ExitProcess still runs some code which can lead to ASan interceptors
running after CHECK failure. This can lead to deadlock if it CHECK fails
again. Avoid that mess by really exiting immediately.

llvm-svn: 286395

This method is particular: it iterates at the top-level and does
not have an enclosing block.

llvm-svn: 286394

The last fix is unlikely to solve the issue. Reverting it to the
original file.

llvm-svn: 286393

llvm-svn: 286392

llvm-svn: 286390

Summary:
Unrolled Loop Size calculations moved to a function.
Constant representing number of optimized instructions
 when "back edge" becomes "fall through" replaced with
 variable.
Some comments added.

Reviewers: mzolotukhin

Differential Revision: http://reviews.llvm.org/D21719

From: Evgeny Stupachenko <evstupac@gmail.com>
llvm-svn: 286389

Suggested in D26438 since I'm touching related code.

llvm-svn: 286388

When placing function name breakpoints on RenderScript Reduction kernel
functions, we were not skipping over the function prologue meaning that
inspection of the arguments could be garbled as the function was not finished
setting up the stack/registers.

In
[122fe8f](https://github.com/llvm-mirror/lldb/commit/122fe8f47255ec850155495d10526fccc51f95fb)
Aidan added the `SkipPrologue` function that allows us to trivially fix up the
kernel's functions' resolved addresses, falling gracefully back to the old
behaviour if we don't know how to handle the prologue or can't resolve its
size.

llvm-svn: 286387

llvm-svn: 286386

Revert r286384, "X86: Introduce the "relocImm" ComplexPattern, which represents a relocatable immediate."

Suspected to be the cause of a sanitizer-windows bot failure:
Assertion failed: isImm() && "Wrong MachineOperand accessor", file C:\b\slave\sanitizer-windows\llvm\include\llvm/CodeGen/MachineOperand.h, line 420

llvm-svn: 286385

A relocatable immediate is either an immediate operand or an operand that
can be relocated by the linker to an immediate, such as a regular symbol
in non-PIC code.

Start using relocImm for 32-bit and 64-bit MOV instructions, and for operands
of type "imm32_su". Remove a number of now-redundant patterns.

Differential Revision: https://reviews.llvm.org/D25812

llvm-svn: 286384

llvm-svn: 286383

Differential Revision: https://reviews.llvm.org/D26439

llvm-svn: 286382

"allow" requires a direct object in this case.

llvm-svn: 286381

llvm-svn: 286380

It was quite confusing that it had SectionKind of Regular, but was not
actually a InputSection.

llvm-svn: 286379

During link of devel/chrpath (FreeBSD port), found next issue:
/usr/bin/ld: error: unclosed comment in a linker script
/usr/bin/ld: error: line 1: unknown directive: �
/usr/bin/ld: error: ��

Problem was not obvious and the reason was that we did not accept
the separate form of -R. While invocation line contained it:
cc -Wl,-R /usr/local/lib -o prog prog.c

CPIO file produced contained /usr/local/lib file. 
Which looks because of reasons above
contained inside the content of whole lib folder, 
and it then was passed as an input and
proccessed as linker script.

llvm-svn: 286378

For pairs of 32-bit registers: isub_lo, isub_hi.
For pairs of vector registers: vsub_lo, vsub_hi.

Add generic subreg indices: ps_sub_lo, ps_sub_hi, and a function
  HexagonRegisterInfo::getHexagonSubRegIndex(RegClass, GenericSubreg)
that returns the appropriate subreg index for RegClass.

llvm-svn: 286377

Differential Revision: https://reviews.llvm.org/D26286

llvm-svn: 286376

It was only used for declaring unused variables. :)

llvm-svn: 286374

llvm-svn: 286373

llvm-svn: 286372

llvm-svn: 286371

llvm-svn: 286370

llvm-svn: 286369

llvm-svn: 286368

Before:
  var i = x!-1;

After:
  var i = x! - 1;

llvm-svn: 286367

llvm-svn: 286366

llvm-svn: 286365

Differential revision: https://reviews.llvm.org/D26445

llvm-svn: 286364

This commit changes the code completion results for block property setters:
The default block property result is now a block invocation rather than a simple
property reference.

rdar://28846196

Differential Revision: https://reviews.llvm.org/D26071

llvm-svn: 286363

Review: U Weigand
llvm-svn: 286362

llvm-svn: 286361

The debug info emitted by clang for static variables improved by
rL286302 and it exposed an incorrect test expactation because now LLDB
able to displays more data 9thanks to better debug info) then before.

llvm-svn: 286360

The default duration constructor does not zero-initialize the object, we need to
do that manually.

llvm-svn: 286359

Summary:
All changes are pretty straight-forward. I chose to use TimePoints with
second precision, as that is all that seems to be required here.

Reviewers: friss, zturner

Subscribers: llvm-commits

Differential Revision: https://reviews.llvm.org/D25908

llvm-svn: 286358

llvm-svn: 286357

Summary: NFCI

Reviewers: benlangmuir, zturner

Subscribers: cfe-commits

Differential Revision: https://reviews.llvm.org/D25948

llvm-svn: 286356

Summary:
r284830 added a summary provider for unique_ptr in libstdc++, whose value printed
the value of the pointee. This is a bit unintuitive as it becomes unobvious that
the value actually is a pointer, and we lose the way to actually obtain the
pointer value.

Change that to print the pointer value instead. The pointee value can still be
obtained through the synthetic children.

Reviewers: tberghammer, granata.enrico

Subscribers: lldb-commits

Differential Revision: https://reviews.llvm.org/D26403

llvm-svn: 286355

by lambdas with an explicit capture list

This commit avoids the -Wshadow warning for variables which shadow variables
that aren't captured by lambdas with an explicit capture list. It provides an
additional note that points to location of the explicit capture.

The old behaviour is preserved with -Wshadow-all or -Wshadow-uncaptured-local.

rdar://17135966

Differential Revision: https://reviews.llvm.org/D26278

llvm-svn: 286354