2006-04-04 |
Chris Lattner | Move isShuffleLegal from TLI to Legalize. |
commit | commitdiff | tree | snapshot |
2006-04-04 |
Chris Lattner | new testcase |
commit | commitdiff | tree | snapshot |
2006-04-04 |
Chris Lattner | Signed shr by a constant is not the same as sdiv by 2^k |
commit | commitdiff | tree | snapshot |
2006-04-04 |
Evan Cheng | cmpps / cmppd encoding bug |
commit | commitdiff | tree | snapshot |
2006-04-04 |
Chris Lattner | Fix the types for these intrinsics. |
commit | commitdiff | tree | snapshot |
2006-04-04 |
Chris Lattner | Constant fold bitconvert(undef) |
commit | commitdiff | tree | snapshot |
2006-04-04 |
Chris Lattner | Allow targets to have fine grained control over which... |
commit | commitdiff | tree | snapshot |
2006-04-04 |
Evan Cheng | Compact some intrinsic definitions. |
commit | commitdiff | tree | snapshot |
2006-04-04 |
Chris Lattner | Plug in the byte and short splats |
commit | commitdiff | tree | snapshot |
2006-04-03 |
Chris Lattner | Revert accidentally committed hunks. |
commit | commitdiff | tree | snapshot |
2006-04-03 |
Chris Lattner | Make sure to mark unsupported SCALAR_TO_VECTOR operatio... |
commit | commitdiff | tree | snapshot |
2006-04-03 |
Evan Cheng | Some SSE1 intrinsics: min, max, sqrt, etc. |
commit | commitdiff | tree | snapshot |
2006-04-03 |
Chris Lattner | revert previous patch |
commit | commitdiff | tree | snapshot |
2006-04-03 |
Evan Cheng | Use movlpd to: store lower f64 extracted from v2f64. |
commit | commitdiff | tree | snapshot |
2006-04-03 |
Chris Lattner | Force use of a frame-pointer if there is anything on... |
commit | commitdiff | tree | snapshot |
2006-04-03 |
Chris Lattner | The stack alignment is now computed dynamically, just... |
commit | commitdiff | tree | snapshot |
2006-04-03 |
Chris Lattner | Remove unused method |
commit | commitdiff | tree | snapshot |
2006-04-03 |
Chris Lattner | Keep track of max stack alignment as objects are added... |
commit | commitdiff | tree | snapshot |
2006-04-03 |
Evan Cheng | - More efficient extract_vector_elt with shuffle and... |
commit | commitdiff | tree | snapshot |
2006-04-03 |
Chris Lattner | Align vectors to the size in bytes, not bits. |
commit | commitdiff | tree | snapshot |
2006-04-03 |
Chris Lattner | Add a missing check, this fixes UnitTests/Vector/sumarray.c |
commit | commitdiff | tree | snapshot |
2006-04-03 |
Chris Lattner | Add a missing check, which broke a bunch of vector... |
commit | commitdiff | tree | snapshot |
2006-04-03 |
Chris Lattner | shrinkify intrinsics more by using some local classes |
commit | commitdiff | tree | snapshot |
2006-04-03 |
Chris Lattner | Add the full set of min/max instructions |
commit | commitdiff | tree | snapshot |
2006-04-03 |
Chris Lattner | Add some classes to make it easier to define intrinsics... |
commit | commitdiff | tree | snapshot |
2006-04-03 |
Andrew Lenharth | support x * (c1 + c2) where c1 and c2 are pow2s. speci... |
commit | commitdiff | tree | snapshot |
2006-04-03 |
Andrew Lenharth | test powers of 2 |
commit | commitdiff | tree | snapshot |
2006-04-03 |
Andrew Lenharth | mul by const conversion sequences. more coming soon |
commit | commitdiff | tree | snapshot |
2006-04-03 |
Andrew Lenharth | back this out |
commit | commitdiff | tree | snapshot |
2006-04-03 |
Andrew Lenharth | test some more mul by constant removal |
commit | commitdiff | tree | snapshot |
2006-04-02 |
Andrew Lenharth | Make sure mul by constant 5 is turned into a s4addq |
commit | commitdiff | tree | snapshot |
2006-04-02 |
Andrew Lenharth | This should be a win of every arch |
commit | commitdiff | tree | snapshot |
2006-04-02 |
Andrew Lenharth | This makes McCat/12-IOtest go 8x faster or so |
commit | commitdiff | tree | snapshot |
2006-04-02 |
Andrew Lenharth | This will be needed soon |
commit | commitdiff | tree | snapshot |
2006-04-02 |
Reid Spencer | For PR722: |
commit | commitdiff | tree | snapshot |
2006-04-02 |
Chris Lattner | add a note |
commit | commitdiff | tree | snapshot |
2006-04-02 |
Chris Lattner | Inform the dag combiner that the predicate compares... |
commit | commitdiff | tree | snapshot |
2006-04-02 |
Chris Lattner | relax assertion |
commit | commitdiff | tree | snapshot |
2006-04-02 |
Chris Lattner | Allow targets to compute masked bits for intrinsics. |
commit | commitdiff | tree | snapshot |
2006-04-02 |
Chris Lattner | Add a little dag combine to compile this: |
commit | commitdiff | tree | snapshot |
2006-04-02 |
Chris Lattner | vector casts of casts are eliminable. Transform this: |
commit | commitdiff | tree | snapshot |
2006-04-02 |
Chris Lattner | vector casts never reinterpret bits |
commit | commitdiff | tree | snapshot |
2006-04-02 |
Chris Lattner | Allow transforming this: |
commit | commitdiff | tree | snapshot |
2006-04-02 |
Chris Lattner | Turn altivec lvx/stvx intrinsics into loads and stores... |
commit | commitdiff | tree | snapshot |
2006-04-02 |
Chris Lattner | Remove done item |
commit | commitdiff | tree | snapshot |
2006-04-02 |
Jeff Cohen | Fix tablegen related dependencies in Visual Studio. |
commit | commitdiff | tree | snapshot |
2006-04-02 |
Chris Lattner | Implement promotion for EXTRACT_VECTOR_ELT, allowing... |
commit | commitdiff | tree | snapshot |
2006-04-02 |
Chris Lattner | add a note |
commit | commitdiff | tree | snapshot |
2006-04-02 |
Chris Lattner | Implement the Expand action for binary vector operation... |
commit | commitdiff | tree | snapshot |
2006-04-02 |
Chris Lattner | Intrinsics that just load from memory can be treated... |
commit | commitdiff | tree | snapshot |
2006-04-02 |
Chris Lattner | Adjust the Intrinsics.gen interface a little bit |
commit | commitdiff | tree | snapshot |
2006-04-02 |
Chris Lattner | Adjust to change in Intrinsics.gen interface. |
commit | commitdiff | tree | snapshot |
2006-04-02 |
Chris Lattner | Constant fold all of the vector binops. This allows... |
commit | commitdiff | tree | snapshot |
2006-04-02 |
Chris Lattner | Add a new -view-legalize-dags command line option |
commit | commitdiff | tree | snapshot |
2006-04-02 |
Chris Lattner | Implement constant folding of bit_convert of arbitrary... |
commit | commitdiff | tree | snapshot |
2006-04-02 |
Chris Lattner | These entries already exist |
commit | commitdiff | tree | snapshot |
2006-04-02 |
Chris Lattner | Add some missing node names |
commit | commitdiff | tree | snapshot |
2006-04-02 |
Chris Lattner | simplify this method |
commit | commitdiff | tree | snapshot |
2006-04-02 |
Chris Lattner | New note |
commit | commitdiff | tree | snapshot |
2006-04-02 |
Chris Lattner | Constant fold casts from things like <4 x int> -> ... |
commit | commitdiff | tree | snapshot |
2006-04-02 |
Chris Lattner | Custom lower all BUILD_VECTOR's so that we can compile... |
commit | commitdiff | tree | snapshot |
2006-04-02 |
Chris Lattner | Prefer larger register classes over smaller ones when... |
commit | commitdiff | tree | snapshot |
2006-04-02 |
Chris Lattner | New testcase that crashes the compiler. |
commit | commitdiff | tree | snapshot |
2006-04-01 |
Chris Lattner | add valuemapper support for inline asm |
commit | commitdiff | tree | snapshot |
2006-04-01 |
Chris Lattner | Implement vnot using VNOR instead of using 'vspltisb... |
commit | commitdiff | tree | snapshot |
2006-04-01 |
Chris Lattner | Fix InstCombine/2006-04-01-InfLoop.ll |
commit | commitdiff | tree | snapshot |
2006-04-01 |
Chris Lattner | New testcase that caused instcombine to infinitely... |
commit | commitdiff | tree | snapshot |
2006-04-01 |
Chris Lattner | Fold A^(B&A) -> (B&A)^A |
commit | commitdiff | tree | snapshot |
2006-04-01 |
Chris Lattner | new testcases |
commit | commitdiff | tree | snapshot |
2006-04-01 |
Chris Lattner | Fix Transforms/IndVarsSimplify/2006-03-31-NegativeStrid... |
commit | commitdiff | tree | snapshot |
2006-04-01 |
Chris Lattner | new testcase for PR726 |
commit | commitdiff | tree | snapshot |
2006-04-01 |
Chris Lattner | ADd a note |
commit | commitdiff | tree | snapshot |
2006-03-31 |
Chris Lattner | If we can look through vector operations to find the... |
commit | commitdiff | tree | snapshot |
2006-03-31 |
Chris Lattner | Shrinkify some more intrinsic definitions. |
commit | commitdiff | tree | snapshot |
2006-03-31 |
Evan Cheng | An entry about packed type alignments. |
commit | commitdiff | tree | snapshot |
2006-03-31 |
Chris Lattner | Pull operand asm string into base class, shrinkifying... |
commit | commitdiff | tree | snapshot |
2006-03-31 |
Evan Cheng | TargetData.cpp::getTypeInfo() was returning alignment... |
commit | commitdiff | tree | snapshot |
2006-03-31 |
Chris Lattner | Delete identity shuffles, implementing CodeGen/Generic... |
commit | commitdiff | tree | snapshot |
2006-03-31 |
Chris Lattner | An identity shuffle shouldn't generate any permute... |
commit | commitdiff | tree | snapshot |
2006-03-31 |
Chris Lattner | Fix 80 column violations :) |
commit | commitdiff | tree | snapshot |
2006-03-31 |
Evan Cheng | Use a X86 target specific node X86ISD::PINSRW instead... |
commit | commitdiff | tree | snapshot |
2006-03-31 |
Chris Lattner | regenerate |
commit | commitdiff | tree | snapshot |
2006-03-31 |
Chris Lattner | Generalize the previous binary operator support and... |
commit | commitdiff | tree | snapshot |
2006-03-31 |
Chris Lattner | new testcase |
commit | commitdiff | tree | snapshot |
2006-03-31 |
Evan Cheng | Added support for SSE3 horizontal ops: haddp{s|d} and... |
commit | commitdiff | tree | snapshot |
2006-03-31 |
Evan Cheng | Added haddp{s|d} and hsubp{s|d} intrinsics. |
commit | commitdiff | tree | snapshot |
2006-03-31 |
Chris Lattner | fix a pasto |
commit | commitdiff | tree | snapshot |
2006-03-31 |
Chris Lattner | Add vperm support for all datatypes |
commit | commitdiff | tree | snapshot |
2006-03-31 |
Chris Lattner | Rearrange code a bit |
commit | commitdiff | tree | snapshot |
2006-03-31 |
Chris Lattner | Add, sub and shuffle are legal for all vector types |
commit | commitdiff | tree | snapshot |
2006-03-31 |
Evan Cheng | Add support to use pextrw and pinsrw to extract and... |
commit | commitdiff | tree | snapshot |
2006-03-31 |
Evan Cheng | Add vector_extract and vector_insert nodes. |
commit | commitdiff | tree | snapshot |
2006-03-31 |
Chris Lattner | add a note |
commit | commitdiff | tree | snapshot |
2006-03-31 |
Chris Lattner | constant fold extractelement with undef operands. |
commit | commitdiff | tree | snapshot |
2006-03-31 |
Chris Lattner | extractelement(undef,x) -> undef |
commit | commitdiff | tree | snapshot |
2006-03-31 |
Chris Lattner | Do not endian swap split vector loads. This fixes... |
commit | commitdiff | tree | snapshot |
2006-03-31 |
Chris Lattner | Do not endian swap the operands to a store if the opera... |
commit | commitdiff | tree | snapshot |
2006-03-31 |
Chris Lattner | Remove dead *extloads. This allows us to codegen vecto... |
commit | commitdiff | tree | snapshot |
2006-03-31 |
Chris Lattner | Delete dead loads in the dag. This allows us to compile |
commit | commitdiff | tree | snapshot |
2006-03-31 |
Chris Lattner | Make this test harder, always check generic x86 and... |
commit | commitdiff | tree | snapshot |
next |