Coping 100MB array (after a warmup) shows that glibc 2.6.1 implementation on
x86-64 (core 2) is 30% faster (from 0.270917s to 0.188079s)

llvm-svn: 41479

changing the interface to allow for future changes.

llvm-svn: 41384

 *) The needed information is propagated to the DAG
 *) The X86-64 backend detects it and aborts

llvm-svn: 41179

hard to catch bugs with retaddr lowering

llvm-svn: 41104

llvm-svn: 41059

Increase efficiency of sign_extend_inreg by using subregisters for truncation. As the README suggests sign_extend_subreg is selected to (sext(trunc)).

llvm-svn: 41010

llvm-svn: 40986

(constants are still not handled).  Adds ConvertActions
to control fp-to-fp conversions (these are currently
defaulted for all other targets, so no changes there).

llvm-svn: 40958

Last x87 bits for full functionality (not
thoroughly tested, and long doubles do not work
in SSE modes at all - use -mcpu=i486 for now)

llvm-svn: 40886

Lots of problems yet but some simple things work.

llvm-svn: 40847

Generalize isPSHUFDMask and add a unary SHUFPD pattern so that SHUFPD's
memory operand alignment can be tested as well, with a fix to avoid
breaking MMX's use of isPSHUFDMask.

llvm-svn: 40756

llvm-svn: 40703

llvm-svn: 40617

llvm-svn: 40566

Make the alignedload and alignedstore patterns always require 16-byte
alignment. This way when they are used in the "Fs" instructions, in which
a vector instruction is used for a scalar purpose, they can still require
the full vector alignment. And add a regression test for this.

llvm-svn: 40555

llvm-svn: 40547

x86 target, replacing them with the new alignment attributes on memory
references.

llvm-svn: 40504

register instead of loading each element individually.

llvm-svn: 40478

llvm-svn: 40443

llvm-svn: 40071

This patch fills the last necessary bits to enable exceptions
handling in LLVM. Currently only on x86-32/linux.

In fact, this patch adds necessary intrinsics (and their lowering) which
represent really weird target-specific gcc builtins used inside unwinder.

After corresponding llvm-gcc patch will land (easy) exceptions should be
more or less workable. However, exceptions handling support should not be 
thought as 'finished': I expect many small and not so small glitches
everywhere.

llvm-svn: 39855

in addition to the intrinsic forms. Add spill-folding entries for these new
instructions, and for the scalar min and max instrinsic instructions which
were missing. And add some preliminary ISelLowering code for using the new
non-intrinsic vector sqrt instruction, and fneg and fabs.

llvm-svn: 38478

llvm-svn: 37923

their names are changed.

llvm-svn: 37876

model to include f32 variants.  Some factoring
improvments forthcoming.

llvm-svn: 37847

llvm-svn: 37786

llvm-svn: 37784

extended vector types. Remove the special SDNode opcodes used for pre-legalize
vector operations, and the special MVT::Vector type used with them. Adjust
lowering and legalize to work with the normal SDNode kinds instead, and to
use the normal MVT functions to work with vector types instead of using the
two special operands that the pre-legalize nodes held.

This allows pre-legalize and post-legalize DAGs, and the code that operates
on them, to be more consistent. Pre-legalize vector operators can be handled
more consistently with scalar operators. And, -view-dag-combine1-dags and
-view-legalize-dags now look prettier for vector code.

llvm-svn: 37719

TargetLowering to SelectionDAG so that they have more convenient
access to the current DAG, in preparation for the ValueType routines
being changed from standalone functions to members of SelectionDAG for
the pre-legalize vector type changes.

llvm-svn: 37704

Anton for half of this patch.

llvm-svn: 37641

Look for VECTOR_SHUFFLE that's identity operation on either LHS or RHS. This can happen before DAGCombiner catches it.

llvm-svn: 37636

define double @test2(i64 %A) {
   %B = bitcast i64 %A to double
   ret double %B
}

$ llvm-as < t.ll | llc -march=x86-64

before:

         .align  4
         .globl  _test2
_test2:
         movd %rdi, %xmm0
         ret

after:

_test2:
         subq $8, %rsp
         movq %rdi, (%rsp)
         movsd (%rsp), %xmm0
         addq $8, %rsp
         ret

llvm-svn: 37617

llvm-svn: 37611

llvm-svn: 37579

llvm-svn: 37534

llvm-svn: 37533

llvm-svn: 37320

llvm-svn: 37234

Fix a bogus check that prevented folding VECTOR_SHUFFLE to UNDEF; add an optimization to fold VECTOR_SHUFFLE to a zero vector.

llvm-svn: 37173

cause other regressions.

llvm-svn: 37160