llvm-svn: 189977

These instructions, such as vmul.f32, require the second source operand to
be in D0-D15 rather than the full D0-D31. When optimizing, make sure to
account for that by constraining the register class of a replacement virtual
register to be compatible with the virtual register(s) it's replacing.

I've been unsuccessful in creating a non-fragile regression test. This issue
was detected by the LLVM nightly test suite running on an A15 (Bullet).

PR17093: http://llvm.org/bugs/show_bug.cgi?id=17093
llvm-svn: 189972

llvm-svn: 189971

No functionality change.

llvm-svn: 189969

llvm-svn: 189967

Unaligned vldm/vstm need more uops and therefore are slower in general on swift.

radar://14522102

llvm-svn: 189961

Fix scheduling for vldm/vstm instructions that load/store more than 32 bytes on Cortex-A9. This also makes the existing code more compact.

llvm-svn: 189958

This reverts commit r189886.

I found a corner case where this optimization is not valid:

Say we have a "linkonce_odr unnamed_addr" in two translation units:
* In TU 1 this optimization kicks in and makes it hidden.
* In TU 2 it gets const merged with a constant that is *not* unnamed_addr,
  resulting in a non unnamed_addr constant with default visibility.
* The static linker rules for combining visibility them produce a hidden
  symbol, which is incorrect from the point of view of the non unnamed_addr
  constant.

The one place we can do this is when we know that the symbol is not used from
another TU in the same shared object, i.e., during LTO. I will move it there.

llvm-svn: 189954

Move generic isPrint and columnWidth implementations to a separate header/source to allow using both generic and system-dependent versions on win32.

Summary:
This is needed so we can use generic columnWidthUTF8 in clang-format on
win32 simultaneously with a separate system-dependent implementations of
isPrint/columnWidth in TextDiagnostic.cpp to avoid attempts to print Unicode
characters using narrow-character interfaces (which is not supported on Windows,
and we'll have to figure out how to handle this).

Reviewers: jordan_rose

Reviewed By: jordan_rose

CC: llvm-commits, klimek

Differential Revision: http://llvm-reviews.chandlerc.com/D1559

llvm-svn: 189952

llvm-svn: 189950

  This assertion is triggered because an integer constant is created with wrong
  type.

llvm-svn: 189948

llvm-svn: 189945

llvm-svn: 189940

MemoryBuffer.cpp: Don't peek the next page if file is multiple of *physical* pagesize(4k) but is not multiple of AllocationGranularity(64k), when a null terminator is required, on cygwin and win32.

For example, r189780's SparcISelLowering.cpp has the size 98304. It crashed clang to touch a null terminator on cygwin.

FIXME: It's not good to hardcode 4096 here. dwPageSize shows 4096.
llvm-svn: 189939

llvm-svn: 189938

This was regression from r134829. When linking we have to be conservative. If
one of the symbols has a significant address, then the result should have it
too.

llvm-svn: 189935

"(icmp op i8 A, B)" is equivalent to "(icmp op i8 (A & 0xff), B)" as a
degenerate case. Allowing this as a "masked" comparison when analysing "(icmp)
&/| (icmp)" allows us to combine them in more cases.

rdar://problem/7625728

llvm-svn: 189931

Even in cases which aren't universally optimisable like "(A & B) != 0 && (A &
C) != 0", the masks can make one of the comparisons completely redundant. In
this case, since we've gone to the effort of spotting masked comparisons we
should combine them.

rdar://problem/7625728

llvm-svn: 189930

      sshr,ushr,ssra,usra,srshr,urshr,srsra,ursra,sri,shl,sli,sqshlu,sqshl,uqshl,shrn,sqrshrun,sqshrn,uqshr,sqrshrn,uqrshrn,sshll,ushll
 and 4 convert instructions:
      scvtf,ucvtf,fcvtzs,fcvtzu

llvm-svn: 189925

Revert "Revert "Remove the darwin gdb option, that version of gdb is now dead and the rest of the compatibility should be done on a dwarf-N level.""

This reverts commit r189913.

Talked with Eric on IRC. I am going to XFAIL the failing test since it
is using what Eric described as "the member hack" which was needed on
that old GDB.

Sorry for the noise!

llvm-svn: 189914

Revert "Remove the darwin gdb option, that version of gdb is now dead and the rest of the compatibility should be done on a dwarf-N level."

This reverts commit r189903.

This commit broke the phase 1 buildbot for a while.

http://lab.llvm.org:8013/builders/clang-x86_64-darwin11-nobootstrap-RAincremental/builds/6684

llvm-svn: 189913

llvm-svn: 189912

Iterator of std::vector may be implemented as a raw pointer. In
this case begin iterators are rvalues and cannot be incremented.
For example, this is the case with STDCXX implementation of vector.

Patch by Konstantin Tokarev <annulen@yandex.ru>.

llvm-svn: 189911

the rest of the compatibility should be done on a dwarf-N level.

llvm-svn: 189903

in the module. Add a FIXME with a comment about darwin's ld.

llvm-svn: 189902

Original message:
If a constant or a function has linkonce_odr linkage and unnamed_addr, mark
hidden. Being linkonce_odr guarantees that it is available in every dso that
needs it. Being a constant/function with unnamed_addr guarantees that the
copies don't have to be merged.

llvm-svn: 189886

Previously, the clang crash handling code would kick in and give a crash
report for these, even though they're not that sort of error.

rdar://14882264

llvm-svn: 189878

llvm-svn: 189870

The reason that I am turning off this optimization is that there is an
additional case where a block can escape that has come up. Specifically, this
occurs when a block is used in a scope outside of its current scope.

This can cause a captured retainable object pointer whose life is preserved by
the objc_retainBlock to be deallocated before the block is invoked.

An example of the code needed to trigger the bug is:

----
\#import <Foundation/Foundation.h>
int main(int argc, const char * argv[]) {
  void (^somethingToDoLater)();

  {
    NSObject *obj = [NSObject new];

    somethingToDoLater = ^{
      [obj self]; // Crashes here
    };
  }

  NSLog(@"test.");

  somethingToDoLater();
  return 0;
}
----

In the next commit, I remove all the dead code that results from this.

Once I put in the fixing commit I will bring back the tests that I deleted in
this commit.

rdar://14802782.
rdar://14868830.

llvm-svn: 189869

hashing the contents of DW_FORM_data1 on top of a type with attributes.

llvm-svn: 189862

llvm-svn: 189861

This patch changes the default setting for the LateVectorization flag that controls where the loop-vectorizer is ran.

Perf gains:
SingleSource/Benchmarks/Shootout/matrix -37.33%
MultiSource/Benchmarks/PAQ8p/paq8p  -22.83%
SingleSource/Benchmarks/Linpack/linpack-pc  -16.22%
SingleSource/Benchmarks/Shootout-C++/ary3 -15.16%
MultiSource/Benchmarks/TSVC/NodeSplitting-flt/NodeSplitting-flt -10.34%
MultiSource/Benchmarks/TSVC/NodeSplitting-dbl/NodeSplitting-dbl -7.12%

Regressions:
SingleSource/Benchmarks/Misc/lowercase  15.10%
MultiSource/Benchmarks/TSVC/Equivalencing-flt/Equivalencing-flt 13.18%
SingleSource/Benchmarks/Shootout-C++/matrix 8.27%
SingleSource/Benchmarks/CoyoteBench/lpbench 7.30%

llvm-svn: 189858

This is another one that doesn't matter much,
but uses the right GEP index types in the first
place.

llvm-svn: 189854

This doesn't actually matter, since alloca is always
0 address space, but this is more consistent.

llvm-svn: 189853

WIP: Refactor some code so that it can be called by more than just one method. No functionality change.

llvm-svn: 189849

This reverts commit r189648.

Fixes for the previously failing clang-side arm_neon_intrinsics test
cases will be checked in separately.

llvm-svn: 189841

This won't affect the kinds of hashes we test for as we actually
do hashing based on form and attribute. Change the fission-hash
testcase one last time to handle DW_AT_comp_dir.

llvm-svn: 189840

1) If the width of vectorization list candidate is bigger than vector reg width, we will break it down to fit the vector reg.
2) We do not vectorize the width which is not power of two.

The performance result shows it will help some spec benchmarks. mesa improved 6.97% and ammp improved 1.54%. 

llvm-svn: 189830

For now this just handles simple comparisons of an ANDed value with zero.
The CC value provides enough information to do any comparison for a
2-bit mask, and some nonzero comparisons with more populated masks,
but that's all future work.

llvm-svn: 189819

llvm-svn: 189796