| 2011-08-23 |
Nate Begeman | Add some useful accessors to c++ api that appear to...
|
commit | commitdiff | tree |
| 2011-02-11 |
Nate Begeman | Implement sdiv & udiv for <4 x i16> and <8 x i8> NEON...
|
commit | commitdiff | tree |
| 2010-12-20 |
Nate Begeman | Implement feedback from Bruno on making pblendvb an...
|
commit | commitdiff | tree |
| 2010-12-17 |
Nate Begeman | Add vector versions of some existing scalar transforms...
|
commit | commitdiff | tree |
| 2010-12-17 |
Nate Begeman | Add support for matching psign & plendvb to the x86...
|
commit | commitdiff | tree |
| 2010-12-10 |
Nate Begeman | Add some missing predicates.
|
commit | commitdiff | tree |
| 2010-12-10 |
Nate Begeman | Formalize the notion that AVX and SSE are non-overlapping...
|
commit | commitdiff | tree |
| 2010-12-09 |
Nate Begeman | Add support for AVX to materialize +0.0 when doing...
|
commit | commitdiff | tree |
| 2010-12-03 |
Nate Begeman | Revert this change since it breaks a couple of the...
|
commit | commitdiff | tree |
| 2010-12-03 |
Nate Begeman | Scalar f32/f64 are also subregs of ymm regs
|
commit | commitdiff | tree |
| 2010-12-03 |
Nate Begeman | Remove SSE1-4 disable when AVX is enabled. While this...
|
commit | commitdiff | tree |
| 2010-11-10 |
Nate Begeman | Fix an issue where we tried to turn a v2f32 build_vector...
|
commit | commitdiff | tree |
| 2010-09-23 |
Nate Begeman | Revert r114596, it's breaking a few tests.
|
commit | commitdiff | tree |
| 2010-09-22 |
Nate Begeman | <rdar://problem/8228022> Wvector-conversions warnings...
|
commit | commitdiff | tree |
| 2010-08-13 |
Nate Begeman | Reapply this transformation now that it is passing...
|
commit | commitdiff | tree |
| 2010-08-13 |
Nate Begeman | Move some code from Verifier into SVI::isValidOperands...
|
commit | commitdiff | tree |
| 2010-08-10 |
Nate Begeman | Add test for recent instcombine vector shuffle enhancement
|
commit | commitdiff | tree |
| 2010-08-10 |
Nate Begeman | Add the minimal amount of smarts necessary to instcombine...
|
commit | commitdiff | tree |
| 2010-08-06 |
Nate Begeman | Add tablegen support for vdup_lane
|
commit | commitdiff | tree |
| 2010-08-03 |
Nate Begeman | Add support for getting & setting the FPSCR application...
|
commit | commitdiff | tree |
| 2010-07-29 |
Nate Begeman | Add builtins for ssat/usat, similar to RealView's __ssat...
|
commit | commitdiff | tree |
| 2010-07-29 |
Nate Begeman | Add intrinsics __builtin_arm_qadd & __builtin_arm_qsub...
|
commit | commitdiff | tree |
| 2010-07-28 |
Nate Begeman | Implement a vectorized algorithm for <16 x i8> << ...
|
commit | commitdiff | tree |
| 2010-07-27 |
Nate Begeman | ~40% faster vector shl <4 x i32> on SSE 4.1 Larger...
|
commit | commitdiff | tree |
| 2010-07-27 |
Nate Begeman | Fix a crash in the dag combiner caused by ConstantFoldBIT_CO...
|
commit | commitdiff | tree |
| 2010-07-22 |
Nate Begeman | Make fast isel win64-aware w.r.t. call-clobbered regs
|
commit | commitdiff | tree |
| 2010-07-21 |
Nate Begeman | Fix a couple issues with Win64 ABI
|
commit | commitdiff | tree |
| 2010-06-20 |
Nate Begeman | Add support for returning multiple vectors via sret...
|
commit | commitdiff | tree |
| 2010-06-17 |
Nate Begeman | Modify tablegen to support generating all NEON code...
|
commit | commitdiff | tree |
| 2010-06-15 |
Nate Begeman | Make VC++ happy
|
commit | commitdiff | tree |
| 2010-06-14 |
Nate Begeman | Add the last of the SemaChecking-gen code.
|
commit | commitdiff | tree |
| 2010-06-13 |
Nate Begeman | Add a helping of comments
|
commit | commitdiff | tree |
| 2010-06-12 |
Nate Begeman | Add generic vector support for bitselect & element...
|
commit | commitdiff | tree |
| 2010-06-10 |
Nate Begeman | Add support for polynomial type, for polynomial multiply
|
commit | commitdiff | tree |
| 2010-06-10 |
Nate Begeman | NEON support for _lane ops, and multiplies by scalar.
|
commit | commitdiff | tree |
| 2010-06-09 |
Nate Begeman | Further refine types for operations which take scalars.
|
commit | commitdiff | tree |
| 2010-06-09 |
Nate Begeman | Specialize I-Class instructions better so that we have...
|
commit | commitdiff | tree |
| 2010-06-09 |
Nate Begeman | Handle instructions which need to be #defines for the...
|
commit | commitdiff | tree |
| 2010-06-08 |
Nate Begeman | Fix a valgrind error.
|
commit | commitdiff | tree |
| 2010-06-08 |
Nate Begeman | Refine BuiltinsARM.def types a bit, we should do a...
|
commit | commitdiff | tree |
| 2010-06-08 |
Nate Begeman | ARM NEON:
|
commit | commitdiff | tree |
| 2010-06-07 |
Nate Begeman | clang codegen support
|
commit | commitdiff | tree |
| 2010-06-04 |
Nate Begeman | Handle multi-vector returns and args.
|
commit | commitdiff | tree |
| 2010-06-04 |
Nate Begeman | Additional fixes to BuiltinsARM.def generator, on to...
|
commit | commitdiff | tree |
| 2010-06-04 |
Nate Begeman | Progress on generating BuiltinsARM.def, still some...
|
commit | commitdiff | tree |
| 2010-06-04 |
Nate Begeman | BuiltinsARM.def emitter, still needs a substantial...
|
commit | commitdiff | tree |
| 2010-06-04 |
Nate Begeman | Mangle __builtin_neon_* names appropriately.
|
commit | commitdiff | tree |
| 2010-06-03 |
Nate Begeman | Add some additional capabilities to the neon emitter
|
commit | commitdiff | tree |
| 2010-06-03 |
Nate Begeman | arm_neon.h now makes it through clang and generates...
|
commit | commitdiff | tree |
| 2010-06-02 |
Nate Begeman | arm_neon.h emitter now mostly complete for the purposes...
|
commit | commitdiff | tree |
| 2010-06-02 |
Nate Begeman | Checkpoint; handle 'int' and 'void' correctly
|
commit | commitdiff | tree |
| 2010-06-02 |
Nate Begeman | Emit full function prototypes. Definitions & typedefs...
|
commit | commitdiff | tree |
| 2010-06-02 |
Nate Begeman | Checkpoint arm_neon.h generation with tablegen
|
commit | commitdiff | tree |
| 2010-05-28 |
Nate Begeman | Comment out some code in prep for actual .td file checkpoint.
|
commit | commitdiff | tree |
| 2010-05-28 |
Nate Begeman | Add support to tablegen for auto-generating arm_neon...
|
commit | commitdiff | tree |
| 2010-03-24 |
Nate Begeman | Per chris's request, add some comments.
|
commit | commitdiff | tree |
| 2010-03-24 |
Nate Begeman | BUILD_VECTOR was missing out on some prime opportunities...
|
commit | commitdiff | tree |
| 2010-03-11 |
Nate Begeman | Whoops this already existed.
|
commit | commitdiff | tree |
| 2010-03-11 |
Nate Begeman | Add a handful of additional useful pass manager things...
|
commit | commitdiff | tree |
| 2010-02-12 |
Nate Begeman | Add a missing pattern for movhps so that we get:
|
commit | commitdiff | tree |
| 2010-02-01 |
Nate Begeman | Kill the Mach-O writer, and temporarily make filetype...
|
commit | commitdiff | tree |
| 2010-01-19 |
Nate Begeman | Remove this fixme, looking at the linker source, all...
|
commit | commitdiff | tree |
| 2010-01-17 |
Nate Begeman | Add a note for the macho streamer and remove a used...
|
commit | commitdiff | tree |
| 2010-01-15 |
Nate Begeman | Hook up llc's -filetype=obj to use MCStreamer if an...
|
commit | commitdiff | tree |
| 2009-12-03 |
Nate Begeman | Don't pull vector sext through both hands of a logical...
|
commit | commitdiff | tree |
| 2009-11-07 |
Nate Begeman | x86 vector shuffle cleanup/fixes:
|
commit | commitdiff | tree |
| 2009-11-03 |
Nate Begeman | Add a couple more target nodes
|
commit | commitdiff | tree |
| 2009-11-03 |
Nate Begeman | Declare sin & cos as readonly so they match the code...
|
commit | commitdiff | tree |
| 2009-10-19 |
Nate Begeman | PR 5245 - The imediate size target flag was not set...
|
commit | commitdiff | tree |
| 2009-10-19 |
Nate Begeman | Add support for matching shuffle patterns with palignr.
|
commit | commitdiff | tree |
| 2009-10-12 |
Nate Begeman | More heuristics for Combiner-AA. Still catches all...
|
commit | commitdiff | tree |
| 2009-09-25 |
Nate Begeman | Fix combiner-aa issue with bases which are different...
|
commit | commitdiff | tree |
| 2009-09-16 |
Nate Begeman | Do not try and sink a load whose chain result has more...
|
commit | commitdiff | tree |
| 2009-09-15 |
Nate Begeman | Do not add the SVOffset to the Node CSE ID. The same...
|
commit | commitdiff | tree |
| 2009-09-15 |
Nate Begeman | Better solution for tracking both the original alignment...
|
commit | commitdiff | tree |
| 2009-09-15 |
Nate Begeman | Remove incorrect CSE code from r81813.
|
commit | commitdiff | tree |
| 2009-09-15 |
Nate Begeman | Substantially speed up combiner-aa in the following...
|
commit | commitdiff | tree |
| 2009-09-15 |
Nate Begeman | Teach the legalizer to propagate the original alignment...
|
commit | commitdiff | tree |
| 2009-09-15 |
Nate Begeman | Add an "original alignment" field to load and store...
|
commit | commitdiff | tree |
| 2009-07-30 |
Nate Begeman | Typo
|
commit | commitdiff | tree |
| 2009-06-05 |
Nate Begeman | Adapt the x86 build_vector dagcombine to the current...
|
commit | commitdiff | tree |
| 2009-04-29 |
Nate Begeman | Fix infinite recursion in the C++ code which handles...
|
commit | commitdiff | tree |
| 2009-04-29 |
Nate Begeman | Update comment, replace theoretically impossible check...
|
commit | commitdiff | tree |
| 2009-04-29 |
Nate Begeman | Implement review feedback for vector shuffle work.
|
commit | commitdiff | tree |
| 2009-04-27 |
Nate Begeman | Revert accidental testcase reduction
|
commit | commitdiff | tree |
| 2009-04-27 |
Nate Begeman | 2nd attempt, fixing SSE4.1 issues and implementing...
|
commit | commitdiff | tree |
| 2009-04-24 |
Nate Begeman | PR2957
|
commit | commitdiff | tree |
| 2009-03-19 |
Nate Begeman | Add support to tablegen for naming the nodes themselves...
|
commit | commitdiff | tree |
| 2009-03-11 |
Nate Begeman | Allow cross-process JIT to handle MachineRelocations...
|
commit | commitdiff | tree |
| 2009-03-07 |
Nate Begeman | Finish cross-process JIT work, and clean up previous...
|
commit | commitdiff | tree |
| 2009-03-05 |
Nate Begeman | When allocating stubs, keep track of which Functions...
|
commit | commitdiff | tree |
| 2009-03-04 |
Nate Begeman | Fix a thinko in the JIT where the address of a GV was...
|
commit | commitdiff | tree |
| 2009-03-02 |
Nate Begeman | Fix the calculation for how big the allocated stub...
|
commit | commitdiff | tree |
| 2009-03-01 |
Nate Begeman | Fix a problem with DAGCombine on 64b targets where...
|
commit | commitdiff | tree |
| 2009-02-23 |
Nate Begeman | Make this test use darwin targe triple, to avoid stack...
|
commit | commitdiff | tree |
| 2009-02-23 |
Nate Begeman | Generate better code for v8i16 shuffles on SSE2
|
commit | commitdiff | tree |
| 2009-02-18 |
Nate Begeman | Add support to the JIT for true non-lazy operation...
|
commit | commitdiff | tree |
| 2009-02-12 |
Nate Begeman | Add suppport for ConstantExprs of shufflevectors whose...
|
commit | commitdiff | tree |
| 2009-02-11 |
Nate Begeman | the two non-mask arguments to a shufflevector must...
|
commit | commitdiff | tree |
| 2009-02-04 |
Nate Begeman | Remove now-incorrect test.
|
commit | commitdiff | tree |
| next |
projects / oota-llvm.git / search