summaryrefslogtreecommitdiff
path: root/lib
Commit message (Expand)AuthorAge
* We have an assertion that checks that we do not encode null values into theChris Lattner2006-04-07
* Constant fold extractelement(zero, x) -> zeroChris Lattner2006-04-07
* Fix inlining of insert/extract element constantexprsChris Lattner2006-04-07
* Add code to RemapOperand() to handle Instruction::ExtractElement andEvan Cheng2006-04-07
* A MOVPS2SSmr, i.e. _mm_store_ss, encoding bug.Evan Cheng2006-04-06
* - movlp{s|d} and movhp{s|d} support.Evan Cheng2006-04-06
* New entries.Evan Cheng2006-04-06
* 1. If both vector operands of a vector_shuffle are undef, turn it into an undef.Evan Cheng2006-04-06
* This may be overconservative, but it lets the new cfe compileAndrew Lenharth2006-04-06
* Add an itemChris Lattner2006-04-06
* Make sure to return the result in the right type.Chris Lattner2006-04-06
* Move this to lib/Analysis.Andrew Lenharth2006-04-06
* Match vpku[hw]um(x,x).Chris Lattner2006-04-06
* Add support for matching vmrg(x,x) patternsChris Lattner2006-04-06
* fix some linking problems with the new gccAndrew Lenharth2006-04-06
* Pattern match vmrg* instructions, which are now lowered by the CFE into shuff...Chris Lattner2006-04-06
* remove two done itemsChris Lattner2006-04-06
* Lower vperm(x,y, mask) -> shuffle(x,y,mask) if mask is constant. This allowsChris Lattner2006-04-06
* Support pattern matching vsldoi(x,y) and vsldoi(x,x), which allows the f.e. toChris Lattner2006-04-06
* Compile the vpkuhum/vpkuwum intrinsics into vpkuhum/vpkuwum instead of intoChris Lattner2006-04-06
* POR encoded as PAND, yikes.Evan Cheng2006-04-06
* An entry about comi / ucomi intrinsics.Evan Cheng2006-04-05
* Support for comi / ucomi intrinsics.Evan Cheng2006-04-05
* Add all of the data stream intrinsics and instructions. wooChris Lattner2006-04-05
* Fix a typoChris Lattner2006-04-05
* Fix CodeGen/PowerPC/2006-04-05-splat-ish.llChris Lattner2006-04-05
* Handle canonical form of e.g.Evan Cheng2006-04-05
* Make a vector live across blocks have the correct Vec type. This fixesChris Lattner2006-04-05
* Bogus assertEvan Cheng2006-04-05
* Fallthrough to expand if a VECTOR_SHUFFLE cannot be custom lowered.Evan Cheng2006-04-05
* Exapnd a VECTOR_SHUFFLE to a BUILD_VECTOR if target asks for it to be expandedEvan Cheng2006-04-05
* revert this, this is safe, if conservative. leave a note to that effectAndrew Lenharth2006-04-05
* Handle v8i16 shuffle that must be broken into a pair of pshufhw / pshuflw.Evan Cheng2006-04-05
* add vslChris Lattner2006-04-05
* add vmladduhmChris Lattner2006-04-05
* Add m[tf]vscr instructions.Chris Lattner2006-04-05
* add a noteChris Lattner2006-04-04
* Add missing byte merges.Chris Lattner2006-04-04
* Add FP -> Int ConversionsChris Lattner2006-04-04
* add average intrinsicsChris Lattner2006-04-04
* add a noteChris Lattner2006-04-04
* Fix some broken logic that would cause us to codegen {2147483647,2147483647,2...Chris Lattner2006-04-04
* Added pslldq and psrldq.Evan Cheng2006-04-04
* Minor fixes + naming changes.Evan Cheng2006-04-04
* PSHUF* encoding bugs.Evan Cheng2006-04-04
* Do not create ZEXTLOAD's unless we are before legalize or the operation isChris Lattner2006-04-04
* Ask legalize to promote all vector shuffles to be v16i8 instead of having toChris Lattner2006-04-04
* * Add supprot for SCALAR_TO_VECTOR operations where the input needs to beChris Lattner2006-04-04
* Signed shr by a constant is not the same as sdiv by 2^kChris Lattner2006-04-04
* cmpps / cmppd encoding bugEvan Cheng2006-04-04