Summary:
Support for OpenCL 2.0 pipe type.
This is a bug-fix version for bader's patch reviews.llvm.org/D14441
Reviewers: pekka.jaaskelainen, Anastasia
Subscribers: bader, Anastasia, cfe-commits
Differential Revision: http://reviews.llvm.org/D15603
llvm-svn: 257254
By storing the instantiated expression back in the ParmVarDecl,
we remove the last need for separately storing the sub-expression
of a CXXDefaultArgExpr. This makes PCH/Modules merging quite
simple: CXXDefaultArgExpr records are serialized as references
to the ParmVarDecl, and we ignore redundant attempts to overwrite
the instantiated expression.
This has some extremely marginal impact on user-facing semantics.
However, the major effect is that it avoids IRGen errors about
conflicting definitions due to lambdas in the argument being
instantiated multiple times while sharing the same mangling.
It should also slightly improve memory usage and module file size.
rdar://23810407
llvm-svn: 256983
This patch attempts to fix the regressions identified when the patch was committed initially.
Thanks to Michael Liao for identifying the fix in the offloading metadata generation
related with side effects in evaluation of function arguments.
llvm-svn: 256933
Summary:
In order to offloading work properly two things need to be in place:
- a descriptor with all the offloading information (device entry functions, and global variable) has to be created by the host and registered in the OpenMP offloading runtime library.
- all the device functions need to be emitted for the device and a convention has to be in place so that the runtime library can easily map the host ID of an entry point with the actual function in the device.
This patch adds support for these two things. However, only entry functions are being registered given that 'declare target' directive is not yet implemented.
About offloading descriptor:
The details of the descriptor are explained with more detail in http://goo.gl/L1rnKJ. Basically the descriptor will have fields that specify the number of devices, the pointers to where the device images begin and end (that will be defined by the linker), and also pointers to a the begin and end of table whose entries contain information about a specific entry point. Each entry has the type:
```
struct __tgt_offload_entry{
void *addr;
char *name;
int64_t size;
};
```
and will be implemented in a pre determined (ELF) section `.omp_offloading.entries` with 1-byte alignment, so that when all the objects are linked, the table is in that section with no padding in between entries (will be like a C array). The code generation ensures that all `__tgt_offload_entry` entries are emitted in the same order for both host and device so that the runtime can have the corresponding entries in both host and device in same index of the table, and efficiently implement the mapping.
The resulting descriptor is registered/unregistered with the runtime library using the calls `__tgt_register_lib` and `__tgt_unregister_lib`. The registration is implemented in a high priority global initializer so that the registration happens always before any initializer (that can potentially include target regions) is run.
The driver flag -omptargets= was created to specify a comma separated list of devices the user wants to support so that the new functionality can be exercised. Each device is specified with its triple.
About target codegen:
The target codegen is pretty much straightforward as it reuses completely the logic of the host version for the same target region. The tricky part is to identify the meaningful target regions in the device side. Unlike other programming models, like CUDA, there are no already outlined functions with attributes that mark what should be emitted or not. So, the information on what to emit is passed in the form of metadata in host bc file. This requires a new option to pass the host bc to the device frontend. Then everything is similar to what happens in CUDA: the global declarations emission is intercepted to check to see if it is an "interesting" declaration. The difference is that instead of checking an attribute, the metadata information in checked. Right now, there is only a form of metadata to pass information about the device entry points (target regions). A class `OffloadEntriesInfoManagerTy` was created to manage all the information and queries related with the metadata. The metadata looks like this:
```
!omp_offload.info = !{!0, !1, !2, !3, !4, !5, !6}
!0 = !{i32 0, i32 52, i32 77426347, !"_ZN2S12r1Ei", i32 479, i32 13, i32 4}
!1 = !{i32 0, i32 52, i32 77426347, !"_ZL7fstatici", i32 461, i32 11, i32 5}
!2 = !{i32 0, i32 52, i32 77426347, !"_Z9ftemplateIiET_i", i32 444, i32 11, i32 6}
!3 = !{i32 0, i32 52, i32 77426347, !"_Z3fooi", i32 99, i32 11, i32 0}
!4 = !{i32 0, i32 52, i32 77426347, !"_Z3fooi", i32 272, i32 11, i32 3}
!5 = !{i32 0, i32 52, i32 77426347, !"_Z3fooi", i32 127, i32 11, i32 1}
!6 = !{i32 0, i32 52, i32 77426347, !"_Z3fooi", i32 159, i32 11, i32 2}
```
The fields in each metadata entry are (in sequence):
Entry 1) an ID of the type of metadata - right now only zero is used meaning "OpenMP target region".
Entry 2) a unique ID of the device where the input source file that contain the target region lives.
Entry 3) a unique ID of the file where the input source file that contain the target region lives.
Entry 4) a mangled name of the function that encloses the target region.
Entries 5) and 6) line and column number where the target region was found.
Entry 7) is the order the entry was emitted.
Entry 2) and 3) are required to distinguish files that have the same function name.
Entry 4) is required to distinguish different instances of the same declaration (usually templated ones)
Entries 5) and 6) are required to distinguish the particular target region in body of the function (it is possible that a given target region is not an entry point - if clause can evaluate always to zero - and therefore we need to identify the "interesting" target regions. )
This patch replaces http://reviews.llvm.org/D12306.
Reviewers: ABataev, hfinkel, tra, rjmccall, sfantao
Subscribers: FBrygidyn, piotr.rak, Hahnfeld, cfe-commits
Differential Revision: http://reviews.llvm.org/D12614
llvm-svn: 256842
than reusing the "overridden buffer" mechanism. This will allow us to make
embedded files and overridden files behave differently in future.
llvm-svn: 254121
https://gcc.gnu.org/onlinedocs/gcc/Typeof.html
Differences from the GCC extension:
* __auto_type is also permitted in C++ (but only in places where
it could appear in C), allowing its use in headers that might
be shared across C and C++, or used from C++98
* __auto_type can be combined with a declarator, as with C++ auto
(for instance, "__auto_type *p")
* multiple variables can be declared in a single __auto_type
declaration, with the C++ semantics (the deduced type must be
the same in each case)
This patch also adds a missing restriction on applying typeof to
a bit-field, which GCC has historically rejected in C (due to
lack of clarity as to whether the operand should be promoted).
The same restriction also applies to __auto_type in C (in both
GCC and Clang).
This also fixes PR25449.
Patch by Nicholas Allegra!
llvm-svn: 252690
This new builtin template allows for incredibly fast instantiations of
templates like std::integer_sequence.
Performance numbers follow:
My work station has 64 GB of ram + 20 Xeon Cores at 2.8 GHz.
__make_integer_seq<std::integer_sequence, int, 90000> takes 0.25
seconds.
std::make_integer_sequence<int, 90000> takes unbound time, it is still
running. Clang is consuming gigabytes of memory.
Differential Revision: http://reviews.llvm.org/D13786
llvm-svn: 252036
Introduce the notion of a module file extension, which introduces
additional information into a module file at the time it is built that
can then be queried when the module file is read. Module file
extensions are identified by a block name (which must be unique to the
extension) and can write any bitstream records into their own
extension block within the module file. When a module file is loaded,
any extension blocks are matched up with module file extension
readers, that are per-module-file and are given access to the input
bitstream.
Note that module file extensions can only be introduced by
programmatic clients that have access to the CompilerInvocation. There
is only one such extension at the moment, which is used for testing
the module file extension harness. As a future direction, one could
imagine allowing the plugin mechanism to introduce new module file
extensions.
llvm-svn: 251955
A 'readonly' Objective-C property declared in the primary class can
effectively be shadowed by a 'readwrite' property declared within an
extension of that class, so long as the types and attributes of the
two property declarations are compatible.
Previously, this functionality was implemented by back-patching the
original 'readonly' property to make it 'readwrite', destroying source
information and causing some hideously redundant, incorrect
code. Simplify the implementation to express how this should actually
be modeled: as a separate property declaration in the extension that
shadows (via the name lookup rules) the declaration in the primary
class. While here, correct some broken Fix-Its, eliminate a pile of
redundant code, clean up the ARC migrator's handling of properties
declared in extensions, and fix debug info's naming of methods that
come from categories.
A wonderous side effect of doing this write is that it eliminates the
"AddedObjCPropertyInClassExtension" method from the AST mutation
listener, which in turn eliminates the last place where we rewrite
entire declarations in a chained PCH file or a module file. This
change (which fixes rdar://problem/18475765) will allow us to
eliminate the rewritten-decls logic from the serialization library,
and fixes a crash (rdar://problem/23247794) illustrated by the
test/PCH/chain-categories.m example.
llvm-svn: 251874
We model predefined declarations as not being from AST files, but in most ways
they act as if they come from some implicit prebuilt module file imported
before all others. Therefore, if we see an update to the predefined 'struct
__va_list_tag' declaration (and we've already loaded any modules), it needs a
corresponding update record, even though it didn't technically come from an AST
file.
llvm-svn: 250134
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
This reapply a variant commit r247179 after post-commit review from
D.Blaikie.
Hopefully I got it right this time: lifetime of initializer list ends
as with any expression, which make invalid the pattern:
ArrayRef<int> Arr = { 1, 2, 3, 4};
Just like StringRef, ArrayRef shouldn't be used to initialize local
variable but only as function argument.
From: Mehdi Amini <mehdi.amini@apple.com>
llvm-svn: 247233
them directly to the control block. These are fairly large, and in a build with
lots of modules / chained PCH, we don't need to read most of them. No
functionality change intended.
llvm-svn: 247055
r246546, with a workaround for an MSVC 2013 miscompile and an MSVC 2015
rejects-valid.
Original commit message:
[modules] Rework serialized DeclContext lookup table management. Instead of
walking the loaded ModuleFiles looking for lookup tables for the context, store
them all in one place, and merge them together if we find we have too many
(currently, more than 4). If we do merge, include the merged form in our
serialized lookup table, so that downstream readers never need to look at our
imports' tables.
This gives a huge performance improvement to builds with very large numbers of
modules (in some cases, more than a 2x speedup was observed).
llvm-svn: 246582
walking the loaded ModuleFiles looking for lookup tables for the context, store
them all in one place, and merge them together if we find we have too many
(currently, more than 4). If we do merge, include the merged form in our
serialized lookup table, so that downstream readers never need to look at our
imports' tables.
This gives a huge performance improvement to builds with very large numbers of
modules (in some cases, more than a 2x speedup was observed).
llvm-svn: 246497
DeclarationName (because all ctor names are considered the same, and so on).
Reflect this in the type used as the lookup table key. As a side-effect, remove
one copy of the duplicated code used to compute the hash of the key.
llvm-svn: 246124
and CompilerInvocation::getFileSystemOpts by renaming it to getFileSystemOpts,
marking the const-returning access method const and adding a non-const version,
making the function prototypes identical to CompilerInstance::getFileSystemOpts.
llvm-svn: 246026
Instead of eagerly deserializing a list of DeclIDs when we load a module file
and doing a binary search to find the redeclarations of a decl, store a list of
redeclarations of each chain before the first declaration and load it directly.
llvm-svn: 245789
all modules and reduce the number of declarations we load when loading a
redeclaration chain.
The new approach is:
* when loading the first declaration of an entity within a module file, we
first load all declarations of the entity that were imported into that
module file, and then load all the other declarations of that entity from
that module file and build a suitable decl chain from them
* when loading any other declaration of an entity, we first load the first
declaration from the same module file
As before, we complete redecl chains through name lookup where necessary.
To make this work, I also had to change the way that template specializations
are stored -- it no longer suffices to track only canonical specializations; we
now emit all "first local" declarations when emitting a list of specializations
for a template.
On one testcase with several thousand imported module files, this reduces the
total runtime by 72%.
llvm-svn: 245779
via a module map found by -fmodule-map-file=, the home directory of the module
is the current working directory, even if that's a different directory on
reload.
llvm-svn: 244988
determine the primary context, rather than sometimes registering the lookup
table on the wrong context.
This exposed a couple of bugs:
* the odr violation check didn't deal properly with mergeable declarations
if the declaration retained by name lookup wasn't in the canonical
definition of the class
* the (broken) RewriteDecl mechanism would emit two name lookup tables for
the same DeclContext into the same module file (one as part of the
rewritten declaration and one as a visible update for the old declaration)
These are both fixed too.
llvm-svn: 244192
chain and fix the cases where it fires.
* Handle the __va_list_tag as a predefined decl. Previously we failed to merge
sometimes it because it's not visible to name lookup. (In passing, remove
redundant __va_list_tag typedefs that we were creating for some ABIs. These
didn't affect the mangling or representation of the type.)
* For Decls derived from Redeclarable that are not in fact redeclarable
(implicit params, function params, ObjC type parameters), remove them from
the list of expected redeclarable decls.
llvm-svn: 243259
the identifier table. This is redundant, since the TU-scope lookups are also
serialized as part of the TU DeclContext, and wasteful in a number of ways. We
still emit the decls for PCH / preamble builds, since for those we want
identical results, not merely semantically equivalent ones.
llvm-svn: 242855
before the first imported declaration.
We don't need to track all formerly-canonical declarations of an entity; it's sufficient to track those ones for which no other formerly-canonical declaration was imported into the same module. We call those ones "key declarations", and use them as our starting points for collecting redeclarations and performing namespace lookups.
llvm-svn: 241999
Xiuli Pan