llvm-svn: 190780

llvm-svn: 190779

The port originally had special patterns for extload, mapping them to the
same instructions as sextload.  It seemed neater to have patterns that
match "an extension that is allowed to be signed" and "an extension that
is allowed to be unsigned".

This was originally meant to be a clean-up, but it does improve the handling
of promoted integers a little, as shown by args-06.ll.

llvm-svn: 190777

llvm-svn: 190775

This is a re-commit of r190764, with an extra check to make sure that we're not
performing the transformation on illegal types (a small test case has been
added for this as well).

Original commit message:

The PPC backend uses a target-specific DAG combine to turn unaligned Altivec
loads into a permutation-based sequence when possible. Unfortunately, the
target-specific DAG combine is not always called on all loads of interest
(sometimes the routines in DAGCombine call CombineTo such that the new node and
users are not added to the worklist); allowing the combine to trigger early
(before type legalization) mitigates this problem. Because the autovectorizers
only create legal vector types, I don't expect a lot of cases where this
optimization is enabled by type legalization in practice.

llvm-svn: 190771

llvm-svn: 190770

This is causing test-suite failures.

Original commit message:

The PPC backend uses a target-specific DAG combine to turn unaligned Altivec
loads into a permutation-based sequence when possible. Unfortunately, the
target-specific DAG combine is not always called on all loads of interest
(sometimes the routines in DAGCombine call CombineTo such that the new node and
users are not added to the worklist); allowing the combine to trigger early
(before type legalization) mitigates this problem. Because the autovectorizers
only create legal vector types, I don't expect a lot of cases where this
optimization is enabled by type legalization in practice.

llvm-svn: 190765

The PPC backend uses a target-specific DAG combine to turn unaligned Altivec
loads into a permutation-based sequence when possible. Unfortunately, the
target-specific DAG combine is not always called on all loads of interest
(sometimes the routines in DAGCombine call CombineTo such that the new node and
users are not added to the worklist); allowing the combine to trigger early
(before type legalization) mitigates this problem. Because the autovectorizers
only create legal vector types, I don't expect a lot of cases where this
optimization is enabled by type legalization in practice.

llvm-svn: 190764

so it can be better used for general interoperability testing between mips32
and mips16.

llvm-svn: 190762

Also assembly/disassembly tests, and for sha256rnds2, aliases with an explicit
xmm0 dependency.

llvm-svn: 190754

llvm-svn: 190749

llvm-svn: 190746

llvm-svn: 190745

llvm-svn: 190744

As it turns out, not a problem in practice, but it should be there.

llvm-svn: 190720

Implements Instruction scheduler latencies for Silvermont,
using latencies from the Intel Silvermont Optimization Guide.

Auto detects SLM.

Turns on post RA scheduler when generating code for SLM.

llvm-svn: 190717

to be more consistent.

llvm-svn: 190692

Patch by Bradley Smith!

llvm-svn: 190683

llvm-svn: 190676

Just a clean-up, no behavioral change intended.

llvm-svn: 190673

E.g. "SRL %r2, 2; TMLL %r2, 1" => "TMLL %r2, 4".

llvm-svn: 190672

Previously we modelled VPR128 and VPR64 as essentially identical
register-classes containing V0-V31 (which had Q0-Q31 as "sub_alias"
sub-registers). This model is starting to cause significant problems
for code generation, particularly writing EXTRACT/INSERT_SUBREG
patterns for converting between the two.

The change here switches to classifying VPR64 & VPR128 as
RegisterOperands, which are essentially aliases for RegisterClasses
with different parsing and printing behaviour. This fits almost
exactly with their real status (VPR128 == FPR128 printed strangely,
VPR64 == FPR64 printed strangely).

llvm-svn: 190665

llvm-svn: 190659

llvm-svn: 190645

llvm-svn: 190644

This move makes possible to correctly handle multiples instructions
from a single pattern.

llvm-svn: 190643

llvm-svn: 190640

When a structure is passed by value, and that structure contains a vector
member, according to the PPC ABI, the structure will receive enhanced alignment
(so that the vector within the structure will always be aligned).

This should resolve PR16641.

llvm-svn: 190636

In fast-math mode sqrt(x) is calculated using the fast expansion of the
reciprocal of the reciprocal sqrt expansion. The reciprocal and reciprocal
sqrt expansions use the associated estimate instructions along with some Newton
iterations. Unfortunately, as a result, sqrt(0) was being calculated as NaN,
which is not correct. Now we explicitly return a result of zero if the input is
zero.

llvm-svn: 190624

FreeBSD kernel.

llvm-svn: 190618

Add basic assembly/disassembly support for the first Intel SHA
instruction 'sha1rnds4'. Also includes feature flag, and test cases.

Support for the remaining instructions will follow in a separate patch.

llvm-svn: 190611

Use the new instruction deprecation feature to mark mftb (now replaced with
mfspr) and dst (along with the other Altivec cache control instructions) as
deprecated when targeting cores supporting at least ISA v2.03.

llvm-svn: 190605

The 'Deprecated' class allows you to specify a SubtargetFeature that the
instruction is deprecated on.

The 'ComplexDeprecationPredicate' class allows you to define a custom
predicate that is called to check for deprecation.
For example:
  ComplexDeprecationPredicate<"MCR">

would mean you would have to define the following function:
  bool getMCRDeprecationInfo(MCInst &MI, MCSubtargetInfo &STI,
                             std::string &Info)

Which returns 'false' for not deprecated, and 'true' for deprecated
and store the warning message in 'Info'.

The MCTargetAsmParser constructor was chaned to take an extra argument of
the MCInstrInfo class, so out-of-tree targets will need to be changed.

llvm-svn: 190598

Added parsing of mask register and "zeroing" semantic, like {%k1} {z}.

llvm-svn: 190595

Aggressive anti-dependency breaking is enabled by default for all PPC cores.
This provides a general speedup on the P7 and other platforms (among other
factors, the instruction group formation for the non-embedded PPC cores is done
during post-RA scheduling). In order to do this safely, the incompatibility
between uses of the MFOCRF instruction and anti-dependency breaking are
resolved by marking MFOCRF with hasExtraSrcRegAllocReq. As noted in the removed
FIXME, the problem was that MFOCRF's output is sensitive to the identify of the
source register, and always paired with a shift to undo this effect. Because
anti-dependency breaking is unaware of this hidden dependency of the shift
amount on the source register of the MFOCRF instruction, changing that register
must be inhibited.

Two test cases were adjusted: The SjLj test was made more insensitive to
register choices and scheduling; the saveCR test disabled anti-dependency
breaking because part of what it is testing is proper register reuse.

llvm-svn: 190587

For _XYZ, the type of VDATA is v4i32, because v3i32 doesn't exist.

The ADDR64 bit is not exposed. A simpler intrinsic that doesn't take
a resource descriptor might be nicer.

The maximum number of input SGPRs is bumped to 17.

Signed-off-by: Marek Olšák <marek.olsak@amd.com>
Reviewed-by: Tom Stellard <thomas.stellard@amd.com>
llvm-svn: 190575

This fixes some regressions in the piglit local memory store tests
introduced by recent commits which made the scheduler aware of the trans
slot.

It's not possible to test this using lit, because there is no way to
determine from the assembly dumps whether or not an instruction is in
the trans slot.

Even if this were possible, the test would be highly sensitive to
changes in the scheduler and might generate confusing false negatives.

Reviewed-by: Vincent Lejeune<vljn at ovi.com>
llvm-svn: 190574

As Andy pointed out to me a long time ago, there are no structural hazards in
the later pipeline stages of the A2, and so modeling them is useless. Also,
modeling the top pre-dispatch stages is deceiving because, when multiple
hardware threads are active, those resources are shared among the threads. The
bypass definitions were mostly wrong, and so those have been removed. The
resulting itinerary is much simpler, and more accurate.

llvm-svn: 190562

For embedded PPC cores (especially the A2 core), using the MI scheduler with AA
is far superior to the other scheduling options.

llvm-svn: 190558

llvm-svn: 190551