> On May 1, 2013, 8:23 a.m., Anthony Gutierrez wrote: > > Are they anymore concerns/objections to this? > > Andreas Hansson wrote: > I guess it's mostly Erik that had outstanding concerns...
Oh, sorry, were you waiting for me to say something? The latest revision fixes the one bug I could see. I'm not sure defining localPredictorSize instead of localHistoryBits makes anything better, but it doesn't make anything worse, so no objections there. As far as changing the actual parameter values, you probably know more about what they should be than me. Some of my preliminary sims a while back showed that increasing the sizes of local predictor structures beyond what they currently are doesn't do anything for performance, but it was far from an exhaustive study. Ship it? - Erik ----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: http://reviews.gem5.org/r/1827/#review4312 ----------------------------------------------------------- On April 19, 2013, 8:26 a.m., Anthony Gutierrez wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > http://reviews.gem5.org/r/1827/ > ----------------------------------------------------------- > > (Updated April 19, 2013, 8:26 a.m.) > > > Review request for Default. > > > Description > ------- > > Changeset 9644:decb4c9e987a > --------------------------- > cpu: remove local/globalHistoryBits params from branch pred > > having separate params for the local/globalHistoryBits and the > local/globalPredictorSize can lead to inconsistencies if they > are not carefully set. this patch dervies the number of bits > necessary to index into the local/global predictors based on > their size. > > the value of the localHistoryTableSize for the ARM O3 CPU has been > increased to 1024 from 64, which is more accurate for an A15 based > on some correlation against A15 hardware. > > > Diffs > ----- > > configs/common/O3_ARM_v7a.py 745e42ffcc802622d54e52d21a494eca9c9ee9c6 > src/cpu/pred/BranchPredictor.py 745e42ffcc802622d54e52d21a494eca9c9ee9c6 > src/cpu/pred/tournament.cc 745e42ffcc802622d54e52d21a494eca9c9ee9c6 > > Diff: http://reviews.gem5.org/r/1827/diff/ > > > Testing > ------- > > > Thanks, > > Anthony Gutierrez > > _______________________________________________ gem5-dev mailing list [email protected] http://m5sim.org/mailman/listinfo/gem5-dev
