llvm-svn: 123048

Instead encode llvm IR level property "HasSideEffects" in an operand (shared
with IsAlignStack). Added MachineInstrs::hasUnmodeledSideEffects() to check
the operand when the instruction is an INLINEASM.

This allows memory instructions to be moved around INLINEASM instructions.

llvm-svn: 123044

llvm-svn: 123042

llvm-svn: 123040

llvm-svn: 123039

Patch by Tim Northover.

llvm-svn: 123035

X = sext x; x >s c ? X : C+1 --> X = sext x; X <s C+1 ? C+1 : X
X = sext x; x <s c ? X : C-1 --> X = sext x; X >s C-1 ? C-1 : X
X = zext x; x >u c ? X : C+1 --> X = zext x; X <u C+1 ? C+1 : X
X = zext x; x <u c ? X : C-1 --> X = zext x; X >u C-1 ? C-1 : X
X = sext x; x >u c ? X : C+1 --> X = sext x; X <u C+1 ? C+1 : X
X = sext x; x <u c ? X : C-1 --> X = sext x; X >u C-1 ? C-1 : X

Instead of calculating this with mixed types promote all to the
larger type. This enables scalar evolution to analyze this
expression. PR8866

llvm-svn: 123034

llvm-svn: 123033

Enable live debug variables pass.

llvm-svn: 123032

llvm-svn: 123031

llvm-svn: 123030

every BranchInst has a fixed number of operands.

llvm-svn: 123027

llvm-svn: 123025

Revert r122955. It seems using movups to lower memcpy can cause massive regression (even on Nehalem) in edge cases. I also didn't see any real performance benefit.

llvm-svn: 123015

Also fix an off-by-one in SelectionDAGBuilder that was preventing shuffle
vectors from being translated to EXTRACT_SUBVECTOR.
Patch by Tim Northover.

The test changes are needed to keep those spill-q tests from testing aligned
spills and restores.  If the only aligned stack objects are spill slots, we
no longer realign the stack frame.  Prior to this patch, an EXTRACT_SUBVECTOR
was legalized by loading from the stack, which created an aligned frame index.
Now, however, there is nothing except the spill slot in the stack frame, so
I added an aligned alloca.

llvm-svn: 122995

llvm-svn: 122994

We were never generating any of these nodes with variable indices, and there
was one legalizer function asserting on a non-constant index.  If we ever have
a need to support variable indices, we can add this back again.

llvm-svn: 122993

we have invokes, so there is no functionality change here.

llvm-svn: 122990

Nadav Rotem.

llvm-svn: 122983

typed atomics. This will lower exclusively to libcalls at the moment.

llvm-svn: 122979

llvm-svn: 122978

llvm-svn: 122977

This fixes PR 8913 crash.

llvm-svn: 122971

llvm-svn: 122970

llvm-svn: 122969

llvm-svn: 122968

llvm-svn: 122966

llvm-svn: 122963

Patch by Richard Simth.

llvm-svn: 122962

This happens when we take the (non-constant) length from a malloc.

llvm-svn: 122961

llvm-svn: 122960

InstCombine: If we call llvm.objectsize on a malloc call we can replace it with the size passed to malloc.

llvm-svn: 122959

llvm-svn: 122958

llvm-svn: 122957

The theory is it's still faster than a pair of movq / a quad of movl. This
will probably hurt older chips like P4 but should run faster on current
and future Intel processors. rdar://8817010

llvm-svn: 122955

llvm-svn: 122954

llvm-svn: 122953

etc. takes an option OptSize. If OptSize is true, it would return
the inline limit for functions with attribute OptSize.

llvm-svn: 122952

  ret i64 ptrtoint (i8* getelementptr ([1000 x i8]* @X, i64 1, i64 sub (i64 0, i64 ptrtoint ([1000 x i8]* @X to i64))) to i64)

to "ret i64 1000".  This allows us to correctly compute the trip count
on a loop in PR8883, which occurs with std::fill on a char array.  This
allows us to transform it into a memset with a constant size.

llvm-svn: 122950

llvm-svn: 122949