Summary:
The range checks performed for the vqrdmulh_lane and vqrdmulh_lane Neon
intrinsics were incorrectly using their return type as the base type for
the range check performed on their 'lane' argument.
This patch updates those intrisics to use the type of the proper reference
argument to perform the range checks.
Reviewers: jmolloy, t.p.northover, rsmith, olista01, dnsampaio
Reviewed By: dnsampaio
Subscribers: dnsampaio, kristof.beyls, cfe-commits
Tags: #clang
Differential Revision: https://reviews.llvm.org/D74766
Summary:
Range checks were not properly performed in the lane arguments of Neon
intrinsics implemented based on splat operations. Calls to those
intrinsics where translated to `__builtin__shufflevector` calls directly
by the pre-processor through the arm_neon.h macros, missing the chance
for the proper range checks.
This patch enables the range check by introducing an auxiliary splat
instruction in arm_neon.td, delaying the translation to shufflevector
calls to CGBuiltin.cpp in clang after the checks were performed.
Reviewers: jmolloy, t.p.northover, rsmith, olista01, ostannard
Reviewed By: ostannard
Subscribers: ostannard, dnsampaio, danielkiss, kristof.beyls, cfe-commits
Tags: #clang
Differential Revision: https://reviews.llvm.org/D74619
Summary:
As multiple versions of the same Neon intrinsic can be created through
the same TableGen definition with the same argument types, the existing
`call` operator is not always able to properly perform overload
resolutions.
As these different intrinsic versions are differentiated later on by the
NeonEmitter through name mangling, this patch introduces a new
`call_mangled` operator to the TableGen definitions, which allows a call
for an otherwise ambiguous intrinsic by matching its mangled name with
the mangled variation of the caller.
Reviewers: jmolloy, t.p.northover, rsmith, olista01, dnsampaio
Reviewed By: dnsampaio
Subscribers: dnsampaio, kristof.beyls, cfe-commits
Tags: #clang
Differential Revision: https://reviews.llvm.org/D74618
Summary:
Some of the `*_laneq` intrinsics defined in arm_neon.td were missing the
setting of the `isLaneQ` attribute. This patch sets the attribute on the
related definitions, as they will be required to properly perform range
checks on their lane arguments.
Reviewers: jmolloy, t.p.northover, rsmith, olista01, dnsampaio
Reviewed By: dnsampaio
Subscribers: dnsampaio, kristof.beyls, cfe-commits
Tags: #clang
Differential Revision: https://reviews.llvm.org/D74616
Summary:
Some of the `*_laneq` intrinsics defined in arm_neon.td were missing the
setting of the `isLaneQ` attribute. This patch sets the attribute on the
related definitions, as they will be required to properly perform range
checks on their lane arguments.
Reviewers: jmolloy, t.p.northover, rsmith, olista01, dnsampaio
Reviewed By: dnsampaio
Subscribers: dnsampaio, kristof.beyls, cfe-commits
Tags: #clang
Differential Revision: https://reviews.llvm.org/D74616
The SVE ACLE allows using a short-form for the intrinsics, e.g.
the following two declarations generate the same code:
svuint32_t svld1(svbool_t, uint32_t const *);
svuint32_t svld1_u32(svbool_t, uint32_t const *);
using the attribute:
__clang_arm_builtin_alias
so that any call to svld1(svbool_t, uint32_t const *) will
map to __builtin_sve_svld1_u32.
Reviewers: SjoerdMeijer, miyuki, efriedma, simon_tatham, rengolin
Reviewed By: SjoerdMeijer
Tags: #clang
Differential Revision: https://reviews.llvm.org/D75861
The only subexpression that is considered an error now is TypoExpr, but
we plan to add expressions with errors to improve editor tooling on broken
code. We intend to use the same mechanism to guard against spurious
diagnostics on those as well.
See the follow-up revision for an actual usage of the flag.
Original patch from Ilya.
Reviewers: sammccall
Reviewed By: sammccall
Tags: #clang
Differential Revision: https://reviews.llvm.org/D65591
Summary:
This is another set of instructions too complicated to be sensibly
expressed in IR by anything short of a target-specific intrinsic.
Given input vectors a,b, the instruction generates intermediate values
2*(a[0]*b[0]+a[1]+b[1]), 2*(a[2]*b[2]+a[3]+b[3]), etc; takes the high
half of each double-width values, and overwrites half the lanes in the
output vector c, which you therefore have to provide the input value
of. Optionally you can swap the elements of b so that the are things
like a[0]*b[1]+a[1]*b[0]; optionally you can round to nearest when
taking the high half; and optionally you can take the difference
rather than sum of the two products. Finally, saturation is applied
when converting back to a single-width vector lane.
Reviewers: dmgreen, MarkMurrayARM, miyuki, ostannard
Reviewed By: miyuki
Subscribers: kristof.beyls, hiraditya, cfe-commits
Tags: #clang
Differential Revision: https://reviews.llvm.org/D76359
Summary:
Copy of https://reviews.llvm.org/D72334, submitting with Ilya's permission.
Handles template declaration of all kinds.
Also builds template declaration nodes for specializations and explicit
instantiations of classes.
Some missing things will be addressed in the follow-up patches:
specializations of functions and variables,
template parameters.
Reviewers: gribozavr2
Reviewed By: gribozavr2
Subscribers: cfe-commits
Tags: #clang
Differential Revision: https://reviews.llvm.org/D76346
Summary:
- https://reviews.llvm.org/D68578 revises the `GlobalDecl` constructors
to ensure all GPU kernels have `ReferenceKenelKind` initialized
properly with an explicit constructor and static one. But, there are
lots of places using the implicit constructor triggering the assertion
on non-GPU kernels. That's found in compilation of many tests and
workloads.
- Fixing all of them may change more code and, more importantly, all of
them assumes the default kernel reference kind. This patch changes
that constructor to tell `CUDAGlobalAttr` and construct `GlobalDecl`
properly.
Reviewers: yaxunl
Subscribers: cfe-commits
Tags: #clang
Differential Revision: https://reviews.llvm.org/D76344
Reworked the patch to avoid sharing a header (SVETypeFlags.h) between
include/clang/Basic and utils/TableGen/SveEmitter.cpp. Now the patch
generates the enum/flags which is included in TargetBuiltins.h.
Also renamed one of the SveEmitter options to be in line with MVE.
Summary:
This is a first patch in a series for the SveEmitter to generate the arm_sve.h
header file and builtins.
I've tried my best to strip down this patch as best as I could, but there
are still a few changes that are not necessarily exercised by the load intrinsics
in this patch, mostly around the SVEType class which has some common logic to
represent types from a type and prototype string. I thought it didn't make
much sense to remove that from this patch and split it up.
Summary:
These are complicated integer multiply+add instructions with extra
saturation, taking the high half of a double-width product, and
optional rounding. There's no sensible way to represent that in
standard IR, so I've converted the clang builtins directly to
target-specific intrinsics.
Reviewers: dmgreen, MarkMurrayARM, miyuki, ostannard
Reviewed By: miyuki
Subscribers: kristof.beyls, hiraditya, cfe-commits
Tags: #clang
Differential Revision: https://reviews.llvm.org/D76123
Summary:
These instructions compute multiply+add in integers, with one of the
operands being a splat of a scalar. (VMLA and VMLAS differ in whether
the splat operand is a multiplier or the addend.)
I've represented these in IR using existing standard IR operations for
the unpredicated forms. The predicated forms are done with target-
specific intrinsics, as usual.
When operating on n-bit vector lanes, only the bottom n bits of the
i32 scalar operand are used. So we have to tell that to isel lowering,
to allow it to remove a pointless sign- or zero-extension instruction
on that input register. That's done in `PerformIntrinsicCombine`, but
first I had to enable `PerformIntrinsicCombine` for MVE targets
(previously all the intrinsics it handled were for NEON), and make it
a method of `ARMTargetLowering` so that it can get at
`SimplifyDemandedBits`.
Reviewers: dmgreen, MarkMurrayARM, miyuki, ostannard
Reviewed By: dmgreen
Subscribers: kristof.beyls, hiraditya, danielkiss, cfe-commits
Tags: #clang
Differential Revision: https://reviews.llvm.org/D76122
Summary:
[Clang] Attribute to allow defining undef global variables
Initializing global variables is very cheap on hosted implementations. The
C semantics of zero initializing globals work very well there. It is not
necessarily cheap on freestanding implementations. Where there is no loader
available, code must be emitted near the start point to write the appropriate
values into memory.
At present, external variables can be declared in C++ and definitions provided
in assembly (or IR) to achive this effect. This patch provides an attribute in
order to remove this reason for writing assembly for performance sensitive
freestanding implementations.
A close analogue in tree is LDS memory for amdgcn, where the kernel is
responsible for initializing the memory after it starts executing on the gpu.
Uninitalized variables in LDS are observably cheaper than zero initialized.
Patch is loosely based on the cuda __shared__ and opencl __local variable
implementation which also produces undef global variables.
Reviewers: kcc, rjmccall, rsmith, glider, vitalybuka, pcc, eugenis, vlad.tsyrklevich, jdoerfert, gregrodgers, jfb, aaron.ballman
Reviewed By: rjmccall, aaron.ballman
Subscribers: Anastasia, aaron.ballman, davidb, Quuxplusone, dexonsmith, cfe-commits
Tags: #clang
Differential Revision: https://reviews.llvm.org/D74361
new-expressions for a type T require sizeof(T) to be computable,
so the SVE ACLE does not allow them for sizeless types. At the moment:
auto f() { return new __SVInt8_t; }
creates a call to operator new with a zero size:
%call = call noalias nonnull i8* @_Znwm(i64 0)
This patch reports an appropriate error instead.
Differential Revision: https://reviews.llvm.org/D76218
This flag is used by avr-gcc (starting with v10) to set the width of the
double type. The double type is by default interpreted as a 32-bit
floating point number in avr-gcc instead of a 64-bit floating point
number as is common on other architectures. Starting with GCC 10, a new
option has been added to control this behavior:
https://gcc.gnu.org/wiki/avr-gcc#Deviations_from_the_Standard
This commit keeps the default double at 32 bits but adds support for the
-mdouble flag (-mdouble=32 and -mdouble=64) to control this behavior.
Differential Revision: https://reviews.llvm.org/D76181
In the current SVE ACLE spec, the usual rules for throwing and
catching incomplete types also apply to sizeless types. However,
throwing pointers to sizeless types should not pose any real difficulty,
so as an extension, the clang implementation allows that.
This patch enforces these rules for catch statements.
Differential Revision: https://reviews.llvm.org/D76090
Summary:
The same rules for throwing and catching incomplete types also apply
to sizeless types. This patch enforces that for throw statements.
It also make sure that we use "sizeless type" rather "incomplete type"
in the associated message. (Both are correct, but "sizeless type" is
more specific and hopefully more user-friendly.)
The SVE ACLE simply extends the rule for incomplete types to
sizeless types. However, throwing pointers to sizeless types
should not pose any real difficulty, so as an extension,
the clang implementation allows that.
Reviewers: sdesmalen, efriedma, rovka, rjmccall
Subscribers: tschuett, rkruppe, psnobl, cfe-commits
Tags: #clang
Differential Revision: https://reviews.llvm.org/D76088
In the current SVE ACLE spec, the usual rules for throwing and
catching incomplete types also apply to sizeless types. However,
throwing pointers to sizeless types should not pose any real difficulty,
so as an extension, the clang implementation allows that.
This patch enforces these rules for explicit exception specs.
Differential Revision: https://reviews.llvm.org/D76087
This patch completes a trio of changes related to arrays of
sizeless types. It rejects various forms of arithmetic on
pointers to sizeless types, in the same way as for other
incomplete types.
Differential Revision: https://reviews.llvm.org/D76086
clang currently accepts:
__SVInt8_t &foo1(__SVInt8_t *x) { return *x; }
__SVInt8_t &foo2(__SVInt8_t *x) { return x[1]; }
The first function is valid ACLE code and generates correct LLVM IR
(and assembly code). But the second function is invalid for the
same reason that arrays of sizeless types are. Trying to code-generate
the function leads to:
llvm/include/llvm/Support/TypeSize.h:126: uint64_t llvm::TypeSize::getFixedSize() const: Assertion `!IsScalable && "Request for a fixed size on a s
calable object"' failed.
Another problem is that:
template<typename T>
constexpr __SIZE_TYPE__ f(T *x) { return &x[1] - x; }
typedef int arr1[f((int *)0) - 1];
typedef int arr2[f((__SVInt8_t *)0) - 1];
produces:
a.cpp:2:48: warning: subtraction of pointers to type '__SVInt8_t' of zero size has undefined behavior [-Wpointer-arith]
constexpr __SIZE_TYPE__ f(T *x) { return &x[1] - x; }
~~~~~ ^ ~
a.cpp:4:18: note: in instantiation of function template specialization 'f<__SVInt8_t>' requested here
typedef int arr2[f((__SVInt8_t *)0) - 1];
This patch reports an appropriate diagnostic instead.
Differential Revision: https://reviews.llvm.org/D76084
To group the code in one place, simplify it and make it easier to add
the containsErrors bit and find existing bugs.
Reviewers: sammccall
Reviewed By: sammccall
Tags: #clang
Differential Revision: https://reviews.llvm.org/D73638
TryAnnotateTypeConstraint could annotate a template-id which doesn't end up being a type-constraint,
in which case control flow would incorrectly flow into ParseImplicitInt.
Reenter the loop in this case.
Enable relevant tests for C++20. This required disabling typo-correction during TryAnnotateTypeConstraint
and changing a test case which is broken due to a separate bug (will be reported and handled separately).
This is a first patch in a series for the SveEmitter to generate the arm_sve.h
header file and builtins.
I've tried my best to strip down this patch as best as I could, but there
are still a few changes that are not necessarily exercised by the load intrinsics
in this patch, mostly around the SVEType class which has some common logic to
represent types from a type and prototype string. I thought it didn't make
much sense to remove that from this patch and split it up.
Reviewers: efriedma, rovka, SjoerdMeijer, rsandifo-arm, rengolin
Reviewed By: SjoerdMeijer
Tags: #clang
Differential Revision: https://reviews.llvm.org/D75470
This patch adds 'q' to mean 'scalable vector' in the builtin
type string, and for SVE will return the matching builtin
type as defined in the C/C++ language extensions for SVE.
This patch also adds some scaffolding to generate the arm_sve.h
header file, and some builtin definitions (+CodeGen) to be able
to implement some simple masked load intrinsics that use the
ACLE types, such as:
svint8_t test_svld1_s8(svbool_t pg, const int8_t *base) {
return svld1_s8(pg, base);
}
Reviewers: efriedma, rjmccall, rovka, rsandifo-arm, rengolin
Reviewed By: efriedma
Tags: #clang
Differential Revision: https://reviews.llvm.org/D75298
The name field is optional if the custom code is supplied, so this updates the
documentation for LangOpt and introduces a tablegen warning if both custom code
and a language option name are supplied.
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.
The SVE ACLE doesn't allow arrays of sizeless types. At the moment
clang accepts the TU:
__SVInt8_t x[2];
but trying to code-generate it triggers the LLVM assertion:
llvm/lib/IR/Type.cpp:588: static llvm::ArrayType* llvm::ArrayType::get(llvm::Type*, uint64_t): Assertion `isValidElementType(ElementType) && "Invalid type for array element!"' failed.
This patch reports an appropriate error instead.
The rules are slightly more restrictive than for general incomplete types.
For example:
struct s;
typedef struct s arr[2];
is valid as far as it goes, whereas arrays of sizeless types are
invalid in all contexts. BuildArrayType therefore needs a specific
check for isSizelessType in addition to the usual handling of
incomplete types.
Differential Revision: https://reviews.llvm.org/D76082
Since fields can't have sizeless type, it also doesn't make sense
to capture sizeless types by value in lambda expressions. This patch
makes sure that we diagnose that and that we use "sizeless type" rather
"incomplete type" in the associated message. (Both are correct, but
"sizeless type" is more specific and hopefully more user-friendly.)
Differential Revision: https://reviews.llvm.org/D75738
The SVE ACLE doesn't allow fields to have sizeless type. At the moment
clang accepts things like:
struct s { __SVInt8_t x; } y;
but trying to code-generate it leads to LLVM asserts like:
llvm/include/llvm/Support/TypeSize.h:126: uint64_t llvm::TypeSize::getFixedSize() const: Assertion `!IsScalable && "Request for a fixed size on a scalable object"' failed.
This patch adds an associated clang diagnostic.
Differential Revision: https://reviews.llvm.org/D75737
Summary:
The parsing of GNU C extended asm statements was a little brittle and
had a few issues:
- It was using Parse::ParseTypeQualifierListOpt to parse the `volatile`
qualifier. That parser is really meant for TypeQualifiers; an asm
statement doesn't really have a type qualifier. This is still maybe
nice to have, but not necessary. We now can check for the `volatile`
token by properly expanding the grammer, rather than abusing
Parse::ParseTypeQualifierListOpt.
- The parsing of `goto` was position dependent, so `asm goto volatile`
wouldn't parse. The qualifiers should be position independent to one
another. Now they are.
- We would warn on duplicate `volatile`, but the parse error for
duplicate `goto` was a generic parse error and wasn't clear.
- We need to add support for the recent GNU C extension `asm inline`.
Adding support to the parser with the above issues highlighted the
need for this refactoring.
Link: https://gcc.gnu.org/onlinedocs/gcc/Extended-Asm.html
Reviewers: aaron.ballman
Reviewed By: aaron.ballman
Subscribers: aheejin, jfb, nathanchance, cfe-commits, echristo, efriedma, rsmith, chandlerc, craig.topper, erichkeane, jyu2, void, srhines
Tags: #clang
Differential Revision: https://reviews.llvm.org/D75563
function and an overridden function until we know whether the overriding
function is deleted.
We previously did these checks when we first built the declaration,
which was too soon in some cases. We now defer all these checks to the
end of the class.
Also add missing check that a consteval function cannot override a
non-consteval function and vice versa.
Lucas Prates