Thanks to Bill Schmidt for finding this in review of r178480.

llvm-svn: 178521

Buffered means a later divide may be executed out-of-order while a
prior divide is sitting (buffered) in a reservation station.

You can tell it's not pipelined, because operations that use it
reserve it for more than one cycle:

def : WriteRes<WriteIDiv, [HWPort0, HWDivider]> {
  let Latency = 25;
  let ResourceCycles = [1, 10];
}

We don't currently distinguish between an unpipeline operation and one
that is split into multiple micro-ops requiring the same unit. Except
that the later may have NumMicroOps > 1 if they also consume
issue/dispatch resources.

llvm-svn: 178519

llvm-svn: 178508

llvm-svn: 178505

llvm-svn: 178504

llvm-svn: 178503

llvm-svn: 178489

The P7 and A2 have additional floating-point conversion instructions which
allow a direct two-instruction sequence (plus load/store) to convert from all
combinations (signed/unsigned i32/i64) <--> (float/double) (on previous cores,
only some combinations were directly available).

llvm-svn: 178480

Code improvement suggested by Jakob (in review of r178450). No functionality
change intended.

llvm-svn: 178473

The popcntw instruction is available whenever the popcntd instruction is
available, and performs a separate popcnt on the lower and upper 32-bits.
Ignoring the high-order count, this can be used for the 32-bit input case
(saving on the explicit zero extension otherwise required to use popcntd).

llvm-svn: 178470

PPCISD::STFIWX is really a memory opcode, and so it should come after
FIRST_TARGET_MEMORY_OPCODE, and we should use DAG.getMemIntrinsicNode to create
nodes using it.

No functionality change intended (although there could be optimization benefits
from preserving the MMO information).

llvm-svn: 178468

llvm-svn: 178462

Reapply r177968:
After commit 178074 we can now have undefined scheduler variants.

Move the CortexA9 resources into the CortexA9 SchedModel namespace. Define
resource mappings under the CortexA9 SchedModel. Define resources and mappings
for the SwiftModel.

Incooperate Andrew's feedback.

llvm-svn: 178460

llvm-svn: 178459

llvm-svn: 178452

been turned off globally.

llvm-svn: 178451

ImmToIdxMap should be a DenseMap (not a std::map) because there
is no ordering requirement. Also, we don't need a separate list
of instructions for noImmForm in eliminateFrameIndex, because this
list is essentially the complement of the keys in ImmToIdxMap.

No functionality change intended.

llvm-svn: 178450

A vector sext + sitofp is a lot cheaper than 8 scalar conversions.

llvm-svn: 178448

This instruction is available on modern PPC64 CPUs, and is now used
to improve the SINT_TO_FP lowering (by eliminating the need for the
separate sign extension instruction and decreasing the amount of
needed stack space).

llvm-svn: 178446

The existing SINT_TO_FP code for i32 -> float/double conversion was disabled
because it relied on broken EXTSW_32/STD_32 instruction definitions. The
original intent had been to enable these 64-bit instructions to be used on CPUs
that support them even in 32-bit mode.  Unfortunately, this form of lying to
the infrastructure was buggy (as explained in the FIXME comment) and had
therefore been disabled.

This re-enables this functionality, using regular DAG nodes, but only when
compiling in 64-bit mode. The old STD_32/EXTSW_32 definitions (which were dead)
are removed.

llvm-svn: 178438

'@SECREL' is what is used by the Microsoft assembler, but GNU as expects '@SECREL32'.
With the patch, the MC-generated code works fine in combination with a recent GNU as (2.23.51.20120920 here).

Patch by David Nadlinger!
Differential Revision: http://llvm-reviews.chandlerc.com/D429

llvm-svn: 178427

llvm-svn: 178417

specific code paths.

This allows us to write code like:

  if (__nvvm_reflect("FOO"))
    // Do something
  else
    // Do something else

and compile into a library, then give "FOO" a value at kernel
compile-time so the check becomes a no-op.

llvm-svn: 178416

Hopefully this resolves any outstanding style issues and gives us
an automated way of ensuring we conform to the style guidelines.

llvm-svn: 178415

llvm-svn: 178408

llvm-svn: 178407

Check that instruction selection can select multiply-add/sub DSP instructions
from a pattern that doesn't have intrinsics.

llvm-svn: 178406

llvm-svn: 178405

derived class MipsSETargetLowering.

We shouldn't be generating madd/msub nodes if target is Mips16, since Mips16
doesn't have support for multipy-add/sub instructions.

llvm-svn: 178404

The new instructions have explicit register output operands and use table-gen
patterns instead of C++ code to do instruction selection.

Mips16's instructions are unaffected by this change.

llvm-svn: 178403

llvm-svn: 178396

llvm-svn: 178395

instructions.

llvm-svn: 178394

called in several places in ScheduleDAGRRList.cpp.

llvm-svn: 178393

to handle accumulator registers.

llvm-svn: 178392

callee-saved scan.

The code makes use of register's scavenger's capability to spill multiple
registers.

llvm-svn: 178391

registers.

llvm-svn: 178390

llvm-svn: 178368

Like nearbyint, rint can be implemented on PPC using the frin instruction. The
complication comes from the fact that rint needs to set the FE_INEXACT flag
when the result does not equal the input value (and frin does not do that). As
a result, we use a custom inserter which, after the rounding, compares the
rounded value with the original, and if they differ, explicitly sets the XX bit
in the FPSCR register (which corresponds to FE_INEXACT).

Once LLVM has better modeling of the floating-point environment we should be
able to (often) eliminate this extra complexity.

llvm-svn: 178362

llvm-svn: 178359