llvm-svn: 146277

Patch by Brendon Cahoon!

This extends the existing LoopUnroll and LoopUnrollPass. Brendon
measured no regressions in the llvm test suite with -unroll-runtime
enabled. This implementation works by using the existing loop
unrolling code to unroll the loop by a power-of-two (default 8). It
generates an if-then-else sequence of code prior to the loop to
execute the extra iterations before entering the unrolled loop.

llvm-svn: 146245

- Walking over pred_begin/pred_end is an expensive operation.
- PHINodes contain a value for each predecessor anyway.
- While it may look like we used to save a few iterations with the set,
  be aware that getIncomingValueForBlock does a linear search on
  the values of the phi node.
- Another -5% on ARMDisassembler.cpp (Release build). This was the last
  entry in the profile that was obviously wasting time.

llvm-svn: 145937

- Calling getUser in a loop is much more expensive than iterating over a few instructions.
- Use it instead of the open-coded loop in AddrModeMatcher.
- 5% speedup on ARMDisassembler.cpp Release builds.

llvm-svn: 145810

Fix quadratic behavior in InlineFunction by fetching the personality function of the callee once and not for every invoke in the caller.

The callee is usually smaller than the caller, too. This reduces the compile
time of ARMDisassembler.cpp by 32% (Release build). It still takes ages to
compile though.

llvm-svn: 145690

InstructionSimplify.cpp.  Other fixups as needed.
Part of rdar://10500969

llvm-svn: 145559

llvm-svn: 145420

weak variable are compiled by different compilers, such as GCC and LLVM, while
LLVM may increase the alignment to the preferred alignment there is no reason to
think that GCC will use anything more than the ABI alignment.  Since it is the
GCC version that might end up in the final program (as the linkage is weak), it
is wrong to increase the alignment of loads from the global up to the preferred
alignment as the alignment might only be the ABI alignment.

Increasing alignment up to the ABI alignment might be OK, but I'm not totally
convinced that it is.  It seems better to just leave the alignment of weak
globals alone.

llvm-svn: 145413

Suggested in code review by Eli.

That code in InstCombine looks kinda suspicious.

llvm-svn: 145013

The right way to check for a binary operation is
cast<BinaryOperator>. The original check: cast<Instruction> &&
numOperands() == 2 would match phi "instructions", leading to an
infinite loop in extreme corner case: a useless phi with operands
[self, constant] that prior optimization passes failed to remove,
being used in the loop by another useless phi, in turn being used by an
lshr or udiv.

Fixes PR11350: runaway iteration assertion.

llvm-svn: 144935

Add support for custom names for library functions in TargetLibraryInfo.  Add a custom name for fwrite and fputs on x86-32 OSX.  Make SimplifyLibCalls honor the custom
names for fwrite and fputs.

Fixes <rdar://problem/9815881>.

llvm-svn: 144876

llvm-svn: 144748

llvm-svn: 144675

llvm-svn: 144536

llvm-svn: 144416

<rdar://problem/10405911>

llvm-svn: 144000

llvm-svn: 143634

Patch by Kostya Serebryany.

llvm-svn: 143405

instructions.

This doesn't introduce any optimizations we weren't doing before (except
potentially due to pass ordering issues), now passes will eliminate them sooner
as part of their own cleanups.

llvm-svn: 142787

llvm-svn: 142684

Refactor code from inlining and globalopt that checks whether a function definition is unused, and enhance it so it can tell that functions which are only used by a blockaddress are in fact dead.  This probably doesn't happen much on most code, but the Linux kernel's _THIS_IP_ can trigger this issue with blockaddress.  (GlobalDCE can also handle the given tescase, but we only run that at -O3.)  Found while looking at PR11180.

llvm-svn: 142572

llvm-svn: 142162

promoting allocas to preferred alignments that exceed the natural
alignment. This avoids some potentially expensive dynamic stack realignments.

The natural stack alignment is set in target data strings via the "S<size>"
option. Size is in bits and must be a multiple of 8. The natural stack alignment
defaults to "unspecified" (represented by a zero value), and the "unspecified"
value does not prevent any alignment promotions. Target maintainers that care
about avoiding promotions should explicitly add the "S<size>" option to their
target data strings.

llvm-svn: 141599

This handles the case in which LSR rewrites an IV user that is a phi and
splits critical edges originating from a switch.
Fixes <rdar://problem/6453893> LSR is not splitting edges "nicely"

llvm-svn: 141059

llvm-svn: 141058

but not load instructions. Noticed by inspection.

llvm-svn: 140966

Clean up uses of switch instructions so they are not dependent on the operand ordering.  Patch by Stepan Dyatkovskiy.

llvm-svn: 140803

Stop emitting instructions with the name "tmp" they eat up memory and have to be uniqued, without any benefit.

If someone prefers %tmp42 to %42, run instnamer.

llvm-svn: 140634

Some passes require breaking critical edges before they're called. Don't
segfault because of that.

llvm-svn: 140196

llvm-svn: 140172

llvm-svn: 140168

llvm-svn: 140164

llvm-svn: 140094

extract the landing pad block. Otherwise, there will be a situation where the
invoke's unwind edge lands on a non-landing pad.

We also forbid the user from extracting the landing pad block by itself. Again,
this is not a valid transformation.

llvm-svn: 140083

llvm-svn: 140026

exception.

llvm-svn: 139117

- Reword comments.
- Allow undefined behavior interfering with undefined behavior.
- Add address space checks.

llvm-svn: 138619

SimplifyCFG: If we have a PHI node that can evaluate to NULL and do a load or store to the address returned by the PHI node then we can consider this incoming value as dead and remove the edge pointing there, unless there are instructions that can affect control flow executed in between.

In theory this could be extended to other instructions, eg. division by zero, but it's likely that it will "miscompile" some code because people depend on div by zero not trapping. NULL pointer dereference usually leads to a crash so we should be on the safe side.

This shrinks the size of a Release clang by 16k on x86_64.

llvm-svn: 138618

getFirstNonPHI so that it will skip over the landingpad instructions as well.

llvm-svn: 138537

dominator information even though dominators were previously computed.

Patch by Nick Sumner.

llvm-svn: 138449