This patch implements codegen for the reduction clause on
any parallel construct for elementary data types. An efficient
implementation requires hierarchical reduction within a
warp and a threadblock. It is complicated by the fact that
variables declared in the stack of a CUDA thread cannot be
shared with other threads.
The patch creates a struct to hold reduction variables and
a number of helper functions. The OpenMP runtime on the GPU
implements reduction algorithms that uses these helper
functions to perform reductions within a team. Variables are
shared between CUDA threads using shuffle intrinsics.
An implementation of reductions on the NVPTX device is
substantially different to that of CPUs. However, this patch
is written so that there are minimal changes to the rest of
OpenMP codegen.
The implemented design allows the compiler and runtime to be
decoupled, i.e., the runtime does not need to know of the
reduction operation(s), the type of the reduction variable(s),
or the number of reductions. The design also allows reuse of
host codegen, with appropriate specialization for the NVPTX
device.
While the patch does introduce a number of abstractions, the
expected use case calls for inlining of the GPU OpenMP runtime.
After inlining and optimizations in LLVM, these abstractions
are unwound and performance of OpenMP reductions is comparable
to CUDA-canonical code.
Patch by Tian Jin in collaboration with Arpith Jacob
Reviewers: ABataev
Differential Revision: https://reviews.llvm.org/D29758
llvm-svn: 295319
Support for CUDA printf is exploited to support printf for
an NVPTX OpenMP device.
To reflect the support of both programming models, the file
CGCUDABuiltin.cpp has been renamed to CGGPUBuiltin.cpp, and
the call EmitCUDADevicePrintfCallExpr has been renamed to
EmitGPUDevicePrintfCallExpr.
Reviewers: jlebar
Differential Revision: https://reviews.llvm.org/D17890
llvm-svn: 293444
in the current lexical scope.
clang currently emits the lifetime.start marker of a variable when the
variable comes into scope even though a variable's lifetime starts at
the entry of the block with which it is associated, according to the C
standard. This normally doesn't cause any problems, but in the rare case
where a goto jumps backwards past the variable declaration to an earlier
point in the block (see the test case added to lifetime2.c), it can
cause mis-compilation.
To prevent such mis-compiles, this commit conservatively disables
emitting lifetime variables when a label has been seen in the current
block.
This problem was discussed on cfe-dev here:
http://lists.llvm.org/pipermail/cfe-dev/2016-July/050066.html
rdar://problem/30153946
Differential Revision: https://reviews.llvm.org/D27680
llvm-svn: 293106
This patch adds support for codegen of 'target teams' on the host.
This combined directive has two captured statements, one for the
'teams' region, and the other for the 'parallel'.
This target teams region is offloaded using the __tgt_target_teams()
call. The patch sets the number of teams as an argument to
this call.
Reviewers: ABataev
Differential Revision: https://reviews.llvm.org/D29084
llvm-svn: 293005
This patch adds support for codegen of 'target teams' on the host.
This combined directive has two captured statements, one for the
'teams' region, and the other for the 'parallel'.
This target teams region is offloaded using the __tgt_target_teams()
call. The patch sets the number of teams as an argument to
this call.
Reviewers: ABataev
Differential Revision: https://reviews.llvm.org/D29084
llvm-svn: 293001
This patch adds support for codegen of 'target parallel' on the host.
It is also the first combined directive that requires two or more
captured statements. Support for this functionality is included in
the patch.
A combined directive such as 'target parallel' has two captured
statements, one for the 'target' and the other for the 'parallel'
region. Two captured statements are required because each has
different implicit parameters (see SemaOpenMP.cpp). For example,
the 'parallel' has 'global_tid' and 'bound_tid' while the 'target'
does not. The patch adds support for handling multiple captured
statements based on the combined directive.
When codegen'ing the 'target parallel' directive, the 'target'
outlined function is created using the outer captured statement
and the 'parallel' outlined function is created using the inner
captured statement.
Reviewers: ABataev
Differential Revision: https://reviews.llvm.org/D28753
llvm-svn: 292419
This patch adds support for codegen of 'target parallel' on the host.
It is also the first combined directive that requires two or more
captured statements. Support for this functionality is included in
the patch.
A combined directive such as 'target parallel' has two captured
statements, one for the 'target' and the other for the 'parallel'
region. Two captured statements are required because each has
different implicit parameters (see SemaOpenMP.cpp). For example,
the 'parallel' has 'global_tid' and 'bound_tid' while the 'target'
does not. The patch adds support for handling multiple captured
statements based on the combined directive.
When codegen'ing the 'target parallel' directive, the 'target'
outlined function is created using the outer captured statement
and the 'parallel' outlined function is created using the inner
captured statement.
Reviewers: ABataev
Differential Revision: https://reviews.llvm.org/D28753
llvm-svn: 292374
This patch refactors code that calls codegen for target regions. Currently
the codebase only supports the 'target' directive. The patch pulls out
common target processing code into a static function that can be called
by codegen for any target directive.
Reviewers: ABataev
Differential Revision: https://reviews.llvm.org/D28752
llvm-svn: 292134
This patch is to implement sema and parsing for 'target teams distribute simd’ pragma.
Differential Revision: https://reviews.llvm.org/D28252
llvm-svn: 291579
Summary:
This patch makes the type_mismatch static data 7 bytes smaller (and it
ends up being 16 bytes smaller due to alignment restrictions, at least
on some x86-64 environments).
It revs up the type_mismatch handler version since we're breaking binary
compatibility. I will soon post a patch for the compiler-rt side.
Reviewers: rsmith, kcc, vitalybuka, pgousseau, gbedwell
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D28242
llvm-svn: 291236
The special case to widen the integer literal zero when passed to
variadic function calls should only apply to variadic functions, not
unprototyped functions. This is consistent with what MSVC does. In this
test case, MSVC uses a 4-byte store to pass the 5th argument to 'kr' and
an 8-byte store to pass the zero to 'v':
void v(int, ...);
void kr();
void f(void) {
v(1, 2, 3, 4, 0);
kr(1, 2, 3, 4, 0);
}
Aaron Ballman discovered this issue in https://reviews.llvm.org/D28166
llvm-svn: 290906
This patch is to implement sema and parsing for 'target teams distribute parallel for simd’ pragma.
Differential Revision: https://reviews.llvm.org/D28202
llvm-svn: 290862
This patch is to implement sema and parsing for 'target teams distribute parallel for’ pragma.
Differential Revision: https://reviews.llvm.org/D28160
llvm-svn: 290725
This patch is to implement sema and parsing for 'target teams distribute' pragma.
Differential Revision: https://reviews.llvm.org/D28015
llvm-svn: 290508
This is a recommit of r290149, which was reverted in r290169 due to msan
failures. msan was failing because we were calling
`isMostDerivedAnUnsizedArray` on an invalid designator, which caused us
to read uninitialized memory. To fix this, the logic of the caller of
said function was simplified, and we now have a `!Invalid` assert in
`isMostDerivedAnUnsizedArray`, so we can catch this particular bug more
easily in the future.
Fingers crossed that this patch sticks this time. :)
Original commit message:
This patch does three things:
- Gives us the alloc_size attribute in clang, which lets us infer the
number of bytes handed back to us by malloc/realloc/calloc/any user
functions that act in a similar manner.
- Teaches our constexpr evaluator that evaluating some `const` variables
is OK sometimes. This is why we have a change in
test/SemaCXX/constant-expression-cxx11.cpp and other seemingly
unrelated tests. Richard Smith okay'ed this idea some time ago in
person.
- Uniques some Blocks in CodeGen, which was reviewed separately at
D26410. Lack of uniquing only really shows up as a problem when
combined with our new eagerness in the face of const.
llvm-svn: 290297
This commit fails MSan when running test/CodeGen/object-size.c in
a confusing way. After some discussion with George, it isn't really
clear what is going on here. We can make the MSan failure go away by
testing for the invalid bit, but *why* things are invalid isn't clear.
And yet, other code in the surrounding area is doing precisely this and
testing for invalid.
George is going to take a closer look at this to better understand the
nature of the failure and recommit it, for now backing it out to clean
up MSan builds.
llvm-svn: 290169
This patch does three things:
- Gives us the alloc_size attribute in clang, which lets us infer the
number of bytes handed back to us by malloc/realloc/calloc/any user
functions that act in a similar manner.
- Teaches our constexpr evaluator that evaluating some `const` variables
is OK sometimes. This is why we have a change in
test/SemaCXX/constant-expression-cxx11.cpp and other seemingly
unrelated tests. Richard Smith okay'ed this idea some time ago in
person.
- Uniques some Blocks in CodeGen, which was reviewed separately at
D26410. Lack of uniquing only really shows up as a problem when
combined with our new eagerness in the face of const.
Differential Revision: https://reviews.llvm.org/D14274
llvm-svn: 290149
copy constructors of classes with array members, instead using
ArrayInitLoopExpr to represent the initialization loop.
This exposed a bug in the static analyzer where it was unable to differentiate
between zero-initialized and unknown array values, which has also been fixed
here.
llvm-svn: 289618
This adds a way for us to version any UBSan handler by itself.
The patch overrides D21289 for a better implementation (we're able to
rev up a single handler).
After this, then we can land a slight modification of D19667+D19668.
We probably don't want to keep all the versions in compiler-rt (maybe we
want to deprecate on one release and remove the old handler on the next
one?), but with this patch we will loudly fail to compile when mixing
incompatible handler calls, instead of silently compiling and then
providing bad error messages.
Reviewers: kcc, samsonov, rsmith, vsk
Subscribers: cfe-commits
Differential Revision: https://reviews.llvm.org/D21695
llvm-svn: 289444
initialization of each array element:
* ArrayInitLoopExpr is a prvalue of array type with two subexpressions:
a common expression (an OpaqueValueExpr) that represents the up-front
computation of the source of the initialization, and a subexpression
representing a per-element initializer
* ArrayInitIndexExpr is a prvalue of type size_t representing the current
position in the loop
This will be used to replace the creation of explicit index variables in lambda
capture of arrays and copy/move construction of classes with array elements,
and also C++17 structured bindings of arrays by value (which inexplicably allow
copying an array by value, unlike all of C++'s other array declarations).
No uses of these nodes are introduced by this change, however.
llvm-svn: 289413
This patch is to implement sema and parsing for 'teams distribute parallel for' pragma.
Differential Revision: https://reviews.llvm.org/D27345
llvm-svn: 289179
This patch is to implement sema and parsing for 'teams distribute parallel for simd' pragma.
Differential Revision: https://reviews.llvm.org/D27084
llvm-svn: 288294
If 'omp cancel' construct is used in a worksharing construct it may
cause hanging of the software in case if reduction clause is used. Patch fixes this problem by avoiding extra reduction processing for branches that were canceled.
llvm-svn: 287227
Summary:
r286944 introduced bugs detected by ASAN as use-after-return.
r287025 have not fixed them completely.
This reverts commit r286944 and r287025.
Reviewers: ABataev
Subscribers: cfe-commits
Differential Revision: https://reviews.llvm.org/D26720
llvm-svn: 287069
If 'omp cancel' construct is used in a worksharing construct it may cause
hanging of the software in case if reduction clause is used. Patch fixes
this problem by avoiding extra reduction processing for branches that
were canceled.
llvm-svn: 286944
can be used to improve the locations when generating remarks for loops.
Depends on the companion LLVM change r286227.
Patch by Florian Hahn.
Differential Revision: https://reviews.llvm.org/D25764
llvm-svn: 286456
abstract information about the callee. NFC.
The goal here is to make it easier to recognize indirect calls and
trigger additional logic in certain cases. That logic will come in
a later patch; in the meantime, I felt that this was a significant
improvement to the code.
llvm-svn: 285258
Summary:
Current generation of lifetime intrinsics does not handle cases like:
```
{
char x;
l1:
bar(&x, 1);
}
goto l1;
```
We will get code like this:
```
%x = alloca i8, align 1
call void @llvm.lifetime.start(i64 1, i8* nonnull %x)
br label %l1
l1:
%call = call i32 @bar(i8* nonnull %x, i32 1)
call void @llvm.lifetime.end(i64 1, i8* nonnull %x)
br label %l1
```
So the second time bar was called for x which is marked as dead.
Lifetime markers here are misleading so it's better to remove them at all.
This type of bypasses are rare, e.g. code detects just 8 functions building
clang (2329 targets).
PR28267
Reviewers: eugenis
Subscribers: beanz, mgorny, cfe-commits
Differential Revision: https://reviews.llvm.org/D24693
llvm-svn: 285176
Summary: D24693 will need access to it from other places
Reviewers: eugenis
Subscribers: cfe-commits
Differential Revision: https://reviews.llvm.org/D24695
llvm-svn: 285158
constexpr variable.
When compiling a constexpr NSString initialized with an objective-c
string literal, CodeGen emits objc_storeStrong on an uninitialized
alloca, which causes a crash.
This patch folds the code in EmitScalarInit into EmitStoreThroughLValue
and fixes the crash by calling objc_retain on the string instead of
using objc_storeStrong.
rdar://problem/28562009
Differential Revision: https://reviews.llvm.org/D25547
llvm-svn: 284516
Summary: _BitScan intrinsics (and some others, for example _Interlocked and _bittest) are supposed to work on both ARM and x86. This is an attempt to isolate them, avoiding repeating their code or writing separate function for each builtin.
Reviewers: hans, thakis, rnk, majnemer
Subscribers: RKSimon, cfe-commits, aemerson
Differential Revision: https://reviews.llvm.org/D25264
llvm-svn: 284060
Summary:
With this commit simple coroutines can be created in plain C using coroutine builtins.
Reviewers: rnk, EricWF, rsmith
Subscribers: modocache, mgorny, mehdi_amini, beanz, cfe-commits
Differential Revision: https://reviews.llvm.org/D24373
llvm-svn: 283155
Instead of ignoring the evaluation order rule, ignore the "destroy parameters
in reverse construction order" rule for the small number of problematic cases.
This only causes incorrect behavior in the rare case where both parameters to
an overloaded operator <<, >>, ->*, &&, ||, or comma are of class type with
non-trivial destructor, and the program is depending on those parameters being
destroyed in reverse construction order.
We could do a little better here by reversing the order of parameter
destruction for those functions (and reversing the argument evaluation order
for all direct calls, not just those with operator syntax), but that is not a
complete solution to the problem, as the same situation can be reached by an
indirect function call.
Approach reviewed off-line by rnk.
llvm-svn: 282777
function correctly when targeting MS ABIs (this appears to have never mattered
prior to this change).
Update test case to always cover both 32-bit and 64-bit Windows ABIs, since
they behave somewhat differently from each other here.
Update test case to also cover operators , && and ||, which it appears are also
affected by P0145R3 (they're not explicitly called out by the design document,
but this is the emergent behavior of the existing wording).
Original commit message:
P0145R3 (C++17 evaluation order tweaks): evaluate the right-hand side of
assignment and compound-assignment operators before the left-hand side. (Even
if it's an overloaded operator.)
This completes the implementation of P0145R3 + P0400R0 for all targets except
Windows, where the evaluation order guarantees for <<, >>, and ->* are
unimplementable as the ABI requires the function arguments are evaluated from
right to left (because parameter destructors are run from left to right in the
callee).
llvm-svn: 282619
assignment and compound-assignment operators before the left-hand side. (Even
if it's an overloaded operator.)
This completes the implementation of P0145R3 + P0400R0 for all targets except
Windows, where the evaluation order guarantees for <<, >>, and ->* are
unimplementable as the ABI requires the function arguments are evaluated from
right to left (because parameter destructors are run from left to right in the
callee).
llvm-svn: 282556
* recurse through intermediate LabelStmts and AttributedStmts when checking
whether a statement inside a switch declares a variable
* if the end of a compound statement is reachable from the chosen case label,
and the compound statement contains a variable declaration, it's not valid
to just emit the contents of the compound statement -- we must emit the
statement itself or we lose the scope (and thus end lifetimes at the wrong
point)
llvm-svn: 281797
This reverts commit r279003 as it breaks some of our buildbots (e.g.
clang-cmake-aarch64-quick, clang-x86_64-linux-selfhost-modules).
The error is in OpenMP/teams_distribute_simd_ast_print.cpp:
clang: /home/buildslave/buildslave/clang-cmake-aarch64-quick/llvm/include/llvm/ADT/DenseMap.h:527:
bool llvm::DenseMapBase<DerivedT, KeyT, ValueT, KeyInfoT, BucketT>::LookupBucketFor(const LookupKeyT&, const BucketT*&) const
[with LookupKeyT = clang::Stmt*; DerivedT = llvm::DenseMap<clang::Stmt*, long unsigned int>;
KeyT = clang::Stmt*; ValueT = long unsigned int;
KeyInfoT = llvm::DenseMapInfo<clang::Stmt*>;
BucketT = llvm::detail::DenseMapPair<clang::Stmt*, long unsigned int>]:
Assertion `!KeyInfoT::isEqual(Val, EmptyKey) && !KeyInfoT::isEqual(Val, TombstoneKey) &&
"Empty/Tombstone value shouldn't be inserted into map!"' failed.
llvm-svn: 279045
This patch is to implement sema and parsing for 'teams distribute simd’ pragma.
This patch is originated by Carlo Bertolli.
Differential Revision: https://reviews.llvm.org/D23528
llvm-svn: 279003
Summary: This patch adds support for the use_device_ptr clause. It includes changes in SEMA that could not be tested without codegen, namely, the use of the first private logic and mappable expressions support.
Reviewers: hfinkel, carlo.bertolli, arpith-jacob, kkwli0, ABataev
Subscribers: caomhin, cfe-commits
Differential Revision: https://reviews.llvm.org/D22691
llvm-svn: 276977
This patch is to implement sema and parsing for 'target parallel for simd' pragma.
Differential Revision: http://reviews.llvm.org/D22096
llvm-svn: 275365
-fxray-instrument: enables XRay annotation of IR
-fxray-instruction-threshold: configures the threshold for function size (looking at IR instructions), and allow LLVM to decide whether to add the nop sleds later on in the process.
Also implements the related xray_always_instrument and xray_never_instrument function attributes.
Patch by Dean Michael Berris.
llvm-svn: 275330
Summary: This patch is an implementation of sema and parsing for the OpenMP composite pragma 'distribute simd'.
Differential Revision: http://reviews.llvm.org/D22007
llvm-svn: 274604
Summary: This patch is an implementation of sema and parsing for the OpenMP composite pragma 'distribute parallel for simd'.
Differential Revision: http://reviews.llvm.org/D21977
llvm-svn: 274530
With all MaterializeTemporaryExprs coming with a ExprWithCleanups, it's
easy to add correct lifetime.end marks into the right RunCleanupsScope.
Differential Revision: http://reviews.llvm.org/D20499
llvm-svn: 274385
Replace inheriting constructors implementation with new approach, voted into
C++ last year as a DR against C++11.
Instead of synthesizing a set of derived class constructors for each inherited
base class constructor, we make the constructors of the base class visible to
constructor lookup in the derived class, using the normal rules for
using-declarations.
For constructors, UsingShadowDecl now has a ConstructorUsingShadowDecl derived
class that tracks the requisite additional information. We create shadow
constructors (not found by name lookup) in the derived class to model the
actual initialization, and have a new expression node,
CXXInheritedCtorInitExpr, to model the initialization of a base class from such
a constructor. (This initialization is special because it performs real perfect
forwarding of arguments.)
In cases where argument forwarding is not possible (for inalloca calls,
variadic calls, and calls with callee parameter cleanup), the shadow inheriting
constructor is not emitted and instead we directly emit the initialization code
into the caller of the inherited constructor.
Note that this new model is not perfectly compatible with the old model in some
corner cases. In particular:
* if B inherits a private constructor from A, and C uses that constructor to
construct a B, then we previously required that A befriends B and B
befriends C, but the new rules require A to befriend C directly, and
* if a derived class has its own constructors (and so its implicit default
constructor is suppressed), it may still inherit a default constructor from
a base class
llvm-svn: 274049
[OpenMP] Initial implementation of parse and sema for composite pragma 'distribute parallel for'
This patch is an initial implementation for #distribute parallel for.
The main differences that affect other pragmas are:
The implementation of 'distribute parallel for' requires blocking of the associated loop, where blocks are "distributed" to different teams and iterations within each block are scheduled to parallel threads within each team. To implement blocking, sema creates two additional worksharing directive fields that are used to pass the team assigned block lower and upper bounds through the outlined function resulting from 'parallel'. In this way, scheduling for 'for' to threads can use those bounds.
As a consequence of blocking, the stride of 'distribute' is not 1 but it is equal to the blocking size. This is returned by the runtime and sema prepares a DistIncrExpr variable to hold that value.
As a consequence of blocking, the global upper bound (EnsureUpperBound) expression of the 'for' is not the original loop upper bound (e.g. in for(i = 0 ; i < N; i++) this is 'N') but it is the team-assigned block upper bound. Sema creates a new expression holding the calculation of the actual upper bound for 'for' as UB = min(UB, PrevUB), where UB is the loop upper bound, and PrevUB is the team-assigned block upper bound.
llvm-svn: 273884
http://reviews.llvm.org/D21564
This patch is an initial implementation for #distribute parallel for.
The main differences that affect other pragmas are:
The implementation of 'distribute parallel for' requires blocking of the associated loop, where blocks are "distributed" to different teams and iterations within each block are scheduled to parallel threads within each team. To implement blocking, sema creates two additional worksharing directive fields that are used to pass the team assigned block lower and upper bounds through the outlined function resulting from 'parallel'. In this way, scheduling for 'for' to threads can use those bounds.
As a consequence of blocking, the stride of 'distribute' is not 1 but it is equal to the blocking size. This is returned by the runtime and sema prepares a DistIncrExpr variable to hold that value.
As a consequence of blocking, the global upper bound (EnsureUpperBound) expression of the 'for' is not the original loop upper bound (e.g. in for(i = 0 ; i < N; i++) this is 'N') but it is the team-assigned block upper bound. Sema creates a new expression holding the calculation of the actual upper bound for 'for' as UB = min(UB, PrevUB), where UB is the loop upper bound, and PrevUB is the team-assigned block upper bound.
llvm-svn: 273705
Summary:
This patch fixes an issue detected when firstprivate variables are passed to an OpenMP outlined function vararg list. Currently they are not compatible with what the runtime library expects causing malfunction in some targets.
This patch fixes the issue by moving the casting logic already in place for offloading to the common code that creates the outline function and arguments and updates the regression tests accordingly.
Reviewers: hfinkel, arpith-jacob, carlo.bertolli, kkwli0, ABataev
Subscribers: cfe-commits, caomhin
Differential Revision: http://reviews.llvm.org/D21150
llvm-svn: 272900
Summary:
This patch is to add parsing and sema support for `target update` directive. Support for the `to` and `from` clauses will be added by a different patch. This patch also adds support for other clauses that are already implemented upstream and apply to `target update`, e.g. `device` and `if`.
This patch is based on the original post by Kelvin Li.
Reviewers: hfinkel, carlo.bertolli, kkwli0, arpith-jacob, ABataev
Subscribers: caomhin, cfe-commits
Differential Revision: http://reviews.llvm.org/D15944
llvm-svn: 270878
Underaligned atomic LValues require libcalls which MSVC doesn't have.
MSVC doesn't seem to consider such operations as requiring a barrier
anyway.
This fixes PR27843.
llvm-svn: 270576
For better performance and to unify code with offloading part we pass
scalar firstprivate values by value, instead of by reference. It will
remove some extra copying operations.
llvm-svn: 269751
schedule modifiers.
Runtime library expects some additional data in schedule argument for
loop-based directives, that have additional schedule modifiers
'monotonic|nonmonotonic'.
llvm-svn: 269035
Currently there is a problem with codegen of inlined directives inside
lambdas, it may cause a crash during codegen because of incorrect
capturing of variables. Patch fixes this problem.
llvm-svn: 267677
The taskloop construct specifies that the iterations of one or more associated loops will be executed in parallel using OpenMP tasks. The iterations are distributed across tasks created by the construct and scheduled to be executed.
The next code will be generated for the taskloop directive:
#pragma omp taskloop num_tasks(N) lastprivate(j)
for( i=0; i<N*GRAIN*STRIDE-1; i+=STRIDE ) {
int th = omp_get_thread_num();
#pragma omp atomic
counter++;
#pragma omp atomic
th_counter[th]++;
j = i;
}
Generated code:
task = __kmpc_omp_task_alloc(NULL,gtid,1,sizeof(struct
task),sizeof(struct shar),&task_entry);
psh = task->shareds;
psh->pth_counter = &th_counter;
psh->pcounter = &counter;
psh->pj = &j;
task->lb = 0;
task->ub = N*GRAIN*STRIDE-2;
task->st = STRIDE;
__kmpc_taskloop(
NULL, // location
gtid, // gtid
task, // task structure
1, // if clause value
&task->lb, // lower bound
&task->ub, // upper bound
STRIDE, // loop increment
0, // 1 if nogroup specified
2, // schedule type: 0-none, 1-grainsize, 2-num_tasks
N, // schedule value (ignored for type 0)
(void*)&__task_dup_entry // tasks duplication routine
);
llvm-svn: 267395
If loop control variable for simd-based directives is explicitly marked
as linear/lastprivate in clauses, codegen for such construct would
crash. Patch fixes this problem.
llvm-svn: 267101
Revert the two changes to thread CodeGenOptions into the TargetInfo allocation
and to fix the layering violation by moving CodeGenOptions into Basic.
Code Generation is arguably not particularly "basic". This addresses Richard's
post-commit review comments. This change purely does the mechanical revert and
will be followed up with an alternate approach to thread the desired information
into TargetInfo.
llvm-svn: 265806
This is a mechanical move of CodeGenOptions from libFrontend to libBasic. This
fixes the layering violation introduced earlier by threading CodeGenOptions into
TargetInfo. It should also fix the modules based self-hosting builds. NFC.
llvm-svn: 265702
Summary: See LLVM change D18775 for details, this change depends on it.
Reviewers: jyknight, reames
Subscribers: cfe-commits
Differential Revision: http://reviews.llvm.org/D18776
llvm-svn: 265569
Solution unifies interface of RegionCodeGenTy type to allow insert
runtime-specific code before/after main codegen action defined in
CGStmtOpenMP.cpp file. Runtime should not define its own RegionCodeGenTy
for general OpenMP directives, but must be allowed to insert its own
(required) code to support target specific codegen.
llvm-svn: 264700
Solution unifies interface of RegionCodeGenTy type to allow insert
runtime-specific code before/after main codegen action defined in
CGStmtOpenMP.cpp file. Runtime should not define its own RegionCodeGenTy
for general OpenMP directives, but must be allowed to insert its own
(required) code to support target specific codegen.
llvm-svn: 264576
Solution unifies interface of RegionCodeGenTy type to allow insert
runtime-specific code before/after main codegen action defined in
CGStmtOpenMP.cpp file. Runtime should not define its own RegionCodeGenTy
for general OpenMP directives, but must be allowed to insert its own
(required) code to support target specific codegen.
llvm-svn: 264569
This reverts commit r263607.
This change caused more objc_retain/objc_release calls in the IR but those
are then incorrectly optimized by the ARC optimizer. Work is going to have
to be done to ensure the ARC optimizer doesn't optimize user written RR, but
that should land before this change.
This change will also need to be updated to take account for any changes required
to ensure that user written calls to RR are distinct from those inserted by ARC.
llvm-svn: 263984
It is faster to directly call the ObjC runtime for methods such as retain/release instead of sending a message to those functions.
This patch adds support for converting messages to retain/release/alloc/autorelease to their equivalent runtime calls.
Tests included for the positive case of applying this transformation, negative tests that we ensure we only convert "alloc" to objc_alloc, not "alloc2", and also a driver test to ensure we enable this only for supported runtime versions.
Reviewed by John McCall.
Differential Revision: http://reviews.llvm.org/D14737
llvm-svn: 263607
OpenMP 4.5 allows privatization of non-static data members in OpenMP
constructs. Patch adds proper codegen support for data members in
'linear' clause
llvm-svn: 263003
This patch provide basic implementation of codegen for teams directive, excluding all clauses except dist_schedule. It also fixes parts of AST reader/writer to enable correct pre-compiled header handling.
http://reviews.llvm.org/D17170
llvm-svn: 262832
This patch provide basic implementation of codegen for teams directive, excluding all clauses except dist_schedule. It also fixes parts of AST reader/writer to enable correct pre-compiled header handling.
http://reviews.llvm.org/D17170
llvm-svn: 262741
We'd lose track of the parent CodeGenFunction, leading us to get
confused with regard to which function a nested finally belonged to.
Differential Revision: http://reviews.llvm.org/D17752
llvm-svn: 262379
This patch introduces the -fwhole-program-vtables flag, which enables the
whole-program vtable optimization feature (D16795) in Clang.
Differential Revision: http://reviews.llvm.org/D16821
llvm-svn: 261767
Expressions inside 'schedule'|'dist_schedule' clause must be captured in
combined directives to avoid possible crash during codegen. Patch
improves handling of such constructs
llvm-svn: 260954
This patch changes cc1 option -fprofile-instr-generate to an enum option
-fprofile-instrument={clang|none}. It also changes cc1 options
-fprofile-instr-generate= to -fprofile-instrument-path=.
The driver level option -fprofile-instr-generate and -fprofile-instr-generate=
remain intact. This change will pave the way to integrate new PGO
instrumentation in IR level.
Review: http://reviews.llvm.org/D16730
llvm-svn: 259811
Codegen for array sections/array subscripts worked only for expressions with arrays as base. Patch fixes codegen for bases with pointer/reference types.
llvm-svn: 259776
Summary:
This patch adds parsing + sema for the target parallel for directive along with testcases.
Reviewers: ABataev
Differential Revision: http://reviews.llvm.org/D16759
llvm-svn: 259654
reclaiming a call result in order to ignore it or assign it
to an __unsafe_unretained variable. This avoids adding
an unwanted retain/release pair when the return value is
not actually returned autoreleased (e.g. when it is returned
from a nonatomic getter or a typical collection accessor).
This runtime function is only available on the latest Apple
OS releases; the backwards-compatibility story is that you
don't get the optimization unless your deployment target is
recent enough. Sorry.
rdar://20530049
llvm-svn: 258962
Summary:
This patch adds parsing + sema for the target parallel directive and its clauses along with testcases.
Reviewers: ABataev
Differential Revision: http://reviews.llvm.org/D16553
Rebased to current trunk and updated test cases.
llvm-svn: 258832
* Runtime diagnostic data for cfi-icall changed to match the rest of
cfi checks
* Layout of all CFI diagnostic data changed to put Kind at the
beginning. There is no ABI stability promise yet.
* Call cfi_slowpath_diag instead of cfi_slowpath when needed.
* Emit __cfi_check_fail function, which dispatches a CFI check
faliure according to trap/recover settings of the current module.
* A tiny driver change to match the way the new handlers are done in
compiler-rt.
llvm-svn: 258745
This is part of a new statistics gathering feature for the sanitizers.
See clang/docs/SanitizerStats.rst for further info and docs.
Differential Revision: http://reviews.llvm.org/D16175
llvm-svn: 257971
Clang-side cross-DSO CFI.
* Adds a command line flag -f[no-]sanitize-cfi-cross-dso.
* Links a runtime library when enabled.
* Emits __cfi_slowpath calls is bitset test fails.
* Emits extra hash-based bitsets for external CFI checks.
* Sets a module flag to enable __cfi_check generation during LTO.
This mode does not yet support diagnostics.
llvm-svn: 255694
`pass_object_size` is our way of enabling `__builtin_object_size` to
produce high quality results without requiring inlining to happen
everywhere.
A link to the design doc for this attribute is available at the
Differential review link below.
Differential Revision: http://reviews.llvm.org/D13263
llvm-svn: 254554
Summary:
This patch implements the 4.5 specification for the implicit data maps. OpenMP 4.5 specification changes the default way data is captured into a target region. All the non-aggregate kinds are passed by value by default. This required activating the capturing by value during SEMA for the target region. All the non-aggregate values that can be encoded in the size of a pointer are properly casted and forwarded to the runtime library. On top of fixing the previous weird behavior for mapping pointers in nested data regions (an explicit map was always required), this also improves performance as the number of allocations/transactions to the device per non-aggregate map are reduced from two to only one - instead of passing a reference and the value, only the value passed.
Explicit maps will be added later on once firstprivate, private, and map clauses' SEMA and parsing are available.
Reviewers: hfinkel, rjmccall, ABataev
Subscribers: cfe-commits, carlo.bertolli
Differential Revision: http://reviews.llvm.org/D14940
llvm-svn: 254521
This patch changes the generation of CGFunctionInfo to contain
the FunctionProtoType if it is available. This enables the code
generation for call instructions to look into this type for
exception information and therefore generate better quality
IR - it will not create invoke instructions for functions that
are know not to throw.
llvm-svn: 253926
When a struct's size is not a power of 2, the corresponding _Atomic() type is
promoted to the nearest. We already correctly handled normal C++ expressions of
this form, but direct calls to the __c11_atomic_whatever builtins ended up
performing dodgy operations on the smaller non-atomic types (e.g. memcpy too
much). Later optimisations removed this as undefined behaviour.
This patch converts EmitAtomicExpr to allocate its temporaries at the full
atomic width, sidestepping the issue.
llvm-svn: 252507
This works around PR25162. The MSVC tables make it very difficult to
correctly inline a C++ destructor that contains try / catch. We've
attempted to address PR25162 in LLVM's backend, but it feels pretty
infeasible. MSVC and ICC both appear to avoid inlining such complex
destructors.
Long term, we want to fix this by making the inliner smart enough to
know when it is inlining into a cleanup, so it can inline simple
destructors (~unique_ptr and ~vector) while avoiding destructors
containing try / catch.
llvm-svn: 251576
match the feature set of the function that they're being called from.
This ensures that we can effectively diagnose some[1] code that would
instead ICE in the backend with a failure to select message.
Example:
__m128d foo(__m128d a, __m128d b) {
return __builtin_ia32_addsubps(b, a);
}
compiled for normal x86_64 via:
clang -target x86_64-linux-gnu -c
would fail to compile in the back end because the normal subtarget
features for x86_64 only include sse2 and the builtin requires sse3.
[1] We're still not erroring on:
__m128i bar(__m128i const *p) { return _mm_lddqu_si128(p); }
where we should fail and error on an always_inline function being
inlined into a function that doesn't support the subtarget features
required.
llvm-svn: 250473
This patch implements the outlining for offloading functions for code
annotated with the OpenMP target directive. It uses a temporary naming
of the outlined functions that will have to be updated later on once
target side codegen and registration of offloading libraries is
implemented - the naming needs to be made unique in the produced
library.
llvm-svn: 249148
Summary:
This change adds support for `__builtin_ms_va_list`, a GCC extension for
variadic `ms_abi` functions. The existing `__builtin_va_list` support is
inadequate for this because `va_list` is defined differently in the Win64
ABI vs. the System V/AMD64 ABI.
Depends on D1622.
Reviewers: rsmith, rnk, rjmccall
CC: cfe-commits
Differential Revision: http://reviews.llvm.org/D1623
llvm-svn: 247941
It seems that there is small bug, and we can't generate assume loads
when some virtual functions have internal visibiliy
This reverts commit 982bb7d966947812d216489b3c519c9825cacbf2.
llvm-svn: 247332
Currently all variables used in OpenMP regions are captured into a record and passed to outlined functions in this record. It may result in some poor performance because of too complex analysis later in optimization passes. Patch makes to emit outlined functions for parallel-based regions with a list of captured variables. It reduces code for 2*n GEPs, stores and loads at least.
Codegen for task-based regions remains unchanged because runtime requires that all captured variables are passed in captured record.
llvm-svn: 247251
Generating call assume(icmp %vtable, %global_vtable) after constructor
call for devirtualization purposes.
For more info go to:
http://lists.llvm.org/pipermail/cfe-dev/2015-July/044227.html
Edit:
Fixed version because of PR24479.
After this patch got reverted because of ScalarEvolution bug (D12719)
Merged after John McCall big patch (Added Address).
http://reviews.llvm.org/D11859
llvm-svn: 247199
Summary:
Currently clang provides no general way to generate nontemporal loads/stores.
There are some architecture specific builtins for doing so (e.g. in x86), but
there is no way to generate non-temporal store on, e.g. AArch64. This patch adds
generic builtins which are expanded to a simple store with '!nontemporal'
attribute in IR.
Differential Revision: http://reviews.llvm.org/D12313
llvm-svn: 247104
Introduce an Address type to bundle a pointer value with an
alignment. Introduce APIs on CGBuilderTy to work with Address
values. Change core APIs on CGF/CGM to traffic in Address where
appropriate. Require alignments to be non-zero. Update a ton
of code to compute and propagate alignment information.
As part of this, I've promoted CGBuiltin's EmitPointerWithAlignment
helper function to CGF and made use of it in a number of places in
the expression emitter.
The end result is that we should now be significantly more correct
when performing operations on objects that are locally known to
be under-aligned. Since alignment is not reliably tracked in the
type system, there are inherent limits to this, but at least we
are no longer confused by standard operations like derived-to-base
conversions and array-to-pointer decay. I've also fixed a large
number of bugs where we were applying the complete-object alignment
to a pointer instead of the non-virtual alignment, although most of
these were hidden by the very conservative approach we took with
member alignment.
Also, because IRGen now reliably asserts on zero alignments, we
should no longer be subject to an absurd but frustrating recurring
bug where an incomplete type would report a zero alignment and then
we'd naively do a alignmentAtOffset on it and emit code using an
alignment equal to the largest power-of-two factor of the offset.
We should also now be emitting much more aggressive alignment
attributes in the presence of over-alignment. In particular,
field access now uses alignmentAtOffset instead of min.
Several times in this patch, I had to change the existing
code-generation pattern in order to more effectively use
the Address APIs. For the most part, this seems to be a strict
improvement, like doing pointer arithmetic with GEPs instead of
ptrtoint. That said, I've tried very hard to not change semantics,
but it is likely that I've failed in a few places, for which I
apologize.
ABIArgInfo now always carries the assumed alignment of indirect and
indirect byval arguments. In order to cut down on what was already
a dauntingly large patch, I changed the code to never set align
attributes in the IR on non-byval indirect arguments. That is,
we still generate code which assumes that indirect arguments have
the given alignment, but we don't express this information to the
backend except where it's semantically required (i.e. on byvals).
This is likely a minor regression for those targets that did provide
this information, but it'll be trivial to add it back in a later
patch.
I partially punted on applying this work to CGBuiltin. Please
do not add more uses of the CreateDefaultAligned{Load,Store}
APIs; they will be going away eventually.
llvm-svn: 246985
Fix processing of shared variables with reference types in OpenMP constructs. Previously, if the variable was not marked in one of the private clauses, the reference to this variable was emitted incorrectly and caused an assertion later.
llvm-svn: 246846
This implements basic support for compiling (though not yet assembling
or linking) for a WebAssembly target. Note that ABI details are not yet
finalized, and may change.
Differential Revision: http://reviews.llvm.org/D12002
llvm-svn: 246814
Added codegen for array section in 'depend' clause of 'task' directive. It emits to pointers, one for the begin of array section and another for the end of array section. Size of the section is calculated as (end + 1 - start) * sizeof(basic_element_type).
llvm-svn: 246422
Added codegen for array section in 'depend' clause of 'task' directive. It emits to pointers, one for the begin of array section and another for the end of array section. Size of the section is calculated as (end + 1 - start) * sizeof(basic_element_type).
llvm-svn: 246278
Summary:
float_cast_overflow is the only UBSan check without a source location attached.
This patch propagates SourceLocations where necessary to get them to the
EmitCheck() call.
Reviewers: rsmith, ABataev, rjmccall
Subscribers: cfe-commits
Differential Revision: http://reviews.llvm.org/D11757
llvm-svn: 244568
The new EH instructions make it possible for LLVM to generate .xdata
tables that the MSVC personality routines will be happy about. Because
this is experimental, hide it behind a -cc1 flag (-fnew-ms-eh).
Differential Revision: http://reviews.llvm.org/D11405
llvm-svn: 243767
When ‘#pragma clang loop vectorize(assume_safety)’ was specified on a loop other loop hints were lost. The problem is that CGLoopInfo attaches metadata differently than EmitCondBrHints in CGStmt. For do-loops CGLoopInfo attaches metadata to the br in the body block and for while and for loops, the inc block. EmitCondBrHints on the other hand always attaches data to the br in the cond block. When specifying assume_safety CGLoopInfo emits an empty llvm.loop metadata shadowing the metadata in the cond block. Loop transformations like rotate and unswitch would then eliminate the cond block and its non-empty metadata.
This patch unifies both approaches for adding metadata and modifies the existing safety tests to include non-assume_safety loop hints.
llvm-svn: 243315
- Make it a proper random access iterator with a little help from iterator_adaptor_base
- Clean up users of magic dereferencing. The iterator should behave like an Expr **.
- Make it an implementation detail of Stmt. This allows inlining of the assertions.
llvm-svn: 242608
When messaging a method that was defined in an Objective-C class (or
category or extension thereof) that has type parameters, substitute
the type arguments for those type parameters. Similarly, substitute
into property accesses, instance variables, and other references.
This includes general infrastructure for substituting the type
arguments associated with an ObjCObject(Pointer)Type into a type
referenced within a particular context, handling all of the
substitutions required to deal with (e.g.) inheritance involving
parameterized classes. In cases where no type arguments are available
(e.g., because we're messaging via some unspecialized type, id, etc.),
we substitute in the type bounds for the type parameters instead.
Example:
@interface NSSet<T : id<NSCopying>> : NSObject <NSCopying>
- (T)firstObject;
@end
void f(NSSet<NSString *> *stringSet, NSSet *anySet) {
[stringSet firstObject]; // produces NSString*
[anySet firstObject]; // produces id<NSCopying> (the bound)
}
When substituting for the type parameters given an unspecialized
context (i.e., no specific type arguments were given), substituting
the type bounds unconditionally produces type signatures that are too
strong compared to the pre-generics signatures. Instead, use the
following rule:
- In covariant positions, such as method return types, replace type
parameters with “id” or “Class” (the latter only when the type
parameter bound is “Class” or qualified class, e.g,
“Class<NSCopying>”)
- In other positions (e.g., parameter types), replace type
parameters with their type bounds.
- When a specialized Objective-C object or object pointer type
contains a type parameter in its type arguments (e.g.,
NSArray<T>*, but not NSArray<NSString *> *), replace the entire
object/object pointer type with its unspecialized version (e.g.,
NSArray *).
llvm-svn: 241543
This reverts commit r241244, but restricts SEH support to Win64.
This way, Chromium builds will still fall back on TUs with SEH, and
Clang developers can work on this incrementally upstream while patching
this small predicate locally. It'll also make it easier to review small
fixes.
llvm-svn: 241533
This is needed to use clang's command line option "-ftrap-function" for LTO and
enable changing the trap function name on a per-call-site basis.
rdar://problem/21225723
Differential Revision: http://reviews.llvm.org/D10831
llvm-svn: 241306
The next code is generated for this construct:
```
if (__kmpc_cancellationpoint(ident_t *loc, kmp_int32 global_tid, kmp_int32 cncl_kind) != 0)
<exit from outer innermost construct>;
```
llvm-svn: 241239
32-bit finally funclets are intended to be called both directly from the
parent function and indirectly from the EH runtime. Because we aren't
contorting LLVM's X86 prologue to match MSVC's, calling the finally
block directly passes in a different value of EBP than the one that the
runtime provides. We need an adapter thunk to adjust EBP to the expected
value. However, WinEHPrepare already has to solve this problem when
cleanups are not pre-outlined, so we can go ahead and rely on it rather
than duplicating work.
Now we only do the llvm.x86.seh.recoverfp dance for 32-bit SEH filter
functions.
llvm-svn: 241187
This re-lands r236052 and adds support for __exception_code().
In 32-bit SEH, the exception code is not available in eax. It is only
available in the filter function, and now we arrange to load it and
store it into an escaped variable in the parent frame.
As a consequence, we have to disable the "catch i8* null" optimization
on 32-bit and always generate a filter function. We can re-enable the
optimization if we detect an __except block that doesn't use the
exception code, but this probably isn't worth optimizing.
Reviewers: majnemer
Differential Revision: http://reviews.llvm.org/D10852
llvm-svn: 241171
The LifetimeExtendedCleanupHeader is carefully fit into 32 bytes,
meaning that cleanups on the LifetimeExtendedCleanupStack are *always*
allocated at a misaligned address and cause undefined behaviour.
There are two ways to solve this - add padding after the header when
we allocated our cleanups, or just simplify the header and let it use
64 bits in the first place. I've opted for the latter, and added a
static assert to avoid the issue in the future.
llvm-svn: 241133
Integer variants are implemented as atomicrmw or cmpxchg instructions.
Atomic add for floating point (__nvvm_atom_add_gen_f()) is implemented
as a call to an overloaded @llvm.nvvm.atomic.load.add.f32.* LVVM
intrinsic.
Differential Revision: http://reviews.llvm.org/D10666
llvm-svn: 240669
This causes programs compiled with this flag to print a diagnostic when
a control flow integrity check fails instead of aborting. Diagnostics are
printed using UBSan's runtime library.
The main motivation of this feature over -fsanitize=vptr is fidelity with
the -fsanitize=cfi implementation: the diagnostics are printed under exactly
the same conditions as those which would cause -fsanitize=cfi to abort the
program. This means that the same restrictions apply regarding compiling
all translation units with -fsanitize=cfi, cross-DSO virtual calls are
forbidden, etc.
Differential Revision: http://reviews.llvm.org/D10268
llvm-svn: 240109
Added parsing, sema analysis and codegen for '#pragma omp taskgroup' directive (OpenMP 4.0).
The code for directive is generated the following way:
#pragma omp taskgroup
<body>
void __kmpc_taskgroup(<loc>, thread_id);
<body>
void __kmpc_end_taskgroup(<loc>, thread_id);
llvm-svn: 240011
Added codegen for combined 'omp for simd' directives, that is a combination of 'omp for' directive followed by 'omp simd' directive. Includes support for all clauses.
llvm-svn: 239990
Previously the last iteration for simd loop-based OpenMP constructs were generated as a separate code. This feature is not required and codegen is simplified.
llvm-svn: 239810
The fact that PGO has a say in how these branch weights are determined
isn't interesting to most of CodeGen, so it makes more sense for this
API to be accessible via CodeGenFunction rather than CodeGenPGO.
llvm-svn: 236380
This is just the clang-side of 32-bit SEH. LLVM still needs work, and it
will determinstically fail to compile until it's feature complete.
On x86, all outlined handlers have no parameters, but they do implicitly
take the EBP value passed in and use it to address locals of the parent
frame. We model this with llvm.frameaddress(1).
This works (mostly), but __finally block inlining can break it. For now,
we apply the 'noinline' attribute. If we really want to inline __finally
blocks on 32-bit x86, we should teach the inliner how to untangle
frameescape and framerecover.
Promote the error diagnostic from codegen to sema. It now rejects SEH on
non-Windows platforms. LLVM doesn't implement SEH on non-x86 Windows
platforms, but there's nothing preventing it.
llvm-svn: 236052
The RegionCounter type does a lot of legwork, but most of it is only
meaningful within the implementation of CodeGenPGO. The uses elsewhere
in CodeGen generally just want to increment or read counters, so do
that directly.
llvm-svn: 235664
Adds codegen for 'atomic capture' constructs with the following forms of expressions/statements:
v = x binop= expr;
v = x++;
v = ++x;
v = x--;
v = --x;
v = x = x binop expr;
v = x = expr binop x;
{v = x; x = binop= expr;}
{v = x; x++;}
{v = x; ++x;}
{v = x; x--;}
{v = x; --x;}
{x = x binop expr; v = x;}
{x binop= expr; v = x;}
{x++; v = x;}
{++x; v = x;}
{x--; v = x;}
{--x; v = x;}
{x = x binop expr; v = x;}
{x = expr binop x; v = x;}
{v = x; x = expr;}
If x and expr are integer and binop is associative or x is a LHS in a RHS of the assignment expression, and atomics are allowed for type of x on the target platform atomicrmw instruction is emitted.
Otherwise compare-and-swap sequence is emitted.
Update of 'v' is not required to be be atomic with respect to the read or write of the 'x'.
bb:
...
atomic load <x>
cont:
<expected> = phi [ <x>, label %bb ], [ <new_failed>, %cont ]
<desired> = <expected> binop <expr>
<res> = cmpxchg atomic &<x>, desired, expected
<new_failed> = <res>.field1;
br <res>field2, label %exit, label %cont
exit:
atomic store <old/new x>, <v>
...
Differential Revision: http://reviews.llvm.org/D9049
llvm-svn: 235573
This reverts commit r234700. It turns out that the lifetime markers
were not the cause of Chromium failing but a bug which was uncovered by
optimizations exposed by the markers.
llvm-svn: 235553
Add codegen for 'ordered' directive:
__kmpc_ordered(ident_t *, gtid);
<associated statement>;
__kmpc_end_ordered(ident_t *, gtid);
Also for 'for' directives with the dynamic scheduling and an 'ordered' clause added a call to '__kmpc_dispatch_fini_(4|8)[u]()' function after increment expression for loop control variable:
while(__kmpc_dispatch_next(&LB, &UB)) {
idx = LB;
while (idx <= UB) { BODY; ++idx;
__kmpc_dispatch_fini_(4|8)[u](); // For ordered loops only.
} // inner loop
}
Differential Revision: http://reviews.llvm.org/D9070
llvm-svn: 235496
Emits the following code for the clause at the beginning of the outlined function for implicit threads:
if (<not a master thread>) {
...
<thread local copy of var> = <master thread local copy of var>;
...
}
<sync point>;
Checking for a non-master thread is performed by comparing of the address of the thread local variable with the address of the master's variable. Master thread always uses original variables, so you always know the address of the variable in the master thread.
Differential Revision: http://reviews.llvm.org/D9026
llvm-svn: 235075
#pragma omp for lastprivate(<var>)
for (i = a; i < b; ++b)
<BODY>;
This construct is translated into something like:
<last_iter> = alloca i32
<lastprivate_var> = alloca <type>
<last_iter> = 0
; No initializer for simple variables or a default constructor is called for objects.
; For arrays perform element by element initialization by the call of the default constructor.
...
OMP_FOR_START(...,<last_iter>, ..); sets <last_iter> to 1 if this is the last iteration.
<BODY>
...
OMP_FOR_END
if (<last_iter> != 0) {
<var> = <lastprivate_var> ; Update original variable with the lastprivate value.
}
call __kmpc_cancel_barrier() ; an implicit barrier to avoid possible data race.
Differential Revision: http://reviews.llvm.org/D8658
llvm-svn: 235074
Adds proper codegen for 'firstprivate' clause in for directive. Initially codegen for 'firstprivate' clause was implemented for 'parallel' directive only.
Also this patch emits sync point only after initialization of firstprivate variables, not all private variables. This sync point is not required for privates, lastprivates etc., only for initialization of firstprivate variables.
Differential Revision: http://reviews.llvm.org/D8660
llvm-svn: 234978
Fixed a bug with codegen of variables with array types specified in 'copyprivate' clause of 'single' directive.
Differential Revision: http://reviews.llvm.org/D8914
llvm-svn: 234856
Even though these symbols are in a comdat group, the Microsoft linker
really wants them to have internal linkage.
I'm planning to tweak the mangling in a follow-up change. This is a
straight revert with a 1-line fix.
llvm-svn: 234613
Now that TailRecursionElimination has been fixed with r222354, the
threshold on size for lifetime marker insertion can be removed. This
only affects named temporary though, as the patch for unnamed temporaries
is still in progress.
My previous commit (r222993) was not handling debuginfo correctly, but
this could only be seen with some asan tests. Basically, lifetime markers
are just instrumentation for the compiler's usage and should not affect
debug information; however, the cleanup infrastructure was assuming it
contained only destructors, i.e. actual code to be executed, and was
setting the breakpoint for the end of the function to the closing '}', and
not the return statement, in order to show some destructors have been
called when leaving the function. This is wrong when the cleanups are only
lifetime markers, and this is now fixed.
llvm-svn: 234581
WinEHPrepare was going to have to pattern match the control flow merge
and split that the old lowering used, and that wasn't really feasible.
Now we can teach WinEHPrepare to pattern match this, which is much
simpler:
%fp = call i8* @llvm.frameaddress(i32 0)
call void @func(iN [01], i8* %fp)
This prototype happens to match the prototype used by the Win64 SEH
personality function, so this is really simple.
llvm-svn: 234532
The test should be fixed. It was failing in NDEBUG builds due to a
missing '*' character in a regex. In asserts builds, the pattern matched
a single digit value, which became a double digit value in NDEBUG
builds. Go figure.
This reverts commit r234261.
llvm-svn: 234447
While capturing filters aren't very common, we'd like to outline
__finally blocks in the frontend to simplify -O0 EH preparation and
reduce code size. Finally blocks are usually have captures, and this is
the first step towards that.
Currently we don't support capturing 'this' or VLAs.
Reviewers: majnemer
Differential Revision: http://reviews.llvm.org/D8825
llvm-svn: 234261
The zEC12 provides the transactional-execution facility. This is exposed
to users via a set of builtin routines on other compilers. This patch
adds clang support to enable those builtins. In partciular, the patch:
- enables the transactional-execution feature by default on zEC12
- allows to override presence of that feature via the -mhtm/-mno-htm options
- adds a predefined macro __HTM__ if the feature is enabled
- adds support for the transactional-execution GCC builtins
- adds Sema checking to verify the __builtin_tabort abort code
- adds the s390intrin.h header file (for GCC compatibility)
- adds s390 sections to the htmintrin.h and htmxlintrin.h header files
Since this is first use of target-specific intrinsics on the platform,
the patch creates the include/clang/Basic/BuiltinsSystemZ.def file and
hooks it up in TargetBuiltins.h and lib/Basic/Targets.cpp.
An associated LLVM patch adds the required LLVM IR intrinsics.
For reference, the transactional-execution instructions are documented
in the z/Architecture Principles of Operation for the zEC12:
http://publibfp.boulder.ibm.com/cgi-bin/bookmgr/download/DZ9ZR009.pdf
The associated builtins are documented in the GCC manual:
http://gcc.gnu.org/onlinedocs/gcc/S_002f390-System-z-Built-in-Functions.html
The htmxlintrin.h intrinsics provided for compatibility with the IBM XL
compiler are documented in the "z/OS XL C/C++ Programming Guide".
llvm-svn: 233804
Adds atomic update codegen for the following forms of expressions:
x binop= expr;
x++;
++x;
x--;
--x;
x = x binop expr;
x = expr binop x;
If x and expr are integer and binop is associative or x is a LHS in a RHS of the assignment expression, and atomics are allowed for type of x on the target platform atomicrmw instruction is emitted.
Otherwise compare-and-swap sequence is emitted:
bb:
...
atomic load <x>
cont:
<expected> = phi [ <x>, label %bb ], [ <new_failed>, %cont ]
<desired> = <expected> binop <expr>
<res> = cmpxchg atomic &<x>, desired, expected
<new_failed> = <res>.field1;
br <res>field2, label %exit, label %cont
exit:
...
Differential Revision: http://reviews.llvm.org/D8536
llvm-svn: 233513
If there is at least one 'copyprivate' clause is associated with the single directive, the following code is generated:
```
i32 did_it = 0; \\ for 'copyprivate' clause
if(__kmpc_single(ident_t *, gtid)) {
SingleOpGen();
__kmpc_end_single(ident_t *, gtid);
did_it = 1; \\ for 'copyprivate' clause
}
<copyprivate_list>[0] = &var0;
...
<copyprivate_list>[n] = &varn;
call __kmpc_copyprivate(ident_t *, gtid, <copyprivate_list_size>,
<copyprivate_list>, <copy_func>, did_it);
...
void<copy_func>(void *LHSArg, void *RHSArg) {
Dst = (void * [n])(LHSArg);
Src = (void * [n])(RHSArg);
Dst[0] = Src[0];
... Dst[n] = Src[n];
}
```
All list items from all 'copyprivate' clauses are gathered into single <copyprivate list> (<copyprivate_list_size> is a size in bytes of this list) and <copy_func> is used to propagate values of private or threadprivate variables from the 'single' region to other implicit threads from outer 'parallel' region.
Differential Revision: http://reviews.llvm.org/D8410
llvm-svn: 232932
This scheme checks that pointer and lvalue casts are made to an object of
the correct dynamic type; that is, the dynamic type of the object must be
a derived class of the pointee type of the cast. The checks are currently
only introduced where the class being casted to is a polymorphic class.
Differential Revision: http://reviews.llvm.org/D8312
llvm-svn: 232241
This is complicated by the fact that we can't simply use side-effecting
calls in an argument list without losing all guarantees about the order
they're emitted. To keep things deterministic we use tuples and brace
initialization, which thankfully guarantees evaluation order.
No functionality change intended.
llvm-svn: 232121
Throwing a C++ exception, under the MS ABI, is implemented using three
components:
- ThrowInfo structure which contains information like CV qualifiers,
what destructor to call and a pointer to the CatchableTypeArray.
- In a significant departure from the Itanium ABI, copying by-value
occurs in the runtime and not at the catch site. This means we need
to enumerate all possible types that this exception could be caught as
and encode the necessary information to convert from the exception
object's type to the catch handler's type. This includes complicated
derived to base conversions and the execution of copy-constructors.
N.B. This implementation doesn't support the execution of a
copy-constructor from within the runtime for now. Adding support for
that functionality is quite difficult due to things like default
argument expressions which may evaluate arbitrary code hiding in the
copy-constructor's parameters.
Differential Revision: http://reviews.llvm.org/D8066
llvm-svn: 231328
This patch introduces the -fsanitize=cfi-vptr flag, which enables a control
flow integrity scheme that checks that virtual calls take place using a vptr of
the correct dynamic type. More details in the new docs/ControlFlowIntegrity.rst
file.
It also introduces the -fsanitize=cfi flag, which is currently a synonym for
-fsanitize=cfi-vptr, but will eventually cover all CFI checks implemented
in Clang.
Differential Revision: http://reviews.llvm.org/D7424
llvm-svn: 230055
The /volatile:ms semantics turn volatile loads and stores into atomic
acquire and release operations. This distinction is important because
volatile memory operations do not form a happens-before relationship
with non-atomic memory. This means that a volatile store is not
sufficient for implementing a mutex unlock routine.
Differential Revision: http://reviews.llvm.org/D7580
llvm-svn: 229082
Previously we would simply double-emit the body of the __finally block,
but that doesn't work when it contains any kind of Decl, which we can't
double emit.
This fixes that by emitting the block once and branching into a shared
code region and then branching back out.
llvm-svn: 228222
Now if you break on a dtor and go 'up' in your debugger (or you get an
asan failure in a dtor) during an exception unwind, you'll have more
context. Instead of all dtors appearing to be called from the '}' of the
function, they'll be attributed to the end of the scope of the variable,
the same as the non-exceptional dtor call.
This doesn't /quite/ remove all uses of CurEHLocation (which might be
nice to remove, for a few reasons) - it's still used to choose the
location for some other work in the landing pad. It'd be nice to
attribute that code to the same location as the exception calls within
the block and to remove CurEHLocation.
llvm-svn: 228181
We would synthesize memcpy intrinsics when emitting calls to trivial C++
constructors but we wouldn't take into account the alignment of the
destination.
llvm-svn: 228061
The lowering looks a lot like normal EH lowering, with the exception
that the exceptions are caught by executing filter expression code
instead of matching typeinfo globals. The filter expressions are
outlined into functions which are used in landingpad clauses where
typeinfo would normally go.
Major aspects that still need work:
- Non-call exceptions in __try bodies won't work yet. The plan is to
outline the __try block in the frontend to keep things simple.
- Filter expressions cannot use local variables until capturing is
implemented.
- __finally blocks will not run after exceptions. Fixing this requires
work in the LLVM SEH preparation pass.
The IR lowering looks like this:
// C code:
bool safe_div(int n, int d, int *r) {
__try {
*r = normal_div(n, d);
} __except(_exception_code() == EXCEPTION_INT_DIVIDE_BY_ZERO) {
return false;
}
return true;
}
; LLVM IR:
define i32 @filter(i8* %e, i8* %fp) {
%ehptrs = bitcast i8* %e to i32**
%ehrec = load i32** %ehptrs
%code = load i32* %ehrec
%matches = icmp eq i32 %code, i32 u0xC0000094
%matches.i32 = zext i1 %matches to i32
ret i32 %matches.i32
}
define i1 zeroext @safe_div(i32 %n, i32 %d, i32* %r) {
%rr = invoke i32 @normal_div(i32 %n, i32 %d)
to label %normal unwind to label %lpad
normal:
store i32 %rr, i32* %r
ret i1 1
lpad:
%ehvals = landingpad {i8*, i32} personality i32 (...)* @__C_specific_handler
catch i8* bitcast (i32 (i8*, i8*)* @filter to i8*)
%ehptr = extractvalue {i8*, i32} %ehvals, i32 0
%sel = extractvalue {i8*, i32} %ehvals, i32 1
%filter_sel = call i32 @llvm.eh.seh.typeid.for(i8* bitcast (i32 (i8*, i8*)* @filter to i8*))
%matches = icmp eq i32 %sel, %filter_sel
br i1 %matches, label %eh.except, label %eh.resume
eh.except:
ret i1 false
eh.resume:
resume
}
Reviewers: rjmccall, rsmith, majnemer
Differential Revision: http://reviews.llvm.org/D5607
llvm-svn: 226760
This workaround was to provide unique call sites to ensure LLVM's inline
debug info handling would properly unique two calls to the same function
on the same line. Instead, this has now been fixed in LLVM (r226736) and
the workaround here can be removed.
Originally committed in r176895, but this isn't a straight revert due to
all the changes since then. I just searched for anything ForcedColumn*
related and removed them.
We could test this - but it didn't strike me as terribly valuable once
we're no longer adding this workaround everything just works as expected
& it's no longer a special case to test for.
llvm-svn: 226738
Several pieces of code were relying on implicit debug location setting
which usually lead to incorrect line information anyway. So I've fixed
those (in r225955 and r225845) separately which should pave the way for
this commit to be cleanly reapplied.
The reason these implicit dependencies resulted in crashes with this
patch is that the debug location would no longer implicitly leak from
one place to another, but be set back to invalid. Once a call with
no/invalid location was emitted, if that call was ever inlined it could
produce invalid debugloc chains and assert during LLVM's codegen.
There may be further cases of such bugs in this patch - they're hard to
flush out with regression testing, so I'll keep an eye out for reports
and investigate/fix them ASAP if they come up.
Original commit message:
Reapply "DebugInfo: Generalize debug info location handling"
Originally committed in r224385 and reverted in r224441 due to concerns
this change might've introduced a crash. Turns out this change fixes the
crash introduced by one of my earlier more specific location handling
changes (those specific fixes are reverted by this patch, in favor of
the more general solution).
Recommitted in r224941 and reverted in r224970 after it caused a crash
when building compiler-rt. Looks to be due to this change zeroing out
the debug location when emitting default arguments (which were meant to
inherit their outer expression's location) thus creating call
instructions without locations - these create problems for inlining and
must not be created. That is fixed and tested in this version of the
change.
Original commit message:
This is a more scalable (fixed in mostly one place, rather than many
places that will need constant improvement/maintenance) solution to
several commits I've made recently to increase source fidelity for
subexpressions.
This resetting had to be done at the DebugLoc level (not the
SourceLocation level) to preserve scoping information (if the resetting
was done with CGDebugInfo::EmitLocation, it would've caused the tail end
of an expression's codegen to end up in a potentially different scope
than the start, even though it was at the same source location). The
drawback to this is that it might leave CGDebugInfo out of sync. Ideally
CGDebugInfo shouldn't have a duplicate sense of the current
SourceLocation, but for now it seems it does... - I don't think I'm
going to tackle removing that just now.
I expect this'll probably cause some more buildbot fallout & I'll
investigate that as it comes up.
Also these sort of improvements might be starting to show a weakness/bug
in LLVM's line table handling: we don't correctly emit is_stmt for
statements, we just put it on every line table entry. This means one
statement split over multiple lines appears as multiple 'statements' and
two statements on one line (without column info) are treated as one
statement.
I don't think we have any IR representation of statements that would
help us distinguish these cases and identify the beginning of each
statement - so that might be something we need to add (possibly to the
lexical scope chain - a scope for each statement). This does cause some
problems for GDB and possibly other DWARF consumers.
llvm-svn: 225956
This reverts commit r225000, r225021, r225083, r225086, r225090.
The root change (r225000) still has several issues where it's caused
calls to be emitted without debug locations. This causes assertion
failures if/when those calls are inlined.
I'll work up some test cases and fixes before recommitting this.
llvm-svn: 225555
Originally committed in r224385 and reverted in r224441 due to concerns
this change might've introduced a crash. Turns out this change fixes the
crash introduced by one of my earlier more specific location handling
changes (those specific fixes are reverted by this patch, in favor of
the more general solution).
Recommitted in r224941 and reverted in r224970 after it caused a crash
when building compiler-rt. Looks to be due to this change zeroing out
the debug location when emitting default arguments (which were meant to
inherit their outer expression's location) thus creating call
instructions without locations - these create problems for inlining and
must not be created. That is fixed and tested in this version of the
change.
Original commit message:
This is a more scalable (fixed in mostly one place, rather than many
places that will need constant improvement/maintenance) solution to
several commits I've made recently to increase source fidelity for
subexpressions.
This resetting had to be done at the DebugLoc level (not the
SourceLocation level) to preserve scoping information (if the resetting
was done with CGDebugInfo::EmitLocation, it would've caused the tail end
of an expression's codegen to end up in a potentially different scope
than the start, even though it was at the same source location). The
drawback to this is that it might leave CGDebugInfo out of sync. Ideally
CGDebugInfo shouldn't have a duplicate sense of the current
SourceLocation, but for now it seems it does... - I don't think I'm
going to tackle removing that just now.
I expect this'll probably cause some more buildbot fallout & I'll
investigate that as it comes up.
Also these sort of improvements might be starting to show a weakness/bug
in LLVM's line table handling: we don't correctly emit is_stmt for
statements, we just put it on every line table entry. This means one
statement split over multiple lines appears as multiple 'statements' and
two statements on one line (without column info) are treated as one
statement.
I don't think we have any IR representation of statements that would
help us distinguish these cases and identify the beginning of each
statement - so that might be something we need to add (possibly to the
lexical scope chain - a scope for each statement). This does cause some
problems for GDB and possibly other DWARF consumers.
llvm-svn: 225000
Originally committed in r224385 and reverted in r224441 due to concerns
this change might've introduced a crash. Turns out this change fixes the
crash introduced by one of my earlier more specific location handling
changes (those specific fixes are reverted by this patch, in favor of
the more general solution).
Original commit message:
This is a more scalable (fixed in mostly one place, rather than many
places that will need constant improvement/maintenance) solution to
several commits I've made recently to increase source fidelity for
subexpressions.
This resetting had to be done at the DebugLoc level (not the
SourceLocation level) to preserve scoping information (if the resetting
was done with CGDebugInfo::EmitLocation, it would've caused the tail end
of an expression's codegen to end up in a potentially different scope
than the start, even though it was at the same source location). The
drawback to this is that it might leave CGDebugInfo out of sync. Ideally
CGDebugInfo shouldn't have a duplicate sense of the current
SourceLocation, but for now it seems it does... - I don't think I'm
going to tackle removing that just now.
I expect this'll probably cause some more buildbot fallout & I'll
investigate that as it comes up.
Also these sort of improvements might be starting to show a weakness/bug
in LLVM's line table handling: we don't correctly emit is_stmt for
statements, we just put it on every line table entry. This means one
statement split over multiple lines appears as multiple 'statements' and
two statements on one line (without column info) are treated as one
statement.
I don't think we have any IR representation of statements that would
help us distinguish these cases and identify the beginning of each
statement - so that might be something we need to add (possibly to the
lexical scope chain - a scope for each statement). This does cause some
problems for GDB and possibly other DWARF consumers.
llvm-svn: 224941
Fixed assertion on type checking for arguments and parameters on function call if arguments are pointers to VLA
Differential Revision: http://reviews.llvm.org/D6655
llvm-svn: 224504
This is a more scalable (fixed in mostly one place, rather than many
places that will need constant improvement/maintenance) solution to
several commits I've made recently to increase source fidelity for
subexpressions.
This resetting had to be done at the DebugLoc level (not the
SourceLocation level) to preserve scoping information (if the resetting
was done with CGDebugInfo::EmitLocation, it would've caused the tail end
of an expression's codegen to end up in a potentially different scope
than the start, even though it was at the same source location). The
drawback to this is that it might leave CGDebugInfo out of sync. Ideally
CGDebugInfo shouldn't have a duplicate sense of the current
SourceLocation, but for now it seems it does... - I don't think I'm
going to tackle removing that just now.
I expect this'll probably cause some more buildbot fallout & I'll
investigate that as it comes up.
Also these sort of improvements might be starting to show a weakness/bug
in LLVM's line table handling: we don't correctly emit is_stmt for
statements, we just put it on every line table entry. This means one
statement split over multiple lines appears as multiple 'statements' and
two statements on one line (without column info) are treated as one
statement.
I don't think we have any IR representation of statements that would
help us distinguish these cases and identify the beginning of each
statement - so that might be something we need to add (possibly to the
lexical scope chain - a scope for each statement). This does cause some
problems for GDB and possibly other DWARF consumers.
llvm-svn: 224385
Currently, if global variable is marked as a private OpenMP variable, the compiler crashes in debug version or generates incorrect code in release version. It happens because in the OpenMP region the original global variable is used instead of the generated private copy. It happens because currently globals variables are not captured in the OpenMP region.
This patch adds capturing of global variables iff private copy of the global variable must be used in the OpenMP region.
Differential Revision: http://reviews.llvm.org/D6259
llvm-svn: 224323
the simplest case, which is used when no chunk_size is specified in
the schedule(static) or no 'schedule' clause is specified - the
iteration space is divided by the library into chunks that are
approximately equal in size, and at most one chunk is distributed
to each thread. In this case, we do not need an outer loop in each
thread - each thread requests once which iterations range it should
handle (using __kmpc_for_static_init runtime call) and then runs the
inner loop on this range.
Differential Revision: http://reviews.llvm.org/D5865
llvm-svn: 224233
Currently clang fires assertions on x86-64 on any atomic operations for long double operands. Patch fixes codegen for such operations.
Differential Revision: http://reviews.llvm.org/D6499
llvm-svn: 224230
The extension has the following syntax:
__builtin_call_with_static_chain(Call, Chain)
where Call must be a function call expression and Chain must be of pointer type
This extension performs a function call Call with a static chain pointer
Chain passed to the callee in a designated register. This is useful for
calling foreign language functions whose ABI uses static chain pointers
(e.g. to implement closures).
Differential Revision: http://reviews.llvm.org/D6332
llvm-svn: 224167
This particularly helps the fidelity of ASan reports (which can occur
even in these examples - if, for example, one uses placement new over a
buffer of insufficient size - now ASan will correctly identify which
member's initialization went over the end of the buffer).
This doesn't cover all types of members - more coming.
llvm-svn: 223726
Create a helper function to construct a value for the ARM hint intrinsic
rather than inling the construction. In order to avoid the use of the sentinel
value, inline the use of intrinsic instruction retrieval. NFC.
llvm-svn: 223338
Consider this program:
struct A {
virtual void operator-() { printf("base\n"); }
};
struct B final : public A {
virtual void operator-() override { printf("derived\n"); }
};
int main() {
B* b = new B;
-static_cast<A&>(*b);
}
Before this patch, clang saw the virtual call to A::operator-(), figured out
that it can be devirtualized, and then just called A::operator-() directly,
without going through the vtable. Instead, it should've looked up which
operator-() the call devirtualizes to and should've called that.
For regular virtual member calls, clang gets all this right already. So
instead of giving EmitCXXOperatorMemberCallee() all the logic that
EmitCXXMemberCallExpr() already has, cut the latter function into two pieces,
call the second piece EmitCXXMemberOrOperatorMemberCallExpr(), and use it also
to generate code for calls to virtual member operators.
This way, virtual overloaded operators automatically don't get devirtualized
if they have covariant returns (like it was done for regular calls in r218602),
etc.
This also happens to fix (or at least improve) codegen for explicit constructor
calls (`A a; a.A::A()`) in MS mode with -fsanitize-address-field-padding=1.
(This adjustment for virtual operator calls seems still wrong with the MS ABI.)
llvm-svn: 223185
Now that TailRecursionElimination has been fixed with r222354, the
threshold on size for lifetime marker insertion can be removed. This
only affects named temporary though, as the patch for unnamed temporaries
is still in progress.
llvm-svn: 222993
Summary:
This change makes CodeGenFunction::EmitCheck() take several
conditions that needs to be checked (all of them need to be true),
together with sanitizer kinds these checks are for. This would allow
to split one call into UBSan runtime into several calls in case
different sanitizer kinds would have different recoverability
settings.
Tests should be fixed accordingly, I'm working on it.
Test Plan: regression test suite.
Reviewers: rsmith
Reviewed By: rsmith
Subscribers: cfe-commits
Differential Revision: http://reviews.llvm.org/D6219
llvm-svn: 221716
Get rid of ugly SanitizerOptions class thrust into LangOptions:
* Make SanitizeAddressFieldPadding a regular language option,
and rely on default behavior to initialize/reset it.
* Make SanitizerBlacklistFile a regular member LangOptions.
* Introduce the helper class "SanitizerSet" to represent the
set of enabled sanitizers and make it a member of LangOptions.
It is exactly the entity we want to cache and modify in CodeGenFunction,
for instance. We'd also be able to reuse SanitizerSet in
CodeGenOptions for storing the set of recoverable sanitizers,
and in the Driver to represent the set of sanitizers
turned on/off by the commandline flags.
No functionality change.
llvm-svn: 221653
Make sure CodeGenFunction::EmitCheck() knows which sanitizer
it emits check for. Make CheckRecoverableKind enum an
implementation detail and move it away from header.
Currently CheckRecoverableKind is determined by the type of
sanitizer ("unreachable" and "return" are unrecoverable,
"vptr" is always-recoverable, all the rest are recoverable).
This will change in future if we allow to specify which sanitizers
are recoverable, and which are not by -fsanitize-recover= flag.
No functionality change.
llvm-svn: 221635
It says there is a narrowing conversion when we assign it to an unsigned
3 bit bitfield.
Also, use unsigned instead of size_t for the Size field of the struct in
question. Otherwise they won't run together in MSVC or clang-cl.
llvm-svn: 221019
Summary:
The Itanium ABI approach of using offset-to-top isn't possible with the
MS ABI, it doesn't have that kind of information lying around.
Instead, we do the following:
- Call the virtual deleting destructor with the "don't delete the object
flag" set. The virtual deleting destructor will return a pointer to
'this' adjusted to the most derived class.
- Call the global delete using the adjusted 'this' pointer.
Reviewers: rnk
Subscribers: cfe-commits
Differential Revision: http://reviews.llvm.org/D5996
llvm-svn: 220993
SanitizerOptions is not even a POD now, so having global variable of
this type, is not nice. Instead, provide a regular constructor and clear()
method, and let each CodeGenFunction has its own copy of SanitizerOptions
it uses.
llvm-svn: 220920
An updated implemnentation of VLA types capturing based on previously committed solution for Lambdas.
This version captures the whole VLA type instead of particular variables which are part of VLA size expression and allows to use previusly calculated size of VLA type in captured regions. Required for OpenMP.
Differential Revision: http://reviews.llvm.org/D5099
llvm-svn: 220850
This patch generates some helper variables which used as a private copies of the corresponding original variables inside an OpenMP 'parallel' directive. These generated variables are initialized by default (with the default constructor, if any). In outlined function references to original variables are replaced by the references to these private helper variables. At the end of the initialization of the private variables and implicit barier is set by calling __kmpc_barrier(...) runtime function to be sure that all threads were initialized using original values of the variables.
Differential Revision: http://reviews.llvm.org/D4752
llvm-svn: 220262
Summary:
The general approach is to add extra paddings after every field
in AST/RecordLayoutBuilder.cpp, then add code to CTORs/DTORs that poisons the paddings
(CodeGen/CGClass.cpp).
Everything is done under the flag -fsanitize-address-field-padding.
The blacklist file (-fsanitize-blacklist) allows to avoid the transformation
for given classes or source files.
See also https://code.google.com/p/address-sanitizer/wiki/IntraObjectOverflow
Test Plan: run SPEC2006 and some of the Chromium tests with -fsanitize-address-field-padding
Reviewers: samsonov, rnk, rsmith
Reviewed By: rsmith
Subscribers: majnemer, cfe-commits
Differential Revision: http://reviews.llvm.org/D5687
llvm-svn: 219961
The functionality contained in CodeGenFunction::EmitAlignmentAssumption has
been moved to IRBuilder (so that it can also be used by LLVM-level code).
Remove this now-duplicate implementation in favor of the IRBuilder code.
llvm-svn: 219877
This change adds UBSan check to upcasts. Namely, when we
perform derived-to-base conversion, we:
1) check that the pointer-to-derived has suitable alignment
and underlying storage, if this pointer is non-null.
2) if vptr-sanitizer is enabled, and we perform conversion to
virtual base, we check that pointer-to-derived has a matching vptr.
llvm-svn: 219642
Moved CGOpenMPRegionInfo from CGOpenMPRuntime.h to CGOpenMPRuntime.cpp file and reworked the code for this change. Also added processing of ThreadID variable passed as an argument in outlined functions in parallel and task directives.
llvm-svn: 219490
This patch makes class OMPPrivateScope a common class for all private variables. Reworked processing of firstprivate variables (now it is based on OMPPrivateScope too).
llvm-svn: 219486
Make it possible to pass NULL through variadic functions on 64-bit
Windows targets. The Visual C++ headers define NULL to 0, when they
should define it to 0LL on Win64 so that NULL is a pointer-sized
integer.
Fixes PR20949.
Reviewers: thakis, rsmith
Differential Revision: http://reviews.llvm.org/D5480
llvm-svn: 219456
Includes parsing and semantic analysis for 'omp teams' directive support from OpenMP 4.0. Adds additional analysis to 'omp target' directive with 'omp teams' directive.
llvm-svn: 219385
This patch generates some helper variables that used as private copies of the corresponding original variables inside an OpenMP 'parallel' directive. These generated variables are initialized by copy using values of the original variables (with the copy constructor, if any). For arrays, initializator is generated for single element and in the codegen procedure this initial value is automatically propagated between all elements of the private copy.
In outlined function, references to original variables are replaced by the references to these private helper variables. At the end of the initialization of the private variables an implicit barier is generated by calling __kmpc_barrier(...) runtime function to be sure that all threads were initialized using original values of the variables.
Differential Revision: http://reviews.llvm.org/D5140
llvm-svn: 219306
This patch generates some helper variables that used as private copies of the corresponding original variables inside an OpenMP 'parallel' directive. These generated variables are initialized by copy using values of the original variables (with the copy constructor, if any). For arrays, initializator is generated for single element and in the codegen procedure this initial value is automatically propagated between all elements of the private copy.
In outlined function, references to original variables are replaced by the references to these private helper variables. At the end of the initialization of the private variables an implicit barier is generated by calling __kmpc_barrier(...) runtime function to be sure that all threads were initialized using original values of the variables.
Differential Revision: http://reviews.llvm.org/D5140
llvm-svn: 219297
This patch generates some helper variables that used as private copies of the corresponding original variables inside an OpenMP 'parallel' directive. These generated variables are initialized by copy using values of the original variables (with the copy constructor, if any). For arrays, initializator is generated for single element and in the codegen procedure this initial value is automatically propagated between all elements of the private copy.
In outlined function, references to original variables are replaced by the references to these private helper variables. At the end of the initialization of the private variables an implicit barier is generated by calling __kmpc_barrier(...) runtime function to be sure that all threads were initialized using original values of the variables.
Differential Revision: http://reviews.llvm.org/D5140
llvm-svn: 219295
Summary:
Previously CodeGen assumed that static locals were emitted before they
could be accessed, which is true for automatic storage duration locals.
However, it is possible to have CodeGen emit a nested function that uses
a static local before emitting the function that defines the static
local, breaking that assumption.
Fix it by creating the static local upon access and ensuring that the
deferred function body gets emitted. We may not be able to emit the
initializer properly from outside the function body, so don't try.
Fixes PR18020. See also previous attempts to fix static locals in
PR6769 and PR7101.
Reviewers: rsmith
Subscribers: cfe-commits
Differential Revision: http://reviews.llvm.org/D4787
llvm-svn: 219265
Includes parsing and semantic analysis for 'omp teams' directive support from OpenMP 4.0. Adds additional analysis to 'omp target' directive with 'omp teams' directive.
llvm-svn: 219197
No functional changes intended.
Renamed EmitOMPSimdLoop to EmitOMPInnerLoop, I plan to re-use
it to emit inner loop in the future patches for CodeGen of the
worksharing loop directives (omp for, omp for simd).
llvm-svn: 219195
Summary:
This add support for the C++11 feature, thread_local global variables.
The ABI Clang implements is an improvement of the MSVC ABI. Sadly,
further improvements could be made but not without sacrificing ABI
compatibility.
The feature is implemented as follows:
- All thread_local initialization routines are pointed to from the
.CRT$XDU section.
- All non-weak thread_local variables have their initialization routines
call from a single function instead of getting their own .CRT$XDU
section entry. This is done to open up optimization opportunities to
the compiler.
- All weak thread_local variables have their own .CRT$XDU section entry.
This entry is in a COMDAT with the global variable it is initializing;
this ensures that we will initialize the global exactly once.
- Destructors are registered in the initialization function using
__tlregdtor.
Differential Revision: http://reviews.llvm.org/D5597
llvm-svn: 219074
This patch implements collapsing of the loops (in particular, in
presense of clause 'collapse'). It calculates number of iterations N
and expressions nesessary to calculate the nested loops counters
values based on new iteration variable (that goes from 0 to N-1)
in Sema. It also adds Codegen for 'omp simd', which uses
(and tests) this feature.
Differential Revision: http://reviews.llvm.org/D5184
llvm-svn: 218743
CodeGen would try to come up with an LLVM IR type for a pointer to
member type on the way to forming an LLVM IR type for a pointer to
pointer to member type.
However, if the pointer to member representation has not been locked in yet,
we would not be able to come up with a pointer to member IR type.
In these cases, make the pointer to member type an incomplete type.
This will make the pointer to pointer to member type a pointer to an
incomplete type. If the class eventually obtains an inheritance model,
we will make the pointer to member type represent the actual inheritance
model.
Differential Revision: http://reviews.llvm.org/D5373
llvm-svn: 218084
Summary:
This patch implements a new UBSan check, which verifies
that function arguments declared to be nonnull with __attribute__((nonnull))
are actually nonnull in runtime.
To implement this check, we pass FunctionDecl to CodeGenFunction::EmitCallArgs
(where applicable) and if function declaration has nonnull attribute specified
for a certain formal parameter, we compare the corresponding RValue to null as
soon as it's calculated.
Test Plan: regression test suite
Reviewers: rsmith
Reviewed By: rsmith
Subscribers: cfe-commits, rnk
Differential Revision: http://reviews.llvm.org/D5082
llvm-svn: 217389
This makes use of the recently-added @llvm.assume intrinsic to implement a
__builtin_assume(bool) intrinsic (to provide additional information to the
optimizer). This hooks up __assume in MS-compatibility mode to mirror
__builtin_assume (the semantics have been intentionally kept compatible), and
implements GCC's __builtin_assume_aligned as assume((p - o) & mask == 0). LLVM
now contains special logic to deal with assumptions of this form.
llvm-svn: 217349
If control falls off the end of a function after an __asm block, MSVC
assumes that the inline assembly filled the EAX and possibly EDX
registers with an appropriate return value. This functionality is used
in inline functions returning 64-bit integers in system headers, so we
need some amount of compatibility.
This is implemented in Clang by adding extra output constraints to every
inline asm block, and storing the resulting output registers into the
return value slot. If we see an asm block somewhere in the function
body, we emit a normal epilogue instead of marking the end of the
function with a return type unreachable.
Normal returns in functions not using this functionality will overwrite
the return value slot, and in most cases LLVM should be able to
eliminate the dead stores.
Fixes PR17201.
Reviewed By: majnemer
Differential Revision: http://reviews.llvm.org/D5177
llvm-svn: 217187
into EmitCXXMemberOrOperatorCall methods. In the end we want
to make declaration visible in EmitCallArgs() method, that
would allow us to alter CodeGen depending on function/parameter
attributes.
No functionality change.
llvm-svn: 216404
for loops introduce two scopes - one for the outer loop variable and its
initialization, and another for the body of the loop, including any
variable declared inside the loop condition.
llvm-svn: 216288
Summary:
This refactoring introduces ClangToLLVMArgMapping class, which
encapsulates the information about the order in which function arguments listed
in CGFunctionInfo should be passed to actual LLVM IR function, such as:
1) positions of sret, if there is any
2) position of inalloca argument, if there is any
3) position of helper padding argument for each call argument
4) positions of regular argument (there can be many if it's expanded).
Simplify several related methods (ConstructAttributeList, EmitFunctionProlog
and EmitCall): now they don't have to maintain iterators over the list
of LLVM IR function arguments, dealing with all the sret/inalloca/this complexities,
and just use expected positions of LLVM IR arguments stored in ClangToLLVMArgMapping.
This may increase the running time of EmitFunctionProlog, as we have to traverse
expandable arguments twice, but in further refactoring we will be able
to speed up EmitCall by passing already calculated CallArgsToIRArgsMapping to
ConstructAttributeList, thus avoiding traversing expandable argument there.
No functionality change.
Test Plan: regression test suite
Reviewers: majnemer, rnk
Reviewed By: rnk
Subscribers: cfe-commits, rjmccall, timurrrr
Differential Revision: http://reviews.llvm.org/D4938
llvm-svn: 216251
Summary:
This is a first small step towards passing generic "Expr" instead of
ArgBeg/ArgEnd pair into EmitCallArgs() family of methods. Having "Expr" will
allow us to get the corresponding FunctionDecl and its ParmVarDecls,
thus allowing us to alter CodeGen depending on the function/parameter
attributes.
No functionality change.
Test Plan: regression test suite
Reviewers: rnk
Reviewed By: rnk
Subscribers: aemerson, cfe-commits
Differential Revision: http://reviews.llvm.org/D4915
llvm-svn: 216214
to instruct the code generator to not enforce a higher alignment
than the given number (of bytes) when accessing memory via an opaque
pointer or reference. Patch reviewed by John McCall (with post-commit
review pending). rdar://16254558
llvm-svn: 214911
This moves some memptr specific code into the generic thunk emission
codepath.
Fixes PR20053.
Reviewers: majnemer
Differential Revision: http://reviews.llvm.org/D4613
llvm-svn: 214004
The target method of the thunk will perform the cleanup. This can't be
tested in 32-bit x86 yet because passing something by value would create
an inalloca, and we refuse to generate broken code for that.
llvm-svn: 213976
This is used to mark the instructions emitted by Clang to implement
variety of UBSan checks. Generally, we don't want to instrument these
instructions with another sanitizers (like ASan).
Reviewed in http://reviews.llvm.org/D4544
llvm-svn: 213291
Originally committed in r211722, this fixed one case of dtor calls being
emitted without locations (this causes problems for debug info if the
call is then inlined), this caught only some of the cases.
Instead of trying to re-enable the location before the cleanup, simply
re-enable the location immediately after the unconditional branches in
question using a scoped device to ensure the no-location state doesn't
leak out arbitrarily.
llvm-svn: 212761
Now CodeGenFunction is responsible for looking at sanitizer blacklist
(in CodeGenFunction::StartFunction) and turning off instrumentation,
if necessary.
No functionality change.
llvm-svn: 212501
This patch removes the dead code, and refines the
getEHResumeBlock() slightly.
The CleanupHackLevel was a hack to the old exception
handling intrinsics, which have several issues with function
inliner.
Since LLVM 3.0, the new landingpad and resume instructions
are added to LLVM IR. With the new exception handling
mechanism, most of the issues are fixed now. We should
always use these instructions to implement the exception
handling code nowadays, and we don't need the hack any more.
Besides, the `CleanupHackLevel` is a compile-time constant,
thus other cases have been considered as dead code for a while.
llvm-svn: 212097
This reverts commit r211096. Looks like it broke the msvc build:
SemaOpenMP.cpp(140) : error C4519: default template arguments are only allowed on a class template
llvm-svn: 211113
to the normal non-placement ::operator new and ::operator delete, but allow
optimizations like new-expressions and delete-expressions do.
llvm-svn: 210137
trailing elements as a single loop, rather than sometimes emitting a nest of
several loops. This fixes a bug where CodeGen would sometimes try to emit an
expression with the wrong type for the element being initialized. Plus various
other minor cleanups to the IR produced for array new initialization.
llvm-svn: 210079
A few (mostly CodeGen) parts of Clang were tightly coupled to the
AArch64 backend. Now that it's gone, they will not even compile.
I've also deduplicated RUN lines in many of the AArch64 tests. This
might improve "make check-all" time noticably: some of those NEON
tests were monsters.
llvm-svn: 209578
It also adds a simple initial version of codegen for pragma omp simd (it will change in the future to support all the clauses).
Differential revision: http://reviews.llvm.org/D3644
llvm-svn: 209411
This patch implements global named registers in Clang, lowering to the just
created intrinsics in LLVM (@llvm.read/write_register). A new type of LValue
had to be created (Register), which just adds support to carry the metadata
node containing the name of the register. Two new methods to emit loads and
stores interoperate with another to emit the named metadata node.
No guarantees are being made and only non-allocatable global variable named
registers are being supported. Local named register support is unchanged.
llvm-svn: 209149
CapturedStmt was being ignored by instrumentation based profiling, and
its counters attributed to the containing function. Instead, we need
to treat this as a top level entity, like we do with blocks.
llvm-svn: 206231
are not associated with any source lines.
Previously, if the Location of a Decl was empty, EmitFunctionStart would
just keep using CurLoc, which would sometimes be correct (e.g., thunks)
but in other cases would just point to a hilariously random location.
This patch fixes this by completely eliminating all uses of CurLoc from
EmitFunctionStart and rather have clients explicitly pass in a
SourceLocation for the function header and the function body.
rdar://problem/14985269
llvm-svn: 205999
This adds Clang support for the ARM64 backend. There are definitely
still some rough edges, so please bring up any issues you see with
this patch.
As with the LLVM commit though, we think it'll be more useful for
merging with AArch64 from within the tree.
llvm-svn: 205100
This extends the intrinsic lookup table format slightly, and adds
entries for use the shared ARM/AArch64 definitions. The benefit is
currently smaller than for the SISD intrinsics (there's more custom
code implementing this set), but a few lines are saved and there's
scope for future expansion.
llvm-svn: 201848
This extracts the table-driven intrinsic lookup phase into a separate
function, to be used by EmitCommonNeonBuiltinExpr soon.
It also simplifies the logic used in that lookup, since VectorCastArgN
and ScalarArgN were actually identical.
llvm-svn: 201847
Previously, we made one traversal of the AST prior to codegen to assign
counters to the ASTs and then propagated the count values during codegen. This
patch now adds a separate AST traversal prior to codegen for the
-fprofile-instr-use option to propagate the count values. The counts are then
saved in a map from which they can be retrieved during codegen.
This new approach has several advantages:
1. It gets rid of a lot of extra PGO-related code that had previously been
added to codegen.
2. It fixes a serious bug. My original implementation (which was mailed to the
list but never committed) used 3 counters for every loop. Justin improved it to
move 2 of those counters into the less-frequently executed breaks and continues,
but that turned out to produce wrong count values in some cases. The solution
requires visiting a loop body before the condition so that the count for the
condition properly includes the break and continue counts. Changing codegen to
visit a loop body first would be a fairly invasive change, but with a separate
AST traversal, it is easy to control the order of traversal. I've added a
testcase (provided by Justin) to make sure this works correctly.
3. It improves the instrumentation overhead, reducing the number of counters for
a loop from 3 to 1. We no longer need dedicated counters for breaks and
continues, since we can just use the propagated count values when visiting
breaks and continues.
To make this work, I needed to make a change to the way we count case
statements, going back to my original approach of not including the fall-through
in the counter values. This was necessary because there isn't always an AST node
that can be used to record the fall-through count. Now case statements are
handled the same as default statements, with the fall-through paths branching
over the counter increments. While I was at it, I also went back to using this
approach for do-loops -- omitting the fall-through count into the loop body
simplifies some of the calculations and make them behave the same as other
loops. Whenever we start using this instrumentation for coverage, we'll need
to add the fall-through counts into the counter values.
llvm-svn: 201528
properties by fixing shouldBindAsLValue to accept arrays
(like record types) because we always manipulate
them in memory. Patch suggested by John MaCall.
// rdar://15610943
llvm-svn: 201428
When a non-trivial parameter is present, clang now gathers up all the
parameters that lack inreg and puts them into a packed struct. MSVC
always aligns each parameter to 4 bytes and no more, so this is a pretty
simple struct to lay out.
On win64, non-trivial records are passed indirectly. Prior to this
change, clang was incorrectly using byval on win64.
I'm able to self-host a working clang with this change and additional
LLVM patches.
Reviewers: rsmith
Differential Revision: http://llvm-reviews.chandlerc.com/D2636
llvm-svn: 200597
I misunderstood the discussion on this. The complexity here is
justified by the malloc overhead it saves.
This reverts commit r199302.
llvm-svn: 199700
Fix a perennial source of confusion in the clang type system: Declarations and
function prototypes have parameters to which arguments are supplied, so calling
these 'arguments' was a stretch even in C mode, let alone C++ where default
arguments, templates and overloading make the distinction important to get
right.
Readability win across the board, especially in the casting, ADL and
overloading implementations which make a lot more sense at a glance now.
Will keep an eye on the builders and update dependent projects shortly.
No functional change.
llvm-svn: 199686
Way back in r129652 we tried to avoid emitting an empty block at -O0
for switch cases that did nothing but break. This led to a poor
debugging experience as reported in PR9796, so we disabled the
optimization for -O0 but left it in for higher optimization levels in
r154420.
Since the whole point of this was to improve -O0, it's silly to keep
the complexity at all.
llvm-svn: 199302
C and C++ don't emit an extra lexical scope for the compound statement
that is the body of an Objective-C method.
rdar://problem/15010825
llvm-svn: 198699
encodes the canonical rules for LLVM's style. I noticed this had drifted
quite a bit when cleaning up LLVM, so wanted to clean up Clang as well.
llvm-svn: 198686
Unlike Itanium's VTTs, the 'most derived' boolean or bitfield is the
last parameter for non-variadic constructors, rather than the second.
For variadic constructors, the 'most derived' parameter comes after the
'this' parameter. This affects constructor calls and constructor decls
in a variety of places.
Reviewers: timurrrr
Differential Revision: http://llvm-reviews.chandlerc.com/D2405
llvm-svn: 197518
Summary:
MSVC destroys arguments in the callee from left to right. Because C++
objects have to be destroyed in the reverse order of construction, Clang
has to construct arguments from right to left and destroy arguments from
left to right.
This patch fixes the ordering by reversing the order of evaluation of
all call arguments under the MS C++ ABI.
Fixes PR18035.
Reviewers: rsmith
Differential Revision: http://llvm-reviews.chandlerc.com/D2275
llvm-svn: 196402
Instead of storing the vtable offset directly in the function pointer and
doing a branch to check for virtualness at each call site, the MS ABI
generates a thunk for calling the function at a specific vtable offset,
and puts that in the function pointer.
This patch adds support for emitting such thunks. However, it doesn't support
pointers to virtual member functions that are variadic, have an incomplete
aggregate return type or parameter, or are overriding a function in a virtual
base class.
Differential Revision: http://llvm-reviews.chandlerc.com/D2104
llvm-svn: 194827
deallocation function (and the corresponding unsized deallocation function has
been declared), emit a weak discardable definition of the function that
forwards to the corresponding unsized deallocation.
This allows a C++ standard library implementation to provide both a sized and
an unsized deallocation function, where the unsized one does not just call the
sized one, for instance by putting both in the same object file within an
archive.
llvm-svn: 194055
This uses function prefix data to store function type information at the
function pointer.
Differential Revision: http://llvm-reviews.chandlerc.com/D1338
llvm-svn: 193058
These IR instructions are undefined when the amount is equal to operand
size, but NEON right shifts support such shifts. Work around that by
emitting a different IR in these cases.
llvm-svn: 191953
The general strategy is to create template versions of the conversion function and static invoker and then during template argument deduction of the conversion function, create the corresponding call-operator and static invoker specializations, and when the conversion function is marked referenced generate the body of the conversion function using the corresponding static-invoker specialization. Similarly, Codegen does something similar - when asked to emit the IR for a specialized static invoker of a generic lambda, it forwards emission to the corresponding call operator.
This patch has been reviewed in person both by Doug and Richard. Richard gave me the LGTM.
A few minor changes:
- per Richard's request i added a simple check to gracefully inform that captures (init, explicit or default) have not been added to generic lambdas just yet (instead of the assertion violation).
- I removed a few lines of code that added the call operators instantiated parameters to the currentinstantiationscope. Not only did it not handle parameter packs, but it is more relevant in the patch for nested lambdas which will follow this one, and fix that problem more comprehensively.
- Doug had commented that the original implementation strategy of using the TypeSourceInfo of the call operator to create the static-invoker was flawed and allowed const as a member qualifier to creep into the type of the static-invoker. I currently kludge around it - but after my initial discussion with Doug, with a follow up session with Richard, I have added a FIXME so that a more elegant solution that involves the use of TrivialTypeSourceInfo call followed by the correct wiring of the template parameters to the functionprototypeloc is forthcoming.
Thanks!
llvm-svn: 191634
This reverts commit r189320.
Alexey Samsonov and Dmitry Vyukov presented some arguments for keeping
these around - though it still seems like those tasks could be solved by
a tool just using the symbol table. In a very small number of cases,
thunks may be inlined & debug info might be able to save profilers &
similar tools from misclassifying those cases as part of the caller.
The extra changes here plumb through the VarDecl for various cases to
CodeGenFunction - this provides better fidelity through a few APIs but
generally just causes the CGF::StartFunction to fallback to using the
name of the IR function as the name in the debug info.
The changes to debug-info-global-ctor-dtor.cpp seem like goodness. The
two names that go missing (in favor of only emitting those names as
linkage names) are names that can be demangled - emitting them only as
the linkage name should encourage tools to do just that.
Again, thanks to Dinesh Dwivedi for investigation/work on this issue.
llvm-svn: 189421
Arpith Chacko Jacob