| Commit message (Expand) | Author | Age | Files | Lines |
| * | Refactor a bunch of includes so that TargetMachine.h doesn't have to include | Owen Anderson | 2006-05-12 | 1 | -0/+1 |
| * | Refactor scheduler code. Move register-reduction list scheduler to a | Evan Cheng | 2006-05-11 | 1 | -907/+20 |
| * | Templatify RegReductionPriorityQueue | Evan Cheng | 2006-05-10 | 1 | -7/+12 |
| * | Add pseudo dependency to force a def&use operand to be scheduled last (unless | Evan Cheng | 2006-05-09 | 1 | -17/+108 |
| * | Fix VC++ compilation error. | Jeff Cohen | 2006-05-05 | 1 | -1/+1 |
| * | Initial support for register pressure aware scheduling. The register reduction | Evan Cheng | 2006-05-04 | 1 | -50/+238 |
| * | Bottom up register pressure reduction work: clean up some hacks and enhanced | Evan Cheng | 2006-05-03 | 1 | -75/+72 |
| * | Dis-favor stores more | Evan Cheng | 2006-05-01 | 1 | -2/+2 |
| * | Bottom up register-pressure reduction scheduler now pushes store operations | Evan Cheng | 2006-05-01 | 1 | -17/+41 |
| * | Didn't mean ScheduleDAGList.cpp to make the last checkin. | Evan Cheng | 2006-05-01 | 1 | -33/+17 |
| * | Remove temp. option -spiller-check-liveout, it didn't cause any failure nor p... | Evan Cheng | 2006-05-01 | 1 | -17/+33 |
| * | Don't advance the hazard recognizer when there are no hazards and no instruct... | Chris Lattner | 2006-03-12 | 1 | -25/+40 |
| * | Chain operands aren't real uses: they don't require the full latency of the | Chris Lattner | 2006-03-12 | 1 | -4/+10 |
| * | As a pending queue data structure to keep track of instructions whose | Chris Lattner | 2006-03-12 | 1 | -36/+62 |
| * | rename priorityqueue -> availablequeue. When a node is scheduled, remember | Chris Lattner | 2006-03-11 | 1 | -34/+37 |
| * | Make CurrCycle a local var instead of an instance var | Chris Lattner | 2006-03-11 | 1 | -19/+20 |
| * | Move some methods around so that BU specific code is together, TD specific code | Chris Lattner | 2006-03-11 | 1 | -236/+245 |
| * | merge preds/chainpreds -> preds set | Chris Lattner | 2006-03-11 | 1 | -93/+61 |
| * | Move some simple-sched-specific instance vars to the simple scheduler. | Chris Lattner | 2006-03-10 | 1 | -1/+1 |
| * | Make EmitNode take a SDNode instead of a NodeInfo* | Chris Lattner | 2006-03-10 | 1 | -8/+3 |
| * | Move the VRBase field from NodeInfo to being a separate, explicit, map. | Chris Lattner | 2006-03-10 | 1 | -2/+3 |
| * | no need to build groups anymore | Chris Lattner | 2006-03-10 | 1 | -2/+0 |
| * | Create SUnits directly from the SelectionDAG. | Chris Lattner | 2006-03-10 | 1 | -87/+87 |
| * | Push PrepareNodeInfo/IdentifyGroups down the inheritance hierarchy | Chris Lattner | 2006-03-10 | 1 | -0/+5 |
| * | Teach the latency scheduler some new tricks. In particular, to break ties, | Chris Lattner | 2006-03-10 | 1 | -9/+156 |
| * | add an aggregate method for reinserting scheduled nodes, add a callback for | Chris Lattner | 2006-03-10 | 1 | -8/+23 |
| * | Fix VC++ build breakage. | Jeff Cohen | 2006-03-10 | 1 | -3/+3 |
| * | remove temporary option | Chris Lattner | 2006-03-09 | 1 | -3/+1 |
| * | yes yes, enabled debug output is bad | Chris Lattner | 2006-03-09 | 1 | -3/+0 |
| * | switch the t-d scheduler to use a really dumb and trivial critical path | Chris Lattner | 2006-03-09 | 1 | -1/+104 |
| * | Pull latency information for target instructions out of the latency tables. :) | Chris Lattner | 2006-03-09 | 1 | -46/+80 |
| * | PriorityQueue is an instance var, use it. | Chris Lattner | 2006-03-09 | 1 | -39/+33 |
| * | add some comments | Chris Lattner | 2006-03-09 | 1 | -8/+13 |
| * | Refactor the priority mechanism one step further: now that it is a separate | Chris Lattner | 2006-03-09 | 1 | -136/+185 |
| * | Split the priority function computation and priority queue management out | Chris Lattner | 2006-03-08 | 1 | -113/+150 |
| * | switch from an explicitly managed list of SUnits to a simple vector of sunits | Chris Lattner | 2006-03-08 | 1 | -35/+28 |
| * | Shrinkify some fields, fit to 80 columns | Chris Lattner | 2006-03-08 | 1 | -11/+11 |
| * | remove "Slot", it is dead | Chris Lattner | 2006-03-08 | 1 | -32/+31 |
| * | Change the interface for getting a target HazardRecognizer to be more clean. | Chris Lattner | 2006-03-08 | 1 | -12/+11 |
| * | Fix some formatting, when looking for hazards, prefer target nodes over | Chris Lattner | 2006-03-07 | 1 | -7/+15 |
| * | update file comment | Chris Lattner | 2006-03-06 | 1 | -3/+8 |
| * | Remove some code that doesn't make sense | Evan Cheng | 2006-03-06 | 1 | -12/+5 |
| * | Remove SUnit::Priority1: it is re-calculated on demand as number of live | Evan Cheng | 2006-03-06 | 1 | -35/+25 |
| * | Hoist the HazardRecognizer out of the ScheduleDAGList.cpp file to where | Chris Lattner | 2006-03-06 | 1 | -108/+22 |
| * | Comment fixes | Chris Lattner | 2006-03-05 | 1 | -2/+2 |
| * | When a hazard recognizer needs noops to be inserted, do so. This represents | Chris Lattner | 2006-03-05 | 1 | -10/+15 |
| * | Implement G5HazardRecognizer as a trivial thing that wants 5 cycles between | Chris Lattner | 2006-03-05 | 1 | -2/+42 |
| * | Add basic hazard recognizer support. noop insertion isn't complete yet though. | Chris Lattner | 2006-03-05 | 1 | -15/+104 |
| * | Split the list scheduler into top-down and bottom-up pieces. The priority | Chris Lattner | 2006-03-05 | 1 | -43/+179 |
| * | Move the available queue to being inside the ListSchedule method, since it | Chris Lattner | 2006-03-05 | 1 | -10/+15 |