It had patterns for zext-loading and extending. This commit adds patterns for loading a wide type, performing a bitcast,
and extending. This is an odd pattern, but it is commonly used when writing code with intrinsics.

rdar://11897677

llvm-svn: 163995

    
- Fix an remaining issue of PR11674 as well

llvm-svn: 163528

llvm-svn: 163473

Use iPTR instead of i32 for extract_subvector/insert_subvector index in lowering and patterns. This makes it consistent with the incoming DAG nodes from the DAG builder.

llvm-svn: 163293

Add patterns for converting stores of subvector_extracts of lower 128-bits of a 256-bit vector to VMOVAPSmr/VMOVUPSmr.

llvm-svn: 163292

Remove some of the patterns added in r163196. Increasing the complexity on insert_subvector into undef accomplishes the same thing.

llvm-svn: 163198

Add patterns for integer forms of VINSERTF128/VINSERTI128 folded with loads. Also add patterns to turn subvector inserts with loads to index 0 of an undef into VMOVAPS.

llvm-svn: 163196

Convert vextracti128/vextractf128 intrinsics to extract_subvector at DAG build time. Similar was previously done for vinserti128/vinsertf128. Add patterns for folding these extract_subvectors with stores.

llvm-svn: 163192

llvm-svn: 163053

llvm-svn: 162973

The assembly string for the VMOVPQIto64rr instruction incorrectly lacked the 'v'
prefix, resulting in mis-assembly of the vanilla movd instruction.

llvm-svn: 162963

- Add 'UseSSEx' to force SSE legacy insn not being selected when AVX is
  enabled.

  As the penalty of inter-mixing SSE and AVX instructions, we need
  prevent SSE legacy insn from being generated except explicitly
  specified through some intrinsics. For patterns supported by both
  SSE and AVX, so far, we force AVX insn will be tried first relying on
  AddedComplexity or position in td file. It's error-prone and
  introduces bugs accidentally.

  'UseSSEx' is disabled when AVX is turned on. For SSE insns inherited
  by AVX, we need this predicate to force VEX encoding or SSE legacy
  encoding only.

  For insns not inherited by AVX, we still use the previous predicates,
  i.e. 'HasSSEx'. So far, these insns fall into the following
  categories:
  * SSE insns with MMX operands
  * SSE insns with GPR/MEM operands only (xFENCE, PREFETCH, CLFLUSH,
    CRC, and etc.)
  * SSE4A insns.
  * MMX insns.
  * x87 insns added by SSE.

2 test cases are modified:

 - test/CodeGen/X86/fast-isel-x86-64.ll
   AVX code generation is different from SSE one. 'vcvtsi2sdq' cannot be
   selected by fast-isel due to complicated pattern and fast-isel
   fallback to materialize it from constant pool.

 - test/CodeGen/X86/widen_load-1.ll
   AVX code generation is different from SSE one after fixing SSE/AVX
   inter-mixing. Exec-domain fixing prefers 'vmovapd' instead of
   'vmovaps'.

llvm-svn: 162919

it here, then a 'register-memory' version would wrongly get the commutative
flag.
<rdar://problem/12180135>

llvm-svn: 162741

llvm-svn: 162740

llvm-svn: 162738

llvm-svn: 162714

Don't allow vextractf128 to be folded with unaligned stores. We don't fold unaligned loads so shouldn't fold unaligned stores as it can cause an alignment fault to occur.

llvm-svn: 162658

Fold some patterns into instruction definitons so tablegen can infer flags removing the need for an explicit 'neverHasSideEffects = 1'

llvm-svn: 162656

Add HasAVX1Only predicate and use it for patterns that have an AVX1 instruction and an AVX2 instruction rather than relying on AddedComplexity.

llvm-svn: 162654

llvm-svn: 162622

They were inserted to silence TableGen's warning about
redundant properties. That warning is now gone.

llvm-svn: 162517

this allows for better code generation.

Added a new DAGCombine transformation to convert FMAX and FMIN to FMANC and
FMINC, which are commutative.

For example:

  movaps  %xmm0, %xmm1
  movsd LC(%rip), %xmm0
  minsd %xmm1, %xmm0

becomes:

  minsd LC(%rip), %xmm0

llvm-svn: 162187

- FP_EXTEND only support extending from vectors with matching elements.
  This results in the scalarization of extending to v2f64 from v2f32,
  which will be legalized to v4f32 not matching with v2f64.
- add X86-specific VFPEXT supproting extending from v4f32 to v2f64.
- add BUILD_VECTOR lowering helper to recover back the original
  extending from v4f32 to v2f64.
- test case is enhanced to include different vector width.

llvm-svn: 161894

Implement proper handling for pcmpistri/pcmpestri intrinsics. Requires custom handling in DAGISelToDAG due to limitations in TableGen's implicit def handling. Fixes PR11305.

llvm-svn: 161318

llvm-svn: 161306

llvm-svn: 161143

llvm-svn: 160942

llvm-svn: 160941

Fix up patterns for VCVTSS2SD. Specifically give it priority over SSE form. Add an OptForSpeed to explicitly pair up with an OptForSize that was already on another pattern.

llvm-svn: 160939

llvm-svn: 160938

llvm-svn: 160937

llvm-svn: 160922

llvm-svn: 160921

llvm-svn: 160914

llvm-svn: 160913

I'll remove these two sub-register indexes shortly.

llvm-svn: 160831

The (COPY_TO_REGCLASS GR32:$src, VR128) pattern looks odd, but
copyPhysReg does the right thing with it. (The old pattern would
eventually produce the same cross-class copy).

llvm-svn: 160830

This gets rid of some more INSERT_SUBREG - IMPLICIT_DEF patterns,
simplifying the emitted code a bit.

llvm-svn: 160820

The SUBREG_TO_REG instruction has magic semantics asserting that the
source value was defined by an instruction that cleared the high half of
the register. Those semantics are never actually exploited for xmm
registers.

llvm-svn: 160818

These idempotent sub-register indices don't do anything --- They simply
map XMM registers to themselves.  They no longer affect register classes
either since the SubRegClasses field has been removed from Target.td.

This patch replaces XMM->XMM EXTRACT_SUBREG and INSERT_SUBREG patterns
with COPY_TO_REGCLASS patterns which simply become COPY instructions.

The number of IMPLICIT_DEF instructions before register allocation is
reduced, and that is the cause of the test case changes.

llvm-svn: 160816