global
>>> -define amdgpu_kernel void @test_image(%opencl.image2d_t
>>> addrspace(1)* %a)
>>> +define amdgpu_kernel void @test_image(%opencl.image2d_t
>>> addrspace(1)* %a) #0
>>> !kernel_arg_addr_space !1 !kernel_arg_access_qual !2
>>> !kernel_arg_type !17
>>> !kernel_arg_base_type !17 !kernel_arg_type_qual !4 {
>>> ret void
>>> @@ -371,7 +371,7 @@ define amdgpu_kernel void @test_image(%o
>>> ; CHECK-NEXT: ValueKind: HiddenPrintfBuffer
>>> ; CHECK-NEXT: ValueType: I8
>>> ; CHECK-NEXT: AddrSpaceQual: Global
>>> -define amdgpu_kernel void @test_sampler(i32 %a)
>>> +define amdgpu_kernel void @test_sampler(i32 %a) #0
>>> !kernel_arg_addr_space !1 !kernel_arg_access_qual !2
>>> !kernel_arg_type !18
>>> !kernel_arg_base_type !18 !kernel_arg_type_qual !4 {
>>> ret void
>>> @@ -407,7 +407,7 @@ define amdgpu_kernel void @test_sampler(
>>> ; CHECK-NEXT: ValueKind: HiddenPrintfBuffer
>>> ; CHECK-NEXT: ValueType: I8
>>> ; CHECK-NEXT: AddrSpaceQual: Global
>>> -define amdgpu_kernel void @test_queue(%opencl.queue_t
>>> addrspace(1)* %a)
>>> +define amdgpu_kernel void @test_queue(%opencl.queue_t
>>> addrspace(1)* %a) #0
>>> !kernel_arg_addr_space !1 !kernel_arg_access_qual !2
>>> !kernel_arg_type !19
>>> !kernel_arg_base_type !19 !kernel_arg_type_qual !4 {
>>> ret void
>>> @@ -443,7 +443,7 @@ define amdgpu_kernel void @test_queue(%o
>>> ; CHECK-NEXT: ValueKind: HiddenPrintfBuffer
>>> ; CHECK-NEXT: ValueType: I8
>>> ; CHECK-NEXT: AddrSpaceQual: Global
>>> -define amdgpu_kernel void @test_struct(%struct.A addrspace(5)*
>>> byval %a)
>>> +define amdgpu_kernel void @test_struct(%struct.A addrspace(5)*
>>> byval %a) #0
>>> !kernel_arg_addr_space !1 !kernel_arg_access_qual !2
>>> !kernel_arg_type !20
>>> !kernel_arg_base_type !20 !kernel_arg_type_qual !4 {
>>> ret void
>>> @@ -478,7 +478,7 @@ define amdgpu_kernel void @test_struct(%
>>> ; CHECK-NEXT: ValueKind: HiddenPrintfBuffer
>>> ; CHECK-NEXT: ValueType: I8
>>> ; CHECK-NEXT: AddrSpaceQual: Global
>>> -define amdgpu_kernel void @test_i128(i128 %a)
>>> +define amdgpu_kernel void @test_i128(i128 %a) #0
>>> !kernel_arg_addr_space !1 !kernel_arg_access_qual !2
>>> !kernel_arg_type !21
>>> !kernel_arg_base_type !21 !kernel_arg_type_qual !4 {
>>> ret void
>>> @@ -527,7 +527,7 @@ define amdgpu_kernel void @test_i128(i12
>>> ; CHECK-NEXT: ValueKind: HiddenPrintfBuffer
>>> ; CHECK-NEXT: ValueType: I8
>>> ; CHECK-NEXT: AddrSpaceQual: Global
>>> -define amdgpu_kernel void @test_multi_arg(i32 %a, <2 x i16> %b,
>>> <3 x i8> %c)
>>> +define amdgpu_kernel void @test_multi_arg(i32 %a, <2 x i16> %b,
>>> <3 x i8> %c) #0
>>> !kernel_arg_addr_space !22 !kernel_arg_access_qual !23
>>> !kernel_arg_type !24
>>> !kernel_arg_base_type !24 !kernel_arg_type_qual !25 {
>>> ret void
>>> @@ -582,7 +582,7 @@ define amdgpu_kernel void @test_multi_ar
>>> ; CHECK-NEXT: AddrSpaceQual: Global
>>> define amdgpu_kernel void @test_addr_space(i32 addrspace(1)* %g,
>>> i32 addrspace(4)* %c,
>>> - i32 addrspace(3)* %l)
>>> + i32 addrspace(3)* %l) #0
>>> !kernel_arg_addr_space !50 !kernel_arg_access_qual !23
>>> !kernel_arg_type !51
>>> !kernel_arg_base_type !51 !kernel_arg_type_qual !25 {
>>> ret void
>>> @@ -640,7 +640,7 @@ define amdgpu_kernel void @test_addr_spa
>>> ; CHECK-NEXT: AddrSpaceQual: Global
>>> define amdgpu_kernel void @test_type_qual(i32 addrspace(1)* %a,
>>> i32 addrspace(1)* %b,
>>> - %opencl.pipe_t addrspace(1)* %c)
>>> + %opencl.pipe_t addrspace(1)* %c) #0
>>> !kernel_arg_addr_space !22 !kernel_arg_access_qual !23
>>> !kernel_arg_type !51
>>> !kernel_arg_base_type !51 !kernel_arg_type_qual !70 {
>>> ret void
>>> @@ -694,7 +694,7 @@ define amdgpu_kernel void @test_type_qua
>>> ; CHECK-NEXT: AddrSpaceQual: Global
>>> 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)
>>> + %opencl.image3d_t addrspace(1)* %rw) #0
>>> !kernel_arg_addr_space !60 !kernel_arg_access_qual !61
>>> !kernel_arg_type !62
>>> !kernel_arg_base_type !62 !kernel_arg_type_qual !25 {
>>> ret void
>>> @@ -731,7 +731,7 @@ define amdgpu_kernel void @test_access_q
>>> ; CHECK-NEXT: ValueKind: HiddenPrintfBuffer
>>> ; CHECK-NEXT: ValueType: I8
>>> ; CHECK-NEXT: AddrSpaceQual: Global
>>> -define amdgpu_kernel void @test_vec_type_hint_half(i32 %a)
>>> +define amdgpu_kernel void @test_vec_type_hint_half(i32 %a) #0
>>> !kernel_arg_addr_space !1 !kernel_arg_access_qual !2
>>> !kernel_arg_type !3
>>> !kernel_arg_base_type !3 !kernel_arg_type_qual !4 !vec_type_hint
>>> !26 {
>>> ret void
>>> @@ -768,7 +768,7 @@ define amdgpu_kernel void @test_vec_type
>>> ; CHECK-NEXT: ValueKind: HiddenPrintfBuffer
>>> ; CHECK-NEXT: ValueType: I8
>>> ; CHECK-NEXT: AddrSpaceQual: Global
>>> -define amdgpu_kernel void @test_vec_type_hint_float(i32 %a)
>>> +define amdgpu_kernel void @test_vec_type_hint_float(i32 %a) #0
>>> !kernel_arg_addr_space !1 !kernel_arg_access_qual !2
>>> !kernel_arg_type !3
>>> !kernel_arg_base_type !3 !kernel_arg_type_qual !4 !vec_type_hint
>>> !27 {
>>> ret void
>>> @@ -805,7 +805,7 @@ define amdgpu_kernel void @test_vec_type
>>> ; CHECK-NEXT: ValueKind: HiddenPrintfBuffer
>>> ; CHECK-NEXT: ValueType: I8
>>> ; CHECK-NEXT: AddrSpaceQual: Global
>>> -define amdgpu_kernel void @test_vec_type_hint_double(i32 %a)
>>> +define amdgpu_kernel void @test_vec_type_hint_double(i32 %a) #0
>>> !kernel_arg_addr_space !1 !kernel_arg_access_qual !2
>>> !kernel_arg_type !3
>>> !kernel_arg_base_type !3 !kernel_arg_type_qual !4 !vec_type_hint
>>> !28 {
>>> ret void
>>> @@ -842,7 +842,7 @@ define amdgpu_kernel void @test_vec_type
>>> ; CHECK-NEXT: ValueKind: HiddenPrintfBuffer
>>> ; CHECK-NEXT: ValueType: I8
>>> ; CHECK-NEXT: AddrSpaceQual: Global
>>> -define amdgpu_kernel void @test_vec_type_hint_char(i32 %a)
>>> +define amdgpu_kernel void @test_vec_type_hint_char(i32 %a) #0
>>> !kernel_arg_addr_space !1 !kernel_arg_access_qual !2
>>> !kernel_arg_type !3
>>> !kernel_arg_base_type !3 !kernel_arg_type_qual !4 !vec_type_hint
>>> !29 {
>>> ret void
>>> @@ -879,7 +879,7 @@ define amdgpu_kernel void @test_vec_type
>>> ; CHECK-NEXT: ValueKind: HiddenPrintfBuffer
>>> ; CHECK-NEXT: ValueType: I8
>>> ; CHECK-NEXT: AddrSpaceQual: Global
>>> -define amdgpu_kernel void @test_vec_type_hint_short(i32 %a)
>>> +define amdgpu_kernel void @test_vec_type_hint_short(i32 %a) #0
>>> !kernel_arg_addr_space !1 !kernel_arg_access_qual !2
>>> !kernel_arg_type !3
>>> !kernel_arg_base_type !3 !kernel_arg_type_qual !4 !vec_type_hint
>>> !30 {
>>> ret void
>>> @@ -916,7 +916,7 @@ define amdgpu_kernel void @test_vec_type
>>> ; CHECK-NEXT: ValueKind: HiddenPrintfBuffer
>>> ; CHECK-NEXT: ValueType: I8
>>> ; CHECK-NEXT: AddrSpaceQual: Global
>>> -define amdgpu_kernel void @test_vec_type_hint_long(i32 %a)
>>> +define amdgpu_kernel void @test_vec_type_hint_long(i32 %a) #0
>>> !kernel_arg_addr_space !1 !kernel_arg_access_qual !2
>>> !kernel_arg_type !3
>>> !kernel_arg_base_type !3 !kernel_arg_type_qual !4 !vec_type_hint
>>> !31 {
>>> ret void
>>> @@ -953,7 +953,7 @@ define amdgpu_kernel void @test_vec_type
>>> ; CHECK-NEXT: ValueKind: HiddenPrintfBuffer
>>> ; CHECK-NEXT: ValueType: I8
>>> ; CHECK-NEXT: AddrSpaceQual: Global
>>> -define amdgpu_kernel void @test_vec_type_hint_unknown(i32 %a)
>>> +define amdgpu_kernel void @test_vec_type_hint_unknown(i32 %a) #0
>>> !kernel_arg_addr_space !1 !kernel_arg_access_qual !2
>>> !kernel_arg_type !3
>>> !kernel_arg_base_type !3 !kernel_arg_type_qual !4 !vec_type_hint
>>> !32 {
>>> ret void
>>> @@ -991,7 +991,7 @@ define amdgpu_kernel void @test_vec_type
>>> ; CHECK-NEXT: ValueKind: HiddenPrintfBuffer
>>> ; CHECK-NEXT: ValueType: I8
>>> ; CHECK-NEXT: AddrSpaceQual: Global
>>> -define amdgpu_kernel void @test_reqd_wgs_vec_type_hint(i32 %a)
>>> +define amdgpu_kernel void @test_reqd_wgs_vec_type_hint(i32 %a) #0
>>> !kernel_arg_addr_space !1 !kernel_arg_access_qual !2
>>> !kernel_arg_type !3
>>> !kernel_arg_base_type !3 !kernel_arg_type_qual !4 !vec_type_hint
>>> !5
>>> !reqd_work_group_size !6 {
>>> @@ -1030,7 +1030,7 @@ define amdgpu_kernel void @test_reqd_wgs
>>> ; CHECK-NEXT: ValueKind: HiddenPrintfBuffer
>>> ; CHECK-NEXT: ValueType: I8
>>> ; CHECK-NEXT: AddrSpaceQual: Global
>>> -define amdgpu_kernel void @test_wgs_hint_vec_type_hint(i32 %a)
>>> +define amdgpu_kernel void @test_wgs_hint_vec_type_hint(i32 %a) #0
>>> !kernel_arg_addr_space !1 !kernel_arg_access_qual !2
>>> !kernel_arg_type !3
>>> !kernel_arg_base_type !3 !kernel_arg_type_qual !4 !vec_type_hint
>>> !7
>>> !work_group_size_hint !8 {
>>> @@ -1067,7 +1067,7 @@ define amdgpu_kernel void @test_wgs_hint
>>> ; CHECK-NEXT: ValueKind: HiddenPrintfBuffer
>>> ; CHECK-NEXT: ValueType: I8
>>> ; CHECK-NEXT: AddrSpaceQual: Global
>>> -define amdgpu_kernel void @test_arg_ptr_to_ptr(i32 addrspace(5)*
>>> addrspace(1)* %a)
>>> +define amdgpu_kernel void @test_arg_ptr_to_ptr(i32 addrspace(5)*
>>> addrspace(1)* %a) #0
>>> !kernel_arg_addr_space !81 !kernel_arg_access_qual !2
>>> !kernel_arg_type !80
>>> !kernel_arg_base_type !80 !kernel_arg_type_qual !4 {
>>> ret void
>>> @@ -1103,7 +1103,7 @@ define amdgpu_kernel void @test_arg_ptr_
>>> ; CHECK-NEXT: ValueKind: HiddenPrintfBuffer
>>> ; CHECK-NEXT: ValueType: I8
>>> ; CHECK-NEXT: AddrSpaceQual: Global
>>> -define amdgpu_kernel void @test_arg_struct_contains_ptr(%struct.B
>>> addrspace(5)* byval %a)
>>> +define amdgpu_kernel void @test_arg_struct_contains_ptr(%struct.B
>>> addrspace(5)* byval %a) #0
>>> !kernel_arg_addr_space !1 !kernel_arg_access_qual !2
>>> !kernel_arg_type !82
>>> !kernel_arg_base_type !82 !kernel_arg_type_qual !4 {
>>> ret void
>>> @@ -1138,7 +1138,7 @@ define amdgpu_kernel void @test_arg_stru
>>> ; CHECK-NEXT: ValueKind: HiddenPrintfBuffer
>>> ; CHECK-NEXT: ValueType: I8
>>> ; CHECK-NEXT: AddrSpaceQual: Global
>>> -define amdgpu_kernel void @test_arg_vector_of_ptr(<2 x i32
>>> addrspace(1)*> %a)
>>> +define amdgpu_kernel void @test_arg_vector_of_ptr(<2 x i32
>>> addrspace(1)*> %a) #0
>>> !kernel_arg_addr_space !1 !kernel_arg_access_qual !2
>>> !kernel_arg_type !83
>>> !kernel_arg_base_type !83 !kernel_arg_type_qual !4 {
>>> ret void
>>> @@ -1175,7 +1175,7 @@ define amdgpu_kernel void @test_arg_vect
>>> ; CHECK-NEXT: ValueType: I8
>>> ; CHECK-NEXT: AddrSpaceQual: Global
>>> define amdgpu_kernel void @test_arg_unknown_builtin_type(
>>> - %opencl.clk_event_t addrspace(1)* %a)
>>> + %opencl.clk_event_t addrspace(1)* %a) #0
>>> !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
>>> @@ -1279,7 +1279,7 @@ define amdgpu_kernel void @test_pointee_
>>> <4 x i8> addrspace(3)* %e,
>>> <8 x i8> addrspace(3)* %f,
>>> <16 x i8> addrspace(3)* %g,
>>> - {} addrspace(3)* %h)
>>> + {} addrspace(3)* %h) #0
>>> !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
>>> @@ -1383,7 +1383,7 @@ define amdgpu_kernel void @test_pointee_
>>> <4 x i8> addrspace(3)* align 256 %e,
>>> <8 x i8> addrspace(3)* align 128 %f,
>>> <16 x i8> addrspace(3)* align 1024 %g,
>>> - {} addrspace(3)* align 16 %h)
>>> + {} addrspace(3)* align 16 %h) #0
>>> !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
>>> @@ -1422,7 +1422,7 @@ define amdgpu_kernel void @test_pointee_
>>> ; CHECK-NEXT: ValueType: I8
>>> ; CHECK-NEXT: AddrSpaceQual: Global
>>> define amdgpu_kernel void @__test_block_invoke_kernel(
>>> - <{ i32, i32, i8*, i8 addrspace(1)*, i8 }> %arg) #0
>>> + <{ i32, i32, i8*, i8 addrspace(1)*, i8 }> %arg) #1
>>> !kernel_arg_addr_space !1 !kernel_arg_access_qual !2
>>> !kernel_arg_type !110
>>> !kernel_arg_base_type !110 !kernel_arg_type_qual !4 {
>>> ret void
>>> @@ -1467,14 +1467,15 @@ define amdgpu_kernel void @__test_block_
>>> ; CHECK-NEXT: ValueKind: HiddenCompletionAction
>>> ; CHECK-NEXT: ValueType: I8
>>> ; CHECK-NEXT: AddrSpaceQual: Global
>>> -define amdgpu_kernel void @test_enqueue_kernel_caller(i8 %a) #1
>>> +define amdgpu_kernel void @test_enqueue_kernel_caller(i8 %a) #2
>>> !kernel_arg_addr_space !1 !kernel_arg_access_qual !2
>>> !kernel_arg_type !9
>>> !kernel_arg_base_type !9 !kernel_arg_type_qual !4 {
>>> ret void
>>> }
>>>
>>> -attributes #0 = {
>>> "runtime-handle"="__test_block_invoke_kernel_runtime_handle" }
>>> -attributes #1 = { "calls-enqueue-kernel" }
>>> +attributes #0 = { "amdgpu-implicitarg-num-bytes"="48" }
>>> +attributes #1 = { "amdgpu-implicitarg-num-bytes"="48"
>>> "runtime-handle"="__test_block_invoke_kernel_runtime_handle" }
>>> +attributes #2 = { "amdgpu-implicitarg-num-bytes"="48"
>>> "calls-enqueue-kernel" }
>>>
>>> !llvm.printf.fmts = !{!100, !101}
>>>
>>> Modified:
>>> llvm/trunk/test/CodeGen/AMDGPU/hsa-metadata-hidden-args.ll
>>> URL:
>>>
>>
> http://llvm.org/viewvc/llvm-project/llvm/trunk/test/CodeGen/AMDGPU/hsa-metadata-hidden-args.ll?rev=336697&r1=336696&r2=336697&view=diff
>>> [7]
>>>
>>
> ==============================================================================
>>> --- llvm/trunk/test/CodeGen/AMDGPU/hsa-metadata-hidden-args.ll
>>> (original)
>>> +++ llvm/trunk/test/CodeGen/AMDGPU/hsa-metadata-hidden-args.ll Tue
>>> Jul 10 09:12:51 2018
>>> @@ -6,8 +6,168 @@
>>> ; CHECK: Version: [ 1, 0 ]
>>> ; CHECK: Kernels:
>>>
>>> -; CHECK: - Name: test
>>> -; CHECK: SymbolName: 'test at kd'
>>> +; CHECK: - Name: test0
>>> +; CHECK: SymbolName: 'test0 at kd'
>>> +; CHECK: Args:
>>> +; CHECK-NEXT: - Name: r
>>> +; CHECK-NEXT: Size: 8
>>> +; CHECK-NEXT: Align: 8
>>> +; CHECK-NEXT: ValueKind: GlobalBuffer
>>> +; CHECK-NEXT: ValueType: F16
>>> +; CHECK-NEXT: AddrSpaceQual: Global
>>> +; CHECK-NEXT: - Name: a
>>> +; CHECK-NEXT: Size: 8
>>> +; CHECK-NEXT: Align: 8
>>> +; CHECK-NEXT: ValueKind: GlobalBuffer
>>> +; CHECK-NEXT: ValueType: F16
>>> +; CHECK-NEXT: AddrSpaceQual: Global
>>> +; CHECK-NEXT: - Name: b
>>> +; CHECK-NEXT: Size: 8
>>> +; CHECK-NEXT: Align: 8
>>> +; CHECK-NEXT: ValueKind: GlobalBuffer
>>> +; CHECK-NEXT: ValueType: F16
>>> +; CHECK-NEXT: AddrSpaceQual: Global
>>> +; CHECK-NEXT: CodeProps:
>>> +define amdgpu_kernel void @test0(
>>> + half addrspace(1)* %r,
>>> + half addrspace(1)* %a,
>>> + half addrspace(1)* %b) {
>>> +entry:
>>> + %a.val = load half, half addrspace(1)* %a
>>> + %b.val = load half, half addrspace(1)* %b
>>> + %r.val = fadd half %a.val, %b.val
>>> + store half %r.val, half addrspace(1)* %r
>>> + ret void
>>> +}
>>> +
>>> +; CHECK: - Name: test8
>>> +; CHECK: SymbolName: 'test8 at kd'
>>> +; CHECK: Args:
>>> +; CHECK-NEXT: - Name: r
>>> +; CHECK-NEXT: Size: 8
>>> +; CHECK-NEXT: Align: 8
>>> +; CHECK-NEXT: ValueKind: GlobalBuffer
>>> +; CHECK-NEXT: ValueType: F16
>>> +; CHECK-NEXT: AddrSpaceQual: Global
>>> +; CHECK-NEXT: - Name: a
>>> +; CHECK-NEXT: Size: 8
>>> +; CHECK-NEXT: Align: 8
>>> +; CHECK-NEXT: ValueKind: GlobalBuffer
>>> +; CHECK-NEXT: ValueType: F16
>>> +; CHECK-NEXT: AddrSpaceQual: Global
>>> +; CHECK-NEXT: - Name: b
>>> +; CHECK-NEXT: Size: 8
>>> +; CHECK-NEXT: Align: 8
>>> +; CHECK-NEXT: ValueKind: GlobalBuffer
>>> +; CHECK-NEXT: ValueType: F16
>>> +; CHECK-NEXT: AddrSpaceQual: Global
>>> +; CHECK-NEXT: - Size: 8
>>> +; CHECK-NEXT: Align: 8
>>> +; CHECK-NEXT: ValueKind: HiddenGlobalOffsetX
>>> +; CHECK-NEXT: ValueType: I64
>>> +; CHECK-NEXT: CodeProps:
>>> +define amdgpu_kernel void @test8(
>>> + half addrspace(1)* %r,
>>> + half addrspace(1)* %a,
>>> + half addrspace(1)* %b) #0 {
>>> +entry:
>>> + %a.val = load half, half addrspace(1)* %a
>>> + %b.val = load half, half addrspace(1)* %b
>>> + %r.val = fadd half %a.val, %b.val
>>> + store half %r.val, half addrspace(1)* %r
>>> + ret void
>>> +}
>>> +
>>> +; CHECK: - Name: test16
>>> +; CHECK: SymbolName: 'test16 at kd'
>>> +; CHECK: Args:
>>> +; CHECK-NEXT: - Name: r
>>> +; CHECK-NEXT: Size: 8
>>> +; CHECK-NEXT: Align: 8
>>> +; CHECK-NEXT: ValueKind: GlobalBuffer
>>> +; CHECK-NEXT: ValueType: F16
>>> +; CHECK-NEXT: AddrSpaceQual: Global
>>> +; CHECK-NEXT: - Name: a
>>> +; CHECK-NEXT: Size: 8
>>> +; CHECK-NEXT: Align: 8
>>> +; CHECK-NEXT: ValueKind: GlobalBuffer
>>> +; CHECK-NEXT: ValueType: F16
>>> +; CHECK-NEXT: AddrSpaceQual: Global
>>> +; CHECK-NEXT: - Name: b
>>> +; CHECK-NEXT: Size: 8
>>> +; CHECK-NEXT: Align: 8
>>> +; CHECK-NEXT: ValueKind: GlobalBuffer
>>> +; CHECK-NEXT: ValueType: F16
>>> +; CHECK-NEXT: AddrSpaceQual: Global
>>> +; CHECK-NEXT: - Size: 8
>>> +; CHECK-NEXT: Align: 8
>>> +; CHECK-NEXT: ValueKind: HiddenGlobalOffsetX
>>> +; CHECK-NEXT: ValueType: I64
>>> +; CHECK-NEXT: - Size: 8
>>> +; CHECK-NEXT: Align: 8
>>> +; CHECK-NEXT: ValueKind: HiddenGlobalOffsetY
>>> +; CHECK-NEXT: ValueType: I64
>>> +; CHECK-NEXT: CodeProps:
>>> +define amdgpu_kernel void @test16(
>>> + half addrspace(1)* %r,
>>> + half addrspace(1)* %a,
>>> + half addrspace(1)* %b) #1 {
>>> +entry:
>>> + %a.val = load half, half addrspace(1)* %a
>>> + %b.val = load half, half addrspace(1)* %b
>>> + %r.val = fadd half %a.val, %b.val
>>> + store half %r.val, half addrspace(1)* %r
>>> + ret void
>>> +}
>>> +
>>> +; CHECK: - Name: test24
>>> +; CHECK: SymbolName: 'test24 at kd'
>>> +; CHECK: Args:
>>> +; CHECK-NEXT: - Name: r
>>> +; CHECK-NEXT: Size: 8
>>> +; CHECK-NEXT: Align: 8
>>> +; CHECK-NEXT: ValueKind: GlobalBuffer
>>> +; CHECK-NEXT: ValueType: F16
>>> +; CHECK-NEXT: AddrSpaceQual: Global
>>> +; CHECK-NEXT: - Name: a
>>> +; CHECK-NEXT: Size: 8
>>> +; CHECK-NEXT: Align: 8
>>> +; CHECK-NEXT: ValueKind: GlobalBuffer
>>> +; CHECK-NEXT: ValueType: F16
>>> +; CHECK-NEXT: AddrSpaceQual: Global
>>> +; CHECK-NEXT: - Name: b
>>> +; CHECK-NEXT: Size: 8
>>> +; CHECK-NEXT: Align: 8
>>> +; CHECK-NEXT: ValueKind: GlobalBuffer
>>> +; CHECK-NEXT: ValueType: F16
>>> +; CHECK-NEXT: AddrSpaceQual: Global
>>> +; CHECK-NEXT: - Size: 8
>>> +; CHECK-NEXT: Align: 8
>>> +; CHECK-NEXT: ValueKind: HiddenGlobalOffsetX
>>> +; CHECK-NEXT: ValueType: I64
>>> +; CHECK-NEXT: - Size: 8
>>> +; CHECK-NEXT: Align: 8
>>> +; CHECK-NEXT: ValueKind: HiddenGlobalOffsetY
>>> +; CHECK-NEXT: ValueType: I64
>>> +; CHECK-NEXT: - Size: 8
>>> +; CHECK-NEXT: Align: 8
>>> +; CHECK-NEXT: ValueKind: HiddenGlobalOffsetZ
>>> +; CHECK-NEXT: ValueType: I64
>>> +; CHECK-NEXT: CodeProps:
>>> +define amdgpu_kernel void @test24(
>>> + half addrspace(1)* %r,
>>> + half addrspace(1)* %a,
>>> + half addrspace(1)* %b) #2 {
>>> +entry:
>>> + %a.val = load half, half addrspace(1)* %a
>>> + %b.val = load half, half addrspace(1)* %b
>>> + %r.val = fadd half %a.val, %b.val
>>> + store half %r.val, half addrspace(1)* %r
>>> + ret void
>>> +}
>>> +
>>> +; CHECK: - Name: test32
>>> +; CHECK: SymbolName: 'test32 at kd'
>>> ; CHECK: Args:
>>> ; CHECK-NEXT: - Name: r
>>> ; CHECK-NEXT: Size: 8
>>> @@ -44,6 +204,52 @@
>>> ; CHECK-NEXT: ValueKind: HiddenNone
>>> ; CHECK-NEXT: ValueType: I8
>>> ; CHECK-NEXT: AddrSpaceQual: Global
>>> +; CHECK-NEXT: CodeProps:
>>> +define amdgpu_kernel void @test32(
>>> + half addrspace(1)* %r,
>>> + half addrspace(1)* %a,
>>> + half addrspace(1)* %b) #3 {
>>> +entry:
>>> + %a.val = load half, half addrspace(1)* %a
>>> + %b.val = load half, half addrspace(1)* %b
>>> + %r.val = fadd half %a.val, %b.val
>>> + store half %r.val, half addrspace(1)* %r
>>> + ret void
>>> +}
>>> +
>>> +; CHECK: - Name: test48
>>> +; CHECK: SymbolName: 'test48 at kd'
>>> +; CHECK: Args:
>>> +; CHECK-NEXT: - Name: r
>>> +; CHECK-NEXT: Size: 8
>>> +; CHECK-NEXT: Align: 8
>>> +; CHECK-NEXT: ValueKind: GlobalBuffer
>>> +; CHECK-NEXT: ValueType: F16
>>> +; CHECK-NEXT: AddrSpaceQual: Global
>>> +; CHECK-NEXT: - Name: a
>>> +; CHECK-NEXT: Size: 8
>>> +; CHECK-NEXT: Align: 8
>>> +; CHECK-NEXT: ValueKind: GlobalBuffer
>>> +; CHECK-NEXT: ValueType: F16
>>> +; CHECK-NEXT: AddrSpaceQual: Global
>>> +; CHECK-NEXT: - Name: b
>>> +; CHECK-NEXT: Size: 8
>>> +; CHECK-NEXT: Align: 8
>>> +; CHECK-NEXT: ValueKind: GlobalBuffer
>>> +; CHECK-NEXT: ValueType: F16
>>> +; CHECK-NEXT: AddrSpaceQual: Global
>>> +; CHECK-NEXT: - Size: 8
>>> +; CHECK-NEXT: Align: 8
>>> +; CHECK-NEXT: ValueKind: HiddenGlobalOffsetX
>>> +; CHECK-NEXT: ValueType: I64
>>> +; CHECK-NEXT: - Size: 8
>>> +; CHECK-NEXT: Align: 8
>>> +; CHECK-NEXT: ValueKind: HiddenGlobalOffsetY
>>> +; CHECK-NEXT: ValueType: I64
>>> +; CHECK-NEXT: - Size: 8
>>> +; CHECK-NEXT: Align: 8
>>> +; CHECK-NEXT: ValueKind: HiddenGlobalOffsetZ
>>> +; CHECK-NEXT: ValueType: I64
>>> ; CHECK-NEXT: - Size: 8
>>> ; CHECK-NEXT: Align: 8
>>> ; CHECK-NEXT: ValueKind: HiddenNone
>>> @@ -54,10 +260,16 @@
>>> ; CHECK-NEXT: ValueKind: HiddenNone
>>> ; CHECK-NEXT: ValueType: I8
>>> ; CHECK-NEXT: AddrSpaceQual: Global
>>> -define amdgpu_kernel void @test(
>>> +; CHECK-NEXT: - Size: 8
>>> +; CHECK-NEXT: Align: 8
>>> +; CHECK-NEXT: ValueKind: HiddenNone
>>> +; CHECK-NEXT: ValueType: I8
>>> +; CHECK-NEXT: AddrSpaceQual: Global
>>> +; CHECK-NEXT: CodeProps:
>>> +define amdgpu_kernel void @test48(
>>> half addrspace(1)* %r,
>>> half addrspace(1)* %a,
>>> - half addrspace(1)* %b) {
>>> + half addrspace(1)* %b) #4 {
>>> entry:
>>> %a.val = load half, half addrspace(1)* %a
>>> %b.val = load half, half addrspace(1)* %b
>>> @@ -66,5 +278,8 @@ entry:
>>> ret void
>>> }
>>>
>>> -!opencl.ocl.version = !{!0}
>>> -!0 = !{i32 2, i32 0}
>>> +attributes #0 = { "amdgpu-implicitarg-num-bytes"="8" }
>>> +attributes #1 = { "amdgpu-implicitarg-num-bytes"="16" }
>>> +attributes #2 = { "amdgpu-implicitarg-num-bytes"="24" }
>>> +attributes #3 = { "amdgpu-implicitarg-num-bytes"="32" }
>>> +attributes #4 = { "amdgpu-implicitarg-num-bytes"="48" }
>>>
>>> _______________________________________________
>>> llvm-commits mailing list
>>> llvm-commits at lists.llvm.org
>>> http://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-commits [8]
>>
>> _______________________________________________
>> llvm-commits mailing list
>> llvm-commits at lists.llvm.org
>> http://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-commits [8]
>
>
>
> Links:
> ------
> [1] http://llvm.org/viewvc/llvm-project?rev=336697&view=rev
> [2] https://reviews.llvm.org/D49096
> [3]
> http://llvm.org/viewvc/llvm-project/llvm/trunk/lib/Target/AMDGPU/MCTargetDesc/AMDGPUHSAMetadataStreamer.cpp?rev=336697&r1=336696&r2=336697&view=diff
> [4]
> http://llvm.org/viewvc/llvm-project/llvm/trunk/lib/Target/AMDGPU/MCTargetDesc/AMDGPUHSAMetadataStreamer.h?rev=336697&r1=336696&r2=336697&view=diff
> [5]
> http://llvm.org/viewvc/llvm-project/llvm/trunk/test/CodeGen/AMDGPU/hsa-metadata-enqueue-kernel.ll?rev=336697&r1=336696&r2=336697&view=diff
> [6]
> http://llvm.org/viewvc/llvm-project/llvm/trunk/test/CodeGen/AMDGPU/hsa-metadata-from-llvm-ir-full.ll?rev=336697&r1=336696&r2=336697&view=diff
> [7]
> http://llvm.org/viewvc/llvm-project/llvm/trunk/test/CodeGen/AMDGPU/hsa-metadata-hidden-args.ll?rev=336697&r1=336696&r2=336697&view=diff
> [8] http://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-commits
> [9]
> http://lab.llvm.org:8011/builders/clang-ppc64le-linux-multistage/builds/7287
> [10]
> http://lab.llvm.org:8011/builders/clang-ppc64le-linux-multistage/builds/7287/steps/build%20stage%201/logs/stdio
> [11] https://reviews.llvm.org/rL336707
More information about the llvm-commits
mailing list