- 30ea478 Have the AutoRegen.sh script prompt the user for the LLVM src and obj by Reid Spencer · 18 years ago
- 746047a Make it possible to default the llvm_src and llvm_obj variables based on by Reid Spencer · 18 years ago
- 44c933e add a bunch of stuff, pieces still missing by Chris Lattner · 18 years ago
- 8b57ff7 Add a warning. by Chris Lattner · 18 years ago
- 6a58e0e Add a warning by Chris Lattner · 18 years ago
- cea2aa7 Use vmladduhm to do v8i16 multiplies which is faster and simpler than doing by Chris Lattner · 18 years ago
- 19a8152 Implement v16i8 multiply with this code: by Chris Lattner · 18 years ago
- a637e58 Add tests for v8i16 and v16i8 by Chris Lattner · 18 years ago
- 4980467 Correct comments by Evan Cheng · 18 years ago
- 72dd9bd Lower v8i16 multiply into this code: by Chris Lattner · 18 years ago
- e7c768e Custom lower v4i32 multiplies into a cute sequence, instead of having legalize by Chris Lattner · 18 years ago
- 843ecd6 new testcase by Chris Lattner · 18 years ago
- 74e955d Another entry by Evan Cheng · 18 years ago
- 309db81 Fix a build failure on Vladimir's tester. by Chris Lattner · 18 years ago
- 7fa094a Another entry. by Evan Cheng · 18 years ago
- cdfc3c8 Use movss to insert_vector_elt(v, s, 0). by Evan Cheng · 18 years ago
- fd6bdf0 Turn x86 unaligned load/store intrinsics into aligned load/store instructions by Chris Lattner · 18 years ago
- 80edfb3 Fix handling of calls in functions that use vectors. This fixes a crash on by Chris Lattner · 18 years ago
- 5edb8d2 Use two pinsrw to insert an element into v4i32 / v4f32 vector. by Evan Cheng · 18 years ago
- 22fcbb1 remove done item by Chris Lattner · 18 years ago
- f9568d8 Don't diddle VRSAVE if no registers need to be added/removed from it. This by Chris Lattner · 18 years ago
- d5c654b New testcase, shouldn't touch vrsave by Chris Lattner · 18 years ago
- 48d7c06 Add a MachineInstr::eraseFromParent convenience method. by Chris Lattner · 18 years ago
- 6b56091 Add some convenience methods. by Chris Lattner · 18 years ago
- 23b7200 Encoding bug by Evan Cheng · 18 years ago
- 402504b Vectors that are known live-in and live-out are clearly already marked in by Chris Lattner · 18 years ago
- 939274f Prefer to allocate V2-V5 before V0,V1. This lets us generate code like this: by Chris Lattner · 18 years ago
- 369503f Move some knowledge about registers out of the code emitter into the register info. by Chris Lattner · 18 years ago
- f7d2372 Use a small table instead of macros to do this conversion. by Chris Lattner · 18 years ago
- c575ca2 Implement v8i16, v16i8 splat using unpckl + pshufd. by Evan Cheng · 18 years ago
- b2be403 implement returns of a vector, testcase here: CodeGen/X86/vec_return.ll by Chris Lattner · 18 years ago
- bddea05 New testcase by Chris Lattner · 18 years ago
- 8d5a894 Codegen insertelement with constant insertion points as scalar_to_vector by Chris Lattner · 18 years ago
- dbce85d Make sure to check splats of every constant we can, handle splat(31) by by Chris Lattner · 18 years ago
- 51c9c43 Incorrect foldMemoryOperand entries by Evan Cheng · 18 years ago
- 083248e Errors in patterns preventing load folding by Evan Cheng · 18 years ago
- 3c280bf Add checks for __OpenBSD__. by Jeff Cohen · 18 years ago
- bdd558c Teach the ppc backend to use rol and vsldoi to generate splatted constants. by Chris Lattner · 18 years ago
- 2f70c8c Some more cases that can be generated with two instructions by Chris Lattner · 18 years ago
- 966083f add a note by Chris Lattner · 18 years ago
- 5001ea1 FP SETOLT, SETOLT, SETUGE, SETUGT conditions were implemented incorrectly by Evan Cheng · 18 years ago
- 6876e66 Make some code more general, adding support for constant formation of several by Chris Lattner · 18 years ago
- 2c4e9c8 New testcases by Chris Lattner · 18 years ago
- c408382 Learn how to make odd splatted constants in range [17,29]. This implements by Chris Lattner · 18 years ago
- 33f24a9 new testcase by Chris Lattner · 18 years ago
- 4a998b9 Pull some code out into a helper function. by Chris Lattner · 18 years ago
- 6e5316f New testcase by Chris Lattner · 18 years ago
- 5913810 Implement a TODO: for any shuffle that can be viewed as a v4[if]32 shuffle, by Chris Lattner · 18 years ago
- c3b5ad9 new testcase, these shuffles can be implemented with discrete instructions, by Chris Lattner · 18 years ago
- cffeb86 Regenerate with adjusted costs by Chris Lattner · 18 years ago
- 8c3e847 Encode a cost of zero as a cost of 1. by Chris Lattner · 18 years ago
- 586d6a8 Regenerate with correct offset by Chris Lattner · 18 years ago
- 195d8ad Really, I can count! by Chris Lattner · 18 years ago
- c74e710 Increase the opcodes by one each to disambiguate COPY from VMRGHW. by Chris Lattner · 18 years ago
- cf1f644 assign stable opcodes to the various altivec ops. by Chris Lattner · 18 years ago
- 2f97623 PPCPerfectShuffle.h is autogenerated, don't include it in the LOC counts. by Chris Lattner · 18 years ago
- 6703461 Check in a table, generated by llvm-PerfectShuffle, of optimal shuffles by Chris Lattner · 18 years ago
- 4ad53bd Rename BuildShuffleTable -> PerfectShuffle by Chris Lattner · 18 years ago
- 3033d4d rename the table by Chris Lattner · 18 years ago
- 27e98aa Initial checking of a perfect shuffle generation program for 4-element by Chris Lattner · 18 years ago
- f0fc7fe Better way to splat v2f64 by Evan Cheng · 18 years ago
- 06aef15 movduprm, movshduprm bugs by Evan Cheng · 18 years ago
- d8e8223 Encoding bugs by Evan Cheng · 18 years ago
- 800f12d Can't fold loads into alias vector SSE ops used for scalar operation. The load by Evan Cheng · 18 years ago
- f3f69de Implement a TODO: have the legalizer canonicalize a bunch of operations to by Chris Lattner · 18 years ago
- 2efce0a Add support for promoting stores from one legal type to another, allowing us by Chris Lattner · 18 years ago
- b17f167 Make the BUILD_VECTOR lowering code much more aggressive w.r.t constant vectors. by Chris Lattner · 18 years ago
- 7f6cc0c Fix a bug in the 'shuffle(undef,x,mask) -> shuffle(x, undef,mask')' xform by Chris Lattner · 18 years ago
- 706126d Canonicalize shuffle(undef,x,mask) -> shuffle(x, undef,mask'). by Chris Lattner · 18 years ago
- 730b456 Fix a crash when faced with a shuffle vector that has an undef in its mask. by Chris Lattner · 18 years ago
- 6e94af7 Add patterns for matching vnots with bit converted inputs. Most of these will by Chris Lattner · 18 years ago
- 1fcee4e Add a new vnot_conv predicate for matching vnot's where the allones vector is by Chris Lattner · 18 years ago
- 547a16f Make these predicates return true for bit_convert(buildvector)'s as well as by Chris Lattner · 18 years ago
- 60d3fa2 More encoding bugs by Evan Cheng · 18 years ago
- 1af1898 pslldrm, psrawrm, etc. encoding bug by Evan Cheng · 18 years ago
- 7076e2d hsubp{s|d} encoding bug by Evan Cheng · 18 years ago
- 57ebe9f Silly bug by Evan Cheng · 18 years ago
- 39fc145 Do not use movs{h|l}dup for a shuffle with a single non-undef node. by Evan Cheng · 18 years ago
- efb4735 significant cleanups to code that uses insert/extractelt heavily. This builds by Chris Lattner · 18 years ago
- 407428e Added SSE (and other) entries to foldMemoryOperand(). by Evan Cheng · 18 years ago
- 9ab1ac5 Some clean up by Evan Cheng · 18 years ago
- b097aa9 Allow undef in a shuffle mask by Chris Lattner · 18 years ago
- f95670f Move these ctors out of line by Chris Lattner · 18 years ago
- 6a56ed4 These instructions always return a packed vector. Improve the class definitions to expose this fact. by Chris Lattner · 18 years ago
- d953947 Last few SSE3 intrinsics. by Evan Cheng · 18 years ago
- de6df88 Teach scalarrepl to promote unions of vectors and floats, producing by Chris Lattner · 18 years ago
- d8f5bd3 New testcase, checking to see we can turn this code: by Chris Lattner · 18 years ago
- 0947429 Adding back vector instructions to keep in mainline. by Tanya Lattner · 18 years ago
- 36ba6d4 Removing unimplemented vector instructions from language referrence. by Tanya Lattner · 18 years ago
- be5e85e Correct the Superclasses list for GlobalVariable and Function to indicate by Reid Spencer · 18 years ago
- 43fdfdc Allow an option, -with-externals, to be specified that provides the location by Reid Spencer · 18 years ago
- f3e1b1d Misc. SSE2 intrinsics: clflush, lfench, mfence by Evan Cheng · 18 years ago
- d9245ca We were not adjusting the frame size to ensure proper alignment when alloca / by Evan Cheng · 18 years ago
- 4f51d85 New entry by Evan Cheng · 18 years ago
- e25fdaf Don't print out the install command for Intrinsics.gen unless VERBOSE mode. by Reid Spencer · 18 years ago
- 3824e50 Make this assertion better by Chris Lattner · 18 years ago
- 619b2d2 Force a specific config, because this test fails in certain configs otherwise. by Chris Lattner · 18 years ago
- 1a635d6 Move the rest of the PPCTargetLowering::LowerOperation cases out into by Chris Lattner · 18 years ago
- f1b4708 Pull the VECTOR_SHUFFLE and BUILD_VECTOR lowering code out into separate by Chris Lattner · 18 years ago
- 0fa07f9 Implement value #'ing for vector operations, implementing by Chris Lattner · 18 years ago