Special handling was added for v4i32 -> v4i64 and v8i16 -> v8i32
extensions.

llvm-svn: 149600

Truncating v4i64 -> v4i32 and v8i32 -> v8i16 may be done with set of shuffles.

llvm-svn: 149485

Move some XOP patterns into instruction definition. Replae VPCMOV intrinsic patterns with custom lowering to a target specific nodes.

llvm-svn: 149216

llvm-svn: 148670

Merge PCMPEQB/PCMPEQW/PCMPEQD/PCMPEQQ and PCMPGTB/PCMPGTW/PCMPGTD/PCMPGTQ X86 ISD node types into only two node types. Simplifying opcode selection and pattern matching.

llvm-svn: 148667

Add target specific ISD node types for SSE/AVX vector shuffle instructions and change all the code that used to create intrinsic nodes to create the new nodes instead.

llvm-svn: 148664

llvm-svn: 148644

llvm-svn: 148466

llvm-svn: 147748

Peephole optimization of ptest-conditioned branch in X86 arch. Performs instruction combining of sequences generated by ptestz/ptestc intrinsics to ptest+jcc pair for SSE and AVX.

Testing: passed 'make check' including LIT tests for all sequences being handled (both SSE and AVX)

Reviewers: Evan Cheng, David Blaikie, Bruno Lopes, Elena Demikhovsky, Chad Rosier, Anton Korobeynikov
llvm-svn: 147601

llvm-svn: 147394

X86ISelLowering C++ code. Because this is lowered via an xor wrapped
around a bsr, we want the dagcombine which runs after isel lowering to
have a chance to clean things up. In particular, it is very common to
see code which looks like:

  (sizeof(x)*8 - 1) ^ __builtin_clz(x)

Which is trying to compute the most significant bit of 'x'. That's
actually the value computed directly by the 'bsr' instruction, but if we
match it too late, we'll get completely redundant xor instructions.

The more naive code for the above (subtracting rather than using an xor)
still isn't handled correctly due to the dagcombine getting confused.

Also, while here fix an issue spotted by inspection: we should have been
expanding the zero-undef variants to the normal variants when there is
an 'lzcnt' instruction. Do so, and test for this. We don't want to
generate unnecessary 'bsr' instructions.

These two changes fix some regressions in encoding and decoding
benchmarks. However, there is still a *lot* to be improve on in this
type of code.

llvm-svn: 147244

llvm-svn: 146833

Don't try to match 'unpackl/h v, v' for 32xi8 and 16xi16 when only AVX1 is supported. Fix 'unpackh v, v' for 256-bit types to understand 128-bit lanes.

llvm-svn: 146726

Remove some remants of the old palign pattern fragment that were still hanging around. Also remove a cast from inside getShuffleVPERM2X128Immediate and getShuffleVPERMILPImmediate since the only caller already had done the cast.

llvm-svn: 146344

llvm-svn: 145926

llvm-svn: 145485

Merge decoding of VPERMILPD and VPERMILPS shuffle masks. Merge X86ISD node type for VPERMILPD/PS. Add instruction selection support for VINSERTI128/VEXTRACTI128.

llvm-svn: 145483

Add X86 instruction selection for VPERM2I128 when AVX2 is enabled. Merge VPERMILPS/VPERMILPD detection since they are pretty similar.

llvm-svn: 145238

Merge 128-bit and 256-bit X86ISD node types for VPERMILPS and VPERMILPD. Simplify some shuffle lowering code since V1 can never be UNDEF due to canonalizing that occurs when shuffle nodes are created.

llvm-svn: 145153

Collapse X86ISD node types for PUNPCKH*, PUNPCKL*, UNPCKLP*, and UNPCKHP* to not be type specific. Now we just have integer high and low and floating point high and low. Pattern matching will choose the correct instruction based on the vector type.

llvm-svn: 145148

Remove 256-bit specific node types for UNPCKHPS/D and instead use the 128-bit versions and let the operand type disinquish. Also fix the load form of the v8i32 patterns for these to realize that the load would be promoted to v4i64.

llvm-svn: 145126

Remove AVX2 specific X86ISD node types for PUNPCKH/L and instead just reuse the 128-bit versions and let the vector type distinguish.

llvm-svn: 145125

llvm-svn: 145028

llvm-svn: 145026

Synthesize SSSE3/AVX 128-bit horizontal integer add/sub instructions from add/sub of appropriate shuffle vectors.

llvm-svn: 144989

llvm-svn: 144988

llvm-svn: 143080

Remove intrinsics for X86 BLSI, BLSMSK, and BLSR intrinsics and replace with custom isel lowering code.

llvm-svn: 142642

llvm-svn: 141947

floating point add/sub of appropriate shuffle vectors.  Does not
synthesize the 256 bit AVX versions because they work differently.

llvm-svn: 140332

Undo the changes from r139285 which added custom lowering to vselect.
Add tablegen lowering for vselect.

llvm-svn: 139479

Implement vector-select support for avx256. Refactor the vblend implementation to have tablegen match the instruction by the node type

llvm-svn: 139400

llvm-svn: 139285

llvm-svn: 139161

with a vector condition); such selects become VSELECT codegen nodes.
This patch also removes VSETCC codegen nodes, unifying them with SETCC
nodes (codegen was actually often using SETCC for vector SETCC already).
This ensures that various DAG combiner optimizations kick in for vector
comparisons.  Passes dragonegg bootstrap with no testsuite regressions
(nightly testsuite as well as "make check-all").  Patch mostly by
Nadav Rotem.

llvm-svn: 139159

init.trampoline and adjust.trampoline intrinsics, into two intrinsics
like in GCC.  While having one combined intrinsic is tempting, it is
not natural because typically the trampoline initialization needs to
be done in one function, and the result of adjust trampoline is needed
in a different (nested) function.  To get around this llvm-gcc hacks the
nested function lowering code to insert an additional parent variable
holding the adjust.trampoline result that can be accessed from the child
function.  Dragonegg doesn't have the luxury of tweaking GCC code, so it
stored the result of adjust.trampoline in the memory GCC set aside for
the trampoline itself (this is always available in the child function),
and set up some new memory (using an alloca) to hold the trampoline.
Unfortunately this breaks Go which allocates trampoline memory on the
heap and wants to use it even after the parent has exited (!).  Rather
than doing even more hacks to get Go working, it seemed best to just use
two intrinsics like in GCC.  Patch mostly by Sanjoy Das.

llvm-svn: 139140

code is inserted to first check if the current stacklet has enough
space. If so, space is allocated by simply decrementing the stack
pointer. Otherwise a runtime routine (__morestack_allocate_stack_space
in libgcc) is called which allocates the required memory from the
heap.

Patch by Sanjoy Das.

llvm-svn: 138818

from DYNAMIC_STACKALLOC.

Two new pseudo instructions (SEG_ALLOCA_32 and SEG_ALLOCA_64) which
will match X86SegAlloca (based on word size) are also added.  They
will be custom emitted to inject the actual stack handling code.

Patch by Sanjoy Das.

llvm-svn: 138814

llvm-svn: 138660