Breaks build with -DBUILD_SHARED_LIBS=ON
```
CMake Error: The inter-target dependency graph contains the following strongly connected component (cycle):
"LLVMFrontendOpenMP" of type SHARED_LIBRARY
depends on "LLVMPasses" (weak)
"LLVMipo" of type SHARED_LIBRARY
depends on "LLVMFrontendOpenMP" (weak)
"LLVMCoroutines" of type SHARED_LIBRARY
depends on "LLVMipo" (weak)
"LLVMPasses" of type SHARED_LIBRARY
depends on "LLVMCoroutines" (weak)
depends on "LLVMipo" (weak)
At least one of these targets is not a STATIC_LIBRARY. Cyclic dependencies are allowed only among static libraries.
CMake Generate step failed. Build files cannot be regenerated correctly.
```
This reverts commit 707ce34b06190e275572c3c46843036db1bab6d1.
Add methods for loop unrolling to the OpenMPIRBuilder class and use them in Clang if `-fopenmp-enable-irbuilder` is enabled. The unrolling methods are:
* `unrollLoopFull`
* `unrollLoopPartial`
* `unrollLoopHeuristic`
`unrollLoopPartial` and `unrollLoopHeuristic` can use compiler heuristics to automatically determine the unroll factor. If possible, that is if no CanonicalLoopInfo is required to pass to another method, metadata for LLVM's LoopUnrollPass is added. Otherwise the unroll factor is determined using the same heurstics as user by LoopUnrollPass. Not requiring a CanonicalLoopInfo, especially with `unrollLoopHeuristic` allows greater flexibility.
With full unrolling and partial unrolling with known unroll factor, instead of duplicating instructions by the OpenMPIRBuilder, the full unroll is still delegated to the LoopUnrollPass. In case of partial unrolling the loop is first tiled using the existing `tileLoops` methods, then the inner loop fully unrolled using the same mechanism.
Reviewed By: jdoerfert, kiranchandramohan
Differential Revision: https://reviews.llvm.org/D107764
The intent of this patch is to add support of -fp-model=[source|double|extended] to allow
the compiler to use a wider type for intermediate floating point calculations. As a side
effect to that, the value of FLT_EVAL_METHOD is changed according to the pragma
float_control.
Unfortunately some issue was uncovered with this change in preprocessing. See details in
https://reviews.llvm.org/D93769 . We are therefore reverting this patch until we find a way
to reconcile the value of FLT_EVAL_METHOD, the pragma and the -E flow.
This reverts commit 66ddac22e2a7f268e91c26d694112970dfa607ae.
This patch implements Clang support for an original OpenMP extension
we have developed to support OpenACC: the `ompx_hold` map type
modifier. The next patch in this series, D106510, implements OpenMP
runtime support.
Consider the following example:
```
#pragma omp target data map(ompx_hold, tofrom: x) // holds onto mapping of x
{
foo(); // might have map(delete: x)
#pragma omp target map(present, alloc: x) // x is guaranteed to be present
printf("%d\n", x);
}
```
The `ompx_hold` map type modifier above specifies that the `target
data` directive holds onto the mapping for `x` throughout the
associated region regardless of any `target exit data` directives
executed during the call to `foo`. Thus, the presence assertion for
`x` at the enclosed `target` construct cannot fail. (As usual, the
standard OpenMP reference count for `x` must also reach zero before
the data is unmapped.)
Justification for inclusion in Clang and LLVM's OpenMP runtime:
* The `ompx_hold` modifier supports OpenACC functionality (structured
reference count) that cannot be achieved in standard OpenMP, as of
5.1.
* The runtime implementation for `ompx_hold` (next patch) will thus be
used by Flang's OpenACC support.
* The Clang implementation for `ompx_hold` (this patch) as well as the
runtime implementation are required for the Clang OpenACC support
being developed as part of the ECP Clacc project, which translates
OpenACC to OpenMP at the directive AST level. These patches are the
first step in upstreaming OpenACC functionality from Clacc.
* The Clang implementation for `ompx_hold` is also used by the tests
in the runtime implementation. That syntactic support makes the
tests more readable than low-level runtime calls can. Moreover,
upstream Flang and Clang do not yet support OpenACC syntax
sufficiently for writing the tests.
* More generally, the Clang implementation enables a clean separation
of concerns between OpenACC and OpenMP development in LLVM. That
is, LLVM's OpenMP developers can discuss, modify, and debug LLVM's
extended OpenMP implementation and test suite without directly
considering OpenACC's language and execution model, which can be
handled by LLVM's OpenACC developers.
* OpenMP users might find the `ompx_hold` modifier useful, as in the
above example.
See new documentation introduced by this patch in `openmp/docs` for
more detail on the functionality of this extension and its
relationship with OpenACC. For example, it explains how the runtime
must support two reference counts, as specified by OpenACC.
Clang recognizes `ompx_hold` unless `-fno-openmp-extensions`, a new
command-line option introduced by this patch, is specified.
Reviewed By: ABataev, jdoerfert, protze.joachim, grokos
Differential Revision: https://reviews.llvm.org/D106509
This change defines a helper function getOpenCLCompatibleVersion()
inside LangOptions class. The function contains mapping between
C++ for OpenCL versions and their corresponding compatible OpenCL
versions. This mapping function should be updated each time a new
C++ for OpenCL language version is introduced. The helper function
is expected to simplify conditions on OpenCL C and C++ for OpenCL
versions inside compiler code.
Code refactoring performed.
Differential Revision: https://reviews.llvm.org/D108693
Some Clang diagnostics could only report OpenCL C version. Because
C++ for OpenCL can be used as an alternative to OpenCL C, the text
for diagnostics should reflect that.
Desrciptions modified for these diagnostics:
`err_opencl_unknown_type_specifier`
`warn_option_invalid_ocl_version`
`err_attribute_requires_opencl_version`
`warn_opencl_attr_deprecated_ignored`
`ext_opencl_ext_vector_type_rgba_selector`
Differential Revision: https://reviews.llvm.org/D107648
Only the bare name is completed, with no args.
For args to be useful we need arg names. These *are* in the tablegen but
not currently emitted in usable form, so left this as future work.
C++11, C2x, GNU, declspec, MS syntax is supported, with the appropriate
spellings of attributes suggested.
`#pragma clang attribute` is supported but not terribly useful as we
only reach completion if parens are balanced (i.e. the line is not truncated)
There's no filtering of which attributes might make sense in this
grammatical context (e.g. attached to a function). In code-completion context
this is hard to do, and will only work in few cases :-(
There's also no filtering by langopts: this is because currently the
only way of checking is to try to produce diagnostics, which requires a
valid ParsedAttr which is hard to get.
This should be fairly simple to fix but requires some tablegen changes
to expose the logic without the side-effect.
Differential Revision: https://reviews.llvm.org/D107696
This patch implements paper P0692R1 from the C++20 standard. Disable usual access checking rules to template argument names in a declaration of partial specializations, explicit instantiation or explicit specialization (C++20 13.7.5/10, 13.9.1/6).
Fixes: https://llvm.org/PR37424
This patch also implements option *A* from this paper P0692R1 from the C++20 standard.
This patch follows the @rsmith suggestion from D78404.
Reviewed By: krisb
Differential Revision: https://reviews.llvm.org/D92024
This patch implements P2092
Simple requirements in requirement body shall not start with requires.
A warning was already in place so we just turn this warning into an error.
In addition, we add tests to make sure typename is optional in
requirement-parameter-list as per the same paper.
Previously we would show an error, but keep the member, and also the
CXXRrecordDecl, valid. This could lead to crashes when attempting to
access the record layout or size.
Differential Revision: https://reviews.llvm.org/D105478
'pipe' keyword is introduced in OpenCL C 2.0: so do checks for OpenCL C version while
parsing and then later on check for language options to construct actual pipe. This feature
requires support of __opencl_c_generic_address_space, so diagnostics for that is provided as well.
This is the same patch as in D106748 but with a tiny fix in checking of diagnostic messages.
Also added tests when program scope global variables are not supported.
Reviewed By: Anastasia
Differential Revision: https://reviews.llvm.org/D107154
'pipe' keyword is introduced in OpenCL C 2.0: so do checks for OpenCL C version while
parsing and then later on check for language options to construct actual pipe. This feature
requires support of __opencl_c_generic_address_space, so diagnostics for that is provided as well.
Reviewed By: Anastasia
Differential Revision: https://reviews.llvm.org/D106748
The Intel compiler ICC supports the option "-fp-model=(source|double|extended)"
which causes the compiler to use a wider type for intermediate floating point
calculations. Also supported is a way to embed this effect in the source
program with #pragma float_control(source|double|extended).
This patch extends pragma float_control syntax, and also adds support
for a new floating point option "-ffp-eval-method=(source|double|extended)".
source: intermediate results use source precision
double: intermediate results use double precision
extended: intermediate results use extended precision
Reviewed By: Aaron Ballman
Differential Revision: https://reviews.llvm.org/D93769
The Intel compiler ICC supports the option "-fp-model=(source|double|extended)"
which causes the compiler to use a wider type for intermediate floating point
calculations. Also supported is a way to embed this effect in the source
program with #pragma float_control(source|double|extended).
This patch extends pragma float_control syntax, and also adds support
for a new floating point option "-ffp-eval-method=(source|double|extended)".
source: intermediate results use source precision
double: intermediate results use double precision
extended: intermediate results use extended precision
Reviewed By: Aaron Ballman
Differential Revision: https://reviews.llvm.org/D93769
No implementation uses the `LocCookie` parameter at all. Errors are
reported from inside that function by `llvm::SourceMgr`, and the
instance of that at the clang call site arranges to pass the error
messages back to a `ClangAsmParserCallback`, which is where the clang
SourceLocation for the error is computed.
(This is part of a patch series working towards the ability to make
SourceLocation into a 64-bit type to handle larger translation units.
But this particular change seems beneficial in its own right.)
Reviewed By: miyuki
Differential Revision: https://reviews.llvm.org/D105490
OpenMP 5.1 added support for writing OpenMP directives using [[]]
syntax in addition to using #pragma and this introduces support for the
new syntax.
In OpenMP, the attributes take one of two forms:
[[omp::directive(...)]] or [[omp::sequence(...)]]. A directive
attribute contains an OpenMP directive clause that is identical to the
analogous #pragma syntax. A sequence attribute can contain either
sequence or directive arguments and is used to ensure that the
attributes are processed sequentially for situations where the order of
the attributes matter (remember:
https://eel.is/c++draft/dcl.attr.grammar#4.sentence-4).
The approach taken here is somewhat novel and deserves mention. We
could refactor much of the OpenMP parsing logic to work for either
pragma annotation tokens or for attribute clauses. It would be a fair
amount of effort to share the logic for both, but it's certainly
doable. However, the semantic attribute system is not designed to
handle the arbitrarily complex arguments that OpenMP directives
contain. Adding support to thread the novel parsed information until we
can produce a semantic attribute would be considerably more effort.
What's more, existing OpenMP constructs are not (often) represented as
semantic attributes. So doing this through Attr.td would be a massive
undertaking that would likely only benefit OpenMP and comes with
additional risks. Rather than walk down that path, I am taking
advantage of the fact that the syntax of the directives within the
directive clause is identical to that of the #pragma form. Once the
parser recognizes that we're processing an OpenMP attribute, it caches
all of the directive argument tokens and then replays them as though
the user wrote a pragma. This reuses the same OpenMP parsing and
semantic logic directly, but does come with a risk if the OpenMP
committee decides to purposefully diverge their pragma and attribute
syntaxes. So, despite this being a novel approach that does token
replay, I think it's actually a better approach than trying to do this
through the declarative syntax in Attr.td.
A user reported an issue to me via email that Clang was accepting some
code that GCC was rejecting. After investigation, it turned out to be a
general problem of us failing to properly reject attributes written in
the type position in C when they don't apply to types. The root cause
was a terminology issue -- we sometimes use "CXX11Attr" to mean [[]] in
C++11 mode and sometimes [[]] in general -- and this came back to bite
us because in this particular case, it really meant [[]] in C++ mode.
I fixed the issue by introducing a new function
AttributeCommonInfo::isStandardAttributeSyntax() to represent [[]] in
either C or C++ mode.
This fix pointed out that we've had the issue in some of our existing
tests, which have all been corrected. This resolves
https://bugs.llvm.org/show_bug.cgi?id=50954.
According to https://eel.is/c++draft/over.literal
> double operator""_Bq(long double); // OK: does not use the reserved identifier _Bq ([lex.name])
> double operator"" _Bq(long double); // ill-formed, no diagnostic required: uses the reserved identifier _Bq ([lex.name])
Obey that rule by keeping track of the operator literal name status wrt. leading whitespace.
Fix: https://bugs.llvm.org/show_bug.cgi?id=50644
Differential Revision: https://reviews.llvm.org/D104299
The parser code assumes building with C++ compiler and asserts when using clang (not clang++) on C file. I made the code dependent on input language. This shows up for amdgpu target.
Reviewed By: ABataev
Differential Revision: https://reviews.llvm.org/D103899
Implementation of the unroll directive introduced in OpenMP 5.1. Follows the approach from D76342 for the tile directive (i.e. AST-based, not using the OpenMPIRBuilder). Tries to use `llvm.loop.unroll.*` metadata where possible, but has to fall back to an AST representation of the outer loop if the partially unrolled generated loop is associated with another directive (because it needs to compute the number of iterations).
Reviewed By: ABataev
Differential Revision: https://reviews.llvm.org/D99459
This implements the 'using enum maybe-qualified-enum-tag ;' part of
1099. It introduces a new 'UsingEnumDecl', subclassed from
'BaseUsingDecl'. Much of the diff is the boilerplate needed to get the
new class set up.
There is one case where we accept ill-formed, but I believe this is
merely an extended case of an existing bug, so consider it
orthogonal. AFAICT in class-scope the c++20 rule is that no 2 using
decls can bring in the same target decl ([namespace.udecl]/8). But we
already accept:
struct A { enum { a }; };
struct B : A { using A::a; };
struct C : B { using A::a;
using B::a; }; // same enumerator
this patch permits mixtures of 'using enum Bob;' and 'using Bob::member;' in the same way.
Differential Revision: https://reviews.llvm.org/D102241
This is a re-application of dc67299 which was reverted in f63adf5b because
it broke the build. The issue should now be fixed.
Attribution note: The original author of this patch is Erik Pilkington.
I'm only trying to land it after rebasing.
Differential Revision: https://reviews.llvm.org/D91630
The original version of this was reverted, and @rjmcall provided some
advice to architect a new solution. This is that solution.
This implements a builtin to provide a unique name that is stable across
compilations of this TU for the purposes of implementing the library
component of the unnamed kernel feature of SYCL. It does this by
running the Itanium mangler with a few modifications.
Because it is somewhat common to wrap non-kernel-related lambdas in
macros that aren't present on the device (such as for logging), this
uniquely generates an ID for all lambdas involved in the naming of a
kernel. It uses the lambda-mangling number to do this, except replaces
this with its own number (starting at 10000 for readabililty reasons)
for lambdas used to name a kernel.
Additionally, this implements itself as constexpr with a slight catch:
if a name would be invalidated by the use of this lambda in a later
kernel invocation, it is diagnosed as an error (see the Sema tests).
Differential Revision: https://reviews.llvm.org/D103112
This makes it possible for targets to define their own MCObjectFileInfo.
This MCObjectFileInfo is then used to determine things like section alignment.
This is a follow up to D101462 and prepares for the RISCV backend defining the
text section alignment depending on the enabled extensions.
Reviewed By: MaskRay
Differential Revision: https://reviews.llvm.org/D101921
Summary:
Call TryAltiVecVectorToken when an identifier is seen in the parser before
annotating the token. This checks the next token where necessary to ensure
that vector is properly handled as the altivec token.
Author: Jamie Schmeiser <schmeise@ca.ibm.com>
Reviewed By: ZarkoCA (Zarko Todorovski)
Differential Revision: https://reviews.llvm.org/D100991
Drop non-conformant extension pragma implementation as
it does not properly disable anything and therefore
enabling non-disabled logic has no meaning.
This simplifies clang code and user interface to the extension
functionality. With this patch extension pragma 'begin'/'end'
and 'enable'/'disable' are only accepted for backward
compatibility and no longer have any default behavior.
Differential Revision: https://reviews.llvm.org/D101043
Currently when including stdbool.h and altivec.h declaration of `vector bool` leads to
errors due to `bool` being expanded to '_Bool`. This patch allows the parser
to recognize `_Bool`.
Reviewed By: hubert.reinterpretcast, Everybody0523
Differential Revision: https://reviews.llvm.org/D102064
This change allows the use of identifiers for image types
from `cl_khr_gl_msaa_sharing` freely in the kernel code if
the extension is not supported since they are not in the
list of the reserved identifiers.
This change also removed the need for pragma for the types
in the extensions since the spec does not require the pragma
uses.
Differential Revision: https://reviews.llvm.org/D100983
This patch fixes various issues with our prior `declare target` handling
and extends it to support `omp begin declare target` as well.
This started with PR49649 in mind, trying to provide a way for users to
avoid the "ref" global use introduced for globals with internal linkage.
From there it went down the rabbit hole, e.g., all variables, even
`nohost` ones, were emitted into the device code so it was impossible to
determine if "ref" was needed late in the game (based on the name only).
To make it really useful, `begin declare target` was needed as it can
carry the `device_type`. Not emitting variables eagerly had a ripple
effect. Finally, the precedence of the (explicit) declare target list
items needed to be taken into account, that meant we cannot just look
for any declare target attribute to make a decision. This caused the
handling of functions to require fixup as well.
I tried to clean up things while I was at it, e.g., we should not "parse
declarations and defintions" as part of OpenMP parsing, this will always
break at some point. Instead, we keep track what region we are in and
act on definitions and declarations instead, this is what we do for
declare variant and other begin/end directives already.
Highlights:
- new diagnosis for restrictions specificed in the standard,
- delayed emission of globals not mentioned in an explicit
list of a declare target,
- omission of `nohost` globals on the host and `host` globals on the
device,
- no explicit parsing of declarations in-between `omp [begin] declare
variant` and the corresponding end anymore, regular parsing instead,
- precedence for explicit mentions in `declare target` lists over
implicit mentions in the declaration-definition-seq, and
- `omp allocate` declarations will now replace an earlier emitted
global, if necessary.
---
Notes:
The patch is larger than I hoped but it turns out that most changes do
on their own lead to "inconsistent states", which seem less desirable
overall.
After working through this I feel the standard should remove the
explicit declare target forms as the delayed emission is horrible.
That said, while we delay things anyway, it seems to me we check too
often for the current status even though that is often not sufficient to
act upon. There seems to be a lot of duplication that can probably be
trimmed down. Eagerly emitting some things seems pretty weak as an
argument to keep so much logic around.
---
Reviewed By: ABataev
Differential Revision: https://reviews.llvm.org/D101030
This untangles the MCContext and the MCObjectFileInfo. There is a circular
dependency between MCContext and MCObjectFileInfo. Currently this dependency
also exists during construction: You can't contruct a MOFI without a MCContext
without constructing the MCContext with a dummy version of that MOFI first.
This removes this dependency during construction. In a perfect world,
MCObjectFileInfo wouldn't depend on MCContext at all, but only be stored in the
MCContext, like other MC information. This is future work.
This also shifts/adds more information to the MCContext making it more
available to the different targets. Namely:
- TargetTriple
- ObjectFileType
- SubtargetInfo
Reviewed By: MaskRay
Differential Revision: https://reviews.llvm.org/D101462
Pipe has not been a reserved keyword in the earlier OpenCL
standards. However we failed to allow its use as an identifier
in the original commit. This issues is fixed now and testing
is improved accordingly.
Differential Revision: https://reviews.llvm.org/D101052
The comment here was introduced in
a3e01cf822f7415337e5424af3c6f4c94a12c1b9 and suggests that we should
handle declaration statements and non-declaration statements the same,
but don't because ProhibitAttributes() can't handle GNU attributes. That
has recently changed, so remove the comment and handle all statements
the same.
Differential Revision: https://reviews.llvm.org/D99936
GCC 8 introduced these new pragmas to control loop unrolling. We should support them for compatibility reasons and the implementation itself requires few lines of code, since everything needed is already implemented for #pragma unroll/nounroll.
Double square bracket attribute arguments can be arbitrarily complex,
and the attribute argument parsing logic recovers by skipping tokens.
As a fallback recovery mechanism, parse recovery stops before reading a
semicolon. This could lead to an infinite loop in the attribute list
parsing logic.
These proposals make the same changes to both C++ and C and remove a
restriction on standard attributes appearing multiple times in the same
attribute list.
We could warn on the duplicate attributes, but do not. This is for
consistency as we do not warn on attributes duplicated within the
attribute specifier sequence. If we want to warn on duplicated
standard attributes, we should do so both for both situations:
[[foo, foo]] and [[foo]][[foo]].
Clang currently has a bug where it allows you to write [[foo bar]] and
both attributes are silently accepted. This patch corrects the comma
parsing rules for such attributes and handles the test case fallout, as
a few tests were accidentally doing this.
