| Commit message (Expand) | Author | Age | Files | Lines |
* | merge a test into store.ll | Chris Lattner | 2009-11-02 | 2 | -37/+45 |
* | convert to filecheck | Chris Lattner | 2009-11-02 | 1 | -2/+6 |
* | Add missing end-tag. | Bill Wendling | 2009-11-02 | 1 | -1/+1 |
* | Some formatting changes. | Bill Wendling | 2009-11-02 | 1 | -57/+54 |
* | Handle splats of undefs properly. This includes the testcase for PR5364 as well. | Anton Korobeynikov | 2009-11-02 | 2 | -0/+23 |
* | Do not infer the target type for COPY_TO_REGCLASS from dest regclass, this wo... | Anton Korobeynikov | 2009-11-02 | 6 | -36/+39 |
* | 64-bit FP loads & stores operate on both NEON and VFP pipelines. | Anton Korobeynikov | 2009-11-02 | 2 | -0/+43 |
* | Use NEON reg-reg moves, where profitable. This reduces "domain-cross" stalls,... | Anton Korobeynikov | 2009-11-02 | 8 | -31/+97 |
* | Add PseudoSourceValue::mayAlias. It returns true if the object can ever alias... | Evan Cheng | 2009-11-01 | 2 | -0/+21 |
* | Line this up as well. | Nick Lewycky | 2009-11-01 | 1 | -1/+1 |
* | Fix whitespace. | Nick Lewycky | 2009-11-01 | 1 | -2/+2 |
* | Fix a couple more places where we are creating ld / st instructions without m... | Evan Cheng | 2009-11-01 | 2 | -4/+36 |
* | Make use of imm12 version of Thumb2 ldr / str instructions more aggressively. | Evan Cheng | 2009-11-01 | 2 | -6/+91 |
* | fix two strange things in the default passmgr: | Chris Lattner | 2009-11-01 | 1 | -2/+8 |
* | merge phi-merge.ll into phi.ll | Chris Lattner | 2009-11-01 | 2 | -31/+33 |
* | when merging two loads, make sure to take the min of their alignment, | Chris Lattner | 2009-11-01 | 2 | -1/+27 |
* | split load sinking out to its own function, like gep sinking. | Chris Lattner | 2009-11-01 | 1 | -66/+101 |
* | fix a bug noticed by inspection: when instcombine sinks loads through | Chris Lattner | 2009-11-01 | 2 | -4/+45 |
* | IPSCCP apparently is not a superset of IPCP, this is bad, | Chris Lattner | 2009-11-01 | 1 | -0/+1 |
* | convert to filecheck. | Chris Lattner | 2009-11-01 | 1 | -20/+49 |
* | Improve the other instance of the comment. | Duncan Sands | 2009-11-01 | 1 | -1/+1 |
* | Add a missing closing parenthesis, and tweak to fit in 80 | Duncan Sands | 2009-11-01 | 1 | -2/+2 |
* | only run GlobalDCE at -O3 and run it late instead of early. | Chris Lattner | 2009-11-01 | 1 | -4/+9 |
* | cleanups, switch GlobalDCE to SmallPtrSet instead of std::set | Chris Lattner | 2009-11-01 | 2 | -16/+10 |
* | We currently only run ipsccp at LTO time, which is silly. It subsumes | Chris Lattner | 2009-11-01 | 2 | -2/+20 |
* | remove a bunch of locking from LLVMContextImpl. Since only one thread | Chris Lattner | 2009-11-01 | 10 | -89/+10 |
* | improve comment. | Chris Lattner | 2009-11-01 | 1 | -1/+1 |
* | add a comment about why we don't allow inlining indbr. | Chris Lattner | 2009-11-01 | 1 | -0/+5 |
* | Fix tests. | Evan Cheng | 2009-11-01 | 2 | -4/+4 |
* | the verifier shouldn't modify the IR. | Chris Lattner | 2009-11-01 | 5 | -12/+20 |
* | Reverting 85714, 85715, 85716, which are breaking the build | Douglas Gregor | 2009-11-01 | 8 | -15/+3 |
* | Add a function to Passes.h to allow clients to create instances | Dan Gohman | 2009-11-01 | 2 | -0/+9 |
* | Don't #include Pass.h from CallGraph.h. | Dan Gohman | 2009-11-01 | 2 | -1/+2 |
* | Remove the #include of Pass.h from PassManager.h. This breaks a significant | Dan Gohman | 2009-11-01 | 4 | -2/+4 |
* | teach ipsccp and ipconstprop that a blockaddress doesn't 'take the address' o... | Chris Lattner | 2009-11-01 | 2 | -2/+7 |
* | change llvm::MergeBlockIntoPredecessor to not merge two blocks BB1->BB2 | Chris Lattner | 2009-11-01 | 1 | -3/+7 |
* | strengthen an assumption: RevectorBlockTo knows that PredBB | Chris Lattner | 2009-11-01 | 1 | -12/+14 |
* | fix an issue where the verifier would reject a function whose entry | Chris Lattner | 2009-11-01 | 3 | -2/+19 |
* | if CostMetrics says to never duplicate some code, don't unswitch a loop. | Chris Lattner | 2009-11-01 | 1 | -1/+2 |
* | constant fold indirectbr(blockaddress(%bb)) -> br label %bb. | Chris Lattner | 2009-11-01 | 1 | -9/+46 |
* | improve x86 codegen support for blockaddress. We now compile | Chris Lattner | 2009-11-01 | 2 | -6/+30 |
* | pull check for return inst out of loop, never inline a callee that contains | Chris Lattner | 2009-11-01 | 1 | -5/+7 |
* | Fix BlockAddress::replaceUsesOfWithOnConstant to correctly | Chris Lattner | 2009-11-01 | 1 | -5/+8 |
* | implement linker support for BlockAddress. | Chris Lattner | 2009-11-01 | 1 | -17/+13 |
* | Revert 85678/85680. The decision is to stay with the current form of | Chris Lattner | 2009-11-01 | 11 | -47/+25 |
* | Use cbz and cbnz instructions. | Evan Cheng | 2009-10-31 | 4 | -11/+96 |
* | vml[as].f32 cause stalls in following advanced SIMD instructions. Avoid using | Jim Grosbach | 2009-10-31 | 4 | -10/+15 |
* | Consolidate test files | Jim Grosbach | 2009-10-31 | 4 | -122/+100 |
* | Change to use FileCheck | Jim Grosbach | 2009-10-31 | 1 | -1/+3 |
* | Make tests more explicit about which instructions are expected. | Jim Grosbach | 2009-10-31 | 1 | -2/+2 |