Model operand restrictions of mul-like instructions on ARMv5 via
earlyclobber stuff. This should fix PRs 2313 and 8157.

Unfortunately, no testcase, since it'd be dependent on register
assignments.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122663 91177308-0d34-0410-b5e6-96231b3b80d8

add a validity check that was missed, fixing a crash on the
new testcase.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122662 91177308-0d34-0410-b5e6-96231b3b80d8

Revert commit 122654 at the request of Chris, who reckons that instsimplify
is the wrong hammer for this nail, and is probably right.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122661 91177308-0d34-0410-b5e6-96231b3b80d8

improve validity check to handle constant-trip-count loops more
aggressively.  In practice, this doesn't help anything though,
see the todo.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122660 91177308-0d34-0410-b5e6-96231b3b80d8

implement the "no aliasing accesses in loop" safety check.  This pass
should be correct now.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122659 91177308-0d34-0410-b5e6-96231b3b80d8

Fix PR8878.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122658 91177308-0d34-0410-b5e6-96231b3b80d8

Correct a bunch of mistakes which meant that the example pass didn't
even compile, let alone work.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122657 91177308-0d34-0410-b5e6-96231b3b80d8

I was unable to get the instructions to work if LLVM was built
using a separate objects directory.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122656 91177308-0d34-0410-b5e6-96231b3b80d8

Clarify that the loadable module turns up in the top-level directory,
not locally.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122655 91177308-0d34-0410-b5e6-96231b3b80d8

Fix a README item by having InstructionSimplify do a mild form of value
numbering, in which it considers (for example) "%a = add i32 %x, %y" and
"%b = add i32 %x, %y" to be equal because the operands are equal and the
result of the instructions only depends on the values of the operands.
This has almost no effect (it removes 4 instructions from gcc-as-one-file),
and perhaps slows down compilation: I measured a 0.4% slowdown on the large
gcc-as-one-file testcase, but it wasn't statistically significant.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122654 91177308-0d34-0410-b5e6-96231b3b80d8

ptx: remove reg-reg addressing mode and st.const

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122653 91177308-0d34-0410-b5e6-96231b3b80d8

ptx: add store instruction

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122652 91177308-0d34-0410-b5e6-96231b3b80d8

Add a reference to the OCamlLangImpl8.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122651 91177308-0d34-0410-b5e6-96231b3b80d8

Add an OCaml tutorial page 8

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122650 91177308-0d34-0410-b5e6-96231b3b80d8

Add to the list of cmake files the object file, not the asm file. This
is necessary for executing the custom command that runs the
assember. Fixes PR8877.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122649 91177308-0d34-0410-b5e6-96231b3b80d8

CMake (MSVC): cmake automatically adds the /EHsc and /GR compiler
options. If we are building with exceptions/rtti disabled, we replace
/EHsc with /EHs-c- and /GR with /GR-, respectively. If we just add the
disabling options we get warnings like this:

cl : Command line warning D9025 : overriding '/EHs' with '/EHs-'

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122648 91177308-0d34-0410-b5e6-96231b3b80d8

Simplify this pass by using a depth-first iterator to ensure that all
operands are visited before the instructions themselves.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122647 91177308-0d34-0410-b5e6-96231b3b80d8

Zap dead instructions harder.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122645 91177308-0d34-0410-b5e6-96231b3b80d8

Make a bunch of symbols internal.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122642 91177308-0d34-0410-b5e6-96231b3b80d8

Add another non-commutable instruction that gas accepts commuted forms for.
Fixes PR8861.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122641 91177308-0d34-0410-b5e6-96231b3b80d8

ptx: add state spaces

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122638 91177308-0d34-0410-b5e6-96231b3b80d8

include the module identifier when emitting this warning, PR8865.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122637 91177308-0d34-0410-b5e6-96231b3b80d8

print the right string, thanks for Frits for noticing.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122636 91177308-0d34-0410-b5e6-96231b3b80d8

Use getVRegDef() instead of def_iterator. This leads to fewer defs being added
with 2-address instructions, for about a 3.5% speedup of StrongPHIElimination on
403.gcc.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122635 91177308-0d34-0410-b5e6-96231b3b80d8

improve warning message to at least say what the triples are.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122632 91177308-0d34-0410-b5e6-96231b3b80d8

Fix stack layout error in MBlaze backend.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122631 91177308-0d34-0410-b5e6-96231b3b80d8

MC/Mach-O/Thumb: Set the thumb bit in the symbol table.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122630 91177308-0d34-0410-b5e6-96231b3b80d8

None of the other pass names in CodeGen have terminating periods.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122628 91177308-0d34-0410-b5e6-96231b3b80d8

Instead of processing every instruction when splitting interferences, only
process those instructions that define phi sources. This is a 47% speedup of
StrongPHIElimination compile time on 403.gcc.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122627 91177308-0d34-0410-b5e6-96231b3b80d8

SPARC backend fix: correctly passing arguments through stack

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122626 91177308-0d34-0410-b5e6-96231b3b80d8

Add a missing word to a comment.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122625 91177308-0d34-0410-b5e6-96231b3b80d8

Correctly encode pcrel|indirect.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122624 91177308-0d34-0410-b5e6-96231b3b80d8

CMake: Add disabling optimization on MSVC8 and MSVC10 as workaround for some files in Target/ARM and Target/X86.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122623 91177308-0d34-0410-b5e6-96231b3b80d8

autoconf: Add --disable-embed-stdcxx to suppress linking libstdc++.a into llvm.dll with --enable-shared on Cygming.

Cygwin has stdc++.dll in it's distribution, and we can assume distro's stdc++.dll might be available.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122622 91177308-0d34-0410-b5e6-96231b3b80d8

autoconf: [PR7874] Add --disable-pthreads to suppress detecting pthreads on certain hosts.

This would be needed to build pthread*.dll-free distribution on recent MinGW-MSYS distros.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122621 91177308-0d34-0410-b5e6-96231b3b80d8

test/Transforms/ConstProp/logicaltest.ll: FileCheck-ize.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122620 91177308-0d34-0410-b5e6-96231b3b80d8

test/CodeGen/X86/negative-sin.ll: FileCheck-ize.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122619 91177308-0d34-0410-b5e6-96231b3b80d8

test/CodeGen/X86/fp-in-intregs.ll: FileCheck-ize.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122618 91177308-0d34-0410-b5e6-96231b3b80d8

Add text explaining an assertion.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122617 91177308-0d34-0410-b5e6-96231b3b80d8

Remove second return.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122616 91177308-0d34-0410-b5e6-96231b3b80d8

Fix bug when trying to output uint16_t or uint32_t.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122615 91177308-0d34-0410-b5e6-96231b3b80d8

Implement cfi_def_cfa. Also don't convert to dwarf reg numbers twice. Looks
like 6 is a fixed point of that and so the previous tests were OK :-)

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122614 91177308-0d34-0410-b5e6-96231b3b80d8

fix PR8867: a crash handling fp128.  Thanks to Nick for the testcase.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122613 91177308-0d34-0410-b5e6-96231b3b80d8

Implement cfi_def_cfa_register.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122612 91177308-0d34-0410-b5e6-96231b3b80d8

Initial .cfi_offset implementation.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122611 91177308-0d34-0410-b5e6-96231b3b80d8

Simplify some code in MachineVerifier that was doing the correct thing, but not
in the most obvious way.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122610 91177308-0d34-0410-b5e6-96231b3b80d8

Don't produce a "DW_CFA_advance_loc 0".

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122609 91177308-0d34-0410-b5e6-96231b3b80d8

Revert the optimization in r122596. It is correct for all current targets, but
it relies on assumptions that may not be true in the future.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122608 91177308-0d34-0410-b5e6-96231b3b80d8

Multiple SPARC backend fixes: added Y register; updated select_cc, subx, subxcc defs/uses;
and fixed CustomInserter.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122607 91177308-0d34-0410-b5e6-96231b3b80d8

Integers are primitive types. Update the documentation to state this

This was done for label, void, floating point, x86mmx, metadata,
just not integer.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122606 91177308-0d34-0410-b5e6-96231b3b80d8

simplify this, isBytewiseValue handles the extra check.  We still
check for "multiple of a byte" in size to make it clear that the
>> 3 below is safe.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122604 91177308-0d34-0410-b5e6-96231b3b80d8

add a note from llvmdev

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122603 91177308-0d34-0410-b5e6-96231b3b80d8

Implement .cfi_remember_state and .cfi_restore_state.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122602 91177308-0d34-0410-b5e6-96231b3b80d8

Clarify that InstructionSimplify only returns values that dominate the
original instruction.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122601 91177308-0d34-0410-b5e6-96231b3b80d8

Small optimization to speed up replacementPreservesLCSSAForm.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122600 91177308-0d34-0410-b5e6-96231b3b80d8

Cast away "comparison between signed and unsigned integer" warnings.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122598 91177308-0d34-0410-b5e6-96231b3b80d8

Fix a signed/unsigned comparison warning.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122597 91177308-0d34-0410-b5e6-96231b3b80d8

Avoid iterating every operand of an instruction in StrongPHIElimination, since
we are only interested in the defs when discovering interferences.

This is a 28% speedup running StrongPHIElimination on 403.gcc.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122596 91177308-0d34-0410-b5e6-96231b3b80d8

Pacify the compiler.  BestWeight cannot in fact be used uninitialized
in this function, but the compiler was warning that it might be when
doing a release build.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122595 91177308-0d34-0410-b5e6-96231b3b80d8

Fix warning about size potentially being used uninitialized
when doing a release build.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122594 91177308-0d34-0410-b5e6-96231b3b80d8

Silence gcc warning about an unused variable when doing a release build.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122593 91177308-0d34-0410-b5e6-96231b3b80d8

Relax address updates in the eh_frame section.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122591 91177308-0d34-0410-b5e6-96231b3b80d8

Start adding basic support for emitting the call frame instructions.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122590 91177308-0d34-0410-b5e6-96231b3b80d8

Support/Path: Deprecate Path::hasMagicNumber and replace all uses with fs::has_magic.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122589 91177308-0d34-0410-b5e6-96231b3b80d8

Fix typo.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122588 91177308-0d34-0410-b5e6-96231b3b80d8

Support/PathV2: Implement has_magic.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122587 91177308-0d34-0410-b5e6-96231b3b80d8

Change an assertion to assert what the code actually relies upon.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122586 91177308-0d34-0410-b5e6-96231b3b80d8

fix some issues Frits noticed, add AliasAnalysis as a dependency

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122585 91177308-0d34-0410-b5e6-96231b3b80d8

Add support for .cfi_lsda.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122584 91177308-0d34-0410-b5e6-96231b3b80d8

MC/Mach-O/Thumb: Select appropriate relocation types for Thumb.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122583 91177308-0d34-0410-b5e6-96231b3b80d8

Land a first cut at StrongPHIElimination. There are only 5 new test failures
when running without the verifier, and I have not yet checked them to see if
the new results are still correct. There are more verifier failures, but they
all seem to be additional occurrences of verifier failures that occur with the
existing PHIElimination pass. There are a few obvious issues with the code:

1) It doesn't properly update the register equivalence classes during copy
insertion, and instead recomputes them before merging live intervals and
renaming registers. I wanted to keep this first patch simple for debugging
purposes, but it shouldn't be very hard to do this.

2) It doesn't mix the renaming and live interval merging with the copy insertion
process, which leads to a lot of virtual register churn. Virtual registers and
live intervals are created, only to later be merged into others. The code should
be smarter and only create a new virtual register if there is no existing
register in the same congruence class.

3) In one place the code uses a DenseMap per basic block, which is unnecessary
heap allocation. There should be an inline storage version of DenseMap.

I did a quick compile-time test of running llc on 403.gcc with and without
StrongPHIElimination. It is slightly slower with StrongPHIElimination, because
the small decrease in the coalescer runtime can't beat the increase in phi
elimination runtime. Perhaps fixing the above performance issues will narrow
the gap.

I also haven't yet run any tests of the quality of the generated code.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122582 91177308-0d34-0410-b5e6-96231b3b80d8

Add knowledge of phi-def and phi-kill valnos to MachineVerifier's predecessor
valno verification. The "Different value live out of predecessor" check is
incorrect in the case of phi-def valnos, so just skip that check for phi-def
valnos and instead check that all of the valnos for predecessors have phi-kill.
Fixes PR8863.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122581 91177308-0d34-0410-b5e6-96231b3b80d8

Support/PathV1: Deprecate GetRootDirectory.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122580 91177308-0d34-0410-b5e6-96231b3b80d8

Handle reloc_riprel_4byte_movq_load. Should make the bots happy.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122579 91177308-0d34-0410-b5e6-96231b3b80d8

Add support for the same encodings of the personality function that gnu as
supports.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122577 91177308-0d34-0410-b5e6-96231b3b80d8

BuildLibCalls: Nuke EmitMemCpy, EmitMemMove and EmitMemSet. They are dead and superseded by IRBuilder.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122576 91177308-0d34-0410-b5e6-96231b3b80d8

SimplifyLibCalls: Use IRBuilder to simplify code.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122575 91177308-0d34-0410-b5e6-96231b3b80d8

have loop-idiom nuke instructions that feed stores that get removed.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122574 91177308-0d34-0410-b5e6-96231b3b80d8

implement enough of the memset inference algorithm to recognize and insert
memsets.  This is still missing one important validity check, but this is enough
to compile stuff like this:

void test0(std::vector<char> &X) {
  for (std::vector<char>::iterator I = X.begin(), E = X.end(); I != E; ++I)
    *I = 0;
}

void test1(std::vector<int> &X) {
  for (long i = 0, e = X.size(); i != e; ++i)
    X[i] = 0x01010101;
}

With:
 $ clang t.cpp -S -o - -O2 -emit-llvm | opt -loop-idiom | opt -O3 | llc

to:

__Z5test0RSt6vectorIcSaIcEE:            ## @_Z5test0RSt6vectorIcSaIcEE
## BB#0:                                ## %entry
subq $8, %rsp
movq (%rdi), %rax
movq 8(%rdi), %rsi
cmpq %rsi, %rax
je LBB0_2
## BB#1:                                ## %bb.nph
subq %rax, %rsi
movq %rax, %rdi
callq ___bzero
LBB0_2:                                 ## %for.end
addq $8, %rsp
ret
...
__Z5test1RSt6vectorIiSaIiEE:            ## @_Z5test1RSt6vectorIiSaIiEE
## BB#0:                                ## %entry
subq $8, %rsp
movq (%rdi), %rax
movq 8(%rdi), %rdx
subq %rax, %rdx
cmpq $4, %rdx
jb LBB1_2
## BB#1:                                ## %for.body.preheader
andq $-4, %rdx
movl $1, %esi
movq %rax, %rdi
callq _memset
LBB1_2:                                 ## %for.end
addq $8, %rsp
ret

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122573 91177308-0d34-0410-b5e6-96231b3b80d8

start using irbuilder to make mem intrinsics in a few passes.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122572 91177308-0d34-0410-b5e6-96231b3b80d8

add methods to IRBuilder to create memcpy/memset/memmove.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122571 91177308-0d34-0410-b5e6-96231b3b80d8

Fix .cfi_personality on 32 bit systems.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122570 91177308-0d34-0410-b5e6-96231b3b80d8

Add support for @note. Patch by Jörg Sonnenberger.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122568 91177308-0d34-0410-b5e6-96231b3b80d8

sketch more of this out.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122567 91177308-0d34-0410-b5e6-96231b3b80d8

Add basic support for .cfi_personality.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122566 91177308-0d34-0410-b5e6-96231b3b80d8

move isBytewiseValue out to ValueTracking.h/cpp

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122565 91177308-0d34-0410-b5e6-96231b3b80d8

actually add the file...

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122563 91177308-0d34-0410-b5e6-96231b3b80d8

Start of a pass for recognizing memset and memcpy idioms.
No functionality yet.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122562 91177308-0d34-0410-b5e6-96231b3b80d8

Simplify code.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122561 91177308-0d34-0410-b5e6-96231b3b80d8

fix some sort of weird pasto

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122560 91177308-0d34-0410-b5e6-96231b3b80d8

add a note

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122559 91177308-0d34-0410-b5e6-96231b3b80d8

Generalize a previous change, fixing PR8855 - an valid large immediate
rejected by the mc assembler.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122557 91177308-0d34-0410-b5e6-96231b3b80d8

don't lose TD info

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122556 91177308-0d34-0410-b5e6-96231b3b80d8

switch the inliner alignment enforcement stuff to use the
getOrEnforceKnownAlignment function, which simplifies the code
and makes it stronger.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122555 91177308-0d34-0410-b5e6-96231b3b80d8

Move getOrEnforceKnownAlignment out of instcombine into Transforms/Utils.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122554 91177308-0d34-0410-b5e6-96231b3b80d8

Support/PathV1: Deprecate makeAbsolute and remove Unix impl because it annoys people.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122553 91177308-0d34-0410-b5e6-96231b3b80d8

Header warning patrol.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122551 91177308-0d34-0410-b5e6-96231b3b80d8

Fix a thinko pointed out by Frits van Bommel: looking through global variables in isBytewiseValue is not safe.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122550 91177308-0d34-0410-b5e6-96231b3b80d8

Merge IsFixupFullyResolved and IsSymbolRefDifferenceFullyResolved. We now
have a single point where targets test if a relocation is needed.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122549 91177308-0d34-0410-b5e6-96231b3b80d8

MemCpyOpt: Turn memcpys from a constant into a memset if possible.

This allows us to compile "int cst[] = {-1, -1, -1};" into
  movl  $-1, 16(%rsp)
  movq  $-1, 8(%rsp)
instead of
  movl  _cst+8(%rip), %eax
  movl  %eax, 16(%rsp)
  movq  _cst(%rip), %rax
  movq  %rax, 8(%rsp)

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122548 91177308-0d34-0410-b5e6-96231b3b80d8