The patch implements header-only support for testure lookups.
The patch has been tested on a source file with all possible combinations of
argument types supported by CUDA headers, compiled and verified that the
generated instructions and their parameters match the code generated by NVCC.
Unfortunately, compiling texture code requires CUDA headers and can't be tested
in clang itself. The test will need to be added to the test-suite later.
While generated code compiles and seems to match NVCC, I do not have any code
that uses textures that I could test correctness of the implementation. Hence
the experimental status.
Differential Revision: https://reviews.llvm.org/D110089
This patch is for fixing potential shufflevector-related bugs like D93818.
As D93818, this patch change shufflevector's default placeholder to poison.
To reduce risk, it was divided into several patches, and this patch is for InstCombineCasts.
Reviewed By: spatel
Differential Revision: https://reviews.llvm.org/D110226
RUN line representing C++ for OpenCL 2021 added to the test. This
should have been done as part of earlier commit fb321c2ea2 but
was missed during rebasing.
Differential Revision: https://reviews.llvm.org/D109492
Following nvptx approach, this patch uses complex function
definitions from complex_cmath.h. With this patch, ovo passes
23/34 complex mathematical test cases.
Reviewed By: JonChesterfield
Differential Revision: https://reviews.llvm.org/D109344
Modifies OpenCL 3.0 optional core feature macro definitions so that
they are set analogously in C++ for OpenCL 2021.
This change aims to achieve compatibility between C++ for OpenCL
2021 and OpenCL 3.0.
Differential Revision: https://reviews.llvm.org/D108704
This change would treat the token `or` in system headers as an
identifier, and elsewhere as an operator. As reported in
llvm.org/pr42427, many users classify their third party library headers
as "system" headers to suppress warnings. There's no clean way to
separate Windows SDK headers from user headers.
Clang is still able to parse old Windows SDK headers if C++ operator
names are disabled. Traditionally this was controlled by
`-fno-operator-names`, but is now also enabled with `/permissive` since
D103773. This change will prevent `clang-cl` from parsing <query.h> from
the Windows SDK out of the box, but there are multiple ways to work
around that:
- Pass `/clang:-fno-operator-names`
- Pass `/permissive`
- Pass `-DQUERY_H_RESTRICTION_PERMISSIVE`
In all of these modes, the operator names will consistently be available
or not available, instead of depending on whether the code is in a
system header.
I added a release note for this, since it may break straightforward
users of the Windows SDK.
Fixes PR42427
Differential Revision: https://reviews.llvm.org/D108720
Partially reverts 85157c0079, which had removed these builtins and intrinsics
in favor of normal codegen patterns. It turns out that it is possible for the
patterns to be split over multiple basic blocks, however, which means that DAG
ISel is not able to select them to the pmin/pmax instructions. To make sure the
SIMD intrinsics generate the correct instructions in these cases, reintroduce
the clang builtins and corresponding LLVM intrinsics, but also keep the normal
pattern matching as well.
Differential Revision: https://reviews.llvm.org/D108387
On some platforms, negative shift values mean to shift in the opposite
direction, but this is not true with WebAssembly. To avoid confusion, make the
shift values in the shift intrinsics unsigned.
Differential Revision: https://reviews.llvm.org/D108415
For each SIMD intrinsic function that takes or returns a scalar signed integer
value, ensure there is a corresponding intrinsic that returns or an
unsigned value. This is a convenience for users who use -Wsign-conversion so
they don't have to insert explicit casts, especially when the intrinsic
arguments are integer literals that fit into the unsigned integer type but not
the signed type.
Differential Revision: https://reviews.llvm.org/D108412
Since they are bitmasks, it will be more common for them to be used and
potentially extended to 64-bit integers as unsigned values rather than signed
values.
Differential Revision: https://reviews.llvm.org/D108401
Fixes miscompile of calls into ocml. Bug 51445.
The stack variable `double __tmp` is moved to dynamically allocated shared
memory by CGOpenMPRuntimeGPU. This is usually fine, but when the variable
is passed to a function that is explicitly annotated address_space(5) then
allocating the variable off-stack leads to a miscompile in the back end,
which cannot decide to move the variable back to the stack from shared.
This could be fixed by removing the AS(5) annotation from the math library
or by explicitly marking the variables as thread_mem_alloc. The cast to
AS(5) is still a no-op once IR is reached.
Reviewed By: jdoerfert
Differential Revision: https://reviews.llvm.org/D107971
With this patch, OpenMP on AMDGCN will use the math functions
provided by ROCm ocml library. Linking device code to the ocml will be
done in the next patch.
Reviewed By: JonChesterfield, jdoerfert, scchan
Differential Revision: https://reviews.llvm.org/D104904
With this patch, OpenMP on AMDGCN will use the math functions
provided by ROCm ocml library. Linking device code to the ocml will be
done in the next patch.
Reviewed By: JonChesterfield, jdoerfert, scchan
Differential Revision: https://reviews.llvm.org/D104904
Replace the clang builtins and LLVM intrinsics for the SIMD extmul instructions
with normal codegen patterns.
Differential Revision: https://reviews.llvm.org/D106724
Replace the clang builtins and LLVM intrinsics for {f32x4,f64x2}.{pmin,pmax}
with standard codegen patterns. Since wasm_simd128.h uses an integer vector as
the standard single vector type, the IR for the pmin and pmax intrinsic
functions contains bitcasts that would not be there otherwise. Add extra codegen
patterns that can still select the pmin and pmax instructions in the presence of
these bitcasts.
Differential Revision: https://reviews.llvm.org/D106612
Add the builtins defined by Section 42 "Integer dot product" in
the OpenCL Extension Specification.
Differential Revision: https://reviews.llvm.org/D106434
Commit db7efcab7d changed the implementations of the wasm_*_extract_lane and
wasm_*_replace_lane intrinsics from using builtin functions to using the
standard vector extensions. This did not change the resulting IR, but it changes
how update_cc_test_checks.py labels values in the IR. This commit simply updates
those labels.
Differential Revision: https://reviews.llvm.org/D106611
With this patch, OpenMP on AMDGCN will use the math functions
provided by ROCm ocml library. Linking device code to the ocml will be
done in the next patch.
Reviewed By: JonChesterfield, jdoerfert, scchan
Differential Revision: https://reviews.llvm.org/D104904
Add the builtins defined by Section 40 "Extended Bit Operations" in
the OpenCL Extension Specification.
Differential Revision: https://reviews.llvm.org/D106267
On Power PC some legacy compilers included a number of builtins in a
builtins.h header file. While this header file is not required to hold
builtins for clang some legacy code does try to include this file and so
this patch provides an empty version of that file.
Differential Revision: https://reviews.llvm.org/D106065
Replace the experimental clang builtins and LLVM intrinsics for these
instructions with normal codegen patterns. Resolves PR50435.
Differential Revision: https://reviews.llvm.org/D106019
Replace the experimental clang builtin and LLVM intrinsics for these
instructions with normal codegen patterns. Resolves PR50433.
Differential Revision: https://reviews.llvm.org/D105950
Replace the clang builtin function and LLVM intrinsic for
f32x4.demote_zero_f64x2 with combines from normal SDNodes. Also add missing
combines for i32x4.trunc_sat_zero_f64x2_{s,u}, which share the same pattern.
Differential Revision: https://reviews.llvm.org/D105755
Replace the clang builtin function and LLVM intrinsic previously used to select
the f64x2.promote_low_f32x4 instruction with custom combines from standard
SelectionDAG nodes. Implement the new combines to share code with the similar
combines for f64x2.convert_low_i32x4_{s,u}. Resolves PR50232.
Differential Revision: https://reviews.llvm.org/D105675
Set the device malloc and free functions as weak,
and move the std headers after device malloc/free
to avoid issues with std malloc/free.
Fixes: SWDEV-293590
Reviewed By: yaxunl
Differential Revision: https://reviews.llvm.org/D105707
This adds a very basic test in `cuda_with_openmp.cu` that just checks whether the CUDA & OpenMP integrated headers do compile, when a CUDA file is compiled with OpenMP (CPU) enabled.
Thus this basically adds the missing test for https://reviews.llvm.org/D90415.
Reviewed By: jdoerfert
Differential Revision: https://reviews.llvm.org/D105322
This ensures that the mangled type names match between C and C++,
which is significant when using -fsanitize=cfi-icall. Ideally we
wouldn't have created this namespace at all, but it's now part of
the ABI (e.g. in mangled names), so we can't change it.
Differential Revision: https://reviews.llvm.org/D104830
This fixes issues with various return types(bool/int) and was already
in place for nvptx headers, adjusted to work for amdgcn. This does
not affect hip as the change is guarded with OPENMP_AMDGCN.
Similar to D85879.
Reviewed By: jdoerfert, JonChesterfield, yaxunl
Differential Revision: https://reviews.llvm.org/D104677
Add runtime functions to detect invalid calls to pure or deleted virtual
functions.
Patch by: Siu Chi Chan
Reviewed by: Yaxun Liu
Differential Revision: https://reviews.llvm.org/D104392
Add a test to verify OpenCL builtin declarations using
OpenCLBuiltins.td.
This test consists of parsing a 60k line generated input file. The
entire test takes about 60s with a debug build on a decent machine.
Admittedly this is not the fastest test, but doesn't seem excessive
compared to other tests in clang/test/Headers (with one of the tests
taking 85s for example).
RFC: https://lists.llvm.org/pipermail/cfe-dev/2021-April/067973.html
Differential Revision: https://reviews.llvm.org/D97869
Add the `memory_scope_all_devices` enum value, which is restricted to
OpenCL 3.0 or newer and the `__opencl_c_atomic_scope_all_devices`
feature. Also guard `memory_scope_all_svm_devices` accordingly, which
is already available in OpenCL 2.0.
The `__opencl_c_atomic_scope_all_devices` feature is header-only, so
set its define to 1 in `opencl-c-base.h`. This is done
unconditionally at the moment, as the mechanism for disabling
header-only options hasn't been decided yet.
This patch only adds a negative test for now. Ideally adding a CL3.0
run line to atomic-ops.cl should suffice as a positive test, but we
cannot do that yet until (at least) generic address spaces and program
scope variables are supported in OpenCL 3.0 mode.
Differential Revision: https://reviews.llvm.org/D103241
To improve hygiene, consistency, and usability, it would be good to replace all
the macro intrinsics in wasm_simd128.h with functions. The reason for using
macros in the first place was to enforce the use of constants for some arguments
using `_Static_assert` with `__builtin_constant_p`. This commit switches to
using functions and uses the `__diagnose_if__` attribute rather than
`_Static_assert` to enforce constantness.
The remaining macro intrinsics cannot be made into functions until the builtin
functions they are implemented with can be replaced with normal code patterns
because the builtin functions themselves require that their arguments are
constants.
This commit also fixes a bug with the const_splat intrinsics in which the f32x4
and f64x2 variants were incorrectly producing integer vectors.
Differential Revision: https://reviews.llvm.org/D102018
The builtins were updated to take signed parameters in 627a526955, but the
intrinsics that use those builtins were not updated as well. The intrinsic test
did not catch this sign mismatch because it is only reported as an error under
-fno-lax-vector-conversions.
This commit fixes the type mismatch and adds -fno-lax-vector-conversions to the
test to catch similar problems in the future.
Differential Revision: https://reviews.llvm.org/D101979
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
We do provide `operator delete(void*)` in `<new>` but it should be
available by default. This is mostly boilerplate to test it and the
unconditional include of `<new>` in the header we always in include
on the device.
Reviewed By: JonChesterfield
Differential Revision: https://reviews.llvm.org/D100620
These intrinsics do not correspond to their own underlying instruction, but are
a convenience for the common case of materializing a constant vector that has
the same value in each lane.
Differential Revision: https://reviews.llvm.org/D101885
Update the SIMD builtin load functions to take pointers to const data and update
the intrinsics themselves to not cast away constness.
Differential Revision: https://reviews.llvm.org/D101884
We previously did not have tests demonstrating that the intrinsics in
wasm_simd128.h lower to reasonable LLVM IR. This commit adds such a test.
Differential Revision: https://reviews.llvm.org/D101805
Artem Belevich