When the declare target variables are emitted for the device,
constructors|destructors for these variables must emitted and registered
by the runtime in the offloading sections.
llvm-svn: 328705
r327219 added wrappers to std::sort which randomly shuffle the container before
sorting. This will help in uncovering non-determinism caused due to undefined
sorting order of objects having the same key.
To make use of that infrastructure we need to invoke llvm::sort instead of
std::sort.
llvm-svn: 328636
If the link clause is used on the declare target directive, the object
should be linked on target or target data directives, not during the
codegen. Patch adds support for this clause.
llvm-svn: 328544
Summary:
Disables certain CMP optimizations to improve fuzzing signal under -O1
and -O2.
Switches all fuzzer tests to -O2 except for a few leak tests where the
leak is optimized out under -O2.
Reviewers: kcc, vitalybuka
Reviewed By: vitalybuka
Subscribers: cfe-commits, llvm-commits
Differential Revision: https://reviews.llvm.org/D44798
llvm-svn: 328384
Need to override convertConstraint to recognise amdgpu specific register names.
Differential Revision: https://reviews.llvm.org/D44533
llvm-svn: 328359
Add two additional implicit arguments for OpenCL for the AMDGPU target using the AMDHSA runtime to support device enqueue.
Differential Revision: https://reviews.llvm.org/D44696
llvm-svn: 328350
- Remove use of the opencl and amdopencl environment member of the target triple for the AMDGPU target.
- Use a function attribute to communicate to the AMDGPU backend.
Differential Revision: https://reviews.llvm.org/D43735
llvm-svn: 328347
The issues was that we were setting hidden visibility if, when
processing a hidden class, we found out that we needed to emit a
reference to a vtable provided by the standard library.
Original message:
Set dso_local on vtables.
llvm-svn: 328288
Putting back the code in commit r327189 that was reverted in r322737. The code is being committed in three stages and this one is the last stage: 1) r327455 fp16 feature flags, 2) r327836 pass half type or i16 based on FullFP16, and 3) the code here which the front-end fp16 vector intrinsic for ARM.
Differential revision https://reviews.llvm.org/D43650
llvm-svn: 328277
The difference between CreateRuntimeFunction and CreateBuiltinFunction
is that CreateBuiltinFunction would not set dllimport or dso_local.
To keep the current semantics, just forward to CreateRuntimeFunction
with Local=true so it doesn't add dllimport.
llvm-svn: 328224
Summary: The workers also need to initialize the global stack. The call to the initialization function needs to happen after the kernel_init() function is called by the master. This ensures that the per-team data structures of the runtime have been initialized.
Reviewers: ABataev, grokos, carlo.bertolli, caomhin
Reviewed By: ABataev
Subscribers: jholewinski, guansong, cfe-commits
Differential Revision: https://reviews.llvm.org/D44749
llvm-svn: 328219
Now that LLVM has support for emitting calling conventions in DWARF (see
r328191) have clang emit them.
Patch by: Adrien Guinet
Differential revision: https://reviews.llvm.org/D42351
llvm-svn: 328196
This is needed for the upcoming implementation of the
new 8x32x16 and 32x8x16 variants of WMMA instructions
introduced in CUDA 9.1.
Differential Revision: https://reviews.llvm.org/D44719
llvm-svn: 328158
Summary:
Libc++'s default allocator uses `__builtin_operator_new` and `__builtin_operator_delete` in order to allow the calls to new/delete to be ellided. However, libc++ now needs to support over-aligned types in the default allocator. In order to support this without disabling the existing optimization Clang needs to support calling the aligned new overloads from the builtins.
See llvm.org/PR22634 for more information about the libc++ bug.
This patch changes `__builtin_operator_new`/`__builtin_operator_delete` to call any usual `operator new`/`operator delete` function. It does this by performing overload resolution with the arguments passed to the builtin to determine which allocation function to call. If the selected function is not a usual allocation function a diagnostic is issued.
One open issue is if the `align_val_t` overloads should be considered "usual" when `LangOpts::AlignedAllocation` is disabled.
In order to allow libc++ to detect this new behavior the value for `__has_builtin(__builtin_operator_new)` has been updated to `201802`.
Reviewers: rsmith, majnemer, aaron.ballman, erik.pilkington, bogner, ahatanak
Reviewed By: rsmith
Subscribers: cfe-commits
Differential Revision: https://reviews.llvm.org/D43047
llvm-svn: 328134
This way we can support address-space specific variants without explicitly
encoding the space in the name of the intrinsic. Less intrinsics to deal with ->
less boilerplate.
Added a bit of tablegen magic to match/replace an intrinsics with a pointer
argument in particular address space with the space-specific instruction
variant.
Updated tests to use non-default address spaces.
Differential Revision: https://reviews.llvm.org/D43268
llvm-svn: 328006
constructs in generic mode.
Fixed codegen for distribute parallel combined constructs. We have to
pass and read the shared lower and upper bound from the distribute
region in the inner parallel region. Patch is for generic mode.
llvm-svn: 327990
If the generic codegen is enabled and private copy of the original
variable escapes the declaration context, this private copy should be
globalized just like it was the original variable.
llvm-svn: 327985
source expressions when iterating over a PseudoObjectExpr's semantic
subexpression list.
Previously the loop in emitPseudoObjectExpr would emit the IR for each
OpaqueValueExpr that was in a PseudoObjectExpr's semantic-form
expression list and use the result when the OpaqueValueExpr later
appeared in other expressions. This caused an assertion failure when
AggExprEmitter tried to copy the result of an OpaqueValueExpr and the
copied type didn't have trivial copy/move constructors or assignment
operators.
This patch adds flag IsUnique to OpaqueValueExpr which indicates it is a
unique reference to its source expression (it is not used in multiple
places). The loop in emitPseudoObjectExpr ignores OpaqueValueExprs that
are unique and CodeGen visitors simply traverse the source expressions
of such OpaqueValueExprs.
rdar://problem/34363596
Differential Revision: https://reviews.llvm.org/D39562
llvm-svn: 327939
The inline assembly generated for the ARC autorelease elision marker
must have a funclet token if it's emitted inside a funclet, otherwise
the inline assembly (and all subsequent code in the funclet) will be
marked unreachable. r324689 fixed this issue for regular inline assembly
blocks.
Note that clang only emits the marker at -O0, so this only fixes that
case. The optimizations case (where the marker is emitted by the
backend) will be fixed in a separate change.
Differential Revision: https://reviews.llvm.org/D44640
llvm-svn: 327892
This patch uses the infrastructure added in r326307 for enabling
non-trivial fields to be declared in C structs to allow __weak fields in
C structs in ARC.
This recommits r327206, which was reverted because it caused
module-enabled builders to fail. I discovered that the
CXXRecordDecl::CanPassInRegisters flag wasn't being set correctly in
some cases after I moved it to RecordDecl.
Thanks to Eric Liu for helping me investigate the bug.
rdar://problem/33599681
https://reviews.llvm.org/D44095
llvm-svn: 327870
For generating NEON intrinsics, this determines the NEON data type, and whether
it should be a half type or an i16 type. I.e., we always pass a half type for
AArch64, this hasn't changed, but now also for ARM but only when FullFP16 is
enabled, and i16 otherwise.
This is intended to be non-functional change, but together with the backend
work in D44538 which adds support for f16 vectors, this enables adding the
AArch32 FP16 (vector) intrinsics.
Differential Revision: https://reviews.llvm.org/D44561
llvm-svn: 327836
Summary:
The codegen for conditions assumes that a normal variable declaration is used in a condition, but this is not the case when a structured binding is used.
This fixes [PR36747](http://llvm.org/pr36747).
Thanks Nicolas Lesser for contributing the patch.
Reviewers: lichray, rsmith
Reviewed By: lichray
Subscribers: cfe-commits
Tags: #clang
Differential Revision: https://reviews.llvm.org/D44534
llvm-svn: 327780
The patch adds nocf_check target independent attribute for disabling checks that were enabled by cf-protection flag.
The attribute can be appertained to functions and function pointers.
Attribute name follows GCC's similar attribute name.
Differential Revision: https://reviews.llvm.org/D41880
llvm-svn: 327768
Summary:
Previously we tried too hard to uphold the fiction that destructor
variants work like they do on Itanium throughout the ABI-neutral parts
of clang. This lead to MS C++ ABI incompatiblities and other bugs. Now,
-mconstructor-aliases will no longer control this ABI detail, and clang
-cc1's LLVM IR output will be this much closer to the clang driver's.
Based on a patch by Zahira Ammarguellat:
https://reviews.llvm.org/D39063
I've tried to move the logic that Zahira added into MicrosoftCXXABI.cpp.
There is only one ABI-specific detail sticking out, and that is in
CodeGenModule::getAddrOfCXXStructor, where we collapse complete dtors to
base dtors in the MS ABI.
This fixes PR32990.
Reviewers: erichkeane, zahiraam, majnemer, rjmccall
Subscribers: cfe-commits
Differential Revision: https://reviews.llvm.org/D44505
llvm-svn: 327732
The compiler complained about
../tools/clang/lib/CodeGen/CGOpenMPRuntimeNVPTX.cpp:184:15: error: unused variable 'CSI' [-Werror,-Wunused-variable]
if (auto *CSI = CGF.CapturedStmtInfo) {
^
1 error generated.
I don't know this code but it seems like an easy fix so I push it anyway
to get rid of the warning.
llvm-svn: 327694
If the variable is captured by value and the corresponding parameter in
the outlined function escapes its declaration context, this parameter
must be globalized. To globalize it we need to get the address of the
original parameter, load the value, store it to the global address and
use this global address instead of the original.
Patch improves globalization for parallel|teams regions + functions in
declare target regions.
llvm-svn: 327654
Added initial codegen for device side of declarations inside `omp
declare target` construct + codegen for implicit `declare target`
functions, which are used in the target regions.
llvm-svn: 327636
In this particular case it would be possible to just add an else with
CGM.setDSOLocal(GV), but it seems better to have as many callers as
possible just call setGVProperties so that we can centralize the logic
there.
This patch then makes setGVProperties able to handle null Decls.
llvm-svn: 327543
Recent change r326946 (https://reviews.llvm.org/D34367) causes regression in Eigen due to increased
memory footprint of CallArg.
This patch reduces LValue size from 112 to 96 bytes and reduces inline argument count of CallArgList
from 16 to 8.
It has been verified that this will let the added deep AST tree test pass with r326946.
In the long run, CallArg or LValue memory footprint should be further optimized.
Differential Revision: https://reviews.llvm.org/D44445
llvm-svn: 327515
David Blaikie