of '#pragma pack' in included files
This commit adds a new -Wpragma-pack warning. It warns in the following cases:
- When a translation unit is missing terminating #pragma pack (pop) directives.
- When entering an included file if the current alignment value as determined
by '#pragma pack' directives is different from the default alignment value.
- When leaving an included file that changed the state of the current alignment
value.
rdar://10184173
Differential Revision: https://reviews.llvm.org/D35484
llvm-svn: 308327
If taskloop directive has no associated nogroup clause, it must emitted
inside implicit taskgroup block. Runtime supports it, but we need to
generate implicit taskgroup block explicitly to support future
reductions codegen.
llvm-svn: 307822
Added checks for the reduction clauses in the taskloop directives:
1. Only addressable items must be used in reduction clauses.
2. Reduction clauses cannot be used with nogroup clauses.
llvm-svn: 307693
Combined directives like 'target parallel' have two captured statements.
Sema has to check the right one from the right direction.
Previously, Sema::IsOpenMPCapturedByRef would return false for mapped
scalars on combined directives. This results in a wrong signature of
the outlined function which triggers an assertion:
void llvm::CallInst::init(llvm::FunctionType *, llvm::Value *, ArrayRef<llvm::Value *>, ArrayRef<OperandBundleDef>, const llvm::Twine &): Assertion `(i >= FTy->getNumParams() || FTy->getParamType(i) == Args[i]->getType()) && "Calling a function with a bad signature!"' failed.
Fixes PR30975 (and PR31985). New function was taken from clang-ykt.
Differential Revision: https://reviews.llvm.org/D34888
llvm-svn: 306956
Currently, if the some of the parameters are captured by value, this
argument is converted to uintptr_t type and thus we loosing the debug
info about real type of the argument (captured variable):
```
void @.outlined_function.(uintptr %par);
...
%a = alloca i32
%a.casted = alloca uintptr
%cast = bitcast uintptr* %a.casted to i32*
%a.val = load i32, i32 *%a
store i32 %a.val, i32 *%cast
%a.casted.val = load uintptr, uintptr* %a.casted
call void @.outlined_function.(uintptr %a.casted.val)
...
```
To resolve this problem, in debug mode a speciall external wrapper
function is generated, that calls the outlined function with the correct
parameters types:
```
void @.wrapper.(uintptr %par) {
%a = alloca i32
%cast = bitcast i32* %a to uintptr*
store uintptr %par, uintptr *%cast
%a.val = load i32, i32* %a
call void @.outlined_function.(i32 %a)
ret void
}
void @.outlined_function.(i32 %par);
...
%a = alloca i32
%a.casted = alloca uintptr
%cast = bitcast uintptr* %a.casted to i32*
%a.val = load i32, i32 *%a
store i32 %a.val, i32 *%cast
%a.casted.val = load uintptr, uintptr* %a.casted
call void @.wrapper.(uintptr %a.casted.val)
...
```
llvm-svn: 306697
According to OpenMP 5.0 at least one 'map' or 'use_device_ptr' clause
must be specified for 'target data' construct. Patch adds support for
this feature.
llvm-svn: 304216
If the variable is marked as TLS variable and target device does not
support TLS, the error is emitted for the variable even if it is not
used in target regions. Patch fixes this and allows to use the values of
the TLS variables in target regions.
llvm-svn: 303768
Currently clang checks for default data sharing attributes only for
variables captured in OpenMP regions by reference. Patch adds checks for
variables captured by value.
llvm-svn: 303077
Summary:
First, getCurFunction looks through blocks and lambdas, which is wrong.
Inside a lambda, va_start should refer to the lambda call operator
prototype. This fixes PR32737.
Second, we shouldn't use any of the getCur* methods, because they look
through contexts that we don't want to look through (EnumDecl,
CapturedStmtDecl). We can use CurContext directly as the calling
context.
Finally, this code assumed that CallExprs would never appear outside of
code contexts (block, function, obj-c method), which is wrong. Struct
member initializers are an easy way to create and parse exprs in a
non-code context.
Reviewers: rsmith
Subscribers: cfe-commits
Differential Revision: https://reviews.llvm.org/D32761
llvm-svn: 302188
https://reviews.llvm.org/D32807
This patch allows the map modifier 'always' to be separated by the map type (to, from, tofrom) only by a whitespace, rather than strictly by a comma as in current trunk.
llvm-svn: 302031
CheckForIntOverflow used to implement a whitelist of top-level expressions to
send to the constant expression evaluator, which handled many more expressions
than the CheckForIntOverflow whitelist did.
llvm-svn: 301742
If some function template is instantiated during handling of OpenMP
code, currently it may cause crash of compiler because of trying of
capturing variables in non-capturing function scopes. Patch fixes this
bug.
llvm-svn: 301416
[OpenMP] Initial implementation of code generation for pragma 'distribute parallel for' on host
https://reviews.llvm.org/D29508
This patch makes the following additions:
It abstracts away loop bound generation code from procedures associated with pragma 'for' and loops in general, in such a way that the same procedures can be used for 'distribute parallel for' without the need for a full re-implementation.
It implements code generation for 'distribute parallel for' and adds regression tests. It includes tests for clauses.
It is important to notice that most of the clauses are implemented as part of existing procedures. For instance, firstprivate is already implemented for 'distribute' and 'for' as separate pragmas. As the implementation of 'distribute parallel for' is based on the same procedures, then we automatically obtain implementation for such clauses without the need to add new code. However, this requires regression tests that verify correctness of produced code.
llvm-svn: 301340
https://reviews.llvm.org/D29508
This patch makes the following additions:
1. It abstracts away loop bound generation code from procedures associated with pragma 'for' and loops in general, in such a way that the same procedures can be used for 'distribute parallel for' without the need for a full re-implementation.
2. It implements code generation for 'distribute parallel for' and adds regression tests. It includes tests for clauses.
It is important to notice that most of the clauses are implemented as part of existing procedures. For instance, firstprivate is already implemented for 'distribute' and 'for' as separate pragmas. As the implementation of 'distribute parallel for' is based on the same procedures, then we automatically obtain implementation for such clauses without the need to add new code. However, this requires regression tests that verify correctness of produced code.
Looking forward to comments.
llvm-svn: 301223
https://reviews.llvm.org/D32237
This patch prepares sema with additional fields to support all those composite and combined constructs of OpenMP that include pragma 'distribute' and 'for', such as 'distribute parallel for'. It also extends the regression tests for 'distribute parallel for' and adds a new one.
llvm-svn: 300802
If the type of the captured variable is a pointer(s) to variably
modified type, this type was not processed correctly. Need to drill into
the type, find the innermost variably modified array type and convert it
to canonical parameter type.
llvm-svn: 299868
checkNestingOfRegions uses CancelRegion to determine whether cancel and
cancellation point are valid in the given nesting. This leads to unuseful
diagnostics if CancelRegion is invalid. The given test case has produced:
region cannot be closely nested inside 'parallel' region
As a solution, introduce checkCancelRegion and call it first to get the
expected error:
one of 'for', 'parallel', 'sections' or 'taskgroup' is expected
Differential Revision: https://reviews.llvm.org/D30135
llvm-svn: 295808
With tasks, the cancel may happen in another task. This has a different
region info which means that we can't find it here.
Differential Revision: https://reviews.llvm.org/D30091
llvm-svn: 295474
This resolves a deadlock with the cancel directive when there is no explicit
cancellation point. In that case, the implicit barrier acts as cancellation
point. After removing the barrier after cancel, the now unmatched barrier for
the explicit cancellation point has to go as well.
This has probably worked before rL255992: With the calls for the explicit
barrier, it was sure that all threads passed a barrier before exiting.
Reported by Simon Convent and Joachim Protze!
Differential Revision: https://reviews.llvm.org/D30088
llvm-svn: 295473
This patch implements codegen for the reduction clause on
any teams construct for elementary data types. It builds
on parallel reductions on the GPU. Subsequently,
the team master writes to a unique location in a global
memory scratchpad. The last team to do so loads and
reduces this array to calculate the final result.
This patch emits two helper functions that are used by
the OpenMP runtime on the GPU to perform reductions across
teams.
Patch by Tian Jin in collaboration with Arpith Jacob
Reviewers: ABataev
Differential Revision: https://reviews.llvm.org/D29879
llvm-svn: 295335
This patch implements codegen for the reduction clause on
any parallel construct for elementary data types. An efficient
implementation requires hierarchical reduction within a
warp and a threadblock. It is complicated by the fact that
variables declared in the stack of a CUDA thread cannot be
shared with other threads.
The patch creates a struct to hold reduction variables and
a number of helper functions. The OpenMP runtime on the GPU
implements reduction algorithms that uses these helper
functions to perform reductions within a team. Variables are
shared between CUDA threads using shuffle intrinsics.
An implementation of reductions on the NVPTX device is
substantially different to that of CPUs. However, this patch
is written so that there are minimal changes to the rest of
OpenMP codegen.
The implemented design allows the compiler and runtime to be
decoupled, i.e., the runtime does not need to know of the
reduction operation(s), the type of the reduction variable(s),
or the number of reductions. The design also allows reuse of
host codegen, with appropriate specialization for the NVPTX
device.
While the patch does introduce a number of abstractions, the
expected use case calls for inlining of the GPU OpenMP runtime.
After inlining and optimizations in LLVM, these abstractions
are unwound and performance of OpenMP reductions is comparable
to CUDA-canonical code.
Patch by Tian Jin in collaboration with Arpith Jacob
Reviewers: ABataev
Differential Revision: https://reviews.llvm.org/D29758
llvm-svn: 295333
This patch implements codegen for the reduction clause on
any parallel construct for elementary data types. An efficient
implementation requires hierarchical reduction within a
warp and a threadblock. It is complicated by the fact that
variables declared in the stack of a CUDA thread cannot be
shared with other threads.
The patch creates a struct to hold reduction variables and
a number of helper functions. The OpenMP runtime on the GPU
implements reduction algorithms that uses these helper
functions to perform reductions within a team. Variables are
shared between CUDA threads using shuffle intrinsics.
An implementation of reductions on the NVPTX device is
substantially different to that of CPUs. However, this patch
is written so that there are minimal changes to the rest of
OpenMP codegen.
The implemented design allows the compiler and runtime to be
decoupled, i.e., the runtime does not need to know of the
reduction operation(s), the type of the reduction variable(s),
or the number of reductions. The design also allows reuse of
host codegen, with appropriate specialization for the NVPTX
device.
While the patch does introduce a number of abstractions, the
expected use case calls for inlining of the GPU OpenMP runtime.
After inlining and optimizations in LLVM, these abstractions
are unwound and performance of OpenMP reductions is comparable
to CUDA-canonical code.
Patch by Tian Jin in collaboration with Arpith Jacob
Reviewers: ABataev
Differential Revision: https://reviews.llvm.org/D29758
llvm-svn: 295319
https://reviews.llvm.org/D29501
It looks like I forgot to remove a FIXME comment with the associated statement. The test does not need it and it gives the wrong impression of being an incomplete test.
llvm-svn: 294195
Alex Lorenz