This patch fixes issues for -fgpu-rdc for Windows MSVC
toolchain:
Fix COFF specific section flags and remove section types
in llvm-mc input file for Windows.
Escape fatbin path in llvm-mc input file.
Add -triple option to llvm-mc.
Put __hip_gpubin_handle in comdat when it has linkonce_odr
linkage.
Reviewed by: Artem Belevich
Differential Revision: https://reviews.llvm.org/D115039
Two identical instantiations of a template function can be emitted by two TU's
with linkonce_odr linkage without causing duplicate symbols in linker. MSVC
also requires these symbols be in comdat sections. Linux does not require
the symbols in comdat sections to be merged by linker but by default
clang puts them in comdat sections.
If a template kernel is instantiated identically in two TU's. MSVC requires
that them to be in comdat sections, otherwise MSVC linker will diagnose them as
duplicate symbols. However, currently clang does not put instantiated template
kernels in comdat sections, which causes link error for MSVC.
This patch allows putting instantiated template kernels into comdat sections.
Reviewed by: Artem Belevich, Reid Kleckner
Differential Revision: https://reviews.llvm.org/D112492
createReplacementInstr was a trivial wrapper around
ConstantExpr::getAsInstruction, which also inserted the new instruction
into a basic block. Implement this directly in getAsInstruction by
adding an InsertBefore parameter and change all callers to use it. NFC.
A follow-up patch will remove createReplacementInstr.
Differential Revision: https://reviews.llvm.org/D112791
Remove uses of to-be-deprecated API. I've fallen back to calling
getPointerElementType() in some cases where the correct type wasn't
immediately obvious to me.
The original version of this was reverted, and @rjmcall provided some
advice to architect a new solution. This is that solution.
This implements a builtin to provide a unique name that is stable across
compilations of this TU for the purposes of implementing the library
component of the unnamed kernel feature of SYCL. It does this by
running the Itanium mangler with a few modifications.
Because it is somewhat common to wrap non-kernel-related lambdas in
macros that aren't present on the device (such as for logging), this
uniquely generates an ID for all lambdas involved in the naming of a
kernel. It uses the lambda-mangling number to do this, except replaces
this with its own number (starting at 10000 for readabililty reasons)
for lambdas used to name a kernel.
Additionally, this implements itself as constexpr with a slight catch:
if a name would be invalidated by the use of this lambda in a later
kernel invocation, it is diagnosed as an error (see the Sema tests).
Differential Revision: https://reviews.llvm.org/D103112
Currently clang does not emit device template variables
instantiated only in host functions, however, nvcc is
able to do that:
https://godbolt.org/z/fneEfferY
This patch fixes this issue by refactoring and extending
the existing mechanism for emitting static device
var ODR-used by host only. Basically clang records
device variables ODR-used by host code and force
them to be emitted in device compilation. The existing
mechanism makes sure these device variables ODR-used
by host code are added to llvm.compiler-used, therefore
they are guaranteed not to be deleted.
It also fixes non-ODR-use of static device variable by host code
causing static device variable to be emitted and registered,
which should not.
Reviewed by: Artem Belevich
Differential Revision: https://reviews.llvm.org/D102237
Add device variables to llvm.compiler.used if they are
ODR-used by either host or device functions.
This is necessary to prevent them from being
eliminated by whole-program optimization
where the compiler has no way to know a device
variable is used by some host code.
Reviewed by: Artem Belevich
Differential Revision: https://reviews.llvm.org/D98814
Add builtin function __builtin_get_device_side_mangled_name
to get device side manged name for functions and global
variables, which can be used to get symbol address of kernels
or variables by mangled name in dynamically loaded
bundled code objects at run time.
Reviewed by: Artem Belevich
Differential Revision: https://reviews.llvm.org/D99301
Currently clang uses stub function to launch kernel. This is inconvenient
to interop with C++ programs since the stub function has different name
as kernel, which is required by ROCm debugger.
This patch emits a variable symbol which has the same name as the kernel
and uses it to register and launch the kernel. This allows C++ program to
launch a kernel by using the original kernel name.
Reviewed by: Artem Belevich
Differential Revision: https://reviews.llvm.org/D86376
For -fgpu-rdc mode, static device vars in different TU's may have the same name.
To support accessing file-scope static device variables in host code, we need to give them
a distinct name and external linkage. This can be done by postfixing each static device variable with
a distinct CUID (Compilation Unit ID) hash.
Since the static device variables have different name across compilation units, now we let
them have external linkage so that they can be looked up by the runtime.
Reviewed by: Artem Belevich, and Jon Chesterfield
Differential Revision: https://reviews.llvm.org/D85223
Currently managed variables are emitted as undefined symbols, which
causes difficulty for diagnosing undefined symbols for non-managed
variables.
This patch transforms managed variables in device compilation so that
they can be emitted as normal variables.
Reviewed by: Artem Belevich
Differential Revision: https://reviews.llvm.org/D96195
For -fgpu-rdc, shadow variables should not be internalized, otherwise
they cannot be accessed by other TUs. This is necessary because
the shadow variable of external device variables are always
emitted as undefined symbols, which need to resolve to a global
symbols.
Managed variables need to be emitted as undefined symbols
in device compilations.
Reviewed by: Artem Belevich
Differential Revision: https://reviews.llvm.org/D95901
- The failures are all cc1-based tests due to the missing `-aux-triple` options,
which is always prepared by the driver in CUDA/HIP compilation.
- Add extra check on the missing aux-targetinfo to prevent crashing.
[hip][cuda] Enable extended lambda support on Windows.
- On Windows, extended lambda has extra issues due to the numbering
schemes are different between the host compilation (Microsoft C++ ABI)
and the device compilation (Itanium C++ ABI. Additional device side
lambda number is required per lambda for the host compilation to
correctly mangle the device-side lambda name.
- A hybrid numbering context `MSHIPNumberingContext` is introduced to
number a lambda for both host- and device-compilations.
Reviewed By: rnk
Differential Revision: https://reviews.llvm.org/D69322
This reverts commit 4874ff0241.
- On Windows, extended lambda has extra issues due to the numbering
schemes are different between the host compilation (Microsoft C++ ABI)
and the device compilation (Itanium C++ ABI. Additional device side
lambda number is required per lambda for the host compilation to
correctly mangle the device-side lambda name.
- A hybrid numbering context `MSHIPNumberingContext` is introduced to
number a lambda for both host- and device-compilations.
Reviewed By: rnk
Differential Revision: https://reviews.llvm.org/D69322
Extract registering device variable to CUDA runtime codegen function since it
will be called in multiple places.
Reviewed by: Artem Belevich
Differential Revision: https://reviews.llvm.org/D95558
This patch implements codegen for __managed__ variable attribute for HIP.
Diagnostics will be added later.
Differential Revision: https://reviews.llvm.org/D94814
Followup to D85191.
This changes getTypeInfoInChars to return a TypeInfoChars
struct instead of a std::pair of CharUnits. This lets the
interface match getTypeInfo more closely.
Reviewed By: efriedma
Differential Revision: https://reviews.llvm.org/D86447
To facilitate faster loading of device binaries and share them among processes,
HIP runtime favors their alignment being 4096 bytes. HIP runtime can load
unaligned device binaries, however, aligning them at 4096 bytes results in
faster loading and less shared memory usage.
This patch adds an option -bundle-align to clang-offload-bundler which allows
bundles to be aligned at specified alignment. By default it is 1, which is NFC
compared to existing format.
This patch then aligns embedded fat binary and device binary inside fat binary
at 4096 bytes.
It has been verified this change does not cause significant overall file size increase
for typical HIP applications (less than 1%).
Differential Revision: https://reviews.llvm.org/D88734
Do not detect device library by default in rocm detector.
Only detect device library in Rocm and HIP toolchain.
Separate detection of HIP runtime and Rocm device library.
Detect rocm path by version file in host toolchains.
Also added detecting rocm version and printing rocm
installation path and version with -v.
Fixed include path and device library detection for
ROCm 3.5.
Added --hip-version option. Renamed --hip-device-lib-path
to --rocm-device-lib-path.
Fixed default value for -fhip-new-launch-api.
Added default -std option for HIP.
Differential Revision: https://reviews.llvm.org/D82930
Summary:
- `RegisterVar` has `void` return type and `size_t` in its variable size
parameter in HIP or CUDA 9.0+.
Reviewers: tra, yaxunl
Subscribers: cfe-commits
Tags: #clang
Differential Revision: https://reviews.llvm.org/D77398
Summary:
- Even though the bindless surface/texture interfaces are promoted,
there are still code using surface/texture references. For example,
[PR#26400](https://bugs.llvm.org/show_bug.cgi?id=26400) reports the
compilation issue for code using `tex2D` with texture references. For
better compatibility, this patch proposes the support of
surface/texture references.
- Due to the absent documentation and magic headers, it's believed that
`nvcc` does use builtins for texture support. From the limited NVVM
documentation[^nvvm] and NVPTX backend texture/surface related
tests[^test], it's believed that surface/texture references are
supported by replacing their reference types, which are annotated with
`device_builtin_surface_type`/`device_builtin_texture_type`, with the
corresponding handle-like object types, `cudaSurfaceObject_t` or
`cudaTextureObject_t`, in the device-side compilation. On the host
side, that global handle variables are registered and will be
established and updated later when corresponding binding/unbinding
APIs are called[^bind]. Surface/texture references are most like
device global variables but represented in different types on the host
and device sides.
- In this patch, the following changes are proposed to support that
behavior:
+ Refine `device_builtin_surface_type` and
`device_builtin_texture_type` attributes to be applied on `Type`
decl only to check whether a variable is of the surface/texture
reference type.
+ Add hooks in code generation to replace that reference types with
the correponding object types as well as all accesses to them. In
particular, `nvvm.texsurf.handle.internal` should be used to load
object handles from global reference variables[^texsurf] as well as
metadata annotations.
+ Generate host-side registration with proper template argument
parsing.
---
[^nvvm]: https://docs.nvidia.com/cuda/pdf/NVVM_IR_Specification.pdf
[^test]: https://raw.githubusercontent.com/llvm/llvm-project/master/llvm/test/CodeGen/NVPTX/tex-read-cuda.ll
[^bind]: See section 3.2.11.1.2 ``Texture reference API` in [CUDA C Programming Guide](https://docs.nvidia.com/cuda/pdf/CUDA_C_Programming_Guide.pdf).
[^texsurf]: According to NVVM IR, `nvvm.texsurf.handle` should be used. But, the current backend doesn't have that supported. We may revise that later.
Reviewers: tra, rjmccall, yaxunl, a.sidorin
Subscribers: cfe-commits
Tags: #clang
Differential Revision: https://reviews.llvm.org/D76365
HIP emits a device stub function for each kernel in host code.
The HIP debugger requires device stub function to have a different unmangled name as the kernel.
Currently the name of the device stub function is the mangled name with a postfix .stub. However,
this does not work with the HIP debugger since the unmangled name is the same as the kernel.
This patch adds prefix __device__stub__ to the unmangled name of the device stub before mangling,
therefore the device stub function has a valid mangled name which is different than the device kernel
name. The device side kernel name is kept unchanged. kernels with extern "C" also gets the prefix added
to the corresponding device stub function.
Differential Revision: https://reviews.llvm.org/D68578
This is how it should've been and brings it more in line with
std::string_view. There should be no functional change here.
This is mostly mechanical from a custom clang-tidy check, with a lot of
manual fixups. It uncovers a lot of minor inefficiencies.
This doesn't actually modify StringRef yet, I'll do that in a follow-up.
Summary:
- Revise the interface to derive the stub name and simplify the
assertion of it.
Reviewers: yaxunl, tra
Subscribers: cfe-commits
Tags: #clang
Differential Revision: https://reviews.llvm.org/D63335
llvm-svn: 363553
Also for CUDA, we need to disable producing these fat binary functions when there is no GPU code.
Reviewers: yaxunl, tra
Differential Revision: https://reviews.llvm.org/D60141
llvm-svn: 357526
Skip producing the fat binary functions for HIP when no device code is present.
Reviewers: yaxunl
Differential Review: https://reviews.llvm.org/D60141
llvm-svn: 357520
Add .stub to kernel stub function name so that it is different from kernel
name in device code. This is necessary to let debugger find correct symbol
for kernel.
Differential Revision: https://reviews.llvm.org/D58518
llvm-svn: 354948
Add .stub to kernel stub function name so that it is different from kernel
name in device code. This is necessary to let debugger find correct symbol
for kernel
Differential Revision: https://reviews.llvm.org/D58518
llvm-svn: 354615
__hipRegisterFunction and __hipRegisterVar need to accept device side kernel and variable names
so that HIP runtime can associate kernel stub functions in host code with kernel symbols in fat binaries,
and associate shadow variables in host code with device variables in fat binaries.
Currently, clang assumes kernel functions and device variables have the same name as the kernel
stub functions and shadow variables. However, when host is compiled in windows with MSVC C++
ABI and device is compiled with Itanium C++ ABI (e.g. AMDGPU), kernels and device symbols in fat
binary are mangled differently than host.
This patch gets the device side kernel and variable name by mangling them in the mangle context
of aux target.
Differential Revision: https://reviews.llvm.org/D58163
llvm-svn: 354004
Emit{Nounwind,}RuntimeCall{,OrInvoke} have been modified to take a
FunctionCallee as an argument, and CreateRuntimeFunction has been
modified to return a FunctionCallee. All callers have been updated.
Additionally, CreateBuiltinFunction is removed, as it was redundant
with CreateRuntimeFunction after some previous changes.
Differential Revision: https://reviews.llvm.org/D57668
llvm-svn: 353184
This argument was added in r254554 in order to support the
pass_object_size attribute. However, in r296076, the attribute's
presence is now also represented in FunctionProtoType's
ExtParameterInfo, and thus it's unnecessary to pass along a separate
FunctionDecl.
The functions modified are:
RequiredArgs::forPrototype{,Plus}, and
CodeGenTypes::ConvertFunctionType.
After this, it's also (again) unnecessary to have a separate
ConvertFunctionType function ConvertType, so convert callers back to
the latter, leaving the former as an internal helper function.
llvm-svn: 352946
Instead of calling CUDA runtime to arrange function arguments,
the new API constructs arguments in a local array and the kernels
are launched with __cudaLaunchKernel().
The old API has been deprecated and is expected to go away
in the next CUDA release.
Differential Revision: https://reviews.llvm.org/D57488
llvm-svn: 352799
to reflect the new license.
We understand that people may be surprised that we're moving the header
entirely to discuss the new license. We checked this carefully with the
Foundation's lawyer and we believe this is the correct approach.
Essentially, all code in the project is now made available by the LLVM
project under our new license, so you will see that the license headers
include that license only. Some of our contributors have contributed
code under our old license, and accordingly, we have retained a copy of
our old license notice in the top-level files in each project and
repository.
llvm-svn: 351636
getGUID() returns an uint64_t and "%x" only prints 32 bits of it.
Use PRIx64 format string to print all 64 bits.
Differential Revision: https://reviews.llvm.org/D52938
llvm-svn: 343875
This patch renames -f{no-}cuda-rdc to -f{no-}gpu-rdc and keeps the original
options as aliases. When -fgpu-rdc is off,
clang will assume the device code in each translation unit does not call
external functions except those in the device library, therefore it is possible
to compile the device code in each translation unit to self-contained kernels
and embed them in the host object, so that the host object behaves like
usual host object which can be linked by lld.
The benefits of this feature is: 1. allow users to create static libraries which
can be linked by host linker; 2. amortized device code linking time.
This patch modifies HIP action builder to insert actions for linking device
code and generating HIP fatbin, and pass HIP fatbin to host backend action.
It extracts code for constructing command for generating HIP fatbin as
a function so that it can be reused by early finalization. It also modifies
codegen of HIP host constructor functions to embed the device fatbin
when it is available.
Differential Revision: https://reviews.llvm.org/D52377
llvm-svn: 343611
Different shared libraries contain different fat binary, which is stored in a global variable
__hip_gpubin_handle. Since different compilation units share the same fat binary, this
variable has linkonce linkage. However, it should not be merged across different shared
libraries.
This patch set the visibility of the global variable to be hidden, which will make it invisible
in the shared library, therefore preventing it from being merged.
Differential Revision: https://reviews.llvm.org/D50596
llvm-svn: 340056
Yaxun (Sam) Liu