Summary:
This patch replaces the CUDA specific action by a generic offload action. The offload action may have multiple dependences classier in “host” and “device”. The way this generic offloading action is used is very similar to what is done today by the CUDA implementation: it is used to set a specific toolchain and architecture to its dependences during the generation of jobs.
This patch also proposes propagating the offloading information through the action graph so that that information can be easily retrieved at any time during the generation of commands. This allows e.g. the "clang tool” to evaluate whether CUDA should be supported for the device or host and ptas to easily retrieve the target architecture.
This is an example of how the action graphs would look like (compilation of a single CUDA file with two GPU architectures)
```
0: input, "cudatests.cu", cuda, (host-cuda)
1: preprocessor, {0}, cuda-cpp-output, (host-cuda)
2: compiler, {1}, ir, (host-cuda)
3: input, "cudatests.cu", cuda, (device-cuda, sm_35)
4: preprocessor, {3}, cuda-cpp-output, (device-cuda, sm_35)
5: compiler, {4}, ir, (device-cuda, sm_35)
6: backend, {5}, assembler, (device-cuda, sm_35)
7: assembler, {6}, object, (device-cuda, sm_35)
8: offload, "device-cuda (nvptx64-nvidia-cuda:sm_35)" {7}, object
9: offload, "device-cuda (nvptx64-nvidia-cuda:sm_35)" {6}, assembler
10: input, "cudatests.cu", cuda, (device-cuda, sm_37)
11: preprocessor, {10}, cuda-cpp-output, (device-cuda, sm_37)
12: compiler, {11}, ir, (device-cuda, sm_37)
13: backend, {12}, assembler, (device-cuda, sm_37)
14: assembler, {13}, object, (device-cuda, sm_37)
15: offload, "device-cuda (nvptx64-nvidia-cuda:sm_37)" {14}, object
16: offload, "device-cuda (nvptx64-nvidia-cuda:sm_37)" {13}, assembler
17: linker, {8, 9, 15, 16}, cuda-fatbin, (device-cuda)
18: offload, "host-cuda (powerpc64le-unknown-linux-gnu)" {2}, "device-cuda (nvptx64-nvidia-cuda)" {17}, ir
19: backend, {18}, assembler
20: assembler, {19}, object
21: input, "cuda", object
22: input, "cudart", object
23: linker, {20, 21, 22}, image
```
The changes in this patch pass the existent regression tests (keeps the existent functionality) and resulting binaries execute correctly in a Power8+K40 machine.
Reviewers: echristo, hfinkel, jlebar, ABataev, tra
Subscribers: guansong, andreybokhanko, tcramer, mkuron, cfe-commits, arpith-jacob, carlo.bertolli, caomhin
Differential Revision: https://reviews.llvm.org/D18171
llvm-svn: 275645
Extend the __declspec(dll*) attribute to cover ObjC interfaces. This was
requested by Microsoft for their ObjC support. Cover both import and export.
This only adds the semantic analysis portion of the support, code-generation
still remains outstanding. Add some basic initial documentation on the
attributes that were previously empty. Tweak the previous tests to use the
relative expected-warnings to make the tests easier to read.
llvm-svn: 275610
This changes the CompilerInstance::createOutputFile function to return
a std::unique_ptr<llvm::raw_ostream>, rather than an llvm::raw_ostream
implicitly owned by the CompilerInstance. This in most cases required that
I move ownership of the output stream to the relevant ASTConsumer.
The motivation for this change is to allow BackendConsumer to be a client
of interfaces such as D20268 which take ownership of the output stream.
Differential Revision: http://reviews.llvm.org/D21537
llvm-svn: 275507
passed on the command line but never actually used. We consider a (top-level)
module to be used if any part of it is imported, either by the current
translation unit, or by any part of a top-level module that is itself used.
(Put another way, a module is used if an implicit modules build would have
loaded its .pcm file.)
llvm-svn: 275481
This patch implements PR#22821.
Taking the address of a packed member is dangerous since the reduced
alignment of the pointee is lost. This can lead to memory alignment
faults in some architectures if the pointer value is dereferenced.
This change adds a new warning to clang emitted when taking the address
of a packed member. A packed member is either a field/data member
declared as attribute((packed)) or belonging to a struct/class
declared as such. The associated flag is -Waddress-of-packed-member.
Conversions (either implicit or via a valid casting) to pointer types
with lower or equal alignment requirements (e.g. void* or char*)
silence the warning.
This change also adds a new error diagnostic when the user attempts to
bind a reference to a packed member, regardless of the alignment.
Differential Revision: https://reviews.llvm.org/D20561
llvm-svn: 275417
This patch is to implement sema and parsing for 'target parallel for simd' pragma.
Differential Revision: http://reviews.llvm.org/D22096
llvm-svn: 275365
-fxray-instrument: enables XRay annotation of IR
-fxray-instruction-threshold: configures the threshold for function size (looking at IR instructions), and allow LLVM to decide whether to add the nop sleds later on in the process.
Also implements the related xray_always_instrument and xray_never_instrument function attributes.
Patch by Dean Michael Berris.
llvm-svn: 275330
This encourages checkers to make logical decisions depending on
value of which region was the symbol under consideration
introduced to denote.
A similar technique is already used in a couple of checkers;
they were modified to call the new method.
Differential Revision: http://reviews.llvm.org/D22242
llvm-svn: 275290
http://reviews.llvm.org/D21904
This patch is similar to the implementation of 'private' clause: it adds a list of private pointers to be used within the target data region to store the device pointers returned by the runtime.
Please refer to the following document for a full description of what the runtime witll return in this case (page 10 and 11):
https://github.com/clang-omp/OffloadingDesign
I am happy to answer any question related to the runtime interface to help reviewing this patch.
llvm-svn: 275271
This is to allow distributed build systems, that do not preserve time stamps, to use PCH files.
Second and last part of the patch proposed at:
Differential Revision: http://reviews.llvm.org/D20867
llvm-svn: 275267
We need to mark the appropriate bits in ThrowInfo and HandlerType so
that the personality routine can correctly handle qualification
conversions.
llvm-svn: 275154
Summary:
return llvm::Expected<> to carry error status and error information.
This is the first step towards introducing "Error" into tooling::Replacements.
Reviewers: djasper, klimek
Subscribers: ioeric, klimek, cfe-commits
Differential Revision: http://reviews.llvm.org/D21601
llvm-svn: 275062
- Changes diagnostics for Blocks to be implicitly
const qualified OpenCL v2.0 s6.12.5.
- Added and unified diagnostics of some OpenCL special types:
blocks, images, samplers, pipes. These types are intended for use
with the OpenCL builtin functions only and, therefore, most regular
uses are not allowed including assignments, arithmetic operations,
pointer dereferencing, etc.
Review: http://reviews.llvm.org/D21989
llvm-svn: 275061
MASM (ML.exe and ML64.exe) and older versions of MSVC (CL.exe) support a
flag called /Zd which is more-or-less -gline-tables-only.
It seems nicer to support this flag instead of exposing
-gline-tables-only.
llvm-svn: 274991
Add OCL option -cl-no-signed-zeros to driver options.
Also added to opencl.cl testcases.
Patch by Aaron En Ye Shi.
Differential Revision: http://reviews.llvm.org/D22067
llvm-svn: 274923
OpenCL s6.6: "Access qualifier must be used with image object arguments
of kernels and of user-defined functions [...] If no qualifier is
provided, read_only is assumed".
This does not define the behavior for image types used in typedef
declaration, but following the spec logic, we should allow access
qualifiers specification in typedefs, e.g.:
typedef write_only image1d_t img1d_wo;
Unlike cv-qualifiers, user cannot add access qualifier to a typedef
type, i.e. this is not allowed:
typedef image1d_t img1d; // note: previously declared 'read_only' here
void foo(write_only img1d im) {} // error: multiple access qualifier
Patch by Andrew Savonichev.
Reviewers: Anastasia Stulova.
Differential revision: http://reviews.llvm.org/D20948
llvm-svn: 274858
Original commit message:
"Add postorder traversal support to the RecursiveASTVisitor.
This feature needs to be explicitly enabled by overriding shouldTraversePostOrder()
as it has performance drawbacks for the iterative Stmt-traversal.
Patch by Raphael Isemann!
Reviewed by Richard Smith and Benjamin Kramer."
llvm-svn: 274830
This proposed patch adds crude handling of atomics to the static analyzer.
Rather than ignore AtomicExprs, as we now do, this patch causes the analyzer
to escape the arguments. This is imprecise -- and we should model the
expressions fully in the future -- but it is less wrong than ignoring their
effects altogether.
This is rdar://problem/25353187
Differential Revision: http://reviews.llvm.org/D21667
llvm-svn: 274816
Summary:
Raise an error if you're using a CUDA installation that's too old for
the requested architectures. In practice, this means that you need a
CUDA 8 install to compile for sm_6*.
Reviewers: tra
Subscribers: cfe-commits
Differential Revision: http://reviews.llvm.org/D21869
llvm-svn: 274781
The builtin was renamed in r274770. But __syncthreads is part of our
user-facing API, so we need to keep the name as-is.
Patch by Justin Bogner.
llvm-svn: 274780
Summary:
Currently our handling of CUDA architectures is scattered all around
clang. This patch centralizes it.
A key advantage of this centralization is that you can now write a C++
switch on e.g. CudaArch and get a compile error if you don't handle one
of the enum values.
Reviewers: tra
Subscribers: cfe-commits
Differential Revision: http://reviews.llvm.org/D21867
llvm-svn: 274681
Samuel Antao