allocas.

llvm-svn: 156458

llvm-svn: 156457

llvm-svn: 156456

My previous change to install llvm-config-host for cross-builds resulted
in that file being installed even when the normal llvm-config was not
installed, e.g., when building the install-clang target.  Daniel suggested
this alternative, which solves the immediate problem and also avoids the gunk
in the top-level makefile.

llvm-svn: 156448

llvm-svn: 156445

PR12731.  Patch by Meador Inge!

llvm-svn: 156444

llvm-svn: 156443

old value after the store but before it is released.
This fixes rdar:/11116986.

llvm-svn: 156442

llvm-svn: 156441

Part of rdar://11352000 and should bring the buildbots back.

llvm-svn: 156421

When a combine twiddles an extract_vector, care should be take to preserve
the type of the index operand. No luck extracting a reasonable testcase,
unfortunately.

rdar://11391009

llvm-svn: 156419

llvm-svn: 156416

llvm-svn: 156414

llvm-svn: 156411

Patch by Jack Carter.

llvm-svn: 156409

Patch by Reed Kotler.

llvm-svn: 156408

Part of rdar://11352000

llvm-svn: 156405

llvm-svn: 156404

patch wasn't ready yet.

llvm-svn: 156395

llvm-svn: 156393

output. Peter Collingbourne also reports that it is showing up in
$(llvm-config --cflags).

Revert this for now since I don't know enough cmake to fix it properly.

This reverts commit 18efed7adc79c1970f307bb5b015d199012ba872.

llvm-svn: 156392

llvm-svn: 156390

 - Currently this leaves us with less build system support (e.g., installing man pages) for the docs than is desired. I'm working on fixing this, but it may take a while. If someone finds this particularly egregious let me know and I will prioritize it.

llvm-svn: 156389

I still left another fixme regarding alignment, because I'm unsure how to remove that code without breaking things

llvm-svn: 156387

llvm-svn: 156386

llvm-svn: 156385

 - The POD versions are slated for execution, but are still around until
   llvm.org machinery is in place.

llvm-svn: 156384

remove TYPE_CODE_FUNCTION_OLD type code. it is no longer in use and it was marked for removal in 3.0

llvm-svn: 156383

Share the CalleeSavedRegs defs between all calling conventions having no
callee-saved registers.

Patch by Yiannis Tsiouris!

llvm-svn: 156382

r145222 "lit/TestRunner.py: [Win32] Introduce WinWaitReleased(f), to wait for file handles to be released by children."
r145223 "lit/TestRunner.py: Use RemoveForce()."
r145381 "lit/TestRunner.py: Try to catch ERROR_FILE_NOT_FOUND, too."
r152916 "lit/TestRunner.py: [Win32] Check all opened_files[] released, rather than (obsoleted) written_files[]."
r153172 "lit/TestRunner.py: [Win32] Rework WinWaitReleased() again! "win32file" from Python Win32 Extensions."

llvm-svn: 156381

Files might be opend by system scanners (eg. file indexer, virus scanner, &c).

llvm-svn: 156380

replace the operands of expressions with only one use with undef and generate
a new expression for the original without using RAUW to update the original.
Thus any copies of the original expression held in a vector may end up
referring to some bogus value - and using a ValueHandle won't help since there
is no RAUW.  There is already a mechanism for getting the effect of recursion
non-recursively: adding the value to be recursed on to RedoInsts.  But it wasn't
being used systematically.  Have various places where recursion had snuck in at
some point use the RedoInsts mechanism instead.  Fixes PR12169.

llvm-svn: 156379

llvm-svn: 156377

llvm-svn: 156375

Added new case-ranges orientated methods for adding/removing cases in SwitchInst. After this patch cases will internally representated as ConstantArray-s instead of ConstantInt, externally cases wrapped within the ConstantRangesSet object.
Old methods of SwitchInst are also works well, but marked as deprecated. So on this stage we have no side effects except that I added support for case ranges in BitcodeReader/Writer, of course test for Bitcode is also added. Old "switch" format is also supported.

llvm-svn: 156374

llvm-svn: 156358

No functional change.

llvm-svn: 156345

llvm-svn: 156342

At least some of them:

  %vreg1:sub_16bit = COPY %vreg2:sub_16bit; GR64:%vreg1, GR32: %vreg2

Previously, we couldn't figure out that the above copy could be
eliminated by coalescing %vreg2 with %vreg1:sub_32bit.

The new getCommonSuperRegClass() hook makes it possible.

This is not very useful yet since the unmodified part of the destination
register usually interferes with the source register. The coalescer
needs to understand sub-register interference checking first.

llvm-svn: 156334

llvm-svn: 156333