aboutsummaryrefslogtreecommitdiff
path: root/lib/CodeGen
AgeCommit message (Expand)Author
2011-11-15Break false dependencies before partial register updates.Jakob Stoklund Olesen
2011-11-15Track register ages more accurately.Jakob Stoklund Olesen
2011-11-14Avoid dereferencing off the beginning of lists.Evan Cheng
2011-11-14At -O0, multiple uses of a virtual registers in the same BB are being markedEvan Cheng
2011-11-14Teach two-address pass to re-schedule two-address instructions (or the killEvan Cheng
2011-11-14Fix early-clobber handling in shrinkToUses.Jakob Stoklund Olesen
2011-11-14It helps to deallocate memory as well as allocate it. =] This actuallyChandler Carruth
2011-11-14Remove an over-eager assert that was firing on one of the ARM regressionChandler Carruth
2011-11-14Begin chipping away at one of the biggest quadratic-ish behaviors inChandler Carruth
2011-11-14Under the hood, MBPI is doing a linear scan of every successor everyChandler Carruth
2011-11-14Reuse the logic in getEdgeProbability within getHotSucc in order toChandler Carruth
2011-11-14Fix an overflow bug in MachineBranchProbabilityInfo. This pass relied onChandler Carruth
2011-11-14Use getVNInfoBefore() when it makes sense.Jakob Stoklund Olesen
2011-11-14Teach machine block placement to cope with unnatural loops. These don'tChandler Carruth
2011-11-13Use kill slots instead of the previous slot in shrinkToUses.Jakob Stoklund Olesen
2011-11-13Cleanup some 80-columns violations and poor formatting. These snuck byChandler Carruth
2011-11-13Terminate all dead defs at the dead slot instead of the 'next' slot.Jakob Stoklund Olesen
2011-11-13Simplify early clobber slots a bit.Jakob Stoklund Olesen
2011-11-13Enhance the assertion mechanisms in place to make it easier to catchChandler Carruth
2011-11-13Rename SlotIndexes to match how they are used.Jakob Stoklund Olesen
2011-11-13Teach MBP to force-merge layout successors for blocks with unanalyzableChandler Carruth
2011-11-13Hoist another gross nested loop into a helper method.Chandler Carruth
2011-11-13Add a missing doxygen comment for a helper method.Chandler Carruth
2011-11-13Hoist a nested loop into its own method.Chandler Carruth
2011-11-13Rewrite #3 of machine block placement. This is based somewhat on theChandler Carruth
2011-11-13Prune more RALinScan. RALinScan was also here!NAKAMURA Takumi
2011-11-13More dead code elimination in VirtRegMap.Jakob Stoklund Olesen
2011-11-13Stop tracking spill slot uses in VirtRegMap.Jakob Stoklund Olesen
2011-11-13Remove dead code and data from VirtRegMap.Jakob Stoklund Olesen
2011-11-13Stop tracking unused registers in VirtRegMap.Jakob Stoklund Olesen
2011-11-13Remove the -color-ss-with-regs option.Jakob Stoklund Olesen
2011-11-13Delete VirtRegRewriter.Jakob Stoklund Olesen
2011-11-13Switch PBQP to VRM's trivial rewriter.Jakob Stoklund Olesen
2011-11-12Delete the old spilling framework from LiveIntervalAnalysis.Jakob Stoklund Olesen
2011-11-12Delete the 'standard' spiller with used the old spilling framework.Jakob Stoklund Olesen
2011-11-12Switch PBQP to the modern InlineSpiller framework.Jakob Stoklund Olesen
2011-11-12Delete the linear scan register allocator.Jakob Stoklund Olesen
2011-11-12The dwarf standard says that the only differences between a out-of-lineRafael Espindola
2011-11-12Don't try to form pre/post-indexed loads/stores until after LegalizeDAG runs....Eli Friedman
2011-11-11Some cleanup and bulletproofing for node replacement in LegalizeDAG. To main...Eli Friedman
2011-11-11Add a custom safepoint method, in order for language implementers to decide w...Nicolas Geoffray
2011-11-11Initialize variable.Eric Christopher
2011-11-11If we have a DIE with an AT_specification use that instead of the normalEric Christopher
2011-11-10Check in getOrCreateSubprogramDIE if a declaration exists and if so outputRafael Espindola
2011-11-10Make types and namespaces take multiple DIEs for the accelerator tablesEric Christopher
2011-11-10Move type handling to make sure we get all created types that aren'tEric Christopher
2011-11-10Rework adding function names to the dwarf accelerator tables, allowEric Christopher
2011-11-10Use a bigger hammer to fix PR11314 by disabling the "forcing two-addressEvan Cheng
2011-11-10Strip old implicit operands after foldMemoryOperand.Jakob Stoklund Olesen
2011-11-09Add check so we don't try to perform an impossible transformation. Fixes issu...Eli Friedman