oota-llvm.git
2006-04-19 Evan ChengMore mov{h|l}p{d|s} patterns.
2006-04-19 Evan Cheng- More mov{h|l}ps patterns.
2006-04-19 Evan ChengAllow "let AddedCost = n in" to increase pattern comple...
2006-04-19 Chris LattnerAlpha too!
2006-04-19 Chris Lattneradd a note
2006-04-19 Andrew LenharthAnother simple case type merge case to try
2006-04-19 Andrew Lenharthdeal with memchr
2006-04-19 Andrew Lenharthfriendlier error message
2006-04-19 Chris Lattneradd a note
2006-04-19 Chris LattnerAdd a note.
2006-04-19 Chris Lattnergrammaro
2006-04-19 Chris LattnerFix a bug owen noticed
2006-04-19 Chris LattnerChange wording
2006-04-19 Chris Lattneradd a note
2006-04-19 Chris Lattneradd some more notes
2006-04-19 Andrew Lenharthstupid stuff
2006-04-18 Andrew Lenharthfix printing call graphs
2006-04-18 Andrew LenharthI understand now. Shoot.
2006-04-18 Evan Cheng- PEXTRW cannot take a memory location as its first...
2006-04-18 Evan ChengSHUFP{S|D}, PSHUF* encoding bugs. Left out the mask...
2006-04-18 Evan ChengName change for clarity sake
2006-04-18 Evan ChengEncoding bug: CMPPSrmi, CMPPDrmi dropped operand 2...
2006-04-18 Evan ChengName change for clarity sake
2006-04-18 Evan ChengLeft a pattern out
2006-04-18 Andrew Lenharthllvm.memc* improvements. helps PA a lot in some specmarks
2006-04-18 Andrew Lenharthllvm.memc* improvements. helps PA a lot in some specmarks
2006-04-18 Chris LattnerThese are correctly encoded by the JIT. I checked :)
2006-04-18 Chris Lattneradd a note
2006-04-18 Chris LattnerFix a crash on:
2006-04-18 Evan ChengFixed an encoding bug: movd from XMM to R32.
2006-04-18 Chris Lattnerpretty print node name
2006-04-18 Chris LattnerImplement an important entry from README_ALTIVEC:
2006-04-18 Chris Lattnernew testcase
2006-04-18 Chris Lattnermove some stuff around, clean things up
2006-04-18 Chris LattnerTeach the codegen about instructions used for SSE spill...
2006-04-18 Nate BegemanFix a copy & paste error from long ago.
2006-04-18 Chris LattnerAdd some more notes, many still missing
2006-04-18 Reid SpencerHave the AutoRegen.sh script prompt the user for the...
2006-04-18 Reid SpencerMake it possible to default the llvm_src and llvm_obj...
2006-04-18 Chris Lattneradd a bunch of stuff, pieces still missing
2006-04-18 Chris LattnerAdd a warning.
2006-04-18 Chris LattnerAdd a warning
2006-04-18 Chris LattnerUse vmladduhm to do v8i16 multiplies which is faster...
2006-04-18 Chris LattnerImplement v16i8 multiply with this code:
2006-04-18 Chris LattnerAdd tests for v8i16 and v16i8
2006-04-18 Evan ChengCorrect comments
2006-04-18 Chris LattnerLower v8i16 multiply into this code:
2006-04-18 Chris LattnerCustom lower v4i32 multiplies into a cute sequence...
2006-04-18 Chris Lattnernew testcase
2006-04-18 Evan ChengAnother entry
2006-04-18 Chris LattnerFix a build failure on Vladimir's tester.
2006-04-18 Evan ChengAnother entry.
2006-04-17 Evan ChengUse movss to insert_vector_elt(v, s, 0).
2006-04-17 Chris LattnerTurn x86 unaligned load/store intrinsics into aligned...
2006-04-17 Chris LattnerFix handling of calls in functions that use vectors...
2006-04-17 Evan ChengUse two pinsrw to insert an element into v4i32 / v4f32...
2006-04-17 Chris Lattnerremove done item
2006-04-17 Chris LattnerDon't diddle VRSAVE if no registers need to be added...
2006-04-17 Chris LattnerNew testcase, shouldn't touch vrsave
2006-04-17 Chris LattnerAdd a MachineInstr::eraseFromParent convenience method.
2006-04-17 Chris LattnerAdd some convenience methods.
2006-04-17 Evan ChengEncoding bug
2006-04-17 Chris LattnerVectors that are known live-in and live-out are clearly...
2006-04-17 Chris LattnerPrefer to allocate V2-V5 before V0,V1. This lets us...
2006-04-17 Chris LattnerMove some knowledge about registers out of the code...
2006-04-17 Chris LattnerUse a small table instead of macros to do this conversion.
2006-04-17 Evan ChengImplement v8i16, v16i8 splat using unpckl + pshufd.
2006-04-17 Chris Lattnerimplement returns of a vector, testcase here: CodeGen...
2006-04-17 Chris LattnerNew testcase
2006-04-17 Chris LattnerCodegen insertelement with constant insertion points...
2006-04-17 Chris LattnerMake sure to check splats of every constant we can...
2006-04-17 Evan ChengIncorrect foldMemoryOperand entries
2006-04-17 Evan ChengErrors in patterns preventing load folding
2006-04-17 Jeff CohenAdd checks for __OpenBSD__.
2006-04-17 Chris LattnerTeach the ppc backend to use rol and vsldoi to generate...
2006-04-17 Chris LattnerSome more cases that can be generated with two instructions
2006-04-17 Chris Lattneradd a note
2006-04-17 Evan ChengFP SETOLT, SETOLT, SETUGE, SETUGT conditions were imple...
2006-04-17 Chris LattnerMake some code more general, adding support for constan...
2006-04-17 Chris LattnerNew testcases
2006-04-17 Chris LattnerLearn how to make odd splatted constants in range ...
2006-04-17 Chris Lattnernew testcase
2006-04-17 Chris LattnerPull some code out into a helper function.
2006-04-17 Chris LattnerNew testcase
2006-04-17 Chris LattnerImplement a TODO: for any shuffle that can be viewed...
2006-04-17 Chris Lattnernew testcase, these shuffles can be implemented with...
2006-04-17 Chris LattnerRegenerate with adjusted costs
2006-04-17 Chris LattnerEncode a cost of zero as a cost of 1.
2006-04-17 Chris LattnerRegenerate with correct offset
2006-04-17 Chris LattnerReally, I can count!
2006-04-17 Chris LattnerIncrease the opcodes by one each to disambiguate COPY...
2006-04-17 Chris Lattnerassign stable opcodes to the various altivec ops.
2006-04-17 Chris LattnerPPCPerfectShuffle.h is autogenerated, don't include...
2006-04-17 Chris LattnerCheck in a table, generated by llvm-PerfectShuffle...
2006-04-17 Chris LattnerRename BuildShuffleTable -> PerfectShuffle
2006-04-17 Chris Lattnerrename the table
2006-04-17 Chris LattnerInitial checking of a perfect shuffle generation progra...
2006-04-16 Evan ChengBetter way to splat v2f64
2006-04-16 Evan Chengmovduprm, movshduprm bugs
2006-04-16 Evan ChengEncoding bugs
next