Because in CUDA types do not have associated address spaces,
globals are declared in their "native" address space, and accessed
by bitcasting the pointer to address space 0. This relies on address
space 0 being a unified address space.
llvm-svn: 157167
that bridging between the two is free. Saves ~4k of code size,
although I don't see any measurable performance difference
(unfortunately).
llvm-svn: 156187
validate that we didn't override the contents of any of such files.
If this is detected, emit a diagnostic error and recover gracefully
by using the contents of the original file that the PCH was built from.
Part of rdar://11305263
llvm-svn: 156107
r155047. See the LLVM log for the primary motivation:
http://llvm.org/viewvc/llvm-project?rev=155047&view=rev
Primary commit r154828:
- Several issues were raised in review, and fixed in subsequent
commits.
- Follow-up commits also reverted, and which should be folded into the
original before reposting:
- r154837: Re-add the 'undef BUILTIN' thing to fix the build.
- r154928: Fix build warnings, re-add (and correct) header and
license
- r154937: Typo fix.
Please resubmit this patch with the relevant LLVM resubmission.
llvm-svn: 155048
This method is very hot, it is called when emitting diagnostics, in -E mode
and for many #pragma handlers. It scans through the whole source file to
count newlines, records and caches them in a vector.
The speedup from vectorization isn't very large, as we fall back to bytewise
scanning when we hit a newline. There might be a way to avoid leaving the sse
loop but everything I tried didn't work out because a call to push_back
clobbers xmm registers.
About 2% speedup on average on "clang -E > /dev/null" of all .cpp files in
clang's lib/Sema.
llvm-svn: 154204
uses Neon instructions for single-precision FP.
-mfpmath=neon is analogous to passing llc -mattr=+neonfp.
-mfpmath=[vfp|vfp2|vfp3|vfp4] is analogous to passing llc -mattr=-neonfp.
rdar://11108618
llvm-svn: 154046
- This is much more important than it appears at first glance...
The intended design of DiagnosticBuilder was that it never escape and that all
its members would get lowered to registers by the compiler. By fixing Emit here,
the compiler can completely eliminate the DiagnosticBuilder object and never
need to push those registers back into it.
Unfortunately, Sema has broken DiagnosticBuilder in other ways (by introducing
SemaDiagnosticBuilder), so we don't get the fill impact of this, but it is still
good for 30k reduction in code size. I'll work on fixing the
SemaDiagnosticBuilder problems next.
llvm-svn: 152669
by ~%.3/~100k in my build -- simply by eliminating the horrible code bloat coming
from the .clear() of the SmallVector<FixItHint>, which does a std::~string, etc.
- My understanding is we don't ever emit arbitrary numbers of fixits, so I just
moved us to using a statically sized array like we do for arguments and
ranges.
llvm-svn: 152639
first codepoint! Also, don't reject empty raw string literals for spurious
"encoding" issues. Also, don't rely on undefined behavior in ConvertUTF.c.
llvm-svn: 152344
If you're using git-svn, the clang and llvm repository will typically
map to a different revision.
Before we had:
clang version 3.1 (trunk 152167 trunk 152162)
After this change:
clang version 3.1 (trunk 152167) (llvm/trunk 152162)
So it's self-descriptive with an extra parens group. Which is more
compatible with version string parsers is probably debatable, but this
style was requested.
llvm-svn: 152183
the new Objective-C NSArray/NSDictionary/NSNumber literal syntax.
This introduces a new library, libEdit, which provides a new way to support
migration of code that improves on the original ARC migrator. We now believe
that most of its functionality can be refactored into the existing libraries,
and thus this new library may shortly disappear.
llvm-svn: 152141
ptrdiff_t on PPC32 on Linux, etc. should be int not long.
This does not matter for C, but it does matter for C++ because of
name mangling.
The preprocessor test has been changed accordingly.
llvm-svn: 151935
Unconditionally define __C99FEATURES__ when using C++ on Solaris. This is a
(hopefully temporary) work around for libc++ exposing C99-but-not-C++98
features in C++98 mode.
llvm-svn: 151889
IndentifierTable::get() and into IdentifierTable's constructor.
This gets a 0.7% reducing on lexing time for Cocoa.h, and
about the same for PCH generation.
llvm-svn: 151854
from the one stored in the PCH/AST, while trying to load a SLocEntry.
We verify that all files of the PCH did not change before loading it but this is not enough because:
- The AST may have been 1) kept around, 2) to do queries on it.
- We may have 1) verified the PCH and 2) started parsing.
Between 1) and 2) files may change and we are going to have crashes because the rest of clang
cannot deal with the ASTReader failing to read a SLocEntry.
Handle this by recovering gracefully in such a case, by initializing the SLocEntry
with the info from the PCH/AST as well as reporting failure by the ASTReader.
rdar://10888929
llvm-svn: 151004
Peter Collingbourne