llvm-svn: 197087

llvm-svn: 197086

llvm-svn: 197085

This adds two additional functions to the hazard recognizer interface. These
are optional (in the sense that the default implementations preserve the
current behavior), and used by the post-RA scheduler. Upcoming commits will use
this functionality in order to improve dispatch-group formation on the POWER7
and related cores. Dispatch groups are an odd construct: sometimes we need to
insert nops to force a new one to start (for performance reasons), and some
instructions need to appear in certain positions within a group, but the groups
are not fundamentally cycle based (they can contain instructions with data
dependencies with non-trivial latencies).

Motivation:

unsigned PreEmitNoops(SUnit *) - Used to force the post-RA scheduler to insert
nops to force a new dispatch group to begin. We already have a NoopHazard, and
this is also still needed. However, NoopHazard only causes a nop to be inserted
if there are no other available instructions, and so is not always sufficient.
The number of nops to insert depends on state that only the hazard recognizer
has, so a general callback is necessary.

bool ShouldPreferAnother(SUnit *) - Used to avoid scheduling instructions that
would start a new dispatch group when others are available that could be part
of the current dispatch group. In this case, we don't want to issue nops,
because the non-preferred instruction will implicitly start a new dispatch
group regardless.

Although the motivation for these functions is driven by the PowerPC backend,
they are completely general.

llvm-svn: 197084

After r196549 there is no need to separate FinalizeCXXLayout and 
FinalizeLayout so they were merged and FinalizeCXXLayout was eliminated.

llvm-svn: 197083

No longer accepting attribute spellings with prefix and suffix underscores except for GNU attributes, or C++11-style attributes in the GNU namespace. This prevents attributes such as __declspec(__dllexport__) or [[__noreturn__]] from being treated as known attributes.

llvm-svn: 197082

exec) and one for PIC codegen (local and general dynamic).

llvm-svn: 197081

The linkers on these systems don't have anything special to do with these
symbols. Since the intent is for them to be absent from the final object,
just treat them as private.

llvm-svn: 197080

Revert "DebugInfo: Move type units into the debug_types section with appropriate comdat grouping and type unit headers"

This reverts commit r197073.

The test seems to be failing on some buildbots for unknown reasons.
Reverting until I can figure that out. If anyone's got a reproduction
(.s and .o together would be great) - I'd really appreciate it.

llvm-svn: 197079

Patch by Todd Fiala

llvm-svn: 197078

llvm-svn: 197077

[objcmt] When emitting a remap file, use a json format with the edit entries, instead of applying the changes
to a temp file directly.

This allows to combine the edits when they can be different based on whether you saw
the implementation or not, e.g. with the designated initializer migration.

llvm-svn: 197076

the ObjC implementation declarations, just don't change implementations for
classes that are not in the whitelisted headers.

For example, if we change a method to return 'instancetype' we should also
update the method definition in the implementation.

llvm-svn: 197075

llvm-svn: 197074

DebugInfo: Move type units into the debug_types section with appropriate comdat grouping and type unit headers

This commit does not complete the type units feature - there are issues
around fission support (skeletal type units, pubtypes/pubnames) and
hashing of some types including those containing references to types in
other type units.

llvm-svn: 197073

llvm-svn: 197072

use f32/f64 types, rather than their vector equivalents.

llvm-svn: 197071

floating-point reciprocal square root step LLVM AArch64 intrinsics to
use f32/f64 types, rather than their vector equivalents.

llvm-svn: 197070

point reciprocal exponent, and floating-point reciprocal square root estimate
LLVM AArch64 intrinsics to use f32/f64 types, rather than their vector
equivalents.

llvm-svn: 197069

use f32/f64 types, rather than their vector equivalents.

llvm-svn: 197068

floating-point reciprocal square root step LLVM AArch64 intrinsics to
use f32/f64 types, rather than their vector equivalents.

llvm-svn: 197067

point reciprocal exponent, and floating-point reciprocal square root estimate
LLVM AArch64 intrinsics to use f32/f64 types, rather than their vector
equivalents.

llvm-svn: 197066

llvm-svn: 197065

llvm-svn: 197064

Fix demangling crasher.  The crasher involved nested <encoding> involving parameter packs, which exposed a logic bug causing an empty vector<string> to be accessed with back().  In addition to fixing the bug, I've inserted numerous preemptive checks for similar bugs in the hopes that if another bug is uncovered, the bug results in an invalid mangled string instead of a demangler crash.  Test suite updated with string that was causing the crash.

llvm-svn: 197063

Replace -arch with -triple so that we can guarantee that
ELF object files can be generated.

llvm-svn: 197062

Move std::begin(array) and std::end(array) out from under an #ifdef that was preventing people from building libc++ using gcc. This corrects a mistake that I introduced in r196058

llvm-svn: 197061

We were mistakengly giving linkonce_odr linkage instead of internal
linkage to the deleting and complete destructor thunks for classes in
anonymous namespaces.

Fixes PR17273.

llvm-svn: 197060

Found by Arthur Yoo!

llvm-svn: 197059

This makes it a little easier to read.

Reviewed-by: Vincent Lejeune <vljn at ovi.com>
llvm-svn: 197058

This enables -print-before-all to dump MachineInstrs after it is run.

Reviewed-by: Vincent Lejeune <vljn at ovi.com>
llvm-svn: 197057

This enables -print-before-all to dump MachineInstrs after it is run.

Reviewed-by: Vincent Lejeune <vljn at ovi.com>
llvm-svn: 197056

Methods are thiscall by default in the MS ABI, and also in MinGW targetting GCC 4.7 or later.

This changes the diagnostic from the technically correct but hard to understand:

  virtual function 'foo' has different calling convention attributes ('void ()') than the function it overrides (which has calling convention 'void () __attribute__((thiscall))')

to the more intuitive and also correct:

  'static' member function 'foo' overrides a virtual function

We already have a test for this. Let's just run it in both ABI modes.

Differential Revision: http://llvm-reviews.chandlerc.com/D2375

llvm-svn: 197055

llvm-svn: 197052

Found by "cppcheck". PR18208.

llvm-svn: 197047

The tests were no longer using fast-isel at all (MachO needs an "ios" rather
than "darwin" triple at the moment and Linux needs ARM mode). Once that was
corrected, the verifier complained about a t2ADDri created for the alloca.

llvm-svn: 197046

Based on a patch by Neil Henning!

llvm-svn: 197045

llvm-svn: 197043

I moved a test from avx512-vbroadcast-crash.ll to avx512-vbroadcast.ll
I defined HasAVX512 predicate as AssemblerPredicate. It means that you should invoke llvm-mc with "-mcpu=knl" to get encoding for AVX-512 instructions. I need this to let AsmMatcher to set different encoding for AVX and AVX-512 instructions that have the same mnemonic and operands (all scalar instructions).

llvm-svn: 197041

llvm-svn: 197040