llvm-svn: 189740

Specify GATHER/SCATTER as heavy instructions.

llvm-svn: 189736

llvm-svn: 189729

llvm-svn: 189728

don't exist in libc. This is really not the right way to solve this problem;
but it's not clear to me at this time exactly what is the right way.
If we create stubs here, they will cause link errors because these functions
do not exist in libc.

llvm-svn: 189727

llvm-svn: 189726

Here are a few miscellaneous things to tidy up the PPC64 fast-isel
implementation.  I corrected a couple of commentary lapses, and added
documentation of future opportunities.  I also implemented
TargetMaterializeAlloca, which I somehow forgot when I split up the
original huge patch.

Finally, I decided to delete SelectCmp.  I hadn't previously hooked it
in to TargetSelectInstruction(), and when I did I realized it wasn't
serving any useful purpose.  This is only useful for compares that
don't feed a branch in the same block, and to handle that we would
have to have logic to interpret i1 as a condition register.  This
could probably be done, but would require Unseemly Hackery, and
honestly does not seem worth the hassle.

This ends the current patch series.

llvm-svn: 189715

This is the last substantive patch I'm planning for fast-isel in the
near future, adding fast selection of integer truncates.  There are
certainly more things that can be improved (many of which are called
out in FIXMEs), but for now we are catching most of the important
cases.

I'll document some of the remaining work in a cleanup patch shortly.

llvm-svn: 189706

llvm-svn: 189705

This patch adds fast-isel support for calls (but not intrinsic calls
or varargs calls).  It also removes a badly-formed assert.  There are
some new tests just for calls, and also for folding loads into
arguments on calls to avoid extra extends.

llvm-svn: 189701

llvm-svn: 189699

Fixed a bug where diassembling an instruction that had a prefix would cause LLVM to identify a 1-byte instruction, but then upon querying it for that 1-byte instruction would cause an undefined opcode.

llvm-svn: 189698

has hard float, when you compile the mips32 code you have to make sure
that it knows to compile any mips32 routines as hard float. I need to clean
up the way mips16 hard float is specified but I need to first think through
all the details. Mips16 always has a form of soft float, the difference being
whether the underlying hardware has floating point. So it's not really
necessary to pass the -soft-float to llvm since soft-float is always true
for mips16 by virtue of the fact that it will not register floating point
registers. By using this fact, I can simplify the way this is all handled.

llvm-svn: 189690

Yet another chunk of fast-isel code.  This one handles various
conversions involving floating-point.  (It also includes some
miscellaneous handling throughout the back end for LWA_32 and LWAX_32
that should have been part of the load-store patch.)

llvm-svn: 189677

llvm-svn: 189673

Currently this is just the atom model with SSE4.2 enabled.

llvm-svn: 189669

llvm-svn: 189656

llvm-svn: 189654

Teach X86 backend to create BMI2 BZHI instructions from (and X, (add (shl 1, Y), -1)). Fixes PR17038.

llvm-svn: 189653

This reverts commit r189619.

The commit was breaking the arm_neon_intrinsic test.

llvm-svn: 189648

Created SUPressureDiffs array to hold the per node PDiff computed during DAG building.

Added a getUpwardPressureDelta API that will soon replace the old
one. Compute PressureDelta here from the precomputed PressureDiffs.

Updating for liveness will come next.

llvm-svn: 189640

Mostly trivial patch adding support for compares.  The meat of the
work was added with the branch support.

llvm-svn: 189639

llvm-svn: 189638

This is the next big chunk of fast-isel code.  The primary purpose is
to implement selection of loads and stores, but there is a lot of
drag-along to support this.  The common code to analyze addresses for
both loads and stores is substantial.  It's also necessary to add the
materialization code for global values.

Related to load-store processing is the code to fold loads into
integer extends, since otherwise we generate lots of redundant
instructions.  We also need to add some overrides to some FastEmit
routines to ensure we don't assign GPR 0 to a virtual register when
this would change the meaning of an instruction.

I added handling selection of a few binary arithmetic instructions, to
enable committing some test cases I wrote a while back.

Finally, ap couple of miscellaneous changes:
 * I cleaned up some poor style from a previous patch in
   PPCISelLowering.cpp, pointed out by David Blaikie.
 * I enlarged the Addr.Offset field to avoid sign problems with 32-bit
   offsets. 

llvm-svn: 189636

In addition to recognizing when the multiply's second argument is
coming from an explicit VDUPLANE, also look for a plain scalar
f32 reference and reference it via the corresponding vector
lane.

rdar://14870054

llvm-svn: 189619

Clean up some usage of Triple.  The base class has methods for determining if the target is iOS and Linux.

llvm-svn: 189604

llvm-svn: 189580

There are several optional (off-by-default) features in CodeGen that can make
use of alias analysis. These features are important for generating code for
some kinds of cores (for example the (in-order) PPC A2 core). This adds a
useAA() function to TargetSubtargetInfo to allow these features to be enabled
by default on a per-subtarget basis.

Here is the first use of this function: To control the default of the
-enable-aa-sched-mi feature.

llvm-svn: 189563

32-bit absolute addressing in instructions likei this:

	mov $_f, %rsi

which is not supported in 64-bit mode.

rdar://8827134

llvm-svn: 189543

Fix a few things in one swoop.

# Add some negative tests.
# Fix some formatting issues.
# Add some missing IsThumb / ARMv8
# Fix some outs / ins mistakes.

llvm-svn: 189490

The usual default of "dmb ish" (inner-shareable) isn't even a valid instruction
on v6M or v7M (well, it does the same thing but software is strongly
discouraged from using it) so we should emit a full-system barrier there.

llvm-svn: 189483

llvm-svn: 189482

Clang is now generating cleaner IR, so this removes the old variants which
should be completely unused.

llvm-svn: 189481

The vqdmlal and vqdmlls instructions are really just a fused pair consisting of
a vqdmull.sN and a vqadd.sN. This adds patterns to LLVM so that we can switch
Clang's CodeGen over to generating these instead of the special vqdmlal
intrinsics.

llvm-svn: 189480

These intrinsics are legalized to V(ALL|ANY)_(NON)?ZERO nodes,
are matched as SN?Z_[BHWDV]_PSEUDO pseudo's, and emitted as
a branch/mov sequence to evaluate to 0 or 1.

Note: The resulting code is sub-optimal since it doesnt seem to be possible
to feed the result of an intrinsic directly into a brcond. At the moment
it uses (SETCC (VALL_ZERO $ws), 0, SETEQ) and similar which unnecessarily
evaluates the boolean twice.

llvm-svn: 189478

llvm-svn: 189476

llvm-svn: 189472

llvm-svn: 189471

For now just handles simple comparisons of an ANDed value with zero.
The CC value provides enough information to do any comparison for a
2-bit mask, and some nonzero comparisons with more populated masks,
but that's all future work.

llvm-svn: 189469

The MSA control registers have been added as reserved registers,
and are only used via ISD::Copy(To|From)Reg. The intrinsics are lowered
into these nodes.

llvm-svn: 189468