| Commit message (Expand) | Author | Age |
* | [LPM] Switch LICM to actively use LCSSA in addition to preserving it. | Chandler Carruth | 2014-02-11 |
* | LoopVectorizer: Keep track of conditional store basic blocks | Arnold Schwaighofer | 2014-02-08 |
* | [Constant Hoisting] Fix insertion point for constant materialization. | Juergen Ributzka | 2014-02-08 |
* | A memcpy out of an fresh alloca is a no-op, delete it. Patch by Patrick Walton! | Nick Lewycky | 2014-02-06 |
* | Set default of inlinecold-threshold to 225. | Manman Ren | 2014-02-06 |
* | Inliner uses a smaller inline threshold for callees with cold attribute. | Manman Ren | 2014-02-05 |
* | SimplifyLibCalls: Push TLI through the exp2->ldexp transform. | Benjamin Kramer | 2014-02-04 |
* | OS X: the correct function is __sincospif_stret, not __sincospi_stretf | Tim Northover | 2014-02-04 |
* | Add strchr(p, 0) -> p + strlen(p) to SimplifyLibCalls | Kai Nacke | 2014-02-04 |
* | inalloca: Don't remove dead arguments in the presence of inalloca args | Reid Kleckner | 2014-02-03 |
* | Lower llvm.expect intrinsic correctly for i1 | Duncan P. N. Exon Smith | 2014-02-02 |
* | LoopVectorizer: Enable unrolling of conditional stores and the load/store | Arnold Schwaighofer | 2014-02-02 |
* | ARMTTI: We don't have 16 allocatable scalar registers | Arnold Schwaighofer | 2014-02-01 |
* | [LPM] Apply a really big hammer to fix PR18688 by recursively reforming | Chandler Carruth | 2014-02-01 |
* | [inliner] Skip debug intrinsics even earlier in computing the inline | Chandler Carruth | 2014-02-01 |
* | Revert "[SLPV] Recognize vectorizable intrinsics during SLP vectorization ..." | Reid Kleckner | 2014-02-01 |
* | [SLPV] Recognize vectorizable intrinsics during SLP vectorization and | Chandler Carruth | 2014-01-31 |
* | [vectorizer] Tweak the way we do small loop runtime unrolling in the | Chandler Carruth | 2014-01-31 |
* | Allow speculating llvm.sqrt, fma and fmuladd | Matt Arsenault | 2014-01-31 |
* | LoopVectorizer: Add a test case for unrolling of small loops that need a runtime | Arnold Schwaighofer | 2014-01-29 |
* | [LPM] Fix PR18643, another scary place where loop transforms failed to | Chandler Carruth | 2014-01-29 |
* | [LPM] Fix PR18642, a pretty nasty bug in IndVars that "never mattered" | Chandler Carruth | 2014-01-29 |
* | Fix pr14893. | Rafael Espindola | 2014-01-28 |
* | [vectorizer] Completely disable the block frequency guidance of the loop | Chandler Carruth | 2014-01-28 |
* | Update optimization passes to handle inalloca arguments | Reid Kleckner | 2014-01-28 |
* | LoopVectorize: Support conditional stores by scalarizing | Arnold Schwaighofer | 2014-01-28 |
* | PGO branch weight: keep halving the weights until they can fit into | Manman Ren | 2014-01-27 |
* | [vectorize] Initial version of respecting PGO in the vectorizer: treat | Chandler Carruth | 2014-01-27 |
* | ConstantHoisting: We can't insert instructions directly in front of a PHI node. | Benjamin Kramer | 2014-01-27 |
* | [vectorizer] Add an override for the target instruction cost and use it | Chandler Carruth | 2014-01-27 |
* | [vectorizer] Teach the loop vectorizer's unroller to only unroll by | Chandler Carruth | 2014-01-27 |
* | [LPM] Make LCSSA a utility with a FunctionPass that applies it to all | Chandler Carruth | 2014-01-25 |
* | InstCombine: Don't try to use aggregate elements of ConstantExprs. | Benjamin Kramer | 2014-01-24 |
* | Fix known typos | Alp Toker | 2014-01-24 |
* | InstSimplify: Make shift, select and GEP simplifications vector-aware. | Benjamin Kramer | 2014-01-24 |
* | Note the PR number. | Rafael Espindola | 2014-01-23 |
* | Remove tail marker when changing an argument to an alloca. | Rafael Espindola | 2014-01-23 |
* | [LPM] Make LoopSimplify no longer a LoopPass and instead both a utility | Chandler Carruth | 2014-01-23 |
* | Add CHECK-LABELs | Matt Arsenault | 2014-01-22 |
* | Handle an addrspacecast case in memcpyopt | Matt Arsenault | 2014-01-22 |
* | Fix all the remaining lost-fast-math-flags bugs I've been able to find. The ... | Owen Anderson | 2014-01-20 |
* | InstCombine: Modernize a bunch of cast combines. | Benjamin Kramer | 2014-01-19 |
* | InstCombine: Replace a hand-rolled version of isKnownToBeAPowerOfTwo with the... | Benjamin Kramer | 2014-01-19 |
* | InstCombine: Teach most integer add/sub/mul/div combines how to deal with vec... | Benjamin Kramer | 2014-01-19 |
* | InstCombine: Refactor fmul/fdiv combines to handle vectors. | Benjamin Kramer | 2014-01-19 |
* | Fix a really nasty SROA bug with how we handled out-of-bounds memcpy | Chandler Carruth | 2014-01-19 |
* | LoopVectorizer: A reduction that has multiple uses of the reduction value is not | Arnold Schwaighofer | 2014-01-19 |
* | Don't refuse to transform constexpr(call(arg, ...)) to call(constexpr(arg), .... | Nick Lewycky | 2014-01-18 |
* | InstCombine: Make the (fmul X, -1.0) -> (fsub -0.0, X) transform handle vecto... | Benjamin Kramer | 2014-01-18 |
* | Fix more instances of dropped fast math flags when optimizing FADD instructio... | Owen Anderson | 2014-01-18 |