aboutsummaryrefslogtreecommitdiff
path: root/lib/CodeGen/SelectionDAG/ScheduleDAGRRList.cpp
AgeCommit message (Expand)Author
2007-10-15One mundane change: Change ReplaceAllUsesOfValueWith to *optionally* Chris Lattner
2007-10-12EXTRACT_SUBREG coalescing support. The coalescer now treats EXTRACT_SUBREG likeEvan Cheng
2007-10-05Fix a typo in a comment.Dan Gohman
2007-10-05Chain producing nodes cannot be moved, not chain reading nodes.Evan Cheng
2007-10-05Oops. Didn't mean to leave this in.Evan Cheng
2007-10-05If a node that defines a physical register that is expensive to copy. TheEvan Cheng
2007-09-28If two instructions are both two-address code, favors (schedule closer toEvan Cheng
2007-09-28Remove a poor scheduling heuristic.Evan Cheng
2007-09-28Trim some unneeded fields.Evan Cheng
2007-09-27Avoid inserting a live register more than once.Evan Cheng
2007-09-27Boogs.Evan Cheng
2007-09-27Be smarter about which node to force schedule. Reduce # of duplications + cop...Evan Cheng
2007-09-27Backtracking only when it won't create a cycle.Evan Cheng
2007-09-26- Move getPhysicalRegisterRegClass() from ScheduleDAG to MRegisterInfo.Evan Cheng
2007-09-25Added major new capabilities to scheduler (only BURR for now) to support phys...Evan Cheng
2007-09-19Use struct SDep instead of std::pair for SUnit pred and succ lists. First stepEvan Cheng
2007-09-13Bug fixes.Evan Cheng
2007-08-20Minor cleanups to reduce some spurious differences between differentDan Gohman
2007-07-13Skeleton of post-RA scheduler; doesn't do anything yet.Dale Johannesen
2007-06-29Remove unused variables.David Greene
2007-06-29Remove unnecessary attributions in comments.David Greene
2007-06-29Remove the "special tie breaker" because it resulted in inconsistentDavid Greene
2007-06-22std::set is really really terrible. Switch to SmallPtrSet to reduce compile t...Evan Cheng
2007-04-26Fix a typo in a comment.Dan Gohman
2007-03-14Estimate a cost using the possible number of scratch registers required and useEvan Cheng
2007-03-13Try schedule def + use closer whne Sethi-Ullman numbers are the same.Evan Cheng
2007-02-03switch the sched unit map over to use a DenseMap instead of std::map. ThisChris Lattner
2007-02-01Fit in 80 columnsChris Lattner
2007-01-08Naming consistency.Evan Cheng
2007-01-08Fix for PR1075: bottom-up register-reduction scheduling actually increases re...Evan Cheng
2006-12-07Removing even more <iostream> includes.Bill Wendling
2006-12-01Match TargetInstrInfo changes.Evan Cheng
2006-11-06Remove dead code; added a missing null ptr check.Evan Cheng
2006-11-04Changes to use operand constraints to process two-address instructions.Evan Cheng
2006-11-03silence warningChris Lattner
2006-11-02For PR786:Reid Spencer
2006-11-01Clean up.Evan Cheng
2006-11-01CopyFromReg starts a live range so its use should not be considered a floater.Evan Cheng
2006-10-14Debug tweak.Evan Cheng
2006-08-27s|llvm/Support/Visibility.h|llvm/Support/Compiler.h|Chris Lattner
2006-08-17switch the SUnit pred/succ sets from being std::sets to being smallvectors.Chris Lattner
2006-08-02Final polish on machine pass registries.Jim Laskey
2006-08-011. Change use of "Cache" to "Default".Jim Laskey
2006-08-01Introducing plugable register allocators and instruction schedulers.Jim Laskey
2006-07-21Use an enumeration to eliminate data relocations.Jim Laskey
2006-06-28Use hidden visibility to make symbols in an anonymous namespace getChris Lattner
2006-06-28Shave another 27K off libllvmgcc.dylib with visibility hiddenChris Lattner
2006-05-30Make sure the register pressure reduction schedulers work for non-uniformEvan Cheng
2006-05-25Turn on -sched-commute-nodes by default.Evan Cheng
2006-05-13Fixing 2006-05-01-SchedCausingSpills.ll; some clean upEvan Cheng