| Commit message (Expand) | Author | Age | Files | Lines |
|---|
| * | fix some issues Frits noticed, add AliasAnalysis as a dependency | Chris Lattner | 2010-12-27 | 1 | -7/+17 |
| * | BuildLibCalls: Nuke EmitMemCpy, EmitMemMove and EmitMemSet. They are dead and... | Benjamin Kramer | 2010-12-27 | 1 | -49/+5 |
| * | SimplifyLibCalls: Use IRBuilder to simplify code. | Benjamin Kramer | 2010-12-27 | 1 | -67/+48 |
| * | have loop-idiom nuke instructions that feed stores that get removed. | Chris Lattner | 2010-12-27 | 1 | -6/+45 |
| * | implement enough of the memset inference algorithm to recognize and insert | Chris Lattner | 2010-12-26 | 1 | -11/+78 |
| * | start using irbuilder to make mem intrinsics in a few passes. | Chris Lattner | 2010-12-26 | 2 | -107/+35 |
| * | sketch more of this out. | Chris Lattner | 2010-12-26 | 1 | -20/+64 |
| * | move isBytewiseValue out to ValueTracking.h/cpp | Chris Lattner | 2010-12-26 | 1 | -68/+1 |
| * | actually add the file... | Chris Lattner | 2010-12-26 | 1 | -0/+103 |
| * | Start of a pass for recognizing memset and memcpy idioms. | Chris Lattner | 2010-12-26 | 2 | -0/+2 |
| * | Simplify code. | Benjamin Kramer | 2010-12-26 | 1 | -1/+1 |
| * | don't lose TD info | Chris Lattner | 2010-12-25 | 2 | -3/+3 |
| * | switch the inliner alignment enforcement stuff to use the | Chris Lattner | 2010-12-25 | 1 | -27/+8 |
| * | Move getOrEnforceKnownAlignment out of instcombine into Transforms/Utils. | Chris Lattner | 2010-12-25 | 4 | -103/+103 |
| * | Fix a thinko pointed out by Frits van Bommel: looking through global variable... | Benjamin Kramer | 2010-12-24 | 1 | -22/+19 |
| * | MemCpyOpt: Turn memcpys from a constant into a memset if possible. | Benjamin Kramer | 2010-12-24 | 1 | -3/+45 |
| * | When determining if we can fold (x >> C1) << C2, the bits that we need to ver... | Owen Anderson | 2010-12-23 | 1 | -1/+2 |
| * | It is possible for SimplifyCFG to cause PHI nodes to become redundant too lat... | Owen Anderson | 2010-12-23 | 1 | -1/+10 |
| * | Preserve the address space when generating bitcasts for MemTransferInst in Co... | Mon P Wang | 2010-12-23 | 1 | -3/+16 |
| * | Change all self assignments X=X to (void)X, so that we can turn on a | Jeffrey Yasskin | 2010-12-23 | 3 | -6/+3 |
| * | InstCombine: creating selects from -1 and 0 is fine, they combine into a sext... | Benjamin Kramer | 2010-12-22 | 1 | -3/+6 |
| * | Add a generic expansion transform: A op (B op' C) -> (A op B) op' (A op C) | Duncan Sands | 2010-12-22 | 5 | -71/+140 |
| * | Add some statistics, good for understanding how much more powerful | Duncan Sands | 2010-12-22 | 1 | -2/+12 |
| * | Give GVN back the ability to perform simple conditional propagation on condit... | Owen Anderson | 2010-12-21 | 1 | -52/+82 |
| * | Remove dead code. | Owen Anderson | 2010-12-21 | 1 | -9/+0 |
| * | GVN's Expression is not POD-like (it contains a SmallVector). Simplify code w... | Benjamin Kramer | 2010-12-21 | 1 | -13/+3 |
| * | Visit instructions deterministically. Use a FIFO so as to approximately | Duncan Sands | 2010-12-21 | 1 | -11/+21 |
| * | If an instruction simplifies, try again to simplify any uses of it. This is | Duncan Sands | 2010-12-21 | 1 | -4/+32 |
| * | Pull a few more simplifications out of instcombine (there are still | Duncan Sands | 2010-12-21 | 1 | -6/+3 |
| * | Oops, forgot to add the pass itself! | Duncan Sands | 2010-12-20 | 1 | -0/+69 |
| * | Add a new convenience pass for testing InstructionSimplify. Previously | Duncan Sands | 2010-12-20 | 2 | -0/+2 |
| * | Add a check missing from my last commit and avoid a potential overflow situat... | Benjamin Kramer | 2010-12-20 | 1 | -3/+3 |
| * | Reduce indentation. | Benjamin Kramer | 2010-12-20 | 1 | -7/+5 |
| * | Teach InstCombine to merge (icmp ult (X + CA), C1) | (icmp eq X, C2) into (ic... | Benjamin Kramer | 2010-12-20 | 1 | -1/+10 |
| * | fix PR8807 by making transformConstExprCastCall aware of byval arguments. | Chris Lattner | 2010-12-20 | 1 | -2/+15 |
| * | various cleanups for transformConstExprCastCall | Chris Lattner | 2010-12-20 | 1 | -13/+10 |
| * | when eliding a byval copy due to inlining a readonly function, we have | Chris Lattner | 2010-12-20 | 1 | -4/+43 |
| * | pull byval processing out to its own helper function. | Chris Lattner | 2010-12-20 | 1 | -56/+72 |
| * | fix PR8769, a miscompilation by inliner when inlining a function with a byval | Chris Lattner | 2010-12-20 | 1 | -6/+10 |
| * | Avoid dropping the address space when InstCombine optimizes memset | Mon P Wang | 2010-12-20 | 1 | -1/+3 |
| * | fix an oversight caught by Frits! | Chris Lattner | 2010-12-19 | 1 | -3/+4 |
| * | tidy up | Chris Lattner | 2010-12-19 | 1 | -18/+17 |
| * | move a transformation to a more logical place, simplifying it. | Chris Lattner | 2010-12-19 | 2 | -16/+7 |
| * | recognize an unsigned add with overflow idiom into uadd. | Chris Lattner | 2010-12-19 | 1 | -5/+50 |
| * | optimize uadd(x, cst) into a comparison when the normal | Chris Lattner | 2010-12-19 | 1 | -0/+16 |
| * | use IC.ReplaceInstUsesWith instead of a raw RAUW so that uses of | Chris Lattner | 2010-12-19 | 1 | -3/+5 |
| * | generalize the sadd creation code to not require that the | Chris Lattner | 2010-12-19 | 1 | -39/+16 |
| * | fix another miscompile in the llvm.sadd formation logic: it wasn't | Chris Lattner | 2010-12-19 | 1 | -4/+39 |
| * | fix a bug (possibly 8816) in the sadd forming xform: it isn't | Chris Lattner | 2010-12-19 | 1 | -0/+10 |
| * | rework the code added in r122072 to pull it out to its own | Chris Lattner | 2010-12-19 | 1 | -61/+64 |
