| Commit message (Expand) | Author | Age | Files | Lines |
* | LICM does get dead instructions input to it. Instead of sinking them | Chris Lattner | 2010-08-29 | 1 | -0/+14 |
* | Make IVUsers iterative instead of recursive. | Dan Gohman | 2010-08-29 | 1 | -1/+1 |
* | Make this test less dependent on register allocation choices. | Dan Gohman | 2010-08-29 | 1 | -1/+1 |
* | Use exec. | Dan Gohman | 2010-08-29 | 1 | -1/+1 |
* | Fix lowering of INSERT_VECTOR_ELT in SPU. | Kalle Raiskila | 2010-08-29 | 1 | -0/+8 |
* | Remove NEON vaddl, vaddw, vsubl, and vsubw intrinsics. Instead, use llvm | Bob Wilson | 2010-08-29 | 4 | -80/+132 |
* | merge a bunch of shuffle tests into sse2.ll | Chris Lattner | 2010-08-29 | 9 | -158/+164 |
* | add some nounwind's | Chris Lattner | 2010-08-29 | 7 | -13/+13 |
* | fixme accomplished | Chris Lattner | 2010-08-28 | 1 | -2/+0 |
* | fix the buildvector->insertp[sd] logic to not always create a redundant | Chris Lattner | 2010-08-28 | 3 | -2/+27 |
* | fix the BuildVector -> unpcklps logic to not do pointless shuffles | Chris Lattner | 2010-08-28 | 1 | -0/+25 |
* | Update ocaml test. | Benjamin Kramer | 2010-08-28 | 1 | -6/+0 |
* | remove unions from LLVM IR. They are severely buggy and not | Chris Lattner | 2010-08-28 | 3 | -59/+0 |
* | remove the ABCD and SSI passes. They don't have any clients that | Chris Lattner | 2010-08-28 | 8 | -182/+0 |
* | handle the constant case of vector insertion. For something | Chris Lattner | 2010-08-28 | 1 | -0/+12 |
* | optimize bitcasts from large integers to vector into vector | Chris Lattner | 2010-08-28 | 1 | -0/+31 |
* | Completely disable tail calls when fast-isel is enabled, as fast-isel | Dan Gohman | 2010-08-28 | 1 | -4/+10 |
* | Add a prototype of a new peephole optimizing pass that uses LazyValue info to... | Owen Anderson | 2010-08-27 | 3 | -0/+45 |
* | Change ARM VFP VLDM/VSTM instructions to use addressing mode #4, just like | Bob Wilson | 2010-08-27 | 1 | -1/+1 |
* | tidy up test. | Chris Lattner | 2010-08-27 | 1 | -1/+2 |
* | no really, fix the test. | Chris Lattner | 2010-08-27 | 1 | -1/+1 |
* | fix this test. It's not clear what it's really testing. | Chris Lattner | 2010-08-27 | 1 | -1/+1 |
* | Enhance the shift propagator to handle the case when you have: | Chris Lattner | 2010-08-27 | 1 | -0/+15 |
* | Implement a pretty general logical shift propagation | Chris Lattner | 2010-08-27 | 1 | -4/+17 |
* | merge and filecheckize test | Chris Lattner | 2010-08-27 | 2 | -42/+57 |
* | merge two tests | Chris Lattner | 2010-08-27 | 2 | -10/+12 |
* | teach the truncation optimization that an entire chain of | Chris Lattner | 2010-08-27 | 1 | -4/+21 |
* | get this test passing on linux builders. | Chris Lattner | 2010-08-27 | 1 | -8/+8 |
* | Add an instcombine to clean up a common pattern produced | Chris Lattner | 2010-08-27 | 1 | -0/+32 |
* | Add alignment arguments to all the NEON load/store intrinsics. | Bob Wilson | 2010-08-27 | 22 | -302/+336 |
* | Use LVI to eliminate conditional branches where we've tested a related condit... | Owen Anderson | 2010-08-27 | 2 | -5/+12 |
* | X86: Fix an encoding issue with LOCK_ADD64mr, which could lead to very hard t... | Daniel Dunbar | 2010-08-27 | 1 | -0/+22 |
* | filecheckize | Chris Lattner | 2010-08-26 | 1 | -7/+19 |
* | rename test. | Chris Lattner | 2010-08-26 | 1 | -0/+0 |
* | optimize "integer extraction out of the middle of a vector" as produced | Chris Lattner | 2010-08-26 | 1 | -0/+25 |
* | optimize bitcast(trunc(bitcast(x))) where the result is a float and 'x' | Chris Lattner | 2010-08-26 | 1 | -0/+22 |
* | filecheckize | Chris Lattner | 2010-08-26 | 1 | -6/+7 |
* | rename test | Chris Lattner | 2010-08-26 | 1 | -0/+0 |
* | Make JumpThreading smart enough to properly thread StrSwitch when it's compil... | Owen Anderson | 2010-08-26 | 1 | -0/+162 |
* | Reapply r112091 and r111922, support for metadata linking, with a | Dan Gohman | 2010-08-26 | 2 | -0/+24 |
* | Add a hackaround for PR7993 which is causing failures on x86 builders that la... | Chris Lattner | 2010-08-26 | 2 | -2/+9 |
* | I think enough general codegen bugs are fixed to allow this to work | Chris Lattner | 2010-08-26 | 1 | -4/+2 |
* | implement SplitVecOp_CONCAT_VECTORS, fixing the included testcase with SSE1. | Chris Lattner | 2010-08-26 | 1 | -0/+6 |
* | Make sure this forces the x86 targets | Chris Lattner | 2010-08-26 | 1 | -1/+2 |
* | fix sse1 only codegen in x86-64 mode, which is something we | Chris Lattner | 2010-08-26 | 1 | -0/+7 |
* | Revert r112091, "Remap metadata attached to instructions when remapping | Daniel Dunbar | 2010-08-26 | 2 | -24/+0 |
* | Enable pre-RA virtual frame base register allocation. rdar://8277890 | Jim Grosbach | 2010-08-26 | 1 | -2/+2 |
* | Revert svn 107892 (with changes to work with trunk). It caused a crash if | Bob Wilson | 2010-08-26 | 1 | -0/+14 |
* | temporarily disable this, which started failing on the llvm-i686-linux | Chris Lattner | 2010-08-25 | 1 | -2/+4 |
* | Change handling of illegal vector types to widen when possible instead of | Chris Lattner | 2010-08-25 | 2 | -16/+49 |