This patch implements the teams directive for the NVPTX backend. It is different from the host code generation path as it:
Does not call kmpc_fork_teams. All necessary teams and threads are started upon touching the target region, when launching a CUDA kernel, and their execution is coordinated through sequential and parallel regions within the target region.
Does not call kmpc_push_num_teams even if a num_teams of thread_limit clause is present. Setting the number of teams and the thread limit is implemented by the nvptx-related runtime.
Please note that I am now passing a Clang Expr * to emitPushNumTeams instead of the originally chosen llvm::Value * type. The reason for that is that I want to avoid emitting expressions for num_teams and thread_limit if they are not needed in the target region.
http://reviews.llvm.org/D17963
llvm-svn: 265304
For better support of some specific GNU extensions some extra
transformation of AST nodes were introduced. These transformations are
very hard to handle. The code is improved in handling of these
extensions by using captured expressions construct.
llvm-svn: 264709
Solution unifies interface of RegionCodeGenTy type to allow insert
runtime-specific code before/after main codegen action defined in
CGStmtOpenMP.cpp file. Runtime should not define its own RegionCodeGenTy
for general OpenMP directives, but must be allowed to insert its own
(required) code to support target specific codegen.
llvm-svn: 264700
Solution unifies interface of RegionCodeGenTy type to allow insert
runtime-specific code before/after main codegen action defined in
CGStmtOpenMP.cpp file. Runtime should not define its own RegionCodeGenTy
for general OpenMP directives, but must be allowed to insert its own
(required) code to support target specific codegen.
llvm-svn: 264576
Solution unifies interface of RegionCodeGenTy type to allow insert
runtime-specific code before/after main codegen action defined in
CGStmtOpenMP.cpp file. Runtime should not define its own RegionCodeGenTy
for general OpenMP directives, but must be allowed to insert its own
(required) code to support target specific codegen.
llvm-svn: 264569
OpenMP 4.0 allows to define custom reduction operations using '#pragma
omp declare reduction' construct. Patch allows to use this custom
defined reduction operations in 'reduction' clauses.
llvm-svn: 263701
OpenMP 4.5 allows privatization of non-static data members in OpenMP
constructs. Patch adds proper codegen support for data members in
'linear' clause
llvm-svn: 263003
This patch provide basic implementation of codegen for teams directive, excluding all clauses except dist_schedule. It also fixes parts of AST reader/writer to enable correct pre-compiled header handling.
http://reviews.llvm.org/D17170
llvm-svn: 262832
This patch provide basic implementation of codegen for teams directive, excluding all clauses except dist_schedule. It also fixes parts of AST reader/writer to enable correct pre-compiled header handling.
http://reviews.llvm.org/D17170
llvm-svn: 262741
Add code generation support for firstprivate and private clauses of teams on the host. Add extensive regression tests including lambda functions and vla testing.
http://reviews.llvm.org/D17582
llvm-svn: 262663
Summary:
This patch implements the launching of a target region in the presence of a nested teams region, i.e calls tgt_target_teams with the required arguments gathered from the enclosed teams directive.
The actual codegen of the region enclosed by the teams construct will be contributed in a separate patch.
Reviewers: hfinkel, arpith-jacob, kkwli0, carlo.bertolli, ABataev
Subscribers: cfe-commits, caomhin, fraggamuffin
Differential Revision: http://reviews.llvm.org/D17019
llvm-svn: 262625
OpenMP 4.5 allows to privatize data members of current class in member
functions. Patch adds initial support for privatization of data members
in 'linear' clause, no codegen support.
llvm-svn: 262578
OpenMP 4.5 allows to privatize non-static data members of current class
in non-static member functions. Patch supports codegen for non-static
data members in 'reduction' clauses.
llvm-svn: 262460
OpenMP 4.5 allows to privatize non-static member decls in non-static
member functions. Patch captures such decls by reference in general (for
bitfields, by value) and then operates with this capture. For bitfields,
at the end of codegen for lastprivates original bitfield is updated with the value of captured copy.
llvm-svn: 261824
Patch fixes bug with codegen for lastprivate loop counters. Also it may
improve performance for lastprivates calculations in some cases.
llvm-svn: 261209
Expressions inside 'schedule'|'dist_schedule' clause must be captured in
combined directives to avoid possible crash during codegen. Patch
improves handling of such constructs
llvm-svn: 260954
Sync barrier will be emitted after generation of firstprivate variables
only if one of the firstprivate vars is used in lastprivate clause.
llvm-svn: 260877
OMPCapturedExprDecl allows caopturing not only of fielddecls, but also
other expressions. It also allows to simplify codegen for several
clauses.
llvm-svn: 260492
Codegen for array sections/array subscripts worked only for expressions with arrays as base. Patch fixes codegen for bases with pointer/reference types.
llvm-svn: 259776
Summary:
This patch adds parsing + sema for the target parallel for directive along with testcases.
Reviewers: ABataev
Differential Revision: http://reviews.llvm.org/D16759
llvm-svn: 259654
Summary:
This patch adds parsing + sema for the target parallel directive and its clauses along with testcases.
Reviewers: ABataev
Differential Revision: http://reviews.llvm.org/D16553
Rebased to current trunk and updated test cases.
llvm-svn: 258832
Summary:
This patch adds parsing + sema for the defaultmap clause associated with the target directive (among others).
Reviewers: ABataev
Differential Revision: http://reviews.llvm.org/D16527
llvm-svn: 258817
If 'sections' directive has only one sub-section, the code for 'single'-based directive was emitted. Removed this codegen, because it causes crashes in different cases.
llvm-svn: 258495
This patch attempts to fix the regressions identified when the patch was committed initially.
Thanks to Michael Liao for identifying the fix in the offloading metadata generation
related with side effects in evaluation of function arguments.
llvm-svn: 256933
Summary:
In order to offloading work properly two things need to be in place:
- a descriptor with all the offloading information (device entry functions, and global variable) has to be created by the host and registered in the OpenMP offloading runtime library.
- all the device functions need to be emitted for the device and a convention has to be in place so that the runtime library can easily map the host ID of an entry point with the actual function in the device.
This patch adds support for these two things. However, only entry functions are being registered given that 'declare target' directive is not yet implemented.
About offloading descriptor:
The details of the descriptor are explained with more detail in http://goo.gl/L1rnKJ. Basically the descriptor will have fields that specify the number of devices, the pointers to where the device images begin and end (that will be defined by the linker), and also pointers to a the begin and end of table whose entries contain information about a specific entry point. Each entry has the type:
```
struct __tgt_offload_entry{
void *addr;
char *name;
int64_t size;
};
```
and will be implemented in a pre determined (ELF) section `.omp_offloading.entries` with 1-byte alignment, so that when all the objects are linked, the table is in that section with no padding in between entries (will be like a C array). The code generation ensures that all `__tgt_offload_entry` entries are emitted in the same order for both host and device so that the runtime can have the corresponding entries in both host and device in same index of the table, and efficiently implement the mapping.
The resulting descriptor is registered/unregistered with the runtime library using the calls `__tgt_register_lib` and `__tgt_unregister_lib`. The registration is implemented in a high priority global initializer so that the registration happens always before any initializer (that can potentially include target regions) is run.
The driver flag -omptargets= was created to specify a comma separated list of devices the user wants to support so that the new functionality can be exercised. Each device is specified with its triple.
About target codegen:
The target codegen is pretty much straightforward as it reuses completely the logic of the host version for the same target region. The tricky part is to identify the meaningful target regions in the device side. Unlike other programming models, like CUDA, there are no already outlined functions with attributes that mark what should be emitted or not. So, the information on what to emit is passed in the form of metadata in host bc file. This requires a new option to pass the host bc to the device frontend. Then everything is similar to what happens in CUDA: the global declarations emission is intercepted to check to see if it is an "interesting" declaration. The difference is that instead of checking an attribute, the metadata information in checked. Right now, there is only a form of metadata to pass information about the device entry points (target regions). A class `OffloadEntriesInfoManagerTy` was created to manage all the information and queries related with the metadata. The metadata looks like this:
```
!omp_offload.info = !{!0, !1, !2, !3, !4, !5, !6}
!0 = !{i32 0, i32 52, i32 77426347, !"_ZN2S12r1Ei", i32 479, i32 13, i32 4}
!1 = !{i32 0, i32 52, i32 77426347, !"_ZL7fstatici", i32 461, i32 11, i32 5}
!2 = !{i32 0, i32 52, i32 77426347, !"_Z9ftemplateIiET_i", i32 444, i32 11, i32 6}
!3 = !{i32 0, i32 52, i32 77426347, !"_Z3fooi", i32 99, i32 11, i32 0}
!4 = !{i32 0, i32 52, i32 77426347, !"_Z3fooi", i32 272, i32 11, i32 3}
!5 = !{i32 0, i32 52, i32 77426347, !"_Z3fooi", i32 127, i32 11, i32 1}
!6 = !{i32 0, i32 52, i32 77426347, !"_Z3fooi", i32 159, i32 11, i32 2}
```
The fields in each metadata entry are (in sequence):
Entry 1) an ID of the type of metadata - right now only zero is used meaning "OpenMP target region".
Entry 2) a unique ID of the device where the input source file that contain the target region lives.
Entry 3) a unique ID of the file where the input source file that contain the target region lives.
Entry 4) a mangled name of the function that encloses the target region.
Entries 5) and 6) line and column number where the target region was found.
Entry 7) is the order the entry was emitted.
Entry 2) and 3) are required to distinguish files that have the same function name.
Entry 4) is required to distinguish different instances of the same declaration (usually templated ones)
Entries 5) and 6) are required to distinguish the particular target region in body of the function (it is possible that a given target region is not an entry point - if clause can evaluate always to zero - and therefore we need to identify the "interesting" target regions. )
This patch replaces http://reviews.llvm.org/D12306.
Reviewers: ABataev, hfinkel, tra, rjmccall, sfantao
Subscribers: FBrygidyn, piotr.rak, Hahnfeld, cfe-commits
Differential Revision: http://reviews.llvm.org/D12614
llvm-svn: 256842
#pragma omp parallel needs an implicit barrier that is currently done by an explicit call to __kmpc_barrier. However, the runtime already ensures a barrier in __kmpc_fork_call which currently leads to two barriers per region per thread.
Differential Revision: http://reviews.llvm.org/D15561
llvm-svn: 255992
OpenMP codegen tried to emit the code for its constructs even if it was detected as a dead-code. Added checks to ensure that the code is emitted if the code is not dead.
llvm-svn: 255990
OpenMP 4.5 adds directives 'taskloop' and 'taskloop simd'. These directives support clause 'num_tasks'. Patch adds parsing/semantic analysis for this clause.
llvm-svn: 255008
