If the file in line directive does not exist on the system we need, to
use the original file to get its file id.
Differential Revision: https://reviews.llvm.org/D97945
If the mapped structure has data members, which have 'default' mappers,
need to map these members individually using their 'default' mappers.
Differential Revision: https://reviews.llvm.org/D92195
OpenMP 5.0 removed a lot of restriction for overlapped mapped items
comparing to OpenMP 4.5. Patch restricts the checks for overlapped data
mappings only for OpenMP 4.5 and less and reorders mapping of the
arguments so, that present and alloc mappings are processed first and
then all others.
Differential Revision: https://reviews.llvm.org/D86119
The tile directive is in OpenMP's Technical Report 8 and foreseeably will be part of the upcoming OpenMP 5.1 standard.
This implementation is based on an AST transformation providing a de-sugared loop nest. This makes it simple to forward the de-sugared transformation to loop associated directives taking the tiled loops. In contrast to other loop associated directives, the OMPTileDirective does not use CapturedStmts. Letting loop associated directives consume loops from different capture context would be difficult.
A significant amount of code generation logic is taking place in the Sema class. Eventually, I would prefer if these would move into the CodeGen component such that we could make use of the OpenMPIRBuilder, together with flang. Only expressions converting between the language's iteration variable and the logical iteration space need to take place in the semantic analyzer: Getting the of iterations (e.g. the overload resolution of `std::distance`) and converting the logical iteration number to the iteration variable (e.g. overload resolution of `iteration + .omp.iv`). In clang, only CXXForRangeStmt is also represented by its de-sugared components. However, OpenMP loop are not defined as syntatic sugar. Starting with an AST-based approach allows us to gradually move generated AST statements into CodeGen, instead all at once.
I would also like to refactor `checkOpenMPLoop` into its functionalities in a follow-up. In this patch it is used twice. Once for checking proper nesting and emitting diagnostics, and additionally for deriving the logical iteration space per-loop (instead of for the loop nest).
Differential Revision: https://reviews.llvm.org/D76342
Even code in target and declare target regions might not be emitted.
With this patch we delay more diagnostics and use laziness and linkage
to determine if a function is emitted (for the device). Note that we
still eagerly emit diagnostics for target regions, unfortunately, see
the TODO for the reason.
This hopefully fixes PR48933.
Reviewed By: JonChesterfield
Differential Revision: https://reviews.llvm.org/D95928
Type errors in function declarations were not (always) diagnosed prior
to this patch. Furthermore, certain remarks did not get associated
properly which caused them to be emitted multiple times.
Reviewed By: JonChesterfield
Differential Revision: https://reviews.llvm.org/D95912
The ability to specify alignment was recently added, and it's an
important property which we should ensure is set as expected by
Clang. (Especially before making further changes to Clang's code in
this area.) But, because it's on the end of the lines, the existing
tests all ignore it.
Therefore, update all the tests to also verify the expected alignment
for atomicrmw and cmpxchg. While I was in there, I also updated uses
of 'load atomic' and 'store atomic', and added the memory ordering,
where that was missing.
Link: https://lists.llvm.org/pipermail/llvm-dev/2020-October/146162.html "[RFC] FileCheck: (dis)allowing unused prefixes"
If a downstream project using lit needs time for transition,
add the following to `lit.local.cfg`:
```
from lit.llvm.subst import ToolSubst
fc = ToolSubst('FileCheck', unresolved='fatal')
config.substitutions.insert(0, (fc.regex, 'FileCheck --allow-unused-prefixes'))
```
Differential Revision: https://reviews.llvm.org/D95849
The number of iterations calculation was failing in some cases with more
than two collpased loops. Now the LoopIterationSpace selected matches
InitDependOnLC and CondDependOnLC.
Differential Revision: https://reviews.llvm.org/D95834
Whenever we enter a new OpenMP data environment we want to enter a
function to simplify reasoning. Later we probably want to remove the
entire specialization wrt. the if clause and pass the result to the
runtime, for now this should fix PR48686.
Reviewed By: ABataev
Differential Revision: https://reviews.llvm.org/D94315
Summary:
The custom mapper API did not previously support the mapping names added previously. This means they were not present if a user requested debugging information while using the mapper functions. This adds basic support for passing the mapped names to the runtime library.
Reviewers: jdoerfert
Differential Revision: https://reviews.llvm.org/D94806
_ExtInt gets stuck in the device-type-checking for __int128 if it is
between 65 and 128 bits inclusive. Anything larger or smaller was
permitted despite this, so this is simply enabling 65-128 bit _ExtInts.
_ExtInt is supported on all our current ABIs, but we stil use the
hasExtIntType in the target info to differentiate here so that it can be
disabled.
Combined with 'da98651 - Revert "DR2064:
decltype(E) is only a dependent', this change (5a391d3) caused verifier
errors when building Chromium. See https://crbug.com/1168494#c1 for a
reproducer.
Additionally it reverts changes that were dependent on this one, see
below.
> Following up on PR48517, fix handling of template arguments that refer
> to dependent declarations.
>
> Treat an id-expression that names a local variable in a templated
> function as being instantiation-dependent.
>
> This addresses a language defect whereby a reference to a dependent
> declaration can be formed without any construct being value-dependent.
> Fixing that through value-dependence turns out to be problematic, so
> instead this patch takes the approach (proposed on the core reflector)
> of allowing the use of pointers or references to (but not values of)
> dependent declarations inside value-dependent expressions, and instead
> treating template arguments as dependent if they evaluate to a constant
> involving such dependent declarations.
>
> This ends up affecting a bunch of OpenMP tests, due to OpenMP
> imprecisely handling instantiation-dependent constructs, bailing out
> early instead of processing dependent constructs to the extent possible
> when handling the template.
>
> Previously committed as 8c1f2d15b8, and
> reverted because a dependency commit was reverted.
This reverts commit 5a391d38ac.
It also restores clang/test/SemaCXX/coroutines.cpp to its state before
da986511fb.
Revert "[c++20] P1907R1: Support for generalized non-type template arguments of scalar type."
> Previously committed as 9e08e51a20, and
> reverted because a dependency commit was reverted. This incorporates the
> following follow-on commits that were also reverted:
>
> 7e84aa1b81 by Simon Pilgrim
> ed13d8c667 by me
> 95c7b6cadb by Sam McCall
> 430d5d8429 by Dave Zarzycki
This reverts commit 4b574008ae.
Revert "[msabi] Mangle a template argument referring to array-to-pointer decay"
> [msabi] Mangle a template argument referring to array-to-pointer decay
> applied to an array the same as the array itself.
>
> This follows MS ABI, and corrects a regression from the implementation
> of generalized non-type template parameters, where we "forgot" how to
> mangle this case.
This reverts commit 18e093faf7.
OMP_MAP_TARGET_PARAM flag is used to mark the data that shoud be passed
as arguments to the target kernels, nothing else. But the compiler still
marks the data with OMP_MAP_TARGET_PARAM flags even if the data is
passed to the data movement directives, like target data, target update
etc. This flag is just ignored for this directives and the compiler does
not need to emit it.
Reviewed By: cchen
Differential Revision: https://reviews.llvm.org/D91261
D94745 rewrites the `deviceRTLs` using OpenMP and compiles it by directly
calling the device compilation. `clang` crashes because entry in
`OffloadEntriesDeviceGlobalVar` is unintialized. Current design supposes the
device compilation can only be invoked after host compilation with the host IR
such that `clang` can initialize `OffloadEntriesDeviceGlobalVar` from host IR.
This avoids us using device compilation directly, especially when we only have
code wrapped into `declare target` which are all device code. The same issue
also exists for `OffloadEntriesInfoManager`.
In this patch, we simply initialized an entry if it is not in the maps. Not sure
we need an option to tell the device compiler that it is invoked standalone.
Reviewed By: jdoerfert
Differential Revision: https://reviews.llvm.org/D94871
to dependent declarations.
Treat an id-expression that names a local variable in a templated
function as being instantiation-dependent.
This addresses a language defect whereby a reference to a dependent
declaration can be formed without any construct being value-dependent.
Fixing that through value-dependence turns out to be problematic, so
instead this patch takes the approach (proposed on the core reflector)
of allowing the use of pointers or references to (but not values of)
dependent declarations inside value-dependent expressions, and instead
treating template arguments as dependent if they evaluate to a constant
involving such dependent declarations.
This ends up affecting a bunch of OpenMP tests, due to OpenMP
imprecisely handling instantiation-dependent constructs, bailing out
early instead of processing dependent constructs to the extent possible
when handling the template.
Previously committed as 8c1f2d15b8, and
reverted because a dependency commit was reverted.
Since isa can cause diagnostics we want it to be evaluated last to avoid
the "unknown isa" warning if the rest of the selector wouldn't match
anyway. That allows us to guard isa with arch properly.
Reviewed By: jhuber6
Differential Revision: https://reviews.llvm.org/D93785
The triple uses a bar "x86-64" instead of an underscore. Since we
have troubles accepting x86-64 as an identifier, we stick with
x86_64 in the frontend and translate it explicitly.
Reviewed By: tianshilei1992
Differential Revision: https://reviews.llvm.org/D93786
[libomptarget][amdgpu] Call into deviceRTL instead of ockl
Amdgpu codegen presently emits a call into ockl. The same functionality
is already present in the deviceRTL. Adds an amdgpu specific entry point
to avoid the dependency. This lets simple openmp code (specifically, that
which doesn't use libm) run without rocm device libraries installed.
Reviewed By: ronlieb
Differential Revision: https://reviews.llvm.org/D93356
ELF -cc1 -mrelocation-model pic will default to no semantic interposition plus
setting dso_local on default visibility external linkage definitions, so that
COFF, Mach-O and ELF output will be similar.
This patch makes tests immune to the differences.
For a default visibility external linkage definition, dso_local is set for ELF
-fno-pic/-fpie and COFF and Mach-O. Since default clang -cc1 for ELF is similar
to -fpic ("PIC Level" is not set), this nuance causes unneeded binary format differences.
To make emitted IR similar, ELF -cc1 -fpic will default to -fno-semantic-interposition,
which sets dso_local for default visibility external linkage definitions.
To make this flip smooth and enable future (dso_local as definition default),
this patch replaces (function) `define ` with `define{{.*}} `,
(variable/constant/alias) `= ` with `={{.*}} `, or inserts appropriate `{{.*}} `.
For a definition (of most linkage types), dso_local is set for ELF -fno-pic/-fpie
and COFF, but not for Mach-O. This nuance causes unneeded binary format differences.
This patch replaces (function) `define ` with `define{{.*}} `,
(variable/constant/alias) `= ` with `={{.*}} `, or inserts appropriate `{{.*}} `
if there is an explicit linkage.
* Clang will set dso_local for Mach-O, which is currently implied by TargetMachine.cpp. This will make COFF/Mach-O and executable ELF similar.
* Eventually I hope we can make dso_local the textual LLVM IR default (write explicit "dso_preemptable" when applicable) and -fpic ELF will be similar to everything else. This patch helps move toward that goal.
to dependent declarations.
Treat an id-expression that names a local variable in a templated
function as being instantiation-dependent.
This addresses a language defect whereby a reference to a dependent
declaration can be formed without any construct being value-dependent.
Fixing that through value-dependence turns out to be problematic, so
instead this patch takes the approach (proposed on the core reflector)
of allowing the use of pointers or references to (but not values of)
dependent declarations inside value-dependent expressions, and instead
treating template arguments as dependent if they evaluate to a constant
involving such dependent declarations.
This ends up affecting a bunch of OpenMP tests, due to OpenMP
imprecisely handling instantiation-dependent constructs, bailing out
early instead of processing dependent constructs to the extent possible
when handling the template.
If a GPU function is externally reachable we give up trying to find the
(unique) kernel it is called from. This can hinder optimizations. Emit a
remark and explain mitigation strategies.
Reviewed By: tianshilei1992
Differential Revision: https://reviews.llvm.org/D93439
The `assumes` directive is an OpenMP 5.1 feature that allows the user to
provide assumptions to the optimizer. Assumptions can refer to
directives (`absent` and `contains` clauses), expressions (`holds`
clause), or generic properties (`no_openmp_routines`, `ext_ABCD`, ...).
The `assumes` spelling is used for assumptions in the global scope while
`assume` is used for executable contexts with an associated structured
block.
This patch only implements the global spellings. While clauses with
arguments are "accepted" by the parser, they will simply be ignored for
now. The implementation lowers the assumptions directly to the
`AssumptionAttr`.
Reviewed By: ABataev
Differential Revision: https://reviews.llvm.org/D91980
The `assumes` directive is an OpenMP 5.1 feature that allows the user to
provide assumptions to the optimizer. Assumptions can refer to
directives (`absent` and `contains` clauses), expressions (`holds`
clause), or generic properties (`no_openmp_routines`, `ext_ABCD`, ...).
The `assumes` spelling is used for assumptions in the global scope while
`assume` is used for executable contexts with an associated structured
block.
This patch only implements the global spellings. While clauses with
arguments are "accepted" by the parser, they will simply be ignored for
now. The implementation lowers the assumptions directly to the
`AssumptionAttr`.
Reviewed By: ABataev
Differential Revision: https://reviews.llvm.org/D91980
Support present modifier in defaultmap by adding an extra dimension
for `ImplicitMap`. Therefore, we now create OMPMapClause in `ActOnOpenMPExecutableDirective`
based on both `maptype` and `maptype-modifier`.
Reviewed By: ABataev
Differential Revision: https://reviews.llvm.org/D92427
The variables used in atomic construct should be captured in outer
task-based regions implicitly. Otherwise, the compiler will crash trying
to find the address of the local variable.
Differential Revision: https://reviews.llvm.org/D92682
Compiler needs to convert some of the loop iteration
variables/conditions to different types for better codegen and it may
lead to spurious warning messages about implicit signed/unsigned
conversions.
Differential Revision: https://reviews.llvm.org/D92655
OpenMPIRBuilder::createParallel outlines the body region of the parallel
construct into a new function that accepts any value previously defined outside
the region as a function argument. This function is called back by OpenMP
runtime function __kmpc_fork_call, which expects trailing arguments to be
pointers. If the region uses a value that is not of a pointer type, e.g. a
struct, the produced code would be invalid. In such cases, make createParallel
emit IR that stores the value on stack and pass the pointer to the outlined
function instead. The outlined function then loads the value back and uses as
normal.
Reviewed By: jdoerfert, llitchev
Differential Revision: https://reviews.llvm.org/D92189
The dependency mechanism for C has been implemented, and we have rolled out
this to all internal users, didn't see crashy issues, we consider it is stable
enough.
Differential Revision: https://reviews.llvm.org/D89046
After fix for PR48174 the base pointer for pointer-based
array-sections/array-subscripts will be emitted as `&ptr[idx]`, but
actually it should be just `ptr`, i.e. the address stored in the ponter
to point correctly to the beginning of the array. Currently it may lead
to a crash in the runtime.
Differential Revision: https://reviews.llvm.org/D91805
Summary:
Add support for passing source locations to libomptarget runtime functions using the ident_t struct present in the rest of the libomp API. This will allow the runtime system to give much more insightful error messages and debugging values.
Reviewers: jdoerfert grokos
Differential Revision: https://reviews.llvm.org/D87946
Summary:
This patch adds support for passing in the original delcaration name in the source file to the libomptarget runtime. This will allow the runtime to provide more intelligent debugging messages. This patch takes the original expression parsed from the OpenMP map / update clause and provides a textual representation if it was explicitly mapped, otherwise it takes the name of the variable declaration as a fallback. The information in passed to the runtime in a global array of strings that matches the existing ident_t source location strings using ";name;filename;column;row;;"
Reviewers: jdoerfert
Differential Revision: https://reviews.llvm.org/D89802
The compiler should treat array subscript with base pointer as a first
pointer in complex data, it is used only for member expression with base
pointer.
Differential Revision: https://reviews.llvm.org/D91660
If the variable is implicitly firstprivatized in the inner task-based
region, it also must be firstprivatized in outer task-based regions.
Previously firstprivates were captured in tasks but later it was
optimized to reduce the memory usage. But still need to mark such
variables as implicit firstprivate in outer tasks.
Differential Revision: https://reviews.llvm.org/D91627
If the data member pointer is mapped, the compiler tries to optimize the
mapping of such data by discarding explicit mapping flags and trying to
emit combined data instead. In some cases, this optimization is not
quite correctly implemented and it leads to a program crash at the
runtime. Instead, if the data member is mapped, just emit it as is and
do not emit combined mapping flags for it.
Differential Revision: https://reviews.llvm.org/D91552
arguments.
* Adds 'nonnull' and 'dereferenceable(N)' to 'this' pointer arguments
* Gates 'nonnull' on -f(no-)delete-null-pointer-checks
* Introduces this-nonnull.cpp and microsoft-abi-this-nullable.cpp tests to
explicitly test the behavior of this change
* Refactors hundreds of over-constrained clang tests to permit these
attributes, where needed
* Updates Clang12 patch notes mentioning this change
Reviewed-by: rsmith, jdoerfert
Differential Revision: https://reviews.llvm.org/D17993
As Richard Smith pointed out in the review of D90123, both the C and C++
standard call it lvalue and rvalue, so let's stick to the same spelling
in Clang.
Need to check if there are map types for the components before trying to
access them when trying to modify type mappings for combined partial
mappings.
Differential Revision: https://reviews.llvm.org/D91370
In order not to modify the `tgt_target_data_update` information but still be
able to pass the extra information for non-contiguous map item (offset,
count, and stride for each dimension), this patch overload `arg` when
the maptype is set as `OMP_MAP_DESCRIPTOR`. The origin `arg` is for
passing the pointer information, however, the overloaded `arg` is an
array of descriptor_dim:
struct descriptor_dim {
int64_t offset;
int64_t count;
int64_t stride
};
and the array size is the same as dimension size. In addition, since we
have count and stride information in descriptor_dim, we can replace/overload the
`arg_size` parameter by using dimension size.
For supporting `stride` in array section, we use a dummy dimension in
descriptor to store the unit size. The formula for counting the stride
in dimension D_n: `unit size * (D_0 * D_1 ... * D_n-1) * D_n.stride`.
Demonstrate how it works:
```
double arr[3][4][5];
D0: { offset = 0, count = 1, stride = 8 } // offset, count, dimension size always be 0, 1, 1 for this extra dimension, stride is the unit size
D1: { offset = 0, count = 2, stride = 8 * 1 * 2 = 16 } // stride = unit size * (product of dimension size of D0) * D1.stride = 4 * 1 * 2 = 8
D2: { offset = 2, count = 2, stride = 8 * (1 * 5) * 1 = 40 } // stride = unit size * (product of dimension size of D0, D1) * D2.stride = 4 * 5 * 1 = 20
D3: { offset = 0, count = 2, stride = 8 * (1 * 5 * 4) * 2 = 320 } // stride = unit size * (product of dimension size of D0, D1, D2) * D3.stride = 4 * 25 * 2 = 200
// X here means we need to offload this data, therefore, runtime will transfer
// data from offset 80, 96, 120, 136, 400, 416, 440, 456
// Runtime patch: https://reviews.llvm.org/D82245
// OOOOO OOOOO OOOOO
// OOOOO OOOOO OOOOO
// XOXOO OOOOO XOXOO
// XOXOO OOOOO XOXOO
```
Reviewed By: ABataev
Differential Revision: https://reviews.llvm.org/D84192
Since C++11, the C++ standard has a forward progress guarantee
[intro.progress], so all such functions must have the `mustprogress`
requirement. In addition, from C11 and onwards, loops without a non-zero
constant conditional or no conditional are also required to make
progress (C11 6.8.5p6). This patch implements these attribute deductions
so they can be used by the optimization passes.
Differential Revision: https://reviews.llvm.org/D86841
Clang now asserts for the below case:
```
void clang::CodeGen::CGOpenMPRuntime::createOffloadEntriesAndInfoMetadata(): Assertion `std::get<0>(E) && "All ordered entries must exist!"' failed.
```
The reason why Clang hit the assert is because in
`emitTargetDataCalls`, both `BeginThenGen` and `BeginElseGen` call
`registerTargetRegionEntryInfo` and try to register the Entry in
OffloadEntriesTargetRegion with same key. If changing the expression in
if clause to any constant expression, then the assert disappear. (https://godbolt.org/z/TW7haj)
The assert itself is to avoid
user from accessing elements out of bound inside `OrderedEntries` in
`createOffloadEntriesAndInfoMetadata`.
In this patch, I add a check in `registerTargetRegionEntryInfo` to avoid
register the target region more than once.
A test case that triggers assert: https://godbolt.org/z/4cnGW8
Reviewed By: ABataev
Differential Revision: https://reviews.llvm.org/D90704
Previously we added support for target nowait, but target data nowait
has not been supported yet. In this patch, target data nowait will also be
wrapped into a task.
Reviewed By: jdoerfert
Differential Revision: https://reviews.llvm.org/D90099
[libomptarget][nvptx] Undef, weak shared variables
Shared variables on nvptx, and LDS on amdgcn, are uninitialized at
the start of kernel execution. Therefore create the variables with
undef instead of zeros, motivated in part by the amdgcn back end
rejecting LDS+initializer.
Common is zero initialized, which seems incompatible with shared. Thus
change them to weak, following the direction of
https://reviews.llvm.org/rG7b3eabdcd215
Reviewed By: jdoerfert
Differential Revision: https://reviews.llvm.org/D90248
Summary:
This patch adds support for passing in the original delcaration name in the
source file to the libomptarget runtime. This will allow the runtime to provide
more intelligent debugging messages. This patch takes the original expression
parsed from the OpenMP map / update clause and provides a textual
representation if it was explicitly mapped, otherwise it takes the name of the
variable declaration as a fallback. The information in passed to the runtime in
a global array of strings that matches the existing ident_t source location
strings using ";name;filename;column;row;;". See
clang/test/OpenMP/target_map_names.cpp for an example of the generated output
for a given map clause.
Reviewers: jdoervert
Differential Revision: https://reviews.llvm.org/D89802
In current implementation, if it requires an outer task, the mapper array will be privatized no matter whether it has mapper. In fact, when there is no mapper, the mapper array only contains number of nullptr. In the libomptarget, the use of mapper array is `if (mappers_array && mappers_array[i])`, which means we can directly set mapper array to nullptr if there is no mapper. This can avoid unnecessary data copy.
In this patch, the data privatization will not be emitted if the mapper array is nullptr. When it comes to the emit of task body, the nullptr will be used directly.
Reviewed By: jdoerfert
Differential Revision: https://reviews.llvm.org/D90101
The changes made in D88594 caused the test OpenMP/driver.c to fail on a 32-bit host becuase it was offloading to a 64-bit architecture by default. The offloading test was moved to a new file and a feature was added to the lit config to check for a 64-bit host.
Reviewed By: daltenty
Differential Revision: https://reviews.llvm.org/D89904
for which it matters.
This is a step towards separating checking for a constant initializer
(in which std::is_constant_evaluated returns true) and any other
evaluation of a variable initializer (in which it returns false).
The changes made in D88594 caused the test OpenMP/driver.c to fail on a 32-bit host becuase it was offloading to a 64-bit architecture by default. The offloading test was moved to a new file and a feature was added to the lit config to check for a 64-bit host.
Reviewed By: jdoerfert
Differential Revision: https://reviews.llvm.org/D89696
callee in constant evaluation.
We previously made a deep copy of function parameters of class type when
passing them, resulting in the destructor for the parameter applying to
the original argument value, ignoring any modifications made in the
function body. This also meant that the 'this' pointer of the function
parameter could be observed changing between the caller and the callee.
This change completely reimplements how we model function parameters
during constant evaluation. We now model them roughly as if they were
variables living in the caller, albeit with an artificially reduced
scope that covers only the duration of the function call, instead of
modeling them as temporaries in the caller that we partially "reparent"
into the callee at the point of the call. This brings some minor
diagnostic improvements, as well as significantly reduced stack usage
during constant evaluation.
callee in constant evaluation.
We previously made a deep copy of function parameters of class type when
passing them, resulting in the destructor for the parameter applying to
the original argument value, ignoring any modifications made in the
function body. This also meant that the 'this' pointer of the function
parameter could be observed changing between the caller and the callee.
This change completely reimplements how we model function parameters
during constant evaluation. We now model them roughly as if they were
variables living in the caller, albeit with an artificially reduced
scope that covers only the duration of the function call, instead of
modeling them as temporaries in the caller that we partially "reparent"
into the callee at the point of the call. This brings some minor
diagnostic improvements, as well as significantly reduced stack usage
during constant evaluation.
callee in constant evaluation.
We previously made a deep copy of function parameters of class type when
passing them, resulting in the destructor for the parameter applying to
the original argument value, ignoring any modifications made in the
function body. This also meant that the 'this' pointer of the function
parameter could be observed changing between the caller and the callee.
This change completely reimplements how we model function parameters
during constant evaluation. We now model them roughly as if they were
variables living in the caller, albeit with an artificially reduced
scope that covers only the duration of the function call, instead of
modeling them as temporaries in the caller that we partially "reparent"
into the callee at the point of the call. This brings some minor
diagnostic improvements, as well as significantly reduced stack usage
during constant evaluation.
References to different declarations of the same entity aren't different
values, so shouldn't have different representations.
Recommit of e6393ee813, most recently
reverted in 9a33f027ac due to a bug caused
by ObjCInterfaceDecls not propagating availability attributes along
their redeclaration chains; that bug was fixed in
e2d4174e9c.
Summary:
Replace the OpenMP Runtime Library functions used in CGOpenMPRuntimeGPU
for OpenMP device code generation with ones in OMPKinds.def and use
OMPIRBuilder for generating runtime calls. This allows us to
consolidate more OpenMP code generation into the OMPIRBuilder. Future
additions to the GPU runtime functions should now go in OMPKinds.def
Reviewers: jdoerfert
Subscribers: aaron.ballman cfe-commits guansong llvm-commits sstefan1 yaxunl
Tags: #OpenMP #LLVM #clang
Differential Revision: https://reviews.llvm.org/D88430
Summary:
This patch adds an error to Clang that detects if OpenMP offloading is used
between two architectures with incompatible pointer sizes. This ensures that
the data mapping can be done correctly and solves an issue in code generation
generating the wrong size pointer.
Reviewer: jdoerfert
Subscribers: cfe-commits delcypher guansong llvm-commits sstefan1 yaxunl
Tags: #OpenMP #Clang
Differential Revision: https://reviews.llvm.org/D88594
Summary:
This patch adds an error to Clang that detects if OpenMP offloading is
used between two architectures with incompatible pointer sizes. This
ensures that the data mapping can be done correctly and solves an issue
in code generation generating the wrong size pointer. This patch adds a
new lit substitution, %omp_powerpc_triple that, if the system is 32-bit or
64-bit, sets the powerpc triple accordingly. This was required to fix
some OpenMP tests that automatically populated the target architecture.
Reviewers: jdoerfert
Subscribers: cfe-commits guansong sstefan1 yaxunl delcypher
Tags: OpenMP clang LLVM
Differential Revision: https://reviews.llvm.org/D88594
Failing tests on Arm due to the tests automatically populating
incomatible pointer width architectures. Reverting until the tests are
updated. Failing tests:
OpenMP/distribute_parallel_for_num_threads_codegen.cpp
OpenMP/distribute_parallel_for_if_codegen.cpp
OpenMP/distribute_parallel_for_simd_if_codegen.cpp
OpenMP/distribute_parallel_for_simd_num_threads_codegen.cpp
OpenMP/target_teams_distribute_parallel_for_if_codegen.cpp
OpenMP/target_teams_distribute_parallel_for_simd_if_codegen.cpp
OpenMP/teams_distribute_parallel_for_if_codegen.cpp
OpenMP/teams_distribute_parallel_for_simd_if_codegen.cpp
This reverts commit 90eaedda9b.
Failing tests on Arm due to the tests automatically populating
incomatible pointer width architectures. Reverting until the tests are
updated. Failing tests:
OpenMP/distribute_parallel_for_num_threads_codegen.cpp
OpenMP/distribute_parallel_for_if_codegen.cpp
OpenMP/distribute_parallel_for_simd_if_codegen.cpp
OpenMP/distribute_parallel_for_simd_num_threads_codegen.cpp
OpenMP/target_teams_distribute_parallel_for_if_codegen.cpp
OpenMP/target_teams_distribute_parallel_for_simd_if_codegen.cpp
OpenMP/teams_distribute_parallel_for_if_codegen.cpp
OpenMP/teams_distribute_parallel_for_simd_if_codegen.cpp
This reverts commit 9d2378b591.
Summary:
Replace the OpenMP Runtime Library functions used in CGOpenMPRuntimeGPU
for OpenMP device code generation with ones in OMPKinds.def and use
OMPIRBuilder for generating runtime calls. This allows us to consolidate
more OpenMP code generation into the OMPIRBuilder. This patch also
invalidates specifying target architectures with conflicting pointer
sizes.
Reviewers: jdoerfert
Subscribers: aaron.ballman cfe-commits guansong llvm-commits sstefan1 yaxunl
Tags: #OpenMP #Clang #LLVM
Differential Revision: https://reviews.llvm.org/D88430
Summary:
This patch adds an error to Clang that detects if OpenMP offloading is used
between two architectures with incompatible pointer sizes. This ensures that
the data mapping can be done correctly and solves an issue in code generation
generating the wrong size pointer.
Reviewer: jdoerfert
Subscribers:
Tags: #OpenMP #Clang
Differential Revision:
References to different declarations of the same entity aren't different
values, so shouldn't have different representations.
Recommit of e6393ee813 with fixed handling
for weak declarations. We now look for attributes on the most recent
declaration when determining whether a declaration is weak. (Second
recommit with further fixes for mishandling of weak declarations. Our
behavior here is fundamentally unsound -- see PR47663 -- but this
approach attempts to not make things worse.)
Previously for nowait target, CG emitted a function call to `__tgt_target_nowait`, etc. However, in OpenMP RTL, these functions just directly call the no-nowait version, which means nowait is not working as expected.
OpenMP specification says a target is acutally a target task, which is an untied and detachable task. It is natural to go to the direction that generates a task for a nowait target. However, OpenMP task has a problem that it must be within to a parallel region; otherwise the task will be executed immediately. As a result, if we directly wrap to a regular task, the `target nowait` outside of a parallel region is still a synchronous version.
In D77609, I added the support for unshackled task in OpenMP RTL. Basically, unshackled task is a task that is not bound to any parallel region. So all nowait target will be tranformed into an unshackled task. In order to distinguish from regular task, a new flag bit is set for unshackled task. This flag will be used by RTL for later process.
Since all target tasks are allocated via `__kmpc_omp_target_task_alloc`, and in current `libomptarget`, `__kmpc_omp_target_task_alloc` just calls `__kmpc_omp_task_alloc`. Therefore, we can modify the flag in `__kmpc_omp_target_task_alloc` so that we don't need to modify the FE too much. If users choose to opt out the feature, they just need to use a RTL w/o support of unshackled threads.
As a result, in this patch, the `target nowait` region is simply wrapped into a regular task. Later once we have RTL support for unshackled tasks, the wrapped tasks can be executed by unshackled threads w/o changes in the FE.
Reviewed By: jdoerfert
Differential Revision: https://reviews.llvm.org/D78075
Need to fix a check for the variable if it is declared in the inner
OpenMP region to be able to firstprivatize it.
Reviewed By: jdoerfert
Differential Revision: https://reviews.llvm.org/D88240
Need to fix a check for the variable if it is declared in the inner
OpenMP region to be able to firstprivatize it.
Differential Revision: https://reviews.llvm.org/D88240
No need to make final copy from the firsptrivate/lastprivate copy to the original item if the item is a data memeber.
Firstprivate copy creates a copy by reference and the original item gets
updated correctly when updating the lastprivate shared variable.
Differential Revision: https://reviews.llvm.org/D88179
This patch fixes the problem that user-defined mapper array is not correctly privatized inside a task. This problem causes openmp/libomptarget/test/offloading/target_depend_nowait.cpp fails.
Differential Revision: https://reviews.llvm.org/D84470
References to different declarations of the same entity aren't different
values, so shouldn't have different representations.
Recommit of e6393ee813 with fixed
handling for weak declarations. We now look for attributes on the most
recent declaration when determining whether a declaration is weak.
Due to `omp begin/end declare variant`, OpenMP context selectors can be
nested. This patch adds initial support for this so we can use it for
target math variants. We should improve the detection of "equivalent"
scores and user conditions, we should also revisit the data structures
of the OMPTraitInfo object, however, both are not pressing issues right
now.
Reviewed By: JonChesterfield
Differential Revision: https://reviews.llvm.org/D85877
The base must be shared between the threads, threadprivates are not
allowed to be bases for array-like reductions.
Reviewed By: jdoerfert
Differential Revision: https://reviews.llvm.org/D85762
Need to map the component as TO instead of the literal, because need to
pass a reference to a component if the pointer is overaligned.
Reviewed By: jdoerfert
Differential Revision: https://reviews.llvm.org/D84887
Local vars, marked with pragma allocate, mustbe allocate by the call of
the runtime function and cannot be allocated as other local variables.
Instead, we allocate a space for the pointer in private record and store
the address, returned by kmpc_alloc call in this pointer.
So, for untied tasks
```
#pragma omp task untied
{
S s;
#pragma omp allocate(s) allocator(allocator)
s = x;
}
```
compiler generates something like this:
```
struct task_with_privates {
S *ptr;
};
void entry(task_with_privates *p) {
S *s = p->s;
switch(partid) {
case 1:
p->s = (S*)kmpc_alloc();
kmpc_omp_task();
br exit;
case 2:
*s = x;
kmpc_omp_task();
br exit;
case 2:
~S(s);
kmpc_free((void*)s);
br exit;
}
exit:
}
```
Reviewed By: jdoerfert
Differential Revision: https://reviews.llvm.org/D86558
NOTE: There is a mailing list discussion on this: http://lists.llvm.org/pipermail/llvm-dev/2019-December/137632.html
Complemantary to the assumption outliner prototype in D71692, this patch
shows how we could simplify the code emitted for an alignemnt
assumption. The generated code is smaller, less fragile, and it makes it
easier to recognize the additional use as a "assumption use".
As mentioned in D71692 and on the mailing list, we could adopt this
scheme, and similar schemes for other patterns, without adopting the
assumption outlining.
This third patch in the series removes version 5.0 string from
test cases making them check for default version. It also add test
cases for version 4.5.
Reviewed By: ABataev
Differential Revision: https://reviews.llvm.org/D85214
Many OpenMP Clang tests do not RUN for version 4.5 and the default
version. This second patch in the series handles test cases which
require updation in CHECK lines along with adding RUN lines for
the default version. It involves updating line number of pragmas.
Reviewed By: ABataev
Differential Revision: https://reviews.llvm.org/D85150
OpenMP 5.0 supports nested declare target regions. So, in general,it is
allow to mark a declarationas declare target with different device_type
or link type. Patch adds support for such kind of nesting.
Differential Revision: https://reviews.llvm.org/D86239
If the function is not marked exlicitly as declare target and it calls
function(s), marked as declare target device_type(host), these host-only
functions should not be dignosed as used in device mode, if the caller
function is not used in device mode too.
Differential Revision: https://reviews.llvm.org/D86164
If the declaration is used in the reduction clause, it is captured by
reference by default. But if the declaration is a pointer and it is a
base for array-like reduction, this declaration can be captured by
value, since the pointee is reduced but not the original declaration.
Differential Revision: https://reviews.llvm.org/D85321
When we implement OpenMP GPU reductions we use type punning a lot during
the shuffle and reduce operations. This is not always compatible with
language rules on aliasing. So far we generated TBAA which later allowed
to remove some of the reduce code as accesses and initialization were
"known to not alias". With this patch we avoid TBAA in this step,
hopefully for all accesses that we need to.
Verified on the reproducer of PR46156 and QMCPack.
Reviewed By: ABataev
Differential Revision: https://reviews.llvm.org/D86037
Many OpenMP Clang tests do not RUN for version 4.5 and the default
version. This first patch in the series only handles test cases
which do not require any modifications in the CHECK lines after
adding RUN lines for default version.
Reviewed By: ABataev
Differential Revision: https://reviews.llvm.org/D84844
Need to call getRawStmt() function instead, when trying to get inner
associated statement for the executable directive. Not all directives
use captured statements.
In untied tasks, need to allocate the space for local variales, declared
in task region, when the memory for task data is allocated. THe function
can be interrupted and we can exit from the function in untied task
switch. Need to keep the state of the local variables in this case.
Also, the compiler should not call cleanup when exiting in untied task
switch until the real exit out of the declaration scope is met during
execution.
Reviewed By: jdoerfert
Differential Revision: https://reviews.llvm.org/D84457
If the arguments are mapped, but are actually not used in the target
region, the compiler still adds attribute TGT_OMP_TARGET_PARAM for such
arguments. It makes the libomptarget to add such parameters to the list
of arguments, passed to the kernel at the runtime, and may lead to
incorrect results/crashes during execution.
Differential Revision: https://reviews.llvm.org/D85755
Summary:
In untied tasks, need to allocate the space for local variales, declared
in task region, when the memory for task data is allocated. THe function
can be interrupted and we can exit from the function in untied task
switch. Need to keep the state of the local variables in this case.
Also, the compiler should not call cleanup when exiting in untied task
switch until the real exit out of the declaration scope is met during
execution.
Reviewers: jdoerfert
Subscribers: yaxunl, guansong, cfe-commits, sstefan1, caomhin
Tags: #clang
Differential Revision: https://reviews.llvm.org/D84457
This test is bottlenecked by heavy regex use (~0.6s per FileCHeck run) with the
content that can't be further fragmented. Instead, the test body is moved into a
common .inc file and test execution has been split into four roughly equal
parts. This reduces wall time for the test from 14s to ~3.5s.
Differential Revision: https://reviews.llvm.org/D85695
Replace the `ident_t` handling in Clang with the methods offered by the
OMPIRBuilder. This cuts down on the clang code as well as the
differences between the two, making further transitions easier. Tests
have changed but there should not be a real functional change. The most
interesting difference is probably that we stop generating local ident_t
allocations for now and just use globals. Given that this happens only
with debug info, the location part of the `ident_t` is probably bigger
than the test anyway. As the location part is already a global, we can
avoid the allocation, memcpy, and store in favor of a constant global
that is slightly bigger. This can be revisited if there are
complications.
Reviewed By: ABataev
Differential Revision: https://reviews.llvm.org/D80735
The test file is the single longest test among clang's tests and ends up about
doubling the wall time of clang tests on machines with high number of cores.
The test appears to consist of multiple independent subtests and does not have
to be in one file. Splitting it into smaller parts reduces test time on my
machine from ~80s down to ~45.
Differential Revision: https://reviews.llvm.org/D85551
Summary:
Introduced OMPChildren class to handle all associated clauses, statement
and child expressions/statements. It allows to represent some directives
more correctly (like flush, depobj etc. with pseudo clauses, ordered
depend directives, which are standalone, and target data directives).
Also, it will make easier to avoid using of CapturedStmt in directives,
if required (atomic, tile etc. directives).
Also, it simplifies serialization/deserialization of the
executable/declarative directives.
Reduces number of allocation operations for mapper declarations.
Reviewers: jdoerfert
Subscribers: yaxunl, guansong, jfb, cfe-commits, sstefan1, aaron.ballman, caomhin
Tags: #clang
Differential Revision: https://reviews.llvm.org/D83261
Without this patch, the following example fails but shouldn't
according to OpenMP TR8:
```
#pragma omp target enter data map(alloc:i)
#pragma omp target data map(present, alloc: i)
{
#pragma omp target exit data map(delete:i)
} // fails presence check here
```
OpenMP TR8 sec. 2.22.7.1 "map Clause", p. 321, L23-26 states:
> If the map clause appears on a target, target data, target enter
> data or target exit data construct with a present map-type-modifier
> then on entry to the region if the corresponding list item does not
> appear in the device data environment an error occurs and the
> program terminates.
There is no corresponding statement about the exit from a region.
Thus, the `present` modifier should:
1. Check for presence upon entry into any region, including a `target
exit data` region. This behavior is already implemented correctly.
2. Should not check for presence upon exit from any region, including
a `target` or `target data` region. Without this patch, this
behavior is not implemented correctly, breaking the above example.
In the case of `target data`, this patch fixes the latter behavior by
removing the `present` modifier from the map types Clang generates for
the runtime call at the end of the region.
In the case of `target`, we have not found a valid OpenMP program for
which such a fix would matter. It appears that, if a program can
guarantee that data is present at the beginning of a `target` region
so that there's no error there, that data is also guaranteed to be
present at the end. This patch adds a comment to the runtime to
document this case.
Reviewed By: grokos, RaviNarayanaswamy, ABataev
Differential Revision: https://reviews.llvm.org/D84422
Provides AMDGCN and NVPTX specific specialization of getGPUWarpSize,
getGPUThreadID, and getGPUNumThreads methods. Adds tests for AMDGCN
codegen for these methods in generic and simd modes. Also changes the
precondition in InitTempAlloca to be slightly more permissive. Useful for
AMDGCN OpenMP codegen where allocas are created with a cast to an
address space.
Reviewed By: ABataev
Differential Revision: https://reviews.llvm.org/D84260
When we use the OpenMPIRBuilder for the parallel region we need to also
use it to get the thread ID (among other things) in the body. This is
because CGOpenMPRuntime::getThreadID() and
CGOpenMPRuntime::emitUpdateLocation implicitly assumes that if they are
called from within a parallel region there is a certain structure to the
code and certain members of the OMPRegionInfo are initialized. It might
make sense to initialize them even if we use the OpenMPIRBuilder but we
would preferably get rid of such state instead.
Bug reported by Anchu Rajendran Sudhakumari.
Depends on D82470.
Reviewed By: anchu-rajendran
Differential Revision: https://reviews.llvm.org/D82822
Need to map the base pointer for all directives, not only target
data-based ones.
The base pointer is mapped for array sections, array subscript, array
shaping and other array-like constructs with the base pointer. Also,
codegen for use_device_ptr clause was modified to correctly handle
mapping combination of array like constructs + use_device_ptr clause.
The data for use_device_ptr clause is emitted as the last records in the
data mapping array.
Reviewed By: ye-luo
Differential Revision: https://reviews.llvm.org/D84767
Need to map the base pointer for all directives, not only target
data-based ones.
The base pointer is mapped for array sections, array subscript, array
shaping and other array-like constructs with the base pointer. Also,
codegen for use_device_ptr clause was modified to correctly handle
mapping combination of array like constructs + use_device_ptr clause.
The data for use_device_ptr clause is emitted as the last records in the
data mapping array.
It applies only for global pointers.
Differential Revision: https://reviews.llvm.org/D84767
This patch implements Clang front end support for the OpenMP TR8
`present` motion modifier for `omp target update` directives. The
next patch in this series implements OpenMP runtime support.
Reviewed By: ABataev
Differential Revision: https://reviews.llvm.org/D84711
It was unclear what `isa` was supposed to mean so we did not provide any
traits for this context selector. With this patch we will allow *any*
string or identifier. We use the target attribute and target info to
determine if the trait matches. In other words, we will check if the
provided value is a target feature that is available (at the call site).
Fixes PR46338
Reviewed By: ABataev
Differential Revision: https://reviews.llvm.org/D83281
This patch implements Clang front end support for the OpenMP TR8
`present` motion modifier for `omp target update` directives. The
next patch in this series implements OpenMP runtime support.
Reviewed By: ABataev
Differential Revision: https://reviews.llvm.org/D84711
Summary:
If the variable is constrcutible, its copy is created by calling a
constructor. Such variables are duplicated and thus, must be captured.
Reviewers: jdoerfert
Subscribers: yaxunl, guansong, cfe-commits, sstefan1, caomhin
Tags: #clang
Differential Revision: https://reviews.llvm.org/D83909
This patch implements Clang front end support for the OpenMP TR8
`present` map type modifier. The next patch in this series implements
OpenMP runtime support.
This patch does not attempt to implement TR8 sec. 2.22.7.1 "map
Clause", p. 319, L14-16:
> If a map clause with a present map-type-modifier is present in a map
> clause, then the effect of the clause is ordered before all other
> map clauses that do not have the present modifier.
Compare to L10-11, which Clang does not appear to implement yet:
> For a given construct, the effect of a map clause with the to, from,
> or tofrom map-type is ordered before the effect of a map clause with
> the alloc, release, or delete map-type.
This patch also does not implement the `present` implicit-behavior for
`defaultmap` or the `present` motion-modifier for `target update`.
Reviewed By: ABataev
Differential Revision: https://reviews.llvm.org/D83061
Summary: This was missed when default version was upgraded to 5.0 (part of D81098)
Reviewers: saiislam, ABataev, jdoerfert
Reviewed By: saiislam
Subscribers: yaxunl, guansong, sstefan1, cfe-commits
Tags: #clang
Differential Revision: https://reviews.llvm.org/D84221
Summary:
Need to avoid an optimization for base pointer mapping for target data
directives.
Reviewers: jdoerfert, ye-luo
Subscribers: yaxunl, guansong, cfe-commits, sstefan1, caomhin
Tags: #clang
Differential Revision: https://reviews.llvm.org/D84182
Summary:
According to OpenMP 5.0, the restrictions for mapping of overlapped data
apply only for explicitly mapped data, there is no restriction for
implicitly mapped data just like in OpenMP 4.5.
Reviewers: jdoerfert
Subscribers: yaxunl, guansong, sstefan1, cfe-commits, caomhin
Tags: #clang
Differential Revision: https://reviews.llvm.org/D83398
Summary:
This patch adds more function attribute information to the runtime function definitions in OMPKinds.def. The goal is to provide sufficient information about OpenMP runtime functions to perform more optimizations on OpenMP code.
Reviewers: jdoerfert
Subscribers: aaron.ballman cfe-commits yaxunl guansong sstefan1 llvm-commits
Tags: #OpenMP #clang #LLVM
Differential Revision: https://reviews.llvm.org/D81031
This patch implements the code generation to use OpenMP 5.0 declare mapper (a.k.a. user-defined mapper) constructs.
Patch written by Lingda Li.
Differential Revision: https://reviews.llvm.org/D67833
llvm function is marked nounwind
This fixes cases where an invoke is emitted, despite the called llvm
function being marked nounwind, because ConstructAttributeList failed to
add the attribute to the attribute list. llvm optimization passes turn
invokes into calls and optimize away the exception handling code, but
it's better to avoid emitting the code in the front-end if the called
function is known not to raise an exception.
Differential Revision: https://reviews.llvm.org/D83906
Summary:
If user-defined reductions with the initializer are used with classes,
the compiler misses the constructor call when trying to create a private
copy of the reduction variable.
Reviewers: jdoerfert
Subscribers: cfe-commits, yaxunl, guansong, caomhin
Tags: #clang
Differential Revision: https://reviews.llvm.org/D83334
Summary:
If the original variable is marked for allocation in the different
address space using #pragma omp allocate, need to cast the allocated
variable to its original type with the original address space.
Otherwise, the compiler may crash trying to bitcast the type of the new
allocated variable to the original type in some cases, like passing this
variable as an argument in function calls.
Reviewers: jdoerfert
Subscribers: jholewinski, cfe-commits, yaxunl, guansong, caomhin
Tags: #clang
Differential Revision: https://reviews.llvm.org/D83696
During code generation we might change/add basic blocks so keeping a
list of them is fairly easy to break. Nested parallel regions were
enough. The new scheme does recompute the list of blocks to be outlined
once it is needed.
Reviewed By: anchu-rajendran
Differential Revision: https://reviews.llvm.org/D82722
Since D83271 we can optimize the GPU state machine to avoid spurious
call edges that increase the register usage of kernels. With this patch
we inform the user why and if this optimization is happening and when it
is not.
Reviewed By: ye-luo
Differential Revision: https://reviews.llvm.org/D83707
Summary:
NOTE: There is a mailing list discussion on this: http://lists.llvm.org/pipermail/llvm-dev/2019-December/137632.html
Complemantary to the assumption outliner prototype in D71692, this patch
shows how we could simplify the code emitted for an alignemnt
assumption. The generated code is smaller, less fragile, and it makes it
easier to recognize the additional use as a "assumption use".
As mentioned in D71692 and on the mailing list, we could adopt this
scheme, and similar schemes for other patterns, without adopting the
assumption outlining.
Reviewers: hfinkel, xbolva00, lebedev.ri, nikic, rjmccall, spatel, jdoerfert, sstefan1
Reviewed By: jdoerfert
Subscribers: thopre, yamauchi, kuter, fhahn, merge_guards_bot, hiraditya, bollu, rkruppe, cfe-commits, llvm-commits
Tags: #clang, #llvm
Differential Revision: https://reviews.llvm.org/D71739
Summary:
Need to privatize addresses of the captured variables when trying to
emit the body of the target data directive in no target codegen mode.
Reviewers: jdoerfert
Subscribers: yaxunl, guansong, cfe-commits, sstefan1, caomhin
Tags: #clang
Differential Revision: https://reviews.llvm.org/D83478
This implements the default(firstprivate) clause as defined in OpenMP
Technical Report 8 (2.22.4).
Reviewed By: jdoerfert, ABataev
Differential Revision: https://reviews.llvm.org/D75591
There are various runtime calls in the device runtime with unused, or
always fixed, arguments. This is bad for all sorts of reasons. Clean up
two before as we match them in OpenMPOpt now.
Reviewed By: JonChesterfield
Differential Revision: https://reviews.llvm.org/D83268
Making -g[no-]column-info opt out reduces the length of a typical CC1 command line.
Additionally, in a non-debug compile, we won't see -dwarf-column-info.
Assume bundle can have more than one entry with the same name,
but at least AlignmentFromAssumptionsPass::extractAlignmentInfo() uses
getOperandBundle("align"), which internally assumes that it isn't the
case, and happily crashes otherwise.
Minimal reduced reproducer: run `opt -alignment-from-assumptions` on
target datalayout = "e-m:e-p270:32:32-p271:32:32-p272:64:64-i64:64-f80:128-n8:16:32:64-S128"
target triple = "x86_64-unknown-linux-gnu"
%0 = type { i64, %1*, i8*, i64, %2, i32, %3*, i8* }
%1 = type opaque
%2 = type { i8, i8, i16 }
%3 = type { i32, i32, i32, i32 }
; Function Attrs: nounwind
define i32 @f(%0* noalias nocapture readonly %arg, %0* noalias %arg1) local_unnamed_addr #0 {
bb:
call void @llvm.assume(i1 true) [ "align"(%0* %arg, i64 8), "align"(%0* %arg1, i64 8) ]
ret i32 0
}
; Function Attrs: nounwind willreturn
declare void @llvm.assume(i1) #1
attributes #0 = { nounwind "reciprocal-estimates"="none" }
attributes #1 = { nounwind willreturn }
This is what we'd have with -mllvm -enable-knowledge-retention
This reverts commit c95ffadb24.
Summary:
Added support for dynamic memory allocation for globalized variables in
case if execution of target regions in parallel is required.
Reviewers: jdoerfert
Subscribers: jholewinski, yaxunl, guansong, sstefan1, cfe-commits, caomhin
Tags: #clang
Differential Revision: https://reviews.llvm.org/D82324
Summary:
NOTE: There is a mailing list discussion on this: http://lists.llvm.org/pipermail/llvm-dev/2019-December/137632.html
Complemantary to the assumption outliner prototype in D71692, this patch
shows how we could simplify the code emitted for an alignemnt
assumption. The generated code is smaller, less fragile, and it makes it
easier to recognize the additional use as a "assumption use".
As mentioned in D71692 and on the mailing list, we could adopt this
scheme, and similar schemes for other patterns, without adopting the
assumption outlining.
Reviewers: hfinkel, xbolva00, lebedev.ri, nikic, rjmccall, spatel, jdoerfert, sstefan1
Reviewed By: jdoerfert
Subscribers: yamauchi, kuter, fhahn, merge_guards_bot, hiraditya, bollu, rkruppe, cfe-commits, llvm-commits
Tags: #clang, #llvm
Differential Revision: https://reviews.llvm.org/D71739
Summary:
When -fopenmp option is specified then version 5.0 will be set as
default.
Reviewers: gregrodgers, jdoerfert, ABataev
Reviewed By: ABataev
Subscribers: pdhaliwal, yaxunl, guansong, sstefan1, cfe-commits
Tags: #clang
Differential Revision: https://reviews.llvm.org/D81098
Otherwise, it's painful to insert new code. There are many existing
examples in the same test file where the line numbers are not
hard-coded.
I intend to do the same for several other OpenMP tests, but I want to
be sure there are no objections before I spend time on it.
Reviewed By: jdoerfert
Differential Revision: https://reviews.llvm.org/D82224
Summary:
Compiler may erroneously treat current context in OpenMP pragmas as the
context where new type declaration/definition is allowed. But the
declartation/definition of the new types in OpenMP pragmas should not be
allowed.
Reviewers: jdoerfert
Subscribers: yaxunl, guansong, sstefan1, cfe-commits, caomhin
Tags: #clang
Differential Revision: https://reviews.llvm.org/D82019
Summary:
According to OpenMP 5.0, nonmonotonic modifier can be used with all
schedule kinds, not only dynamic and guided as in OpenMP 4.5.
Reviewers: jdoerfert
Subscribers: yaxunl, guansong, sstefan1, cfe-commits, caomhin
Tags: #clang
Differential Revision: https://reviews.llvm.org/D82026
Summary:
The OpenMP loops are normalized and transformed into the loops from 0 to
max number of iterations. In some cases, original scheme may lead to
overflow during calculation of number of iterations. If it is unknown,
if we can end up with overflow or not (the bounds are not constant and
we cannot define if there is an overflow), cast original type to the
unsigned.
Reviewers: jdoerfert
Subscribers: yaxunl, guansong, sstefan1, openmp-commits, cfe-commits, caomhin
Tags: #clang, #openmp
Differential Revision: https://reviews.llvm.org/D81881
Summary:
SYCL and OpenMP prohibits thread local storage in device code,
so this commit ensures that error is emitted for device code and not
emitted for host code when host target supports it.
Reviewers: jdoerfert, erichkeane, bader
Reviewed By: jdoerfert, erichkeane
Subscribers: guansong, riccibruno, ABataev, yaxunl, ebevhan, Anastasia, sstefan1, cfe-commits
Tags: #clang
Differential Revision: https://reviews.llvm.org/D81641
Summary:
According to OpenMP, During execution of an iteration of a worksharing-loop or a loop nest within a worksharing-loop, simd, or worksharing-loop SIMD region, a thread must not execute more than one ordered region corresponding to an ordered construct without a depend clause.
Need to report an error in this case.
Reviewers: jdoerfert
Subscribers: yaxunl, guansong, sstefan1, cfe-commits, caomhin
Tags: #clang
Differential Revision: https://reviews.llvm.org/D81951
Reland https://reviews.llvm.org/D76696
All known crashes have been fixed, another attemption.
We have rolled out this to all internal users for a while, didn't see
big issues, we consider it is stable enough.
Reviewed By: sammccall
Subscribers: rsmith, hubert.reinterpretcast, ebevhan, jkorous, arphaman, kadircet, usaxena95, cfe-commits
Tags: #clang
Differential Revision: https://reviews.llvm.org/D78350
Summary:
Added codegen for use_device_addr clause. The components of the list
items are mapped as a kind of RETURN components and then the returned
base address is used instead of the real address of the base declaration
used in the use_device_addr expressions.
Reviewers: jdoerfert
Subscribers: yaxunl, guansong, sstefan1, cfe-commits, caomhin
Tags: #clang
Differential Revision: https://reviews.llvm.org/D80730
Summary:
If the variables must be globalized in OpenMP mode (local automatic
variable, GPU compilation mode, the variable may escape its declaration
context by the reference or by the pointer), it should not be considered
as the NRVO candidate. Otherwise, incorrect the return value of the
function might not be updated.
Reviewers: jdoerfert
Subscribers: yaxunl, guansong, sstefan1, cfe-commits, caomhin
Tags: #clang
Differential Revision: https://reviews.llvm.org/D80936
Summary:
If the array subscript expression is type depent, its analysis must be
delayed before its instantiation.
Reviewers: jdoerfert
Subscribers: yaxunl, guansong, caomhin, cfe-commits
Tags: #clang
Differential Revision: https://reviews.llvm.org/D78637
Summary:
Nvidia PTX does not allow `.` to appear in identifiers, so OpenMP variant mangling now uses `$` to separate segments of the mangled name for variants of functions declared via `declare variant`.
Reviewers: jdoerfert, Hahnfeld
Reviewed By: jdoerfert
Subscribers: yaxunl, guansong, sstefan1, cfe-commits
Tags: #openmp, #clang
Differential Revision: https://reviews.llvm.org/D80439
Summary:
If the data member is mapped as an array section, need to emit the
pointer to the last element of this array section and use this pointer
as the highest element in partial struct data.
Reviewers: jdoerfert
Subscribers: yaxunl, guansong, sstefan1, cfe-commits, caomhin
Tags: #clang
Differential Revision: https://reviews.llvm.org/D81037
Summary:
Added initial codegen for 'affinity' clauses on task directives.
Emits next code:
```
kmp_task_affinity_info_t affs[<num_elems>];
void *td = __kmpc_task_alloc(..);
affs[<i>].base = &data_i;
affs[<i>].size = sizeof(data_i);
__kmpc_omp_reg_task_with_affinity(&loc, <gtid>, td, <num_elems>, affs);
```
The result returned by the call of `__kmpc_omp_reg_task_with_affinity`
function is ignored currently sincethe runtime currently ignores args
and returns 0 uncoditionally.
Reviewers: jdoerfert
Subscribers: yaxunl, guansong, sstefan1, llvm-commits, cfe-commits, caomhin
Tags: #clang, #llvm
Differential Revision: https://reviews.llvm.org/D80240
Summary: This changes Clang's generation of OpenMP runtime functions to use the types and functions defined in OpenMPKinds and OpenMPConstants. New OpenMP runtime function information should now be added to OMPKinds.def. This patch also changed the definitions of __kmpc_push_num_teams and __kmpc_copyprivate to match those found in the runtime.
Reviewers: jdoerfert
Reviewed By: jdoerfert
Subscribers: jfb, AndreyChurbanov, openmp-commits, fghanim, hiraditya, sstefan1, cfe-commits, llvm-commits
Tags: #openmp, #clang, #llvm
Differential Revision: https://reviews.llvm.org/D80222
Summary:
Do not ask size of type if it is dependent. ASTContext doesn't seem expecting
this.
Reviewers: jdoerfert, ABataev, bader
Reviewed By: ABataev
Subscribers: yaxunl, guansong, ebevhan, Anastasia, sstefan1, cfe-commits
Tags: #clang
Differential Revision: https://reviews.llvm.org/D80829
Summary:
Diagnostic is emitted if some declaration of unsupported type
declaration is used inside device code.
Memcpy operations for structs containing member with unsupported type
are allowed. Fixed crash on attempt to emit diagnostic outside of the
functions.
The approach is generalized between SYCL and OpenMP.
CUDA/OMP deferred diagnostic interface is going to be used for SYCL device.
Reviewers: rsmith, rjmccall, ABataev, erichkeane, bader, jdoerfert, aaron.ballman
Reviewed By: jdoerfert
Subscribers: guansong, sstefan1, yaxunl, mgorny, bader, ebevhan, Anastasia, cfe-commits
Tags: #clang
Differential Revision: https://reviews.llvm.org/D74387
We currently diagnose static data members directly contained in unnamed classes,
but we should also diagnose when they're in a class that is nested (directly or
indirectly) in an unnamed class. Do this by iterating up the list of parent
DeclContexts and checking if any is an unnamed class.
Similarly also check for function or method DeclContexts (which includes things
like blocks and openmp captured statements) as then the class is considered to
be a local class, which means static data members aren't allowed.
Differential Revision: https://reviews.llvm.org/D80295
Fixes PR45753
When a program that contains a loop to which both `omp parallel for`
pragma and `clang loop` pragma are associated is compiled with the
-fopenmp option, `clang loop` pragma did not take effect. The example
below should not be vectorized by the `clang loop` pragma but it was
actually vectorized. The cause is that `llvm.loop.vectorize.width`
was not output to the IR when -fopenmp is specified.
The fix attaches attributes if they exist for the loop.
[example.c]
```
int a[100], b[100];
void foo() {
#pragma omp parallel for
#pragma clang loop vectorize(disable)
for (int i = 0; i < 100; i++)
a[i] += b[i] * i;
}
```
[compile]
```
$ clang -O2 -fopenmp example.c -c -Rpass=vect
example.c:3:11: remark: vectorized loop (vectorization width: 4, interleaved count: 2) [-Rpass=loop-vectorize]
#pragma omp parallel for
^
```
[IR with -fopenmp]
```
$ clang -O2 exmaple.c -S -emit-llvm -mllvm -disable-llvm-optzns -o - -fopenmp | grep 'vectorize\.width'
```
[IR with -fno-openmp]
```
$ clang -O2 example.c -S -emit-llvm -mllvm -disable-llvm-optzns -o - -fno-openmp | grep 'vectorize\.width'
!7 = !{!"llvm.loop.vectorize.width", i32 1}
```
Differential Revision: https://reviews.llvm.org/D79921
Summary:
No need to generate inlined OpenMP region for variables captured in
lambdas or block decls, only for implicitly captured variables in the
OpenMP region.
Reviewers: jdoerfert
Subscribers: yaxunl, guansong, cfe-commits, caomhin
Tags: #clang
Differential Revision: https://reviews.llvm.org/D79966
If we're going to assume references are dereferenceable, we should also
assume they're aligned: otherwise, we can't actually dereference them.
See also D80072.
Differential Revision: https://reviews.llvm.org/D80166
Summary:
With recovery expr, it is possible that we have a value-dependent expr
within non-dependent context.
Reviewers: sammccall, jdoerfert
Subscribers: yaxunl, guansong, cfe-commits
Tags: #clang
Differential Revision: https://reviews.llvm.org/D80200
Summary:
Predefined allocators should not be mapped at all (they are just enumeric
constants). FOr user-defined allocators need to map the traits only as
firstprivates, the allocator itself is private.
At the beginning of the target region the user-defined allocatores must
be created and then destroyed at the end of the target region:
```
omp_allocator_handle_t my_allocator = __kmpc_init_allocator(<gtid>,
/*default memhandle*/ 0, <number_of_traits>, &<traits>);
...
call void @__kmpc_destroy_allocator(<gtid>, my_allocator);
```
Reviewers: jdoerfert, aaron.ballman
Subscribers: jholewinski, yaxunl, guansong, cfe-commits, caomhin
Tags: #clang
Differential Revision: https://reviews.llvm.org/D79257
Summary:
omp.h header file defines omp_null_allocator as a predefined allocator,
need to consider it also as a predefined allocator.
Reviewers: jdoerfert
Subscribers: jholewinski, yaxunl, guansong, cfe-commits, caomhin
Tags: #clang
Differential Revision: https://reviews.llvm.org/D79186
Summary:
This change fixes an aarch64-specific bug in the generation of the NDS and WDS values used to compute the signature of the vector functions out of OpenMP directives like `declare simd`. When the directive is used in conjunction with the `linear` clause, the size of the pointee must be used instead of the size of the pointer to compute NDS and WDS.
The code-fix is strictly related to the behavior for `linear`, but given that the only way we have to test the NDS and WDS values is to check the resulting `<vlen>` token in the mangled name of the vector function, the tests have been extended to cover all the possible values of WDS and NDS as defined in the ABI at https://github.com/ARM-software/abi-aa/tree/master/vfabia64.
Reviewers: ABataev, jdoerfert, andwar
Reviewed By: jdoerfert
Subscribers: yaxunl, kristof.beyls, guansong, danielkiss, cfe-commits
Tags: #clang
Differential Revision: https://reviews.llvm.org/D78969
Summary:
The linear parameter token in the mangling function must be multiplied
by the pointee size in bytes when the parameter is a pointer.
Reviewers: ABataev, andwar, jdoerfert
Subscribers: yaxunl, guansong, cfe-commits
Tags: #clang
Differential Revision: https://reviews.llvm.org/D78965
Summary:
Added basic support for 'task' modifier in the reduction clauses in
non-simd parallel and worksharing constructs.
Reviewers: jdoerfert
Subscribers: yaxunl, guansong, cfe-commits, caomhin
Tags: #clang
Differential Revision: https://reviews.llvm.org/D78738
Summary:
Patch forces codegen to use the new runtime functions for task reductions where
the issue with passing the address of the original variables to the UDR
initializers is fixed. Also, this patch is required for upcoming
support of task modifier inreduction clause.
Reviewers: jdoerfert
Subscribers: yaxunl, guansong, cfe-commits, caomhin
Tags: #clang
Differential Revision: https://reviews.llvm.org/D78733
Summary:
This patch contains 2 separate changes:
1) the initializer of a variable should play no part in decl "invalid" bit;
2) preserve the invalid initializer via recovery exprs;
With 1), we will regress the diagnostics (one big regression is that we loose
the "selected 'begin' function with iterator type" diagnostic in for-range stmt;
but with 2) together, we don't have regressions (the new diagnostics seems to be
improved).
Reviewers: sammccall
Reviewed By: sammccall
Subscribers: cfe-commits
Tags: #clang
Differential Revision: https://reviews.llvm.org/D78116
Summary:
The global variable should be captured in the region only if it was
privitized in the region or in any of the outer regions. Otherwise, it
should not be captured.
Reviewers: jdoerfert
Subscribers: yaxunl, guansong, cfe-commits, caomhin
Tags: #clang
Differential Revision: https://reviews.llvm.org/D77731
Summary:
According to the standard, variable-category is the optional part of the
defaultmap clause while the compiler always requires it. Turned it into
optional part.
Reviewers: jdoerfert
Subscribers: yaxunl, guansong, cfe-commits, caomhin
Tags: #clang
Differential Revision: https://reviews.llvm.org/D77751
By default, all traits in the OpenMP context selector have to match for
it to be acceptable. Though, we sometimes want a single property out of
multiple to match (=any) or no match at all (=none). We offer these
choices as extensions via
`implementation={extension(match_{all,any,none})}`
to the user. The choice will affect the entire context selector not only
the traits following the match property.
The first user will be D75788. There we can replace
```
#pragma omp begin declare variant match(device={arch(nvptx64)})
#define __CUDA__
#include <__clang_cuda_cmath.h>
// TODO: Hack until we support an extension to the match clause that allows "or".
#undef __CLANG_CUDA_CMATH_H__
#undef __CUDA__
#pragma omp end declare variant
#pragma omp begin declare variant match(device={arch(nvptx)})
#define __CUDA__
#include <__clang_cuda_cmath.h>
#undef __CUDA__
#pragma omp end declare variant
```
with the much simpler
```
#pragma omp begin declare variant match(device={arch(nvptx, nvptx64)}, implementation={extension(match_any)})
#define __CUDA__
#include <__clang_cuda_cmath.h>
#undef __CUDA__
#pragma omp end declare variant
```
Reviewed By: mikerice
Differential Revision: https://reviews.llvm.org/D77414
Implemented codegen for the iterator expression in the depend clauses.
Iterator construct is emitted the following way:
iterator(cnt1, cnt2, ...), in : <dep>
<TotalNumDeps> = <cnt1_size> * <cnt2_size> * ...;
kmp_depend_t deps[<TotalNumDeps>];
deps_counter = 0;
for (cnt1) {
for (cnt2) {
...
deps[deps_counter].base_addr = &<dep>;
deps[deps_counter].size = sizeof(<dep>);
deps[deps_counter].flags = in;
deps_counter += 1;
...
}
}
For depobj construct the codegen is very similar, but the memory is
allocated dynamically and added extra first item reserved for internal use.
Currently deferred diagnostic emitter checks variable decl in DeclRefExpr, which
causes infinite recursion for cases like long a = (long)&a;.
Deferred diagnostic emitter does not need check variable decls in DeclRefExpr
since reference of a variable does not cause emission of functions directly or
indirectly. Therefore there is no need to check variable decls in DeclRefExpr.
Differential Revision: https://reviews.llvm.org/D76937
Need to allow arrayshaping expression in a list of expressions, so use
ParseAssignmentExpression() when try to parse the base of the shaping
operation.
clauses.
Implemented codegen for array shaping operation in depend clauses. The
begin of the expression is the pointer itself, while the size of the
dependence data is the mukltiplacation of all dimensions in the array
shaping expression.
Summary:
Added basic representation and parsing/sema handling of array-shaping
operations. Array shaping expression is an expression of form ([s0]..[sn])base,
where s0, ..., sn must be a positive integer, base - a pointer. This
expression is a kind of cast operation that converts pointer expression
into an array-like kind of expression.
Reviewers: rjmccall, rsmith, jdoerfert
Subscribers: guansong, arphaman, cfe-commits, caomhin, kkwli0
Tags: #clang
Differential Revision: https://reviews.llvm.org/D74144
This is the second part loosely extracted from D71179 and cleaned up.
This patch provides semantic analysis support for `omp begin/end declare
variant`, mostly as defined in OpenMP technical report 8 (TR8) [0].
The sema handling makes code generation obsolete as we generate "the
right" calls that can just be handled as usual. This handling also
applies to the existing, albeit problematic, `omp declare variant
support`. As a consequence a lot of unneeded code generation and
complexity is removed.
A major purpose of this patch is to provide proper `math.h`/`cmath`
support for OpenMP target offloading. See PR42061, PR42798, PR42799. The
current code was developed with this feature in mind, see [1].
The logic is as follows:
If we have seen a `#pragma omp begin declare variant match(<SELECTOR>)`
but not the corresponding `end declare variant`, and we find a function
definition we will:
1) Create a function declaration for the definition we were about to generate.
2) Create a function definition but with a mangled name (according to
`<SELECTOR>`).
3) Annotate the declaration with the `OMPDeclareVariantAttr`, the same
one used already for `omp declare variant`, using and the mangled
function definition as specialization for the context defined by
`<SELECTOR>`.
When a call is created we inspect it. If the target has an
`OMPDeclareVariantAttr` attribute we try to specialize the call. To this
end, all variants are checked, the best applicable one is picked and a
new call to the specialization is created. The new call is used instead
of the original one to the base function. To keep the AST printing and
tooling possible we utilize the PseudoObjectExpr. The original call is
the syntactic expression, the specialized call is the semantic
expression.
[0] https://www.openmp.org/wp-content/uploads/openmp-TR8.pdf
[1] https://reviews.llvm.org/D61399#change-496lQkg0mhRN
Reviewers: kiranchandramohan, ABataev, RaviNarayanaswamy, gtbercea, grokos, sdmitriev, JonChesterfield, hfinkel, fghanim, aaron.ballman
Subscribers: bollu, guansong, openmp-commits, cfe-commits
Tags: #clang
Differential Revision: https://reviews.llvm.org/D75779
This is the first part extracted from D71179 and cleaned up.
This patch provides parsing support for `omp begin/end declare variant`,
as defined in OpenMP technical report 8 (TR8) [0].
A major purpose of this patch is to provide proper math.h/cmath support
for OpenMP target offloading. See PR42061, PR42798, PR42799. The current
code was developed with this feature in mind, see [1].
[0] https://www.openmp.org/wp-content/uploads/openmp-TR8.pdf
[1] https://reviews.llvm.org/D61399#change-496lQkg0mhRN
Reviewed By: aaron.ballman
Differential Revision: https://reviews.llvm.org/D74941
This reverts commit 0788acbccb.
This reverts commit c2d7a1f79cedfc9fcb518596aa839da4de0adb69: Revert "[clangd] Add test for FindTarget+RecoveryExpr (which already works). NFC"
It causes a crash on invalid code:
class X {
decltype(unresolved()) foo;
};
constexpr int s = sizeof(X);
region.
According to OpenMP 5.0, exactly one scan directive must appear in the loop body of an enclosing worksharing-loop, worksharing-loop SIMD, or simd construct on which a reduction clause with the inscan modifier is present.
If the ancestor device modifier is used and the value of the device
clause is evaluated to 1, the ancestor device shall be used for the
execution.
Since the reverse offloading is not supported yet, the target construct
execution is always initiated from the host, not from the device. So, if
the ancestor modifier is specified, just execute target region on the
host.
Avoid copying of the orignal variable if it is going to be marked as
firstprivate in task regions. For taskloops, still need to copy the
non-trvially copyable variables to correctly construct them upon task
creation.
Alexey Bataev