2008-07-07 |
Dan Gohman | Pool-allocation for MachineInstrs, MachineBasicBlocks... |
blob | commitdiff | raw |
2008-07-03 |
Evan Cheng | - Remove calls to copyKillDeadInfo which is an N^2... |
blob | commitdiff | raw | diff to current |
2008-06-25 |
Owen Anderson | Remember which MachineOperand we were processing, so... |
blob | commitdiff | raw | diff to current |
2008-06-21 |
Evan Cheng | Undo spill weight tweak. Need to investigate the perfor... |
blob | commitdiff | raw | diff to current |
2008-06-19 |
Owen Anderson | Revert my last patch, which was causing regression... |
blob | commitdiff | raw | diff to current |
2008-06-19 |
Evan Cheng | Minor spiller tweak to unfavor reload into load/store... |
blob | commitdiff | raw | diff to current |
2008-06-19 |
Owen Anderson | Insert empty slots into the instruction numbering in... |
blob | commitdiff | raw | diff to current |
2008-06-17 |
Evan Cheng | Live-through live interval is [mbb start, mbb end+1]. |
blob | commitdiff | raw | diff to current |
2008-06-16 |
Owen Anderson | Remove special case handling of empty MBBs now that... |
blob | commitdiff | raw | diff to current |
2008-06-16 |
Owen Anderson | Re-enable empty block indexing by default, since it... |
blob | commitdiff | raw | diff to current |
2008-06-16 |
Owen Anderson | Make indexing empty basic blocks an option for the... |
blob | commitdiff | raw | diff to current |
2008-06-16 |
Owen Anderson | Assign indices to empty basic blocks. This will be... |
blob | commitdiff | raw | diff to current |
2008-06-06 |
Evan Cheng | Refine stack slot interval weight computation. |
blob | commitdiff | raw | diff to current |
2008-06-05 |
Owen Anderson | Add a helper for constructing new live ranges that... |
blob | commitdiff | raw | diff to current |
2008-06-04 |
Evan Cheng | Add a stack slot coloring pass. Not yet enabled. |
blob | commitdiff | raw | diff to current |
2008-06-02 |
Owen Anderson | Correctly handle removed instructions at the beginning... |
blob | commitdiff | raw | diff to current |
2008-05-29 |
Owen Anderson | Make the renumbering correct in the face of deleted... |
blob | commitdiff | raw | diff to current |
2008-05-29 |
Bill Wendling | Remove <iostream>. |
blob | commitdiff | raw | diff to current |
2008-05-29 |
Owen Anderson | Revert part of my last patch that I didn't intend to... |
blob | commitdiff | raw | diff to current |
2008-05-29 |
Owen Anderson | Renumbering needs to account for instruction slot offse... |
blob | commitdiff | raw | diff to current |
2008-05-28 |
Owen Anderson | Remap VNInfo data as well when doing renumbering. |
blob | commitdiff | raw | diff to current |
2008-05-28 |
Owen Anderson | Factor the numbering computation into a separate method... |
blob | commitdiff | raw | diff to current |
2008-05-23 |
Evan Cheng | Revert 51440 as it breaks a bunch of PIC tests. |
blob | commitdiff | raw | diff to current |
2008-05-22 |
David Greene | When rewriting defs and uses after spilling, don't... |
blob | commitdiff | raw | diff to current |
2008-05-20 |
Evan Cheng | Don't spill dead def. |
blob | commitdiff | raw | diff to current |
2008-05-13 |
Dan Gohman | Clean up the use of static and anonymous namespaces... |
blob | commitdiff | raw | diff to current |
2008-05-06 |
Dan Gohman | Make several variable declarations static. |
blob | commitdiff | raw | diff to current |
2008-04-16 |
Evan Cheng | Empty basic block should have an empty range. |
blob | commitdiff | raw | diff to current |
2008-04-11 |
Evan Cheng | Use of implicit_def is not part of live interval. Creat... |
blob | commitdiff | raw | diff to current |
2008-04-09 |
Evan Cheng | - More aggressively coalescing away copies whose source... |
blob | commitdiff | raw | diff to current |
2008-04-03 |
Evan Cheng | - Treat a live range defined by an implicit_def as... |
blob | commitdiff | raw | diff to current |
2008-04-01 |
Evan Cheng | Re-materialization is for uses only. |
blob | commitdiff | raw | diff to current |
2008-03-31 |
Evan Cheng | It's not safe to fold a load from GV stub or constantpo... |
blob | commitdiff | raw | diff to current |
2008-03-31 |
Evan Cheng | The support for remat of instructions with a register... |
blob | commitdiff | raw | diff to current |
2008-03-15 |
Evan Cheng | Remove isImplicitDef TargetInstrDesc flag. |
blob | commitdiff | raw | diff to current |
2008-03-11 |
Evan Cheng | Transfer physical register spill info when load / store... |
blob | commitdiff | raw | diff to current |
2008-03-11 |
Evan Cheng | When the register allocator runs out of registers,... |
blob | commitdiff | raw | diff to current |
2008-03-05 |
Evan Cheng | Refactor code. Remove duplicated functions that basical... |
blob | commitdiff | raw | diff to current |
2008-02-27 |
Evan Cheng | Spiller now remove unused spill slots. |
blob | commitdiff | raw | diff to current |
2008-02-26 |
Bill Wendling | Rename PrintableName to Name. |
blob | commitdiff | raw | diff to current |
2008-02-26 |
Bill Wendling | Change "Name" to "AsmName" in the target register info... |
blob | commitdiff | raw | diff to current |
2008-02-25 |
Evan Cheng | All remat'ed loads cannot be folded into two-address... |
blob | commitdiff | raw | diff to current |
2008-02-25 |
Evan Cheng | Correctly determine whether a argument load can be... |
blob | commitdiff | raw | diff to current |
2008-02-23 |
Evan Cheng | Rematerialization logic was overly conservative when... |
blob | commitdiff | raw | diff to current |
2008-02-23 |
Evan Cheng | If remating a machine instr with virtual register opera... |
blob | commitdiff | raw | diff to current |
2008-02-23 |
Evan Cheng | Recognize loads of arguments as re-materializable first... |
blob | commitdiff | raw | diff to current |
2008-02-23 |
Evan Cheng | Fix spill weight updating bug. |
blob | commitdiff | raw | diff to current |
2008-02-22 |
Evan Cheng | Enable re-materialization of instructions which have... |
blob | commitdiff | raw | diff to current |
2008-02-21 |
Evan Cheng | Clean up some spilling code using MachineRegisterInfo. |
blob | commitdiff | raw | diff to current |
2008-02-18 |
Roman Levenstein | New helper function getMBBFromIndex() that given an... |
blob | commitdiff | raw | diff to current |
2008-02-15 |
Evan Cheng | - Removing the infamous r2rMap_ and rep() method. Now... |
blob | commitdiff | raw | diff to current |
2008-02-13 |
Evan Cheng | Fix a potential serious problem where kills belonging... |
blob | commitdiff | raw | diff to current |
2008-02-10 |
Dan Gohman | Rename MRegisterInfo to TargetRegisterInfo. |
blob | commitdiff | raw | diff to current |
2008-02-08 |
Evan Cheng | Forgot these files. |
blob | commitdiff | raw | diff to current |
2008-01-24 |
Owen Anderson | Move some functionality for adding flags to MachineInst... |
blob | commitdiff | raw | diff to current |
2008-01-10 |
Evan Cheng | Only remat loads from immutable stack slots. |
blob | commitdiff | raw | diff to current |
2008-01-10 |
Evan Cheng | Simplify some code. |
blob | commitdiff | raw | diff to current |
2008-01-10 |
Owen Anderson | Don't use LiveVariables::VarInfo::DefInst. |
blob | commitdiff | raw | diff to current |
2008-01-07 |
Chris Lattner | rename TargetInstrDescriptor -> TargetInstrDesc. |
blob | commitdiff | raw | diff to current |
2008-01-07 |
Chris Lattner | simplify some code using new predicates |
blob | commitdiff | raw | diff to current |
2008-01-07 |
Chris Lattner | Rename MachineInstr::getInstrDescriptor -> getDesc... |
blob | commitdiff | raw | diff to current |
2008-01-07 |
Owen Anderson | Update CodeGen for MRegisterInfo --> TargetInstrInfo... |
blob | commitdiff | raw | diff to current |
2008-01-06 |
Chris Lattner | rename isLoad -> isSimpleLoad due to evan's desire... |
blob | commitdiff | raw | diff to current |
2008-01-04 |
Bill Wendling | Don't recalculate the loop info and loop dominators... |
blob | commitdiff | raw | diff to current |
2007-12-31 |
Chris Lattner | Rename SSARegMap -> MachineRegisterInfo in keeping... |
blob | commitdiff | raw | diff to current |
2007-12-30 |
Chris Lattner | More cleanups for MachineOperand: |
blob | commitdiff | raw | diff to current |
2007-12-29 |
Chris Lattner | Remove attribution from file headers, per discussion... |
blob | commitdiff | raw | diff to current |
2007-12-12 |
Evan Cheng | Implicit def instructions, e.g. X86::IMPLICIT_DEF_GR32... |
blob | commitdiff | raw | diff to current |
2007-12-11 |
Evan Cheng | Switch over to MachineLoopInfo. |
blob | commitdiff | raw | diff to current |
2007-12-08 |
Bill Wendling | Reverting 44702. It wasn't correct to rename them. |
blob | commitdiff | raw | diff to current |
2007-12-08 |
Bill Wendling | Renaming: |
blob | commitdiff | raw | diff to current |
2007-12-06 |
Evan Cheng | Turning simple splitting on. Start testing new coalesce... |
blob | commitdiff | raw | diff to current |
2007-12-06 |
Evan Cheng | Fix for PR1831: if all defs of an interval are re-mater... |
blob | commitdiff | raw | diff to current |
2007-12-05 |
Evan Cheng | MachineInstr can change. Store indexes instead. |
blob | commitdiff | raw | diff to current |
2007-12-05 |
Evan Cheng | If a split live interval is spilled again, remove the... |
blob | commitdiff | raw | diff to current |
2007-12-05 |
Evan Cheng | Clobber more bugs. |
blob | commitdiff | raw | diff to current |
2007-12-05 |
Evan Cheng | Fix kill info for split intervals. |
blob | commitdiff | raw | diff to current |
2007-12-05 |
Evan Cheng | - Mark last use of a split interval as kill instead... |
blob | commitdiff | raw | diff to current |
2007-12-04 |
Evan Cheng | Discard split intervals made empty due to folding. |
blob | commitdiff | raw | diff to current |
2007-12-03 |
Evan Cheng | Typo |
blob | commitdiff | raw | diff to current |
2007-12-03 |
Evan Cheng | Update kill info for uses of split intervals. |
blob | commitdiff | raw | diff to current |
2007-12-02 |
Evan Cheng | Remove redundant foldMemoryOperand variants and other... |
blob | commitdiff | raw | diff to current |
2007-12-01 |
Evan Cheng | Fix a bug where splitting cause some unnecessary spilling. |
blob | commitdiff | raw | diff to current |
2007-12-01 |
Evan Cheng | Allow some reloads to be folded in multi-use cases... |
blob | commitdiff | raw | diff to current |
2007-11-30 |
Evan Cheng | Do not fold reload into an instruction with multiple... |
blob | commitdiff | raw | diff to current |
2007-11-29 |
Evan Cheng | Do not lose rematerialization info when spilling alread... |
blob | commitdiff | raw | diff to current |
2007-11-29 |
Evan Cheng | Fix a major performance issue with splitting. If there... |
blob | commitdiff | raw | diff to current |
2007-11-29 |
Evan Cheng | Replace the odd kill# hack with something less fragile. |
blob | commitdiff | raw | diff to current |
2007-11-29 |
Evan Cheng | Fixed various live interval splitting bugs / compile... |
blob | commitdiff | raw | diff to current |
2007-11-28 |
Evan Cheng | Recover compile time regression. |
blob | commitdiff | raw | diff to current |
2007-11-17 |
Evan Cheng | Live interval splitting: |
blob | commitdiff | raw | diff to current |
2007-11-15 |
Evan Cheng | Fix a thinko in post-allocation coalescer. |
blob | commitdiff | raw | diff to current |
2007-11-14 |
Evan Cheng | Clean up sub-register implementation by moving subReg... |
blob | commitdiff | raw | diff to current |
2007-11-12 |
Evan Cheng | Refactor some code. |
blob | commitdiff | raw | diff to current |
2007-11-07 |
Evan Cheng | Simplify my (il)logic. |
blob | commitdiff | raw | diff to current |
2007-11-06 |
Evan Cheng | When the allocator rewrite a spill register with new... |
blob | commitdiff | raw | diff to current |
2007-11-06 |
Evan Cheng | Fix a bug where a def use operand isn't being detected... |
blob | commitdiff | raw | diff to current |
2007-11-05 |
Evan Cheng | Fix PR1187. |
blob | commitdiff | raw | diff to current |
2007-11-03 |
Evan Cheng | There are times when the coalescer would not coalesce... |
blob | commitdiff | raw | diff to current |
2007-10-17 |
Evan Cheng | Apply Chris' suggestions. |
blob | commitdiff | raw | diff to current |
next |