- caad163 Add support for matching vmrg(x,x) patterns by Chris Lattner · 18 years ago
- 253b9e7 fix some linking problems with the new gcc by Andrew Lenharth · 18 years ago
- 116cc48 Pattern match vmrg* instructions, which are now lowered by the CFE into shuffles. by Chris Lattner · 18 years ago
- 58d665c remove two done items by Chris Lattner · 18 years ago
- e2ed057 Lower vperm(x,y, mask) -> shuffle(x,y,mask) if mask is constant. This allows by Chris Lattner · 18 years ago
- d0608e1 Support pattern matching vsldoi(x,y) and vsldoi(x,x), which allows the f.e. to by Chris Lattner · 18 years ago
- ddb739e Compile the vpkuhum/vpkuwum intrinsics into vpkuhum/vpkuwum instead of into by Chris Lattner · 18 years ago
- c6cb5bb POR encoded as PAND, yikes. by Evan Cheng · 18 years ago
- 8af5ef9 An entry about comi / ucomi intrinsics. by Evan Cheng · 18 years ago
- 6be2c58 Support for comi / ucomi intrinsics. by Evan Cheng · 18 years ago
- d8242b4 Add all of the data stream intrinsics and instructions. woo by Chris Lattner · 18 years ago
- 99bdc65 Fix a typo by Chris Lattner · 18 years ago
- 9b42bdd Fix CodeGen/PowerPC/2006-04-05-splat-ish.ll by Chris Lattner · 18 years ago
- 1d5a8cc Handle canonical form of e.g. vector_shuffle v1, v1, <0, 4, 1, 5, 2, 6, 3, 7> by Evan Cheng · 18 years ago
- 2e2ef95 Make a vector live across blocks have the correct Vec type. This fixes by Chris Lattner · 18 years ago
- 865f060 Bogus assert by Evan Cheng · 18 years ago
- 278158b Fallthrough to expand if a VECTOR_SHUFFLE cannot be custom lowered. by Evan Cheng · 18 years ago
- 18dd6d0 Exapnd a VECTOR_SHUFFLE to a BUILD_VECTOR if target asks for it to be expanded by Evan Cheng · 18 years ago
- aed967d revert this, this is safe, if conservative. leave a note to that effect by Andrew Lenharth · 18 years ago
- c21a053 Handle v8i16 shuffle that must be broken into a pair of pshufhw / pshuflw. by Evan Cheng · 18 years ago
- 3827f71 add vsl by Chris Lattner · 18 years ago
- 0d2cf6b add vmladduhm by Chris Lattner · 18 years ago
- 4d9100d Add m[tf]vscr instructions. by Chris Lattner · 18 years ago
- 8b46842 add a note by Chris Lattner · 18 years ago
- 72e241c Add missing byte merges. by Chris Lattner · 18 years ago
- a046d4a Add FP -> Int Conversions by Chris Lattner · 18 years ago
- 3f0b7ff add average intrinsics by Chris Lattner · 18 years ago
- 298b684 add a note by Chris Lattner · 18 years ago
- 098e699 Fix some broken logic that would cause us to codegen {2147483647,2147483647,2147483647,2147483647} as 'vspltisb v0, -1'. by Chris Lattner · 18 years ago
- ff65e38 Added pslldq and psrldq. by Evan Cheng · 18 years ago
- 8703be4 Minor fixes + naming changes. by Evan Cheng · 18 years ago
- 5333b7b PSHUF* encoding bugs. by Evan Cheng · 18 years ago
- e44be60 Do not create ZEXTLOAD's unless we are before legalize or the operation is by Chris Lattner · 18 years ago
- 7ff7e67 Ask legalize to promote all vector shuffles to be v16i8 instead of having to by Chris Lattner · 18 years ago
- 4352cc9 * Add supprot for SCALAR_TO_VECTOR operations where the input needs to be by Chris Lattner · 18 years ago
- 1f23930 Signed shr by a constant is not the same as sdiv by 2^k by Chris Lattner · 18 years ago
- 2176046 cmpps / cmppd encoding bug by Evan Cheng · 18 years ago
- 08da55e Constant fold bitconvert(undef) by Chris Lattner · 18 years ago
- 6e96740 Compact some intrinsic definitions. by Evan Cheng · 18 years ago
- 684ad77 Plug in the byte and short splats by Chris Lattner · 18 years ago
- b683144 Revert accidentally committed hunks. by Chris Lattner · 18 years ago
- 01cae07 Make sure to mark unsupported SCALAR_TO_VECTOR operations as expand. by Chris Lattner · 18 years ago
- 97ac5fa Some SSE1 intrinsics: min, max, sqrt, etc. by Evan Cheng · 18 years ago
- 0aab36f revert previous patch by Chris Lattner · 18 years ago
- 20e3ed1 Use movlpd to: store lower f64 extracted from v2f64. by Evan Cheng · 18 years ago
- 4f91a4c Force use of a frame-pointer if there is anything on the stack that is aligned by Chris Lattner · 18 years ago
- 4672f71 The stack alignment is now computed dynamically, just verify it is correct. by Chris Lattner · 18 years ago
- 7dca8e2 Remove unused method by Chris Lattner · 18 years ago
- 11e15b3 - More efficient extract_vector_elt with shuffle and movss, movsd, movd, etc. by Evan Cheng · 18 years ago
- 630ebaf Align vectors to the size in bytes, not bits. by Chris Lattner · 18 years ago
- 97c2073 Add a missing check, this fixes UnitTests/Vector/sumarray.c by Chris Lattner · 18 years ago
- a4c5d8c Add a missing check, which broke a bunch of vector tests. by Chris Lattner · 18 years ago
- c461a51 Add the full set of min/max instructions by Chris Lattner · 18 years ago
- f87e793 support x * (c1 + c2) where c1 and c2 are pow2s. special case for c2 == 4 by Andrew Lenharth · 18 years ago
- afe3f49 mul by const conversion sequences. more coming soon by Andrew Lenharth · 18 years ago
- cf4fb61 back this out by Andrew Lenharth · 18 years ago
- 50a0d42 This should be a win of every arch by Andrew Lenharth · 18 years ago
- ccd9f98 This makes McCat/12-IOtest go 8x faster or so by Andrew Lenharth · 18 years ago
- ed5d1af This will be needed soon by Andrew Lenharth · 18 years ago
- 220d2b8 add a note by Chris Lattner · 18 years ago
- bbe77de Inform the dag combiner that the predicate compares only return a low bit. by Chris Lattner · 18 years ago
- 1b5232a relax assertion by Chris Lattner · 18 years ago
- 1482b5f Allow targets to compute masked bits for intrinsics. by Chris Lattner · 18 years ago
- 350bec0 Add a little dag combine to compile this: by Chris Lattner · 18 years ago
- 4132afb vector casts of casts are eliminable. Transform this: by Chris Lattner · 18 years ago
- a3124a3 vector casts never reinterpret bits by Chris Lattner · 18 years ago
- a1c3538 Allow transforming this: by Chris Lattner · 18 years ago
- 82ed58f Turn altivec lvx/stvx intrinsics into loads and stores. This allows the by Chris Lattner · 18 years ago
- 02b7255 Remove done item by Chris Lattner · 18 years ago
- 4aab2f4 Implement promotion for EXTRACT_VECTOR_ELT, allowing v16i8 multiplies to work with PowerPC. by Chris Lattner · 18 years ago
- c55bfd0 add a note by Chris Lattner · 18 years ago
- bc70cf8 Implement the Expand action for binary vector operations to break the binop by Chris Lattner · 18 years ago
- e58a780 Intrinsics that just load from memory can be treated like loads: they don't by Chris Lattner · 18 years ago
- 7224f84 Adjust to change in Intrinsics.gen interface. by Chris Lattner · 18 years ago
- edab1b9 Constant fold all of the vector binops. This allows us to compile this: by Chris Lattner · 18 years ago
- 5e46a19 Add a new -view-legalize-dags command line option by Chris Lattner · 18 years ago
- 6258fb2 Implement constant folding of bit_convert of arbitrary constant vbuild_vector nodes. by Chris Lattner · 18 years ago
- 5e08a11 These entries already exist by Chris Lattner · 18 years ago
- 97d2333 Add some missing node names by Chris Lattner · 18 years ago
- 5295122 New note by Chris Lattner · 18 years ago
- 4460f40 Constant fold casts from things like <4 x int> -> <4 x uint>, likewise int<->fp. by Chris Lattner · 18 years ago
- 541f91b Custom lower all BUILD_VECTOR's so that we can compile vec_splat_u8(8) into by Chris Lattner · 18 years ago
- f8814cf Prefer larger register classes over smaller ones when a register occurs in by Chris Lattner · 18 years ago
- 3805dea add valuemapper support for inline asm by Chris Lattner · 18 years ago
- c3837d4 Implement vnot using VNOR instead of using 'vspltisb v0, -1' and vxor by Chris Lattner · 18 years ago
- ae1ab39 Fix InstCombine/2006-04-01-InfLoop.ll by Chris Lattner · 18 years ago
- 64daab5 Fold A^(B&A) -> (B&A)^A Fold (B&A)^A == ~B & A by Chris Lattner · 18 years ago
- 60a05cc Fix Transforms/IndVarsSimplify/2006-03-31-NegativeStride.ll and by Chris Lattner · 18 years ago
- eaa7c06 ADd a note by Chris Lattner · 18 years ago
- 6e6b0da If we can look through vector operations to find the scalar version of an by Chris Lattner · 18 years ago
- 348ba3f Shrinkify some more intrinsic definitions. by Chris Lattner · 18 years ago
- 67d3d4c An entry about packed type alignments. by Evan Cheng · 18 years ago
- 6cea814 Pull operand asm string into base class, shrinkifying intrinsic definitions. by Chris Lattner · 18 years ago
- e668bda TargetData.cpp::getTypeInfo() was returning alignment of element type as the by Evan Cheng · 18 years ago
- f1d0c62 Delete identity shuffles, implementing CodeGen/Generic/vector-identity-shuffle.ll by Chris Lattner · 18 years ago
- b5c4d17 Fix 80 column violations :) by Chris Lattner · 18 years ago
- 653159f Use a X86 target specific node X86ISD::PINSRW instead of a mal-formed by Evan Cheng · 18 years ago
- 4b1734f Added support for SSE3 horizontal ops: haddp{s|d} and hsub{s|d}. by Evan Cheng · 18 years ago
- 7376a5e fix a pasto by Chris Lattner · 18 years ago
- a9cb441 Add vperm support for all datatypes by Chris Lattner · 18 years ago