Move the loop vectorizer from O2 to O3. It looks like the increase in code size actually hurts the performance on many programs.

llvm-svn: 171471

1. Add code to estimate register pressure.
2. Add code to select the unroll factor based on register pressure.
3. Add bits to TargetTransformInfo to provide the number of registers.

llvm-svn: 171469

Changed two debug statements that state that a queue had finished being processed when said queue was really a list to state a list had finished being processed.

llvm-svn: 171465

llvm-svn: 171464

llvm-svn: 171463

Added DEBUG message for ObjCARC when an inline asm marker is inserted for architectures where this is required to perform a retainAutoreleasedReturnValue optimization.

llvm-svn: 171462

LoopVectorizer: Add support for loop-unrolling during vectorization for increasing the ILP. At the moment this feature is disabled by default and this commit should not cause any functional changes.

llvm-svn: 171436

llvm-svn: 171429

- Fix a typo

llvm-svn: 171399

code that includes Intrinsics.gen directly.

This never showed up in my testing because the old Intrinsics.gen was
still kicking around in the make build system and was correct there. =[
Thankfully, some of the bots to clean rebuilds and that caught this.

llvm-svn: 171373

into their new header subdirectory: include/llvm/IR. This matches the
directory structure of lib, and begins to correct a long standing point
of file layout clutter in LLVM.

There are still more header files to move here, but I wanted to handle
them in separate commits to make tracking what files make sense at each
layer easier.

The only really questionable files here are the target intrinsic
tablegen files. But that's a battle I'd rather not fight today.

I've updated both CMake and Makefile build systems (I think, and my
tests think, but I may have missed something).

I've also re-sorted the includes throughout the project. I'll be
committing updates to Clang, DragonEgg, and Polly momentarily.

llvm-svn: 171366

utils/sort_includes.py script.

Most of these are updating the new R600 target and fixing up a few
regressions that have creeped in since the last time I sorted the
includes.

llvm-svn: 171362

llvm-svn: 171349

No functionality change.

llvm-svn: 171348

Added DEBUG message when ObjCARC replaces a call which returns its argument verbatim with its argument to temporarily undo an optimization.

Specifically these calls return their argument verbatim, as a low-level
optimization. However, this makes high-level optimizations
harder. We undo any uses of this optimization that the front-end
emitted. We redo them later in the contract pass.

llvm-svn: 171346

Added DEBUG messages to the top of several processing loops in ObjCARC.cpp that emit what instructions are being visited.

This is a part of a larger effort of adding DEBUG messages to the ARC
Optimizer Backend.

llvm-svn: 171345

Also add an assert to avoid confusion in the code where is known that C1 <= C2.

llvm-svn: 171310

promoting a store in a loop.  This was noticed when working on PR14753,
but isn't directly related.

llvm-svn: 171281

PR14753

llvm-svn: 171279

llvm-svn: 171272

Remove the getAttributesAtIndex and getNumAttrs methods in favor of using the getAttrSomewhere predicate. This prevents the uses of 'Attribute' as a collection of attributes.

llvm-svn: 171271

if C1 and C2 differ only with one bit.
Fixes PR14708.

llvm-svn: 171270

The later API is nicer than the former, and is correct regarding wrap-around offsets (if anyone cares).
There are a few more places left with duplicated code, which I'll remove soon.

llvm-svn: 171259

llvm-svn: 171255

directly.

This is in preparation for removing the use of the 'Attribute' class as a
collection of attributes. That will shift to the AttributeSet class instead.

llvm-svn: 171253

LCSSA PHIs may have undef values. The vectorizer updates values that are used by outside users such as PHIs.
The bug happened because undefs are not loop values. This patch handles these PHIs.

PR14725

llvm-svn: 171251

llvm-svn: 171183

getAnalysisUsage implementations.

llvm-svn: 171157

[ASan] Fix lifetime intrinsics handling. Now for each intrinsic we check if it describes one of 'interesting' allocas. Assume that allocas can go through casts and phi-nodes before apperaring as llvm.lifetime arguments

llvm-svn: 171153

If all of the write objects are identified then we can vectorize the loop even if the read objects are unidentified.

PR14719.

llvm-svn: 171124

such as by a compiler warning, a check in clang -fsanitizer=undefined, being
optimized to unreachable, or a combination of the above. PR14722.

llvm-svn: 171119

LoopVectorizer: Optimize the vectorization of consecutive memory access when the iteration step is -1

llvm-svn: 171114

Origin alignment is as high as the alignment of the corresponding application
location, but never less than 4.

llvm-svn: 171110

llvm-svn: 171109

For the time being this includes only some dummy test cases. Once the
generic implementation of the intrinsics cost function does something other
than assuming scalarization in all cases, or some target specializes the
interface, some real test cases can be added.

Also, for consistency, I changed the type of IID from unsigned to Intrinsic::ID
in a few other places.

llvm-svn: 171079

llvm-svn: 171076

llvm-svn: 171075

VectorType::getInteger() can not be used with them, because pointer size
depends on the target.

llvm-svn: 171070

llvm-svn: 171069

llvm-svn: 171061