oota-llvm.git
2006-03-25 Chris LattnerImplement the ISD::isBuildVectorAllOnesInteger predicate
2006-03-25 Chris LattnerAdd a predicate
2006-03-25 Chris LattnerDon't call SimplifyDemandedBits on vectors
2006-03-25 Chris LattnerAdd some logical operations
2006-03-25 Chris LattnerAdd a missing check which cause an invalid .td file...
2006-03-25 Chris LattnerDon't crash on packed logical ops
2006-03-25 Chris LattnerTeach BinaryOperator::createNot to work with packed...
2006-03-25 Jim LaskeyUnused function - easier to throw away than fix.
2006-03-25 Jim LaskeyCast instruction not inserted into basic block.
2006-03-25 Evan ChengAdd a BUILD_VECTOR with unpack and interleave testcase.
2006-03-25 Evan ChengAdded missing (any_extend (load ...)) patterns.
2006-03-25 Evan ChengBuild arbitrary vector with more than 2 distinct scalar...
2006-03-25 Chris Lattnerimplement a bunch of intrinsics
2006-03-25 Chris LattnerMove all Altivec stuff out into a new PPCInstrAltivec...
2006-03-25 Chris LattnerAdd some basic patterns for other datatypes
2006-03-25 Chris Lattneradd all supported formats to the vector register file
2006-03-25 Chris Lattnerremove extraneous lets
2006-03-25 Chris LattnerAdd a bunch of simple altivec intrinsics
2006-03-25 Chris LattnerAdd support for __builtin_altivec_vnmsubfp /vmaddfp
2006-03-25 Chris LattnerAdd support for __builtin_altivec_vnmsubfp
2006-03-25 Chris LattnerWhen failing selection for an intrinsic, print this:
2006-03-25 Chris Lattner#include Intrinsics.h into all dag isels
2006-03-25 Chris LattnerImplement Intrinsic::getName
2006-03-25 Chris LattnerAdd a programatic interface to intrinsic names.
2006-03-25 Chris LattnerCodegen things like:
2006-03-25 Chris LattnerNew tests for vsplti*
2006-03-25 Evan ChengX86 SSE1 cacheability support ops intrinsics
2006-03-25 Evan ChengAdded SSE cachebility ops
2006-03-25 Evan ChengInstruction encoding bug
2006-03-25 Evan ChengAdded a scalar to vector with zero extension testcase
2006-03-25 Chris LattnerAdd new intrinsic node definitions for tblgen use
2006-03-25 Evan ChengX86 SSE1 SIMD store intrinsics.
2006-03-25 Evan ChengX86 SSE1 SIMD load intrinsics (movhps, movlps, and...
2006-03-25 Evan ChengX86 SSE1 conversion operations intrinsics.
2006-03-25 Evan ChengAdded 128-bit packed integer subtraction.
2006-03-25 Evan ChengAdded CVTTPS2PI.
2006-03-25 Evan ChengAdded CVTSS2SI.
2006-03-25 Evan ChengX86 SSE1 comparison intrinsics.
2006-03-25 Evan ChengX86 SSE1 arithmetic and logical operation intrinsics.
2006-03-24 Evan ChengSupport for scalar to vector with zero extension.
2006-03-24 Chris LattnerChange approach so that we get codegen for free for...
2006-03-24 Chris Lattnerfix inverted conditional
2006-03-24 Jim LaskeyD'oh - should be even numbered.
2006-03-24 Evan ChengAdded LDMXCSR
2006-03-24 Evan Chengldmxcsr is a SSE instruction.
2006-03-24 Evan ChengAdded ldmxcsr intrinsic.
2006-03-24 Chris Lattnerfix 80 column violations
2006-03-24 Chris Lattnerplug the intrinsics into the patterns for movmsk*
2006-03-24 Chris LattnerParse intrinsics correctly and perform type propagation...
2006-03-24 Jim LaskeyAdd dwarf register numbering to register data.
2006-03-24 Jim LaskeyAdd support for dwarf register numbering.
2006-03-24 Jim LaskeyHack no more.
2006-03-24 Chris LattnerMake sure to initialize the TheDef field!
2006-03-24 Chris Lattneradd another note
2006-03-24 Chris Lattneradd a note
2006-03-24 Chris LattnerMove CodeGenIntrinsic implementation to CodeGenTarget...
2006-03-24 Chris LattnerSpecify the value type for each llvm type. This needs...
2006-03-24 Chris LattnerShuffle some includes around
2006-03-24 Chris LattnerPull in valuetypes.td here, we will use it shortly.
2006-03-24 Evan ChengOnly to vector shuffle for {x,x,y,y} cases when SCALAR_...
2006-03-24 Chris Lattnerexpose intrinsic info to the targets.
2006-03-24 Chris LattnerFix a bad JIT encoding of VPERM. Why is VPERM D,A...
2006-03-24 Chris LattnerLike the comment says, prefer to use the implicit add...
2006-03-24 Chris LattnerNew testcase
2006-03-24 Jim LaskeyTweak a comment.
2006-03-24 Jim LaskeyFix indent.
2006-03-24 Jim LaskeyClean up some commentary.
2006-03-24 Jim LaskeyRename for truth in advertising.
2006-03-24 Jim LaskeyFixed some grammer and spelling.
2006-03-24 Evan ChengZero vector testcase
2006-03-24 Chris LattnerDisable the i32->float G5 optimization. It is unsafe...
2006-03-24 Chris Lattneradd support for using vxor to build zero vectors. ...
2006-03-24 Chris Lattnernew testcase
2006-03-24 Reid SpencerAdd a facility for invoking the llvm-config tool when...
2006-03-24 Evan ChengHandle BUILD_VECTOR with all zero elements.
2006-03-24 Chris Lattnerprefer to generate constant pool loads over splats...
2006-03-24 Chris LattnerFix a typo, thanks to Gabor Greif
2006-03-24 Chris Lattnerfix spello
2006-03-24 Chris LattnerFix spello
2006-03-24 Chris Lattneradd the actual cost to the debug info
2006-03-24 Chris Lattnerfix two spellos
2006-03-24 Chris LattnerGabor points out that we can't spell. :)
2006-03-24 Reid SpencerRemove a spurious <table> tag.
2006-03-24 Evan ChengAll v2f64 shuffle cases can be handled.
2006-03-24 Jeff CohenFix headers
2006-03-24 Jeff CohenMinor corrections.
2006-03-24 Jeff CohenMinor corrections.
2006-03-24 Evan ChengMore efficient v2f64 shuffle using movlhps, movhlps...
2006-03-24 Evan ChengA new entry
2006-03-24 Evan ChengAdd a v2f64 shuffle case.
2006-03-24 Jeff CohenGet JIT/Interpreter working on Windows again.
2006-03-24 Chris Lattnerlegalize vbit_convert nodes whose result is a legal...
2006-03-24 Chris LattnerLower target intrinsics into an INTRINSIC node
2006-03-24 Reid SpencerIgnore the lexer output.
2006-03-24 Reid SpencerIgnore the burg output files.
2006-03-24 Chris Lattnerfix some bogus assertions: noop bitconverts are legal
2006-03-24 Jeff CohenGet Visual Studio happy again.
2006-03-24 Evan ChengAdd a vector shuffle test case
2006-03-24 Evan ChengHandle more shuffle cases with SHUFP* instructions.
2006-03-24 Evan ChengLower BUILD_VECTOR to VECTOR_SHUFFLE if there are two...
next