Lines Matching full:carried

20 // represent loop carried dependences in the DAG as order edges to the Phi
151 /// A command line option to disable the pruning of loop carried order
154     SwpPruneLoopCarried("pipeliner-prune-loop-carried",
155                         cl::desc("Prune loop carried order dependences."),
835 /// alias of the load on a subsequent iteration, i.e., a loop carried
837 /// but that code doesn't create loop carried dependences.
871           // offset, then mark the dependence as loop carried potentially.  in addLoopCarriedDependences()
892           // the store offset, the mark the dependence as loop carried.  in addLoopCarriedDependences()
930 /// from a PHI to a use, and a loop carried dependence from the use to the
931 /// PHI. The loop carried dependence is represented as an anti dependence
1373   //   - by a phi instruction of the next iteration (loop carried value)
1780 // is loop-carried to the USE in next iteration. This will help pipeliner avoid
1784 // PHI-------Anti-Dep-----> COPY/REG_SEQUENCE (loop-carried)
1803     // Record the loop carried PHIs.  in apply()
1812       // Save the loop carried PHI.  in apply()
2498         // loop-carried output/order dependencies. Empirically, there are also  in schedulePipeline()
2718 /// Return true for an order or output dependence that is loop carried
2719 /// potentially. A dependence is loop carried if the destination defines a value
2739   // Assume ordered loads and stores may have a loop carried dependence.  in isLoopCarriedDep()
2749   // may be loop carried. The following code checks when it can be proved that  in isLoopCarriedDep()
2750   // there is no loop carried dependence.  in isLoopCarriedDep()
2796   // increment value to determine if the accesses may be loop carried.  in isLoopCarriedDep()
3075   // to a loop-carried dependence.  in orderDependence()
3104 /// Return true if the scheduled Phi has a loop carried operand.
3127 /// Return true if the instruction is a definition that is loop carried
3154 /// Return true if all scheduled predecessors are loop-carried output/order