DwarfDebug: Store the filename/dirname pair as a zero-separated string in a stringmap, instead of using a highly inefficient std::map of a pair of std::strings.

llvm-svn: 152541

llvm-svn: 152538

llvm-svn: 152537

llvm-svn: 152535

Renamed methods caseBegin, caseEnd and caseDefault with case_begin, case_end, and case_default.
Added some notes relative to case iterators.

llvm-svn: 152532

it would fail with {,u}int64_t on x86-64 Linux.

This also removes code duplication.

llvm-svn: 152517

llvm-svn: 152515

llvm-svn: 152507

llvm-svn: 152499

The 'CmpInst::isFalseWhenEqual' function returns 'false' for values other than
simply equality. For instance, it returns 'false' for <= or >=. This isn't the
correct behavior for this transformation, which is checking for strict equality
and non-equality. It was causing the gcc.c-torture/execute/frame-address.c test
to fail because it would completely (and incorrectly) optimize a whole function
into a 'ret i32 0'.

llvm-svn: 152497

llvm-svn: 152495

llvm-svn: 152492

a common collection of methods on Value, and share their implementation.
We had two variations in two different places already, and I need the
third variation for inline cost estimation.

Reviewed by Duncan Sands on IRC, but further comments here welcome.

llvm-svn: 152490

Patch by Kay Tiong Khoo!

llvm-svn: 152487

The old way of determine when and where to spill a value that was used inside of
a landing pad resulted in spilling that value everywhere and not just at the
invoke edge.

This algorithm determines which values are used within a landing pad. It then
spills those values before the invoke and reloads them before the uses. This
should prevent excessive spilling in many cases, e.g. inside of loops.
<rdar://problem/10609139>

llvm-svn: 152486

llvm-svn: 152460

Somehow we never verified SSA dominance before.

llvm-svn: 152458

llvm-svn: 152457

llvm-svn: 152454

operation action of nodes.

llvm-svn: 152452

llvm-svn: 152450

All uses of a virtual register must be dominated by its def.

llvm-svn: 152449

llvm-svn: 152447

And add comments, since this is obviously confusing.

llvm-svn: 152445

llvm-svn: 152443

Make Binary::TypeID more granular, to distinguish between ELF 32/64 little/big

llvm-svn: 152435

to count the number of phis changed, not the number visited.

llvm-svn: 152425

traversal, consider nodes for which the only successors are backedges
which the traversal is ignoring to be exit nodes. This fixes a problem
where the bottom-up traversal was failing to visit split blocks along
split loop backedges. This fixes rdar://10989035.

llvm-svn: 152421

prefix.  Added a FIXME to remind us this still does not work when it is not the
first prefix.

llvm-svn: 152414

negative switch cases if the branch condition is known to be positive.
Inspired by a recent improvement to GCC's VRP.

llvm-svn: 152405

Patch by Tom Stellard!

llvm-svn: 152400

it to analyze extractvalue(llvm.[us](add|sub).with.overflow.*) intrinsics!

llvm-svn: 152398

llvm-svn: 152393

llvm-svn: 152391

llvm-svn: 152390

llvm-svn: 152382

llvm-svn: 152374

llvm-svn: 152373

introduced. Specifically, there are cost reductions for all
constant-operand icmp instructions against an alloca, regardless of
whether the alloca will in fact be elligible for SROA. That means we
don't want to abort the icmp reduction computation when we abort the
SROA reduction computation. That in turn frees us from the need to keep
a separate worklist and defer the ICmp calculations.

Use this new-found freedom and some judicious function boundaries to
factor the innards of computing the cost factor of any given instruction
out of the loop over the instructions and into static helper functions.
This greatly simplifies the code, and hopefully makes it more clear what
is happening here.

Reviewed by Eric Christopher. There is some concern that we'd like to
ensure this doesn't get out of hand, and I plan to benchmark the effects
of this change over the next few days along with some further fixes to
the inline cost.

llvm-svn: 152368

Original commit message from r147481:
DAGCombine for transforming 128->256 casts into a vmovaps, rather
then a vxorps + vinsertf128 pair if the original vector came from a load.

Fix:
Unaligned loads need to generate a vmovups.
rdar://10974078

llvm-svn: 152366