Fixes wrong lighting in some corner cases with r600g and radeonsi, e.g.
manifested by failure of two piglit/glean tests and intermittent black
patches in many apps.

Tested on SI and RS880.

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=62012 [radeonsi]
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=58150

 [r600g]

NOTE: This is a candidate for the Mesa stable branch.

Reviewed-by: Christian König <christian.koenig@amd.com>
llvm-svn: 177730

For mips a branch an 18-bit signed offset (the 16-bit 
offset field shifted left 2 bits) is added to the 
address of the instruction following the branch 
(not the branch itself), in the branch delay slot, 
to form a PC-relative effective target address. 

Previously, the code generator did not perform the 
shift of the immediate branch offset which resulted 
in wrong instruction opcode. This patch fixes the issue.

Contributor: Vladimir Medic
llvm-svn: 177687

This patch uses the generated instruction info tables to 
identify memory/load store instructions.
After successful matching and based on the operand type 
and size, it generates additional instructions to the output.

Contributor: Vladimir Medic
llvm-svn: 177685

As Jakob pointed out in his review of r177423, having a shared ZERO
register between the 32- and 64-bit register classes causes this
odd G8RC_NOX0_and_GPRC_NOR0 class to be created. As recommended,
this adds a ZERO8 register which differentiates the 32- and 64-bit
zeros.

No functionality change intended.

llvm-svn: 177683

Thanks to Jakob for isolating the underlying problem from the
test case in r177423. The original commit had introduced
asymmetric copy operations, but these turned out to be a work-around
to the real problem (the use of == instead of hasSubClassEq in PPCCTRLoops).

llvm-svn: 177679

The .set directive in the Mips the assembler can be 
used to set the value of a symbol to an expression. 
This changes the symbol's value and type to conform 
to the expression's.

Syntax: .set symbol, expression

This patch implements the parsing of the above syntax 
and enables the parser to use defined symbols when 
parsing operands.

Contributor: Vladimir Medic
llvm-svn: 177667

This implements SJLJ lowering on PPC, making the Clang functions
__builtin_{setjmp/longjmp} functional on PPC platforms. The implementation
strategy is similar to that on X86, with the exception that a branch-and-link
variant is used to get the right jump address. Credit goes to Bill Schmidt for
suggesting the use of the unconditional bcl form (instead of the regular bl
instruction) to limit return-address-cache pollution.

Benchmarking the speed at -O3 of:

static jmp_buf env_sigill;

void foo() {
                __builtin_longjmp(env_sigill,1);
}

main() {
	...

        for (int i = 0; i < c; ++i) {
                if (__builtin_setjmp(env_sigill)) {
                        goto done;
                } else {
                        foo();
                }

done:;
        }

	...
}

vs. the same code using the libc setjmp/longjmp functions on a P7 shows that
this builtin implementation is ~4x faster with Altivec enabled and ~7.25x
faster with Altivec disabled. This comparison is somewhat unfair because the
libc version must also save/restore the VSX registers which we don't yet
support.

llvm-svn: 177666

Although there is only one Altivec VRSAVE register, it is a member of
a register class, and we need the ability to spill it. Because this
register is normally callee-preserved and handled by special code this
has never before been necessary. However, this capability will be required by
a forthcoming commit adding SjLj support.

llvm-svn: 177654

The old code used to lower FRAMEADDR tried to replicate the logic in the real
frame-lowering code that determines whether or not the frame pointer (r31) will
be used. When it seemed as through the frame pointer would not be used, the
stack pointer (r1) was used instead. Unfortunately, because the stack size is
not yet known, this does not work. Instead, this change introduces new
always-reserved pseudo-registers (FP and FP8) that are replaced during prologue
insertion with the real frame-pointer register (either r1 or r31).

It is important that this intrinsic always return a valid frame address because
it is used by Clang to store the frame address as part of code generation for
__builtin_setjmp.

llvm-svn: 177653

NEON is not IEEE 754 compliant, so we should avoid lowering single-precision
floating point operations with NEON unless unsafe-math is turned on. The
equivalent VFP instructions are IEEE 754 compliant, but in some cores they're
much slower, so some archs/OSs might still request it to be on by default,
such as Swift and Darwin.

llvm-svn: 177651

llvm-svn: 177611

It's not yet clear if these instructions need a more careful model.

llvm-svn: 177599

This is used for all the expensive system instructions.

llvm-svn: 177598

llvm-svn: 177592

llvm-svn: 177591

- After moving logic recognizing vector shift with scalar amount from
  DAG combining into DAG lowering, we declare to customize all vector
  shifts even vector shift on AVX is legal. As a result, the cost model
  needs special tuning to identify these legal cases.

llvm-svn: 177586

llvm-svn: 177540

llvm-svn: 177539

- Move SRA/SRL/SHL lowering support from DAG combination to DAG lowering
  to support extended 256-bit integer in AVX but not AVX2.

llvm-svn: 177478

- Prepare moving logic from DAG combining into DAG lowering. There's no
  functionality change.

llvm-svn: 177477

- no functionality change

llvm-svn: 177476

Patch by Stepan Dyatkovskiy <stpworld@narod.ru>
rdar://13457826

llvm-svn: 177463

llvm-svn: 177461

llvm-svn: 177460

llvm-svn: 177459

logic as a QOI cleanup.  No functional change.  Tests already in place.
rdar://13456414

llvm-svn: 177446

Add a new WriteZero SchedWrite type for the common dependency-breaking
instructions that clear a register.

llvm-svn: 177442

an X86Operand, but also performs a Sema lookup and adds the sizing directive
when appropriate.  Use this when parsing a bracketed statement.  This is
necessary to get the instruction matching correct as well.  Test case coming
on clang side.
rdar://13455408

llvm-svn: 177439

All pre-increment load patterns need to set the mayLoad flag (since
they don't provide a DAG pattern).

This was missing for LHAUX8 and LWAUX, which is added by this patch.

llvm-svn: 177431

As opposed to to pre-increment store patterns, the pre-increment
load patterns were already using standard memory operands, with
the sole exception of LHAU8.

As there's no real reason why LHAU8 should be different here,
this patch simply rewrites the pattern to also use a memri
operand, just like all the other patterns.

llvm-svn: 177430

Currently, pre-increment store patterns are written to use two separate
operands to represent address base and displacement:

  stwu $rS, $ptroff($ptrreg)

This causes problems when implementing the assembler parser, so this
commit changes the patterns to use standard (complex) memory operands
like in all other memory access instruction patterns:

  stwu $rS, $dst

To still match those instructions against the appropriate pre_store
SelectionDAG nodes, the patch uses the new feature that allows a Pat
to match multiple DAG operands against a single (complex) instruction
operand.

Approved by Hal Finkel.

llvm-svn: 177429

The tocentry operand class refers to 64-bit values (it is only used in 64-bit,
where iPTR is a 64-bit type), but its sole suboperand is designated as 32-bit
type.  This causes a mismatch to be detected at compile-time with the TableGen
patch I'll check in shortly.

To fix this, this commit changes the suboperand to a 64-bit type as well.

llvm-svn: 177427

  def : Pat<(load (i64 (X86Wrapper tglobaltlsaddr :$dst))),
            (MOV64rm tglobaltlsaddr :$dst)>;

This pattern is invalid because the MOV64rm instruction expects a
source operand of type "i64mem", which is a subclass of X86MemOperand
and thus actually consists of five MI operands, but the Pat provides
only a single MI operand ("tglobaltlsaddr" matches an SDnode of
type ISD::TargetGlobalTLSAddress and provides a single output).

Thus, if the pattern were ever matched, subsequent uses of the MOV64rm
instruction pattern would access uninitialized memory.  In addition,
with the TableGen patch I'm about to check in, this would actually be
reported as a build-time error.

Fortunately, the pattern does in fact never match, for at least two
independent reasons.

First, the code generator actually never generates a pattern of the
form (load (X86Wrapper (tglobaltlsaddr))).  For most combinations of
TLS and code models, (tglobaltlsaddr) represents just an offset that
needs to be added to some base register, so it is never directly
dereferenced.  The only exception is the initial-exec model, where
(tglobaltlsaddr) refers to the (pc-relative) address of a GOT slot,
which *is* in fact directly dereferenced: but in that case, the
X86WrapperRIP node is used, not X86Wrapper, so the Pat doesn't match.

Second, even if some patterns along those lines *were* ever generated,
we should not need an extra Pat pattern to match it.  Instead, the
original MOV64rm instruction pattern ought to match directly, since
it uses an "addr" operand, which is implemented via the SelectAddr
C++ routine; this routine is supposed to accept the full range of
input DAGs that may be implemented by a single mov instruction,
including those cases involving ISD::TargetGlobalTLSAddress (and
actually does so e.g. in the initial-exec case as above).

To avoid build breaks (due to the above-mentioned error) after the
TableGen patch is checked in, I'm removing this Pat here.

llvm-svn: 177426

Currently the PPC r0 register is unconditionally reserved. There are two reasons
for this:

 1. r0 is treated specially (as the constant 0) by certain instructions, and so
    cannot be used with those instructions as a regular register.

 2. r0 is used as a temporary register in the CR-register spilling process
    (where, under some circumstances, we require two GPRs).

This change addresses the first reason by introducing a restricted register
class (without r0) for use by those instructions that treat r0 specially. These
register classes have a new pseudo-register, ZERO, which represents the r0-as-0
use. This has the side benefit of making the existing target code simpler (and
easier to understand), and will make it clear to the register allocator that
uses of r0 as 0 don't conflict will real uses of the r0 register.

Once the CR spilling code is improved, we'll be able to allocate r0.

Adding these extra register classes, for some reason unclear to me, causes
requests to the target to copy 32-bit registers to 64-bit registers. The
resulting code seems correct (and causes no test-suite failures), and the new
test case covers this new kind of asymmetric copy.

As r0 is still reserved, no functionality change intended.

llvm-svn: 177423

Patch by Ahmad, Muhammad T <muhammad.t.ahmad@intel.com>

llvm-svn: 177421

llvm-svn: 177418

llvm-svn: 177417

logic as a QOI cleanup.
rdar://13445327

llvm-svn: 177413

Remove an accidentally-added instruction definition and add a comment in the
test case. This is in response to a post-commit review by Bill Schmidt.

No functionality change intended.

llvm-svn: 177404

The ARM backend currently has poor codegen for long sext/zext
operations, such as v8i8 -> v8i32. This patch addresses this
by performing a custom expansion in ARMISelLowering. It also
adds/changes the cost of such lowering in ARMTTI.

This partially addresses PR14867.

Patch by Pete Couperus

llvm-svn: 177380