----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: http://reviews.gem5.org/r/693/#review4513 -----------------------------------------------------------
Ship it! It seems like the old code was being very cautious and limiting the number of loads that could be sent to the lsq to the current free - the number dispatched. This change seems to actually check how many memory references were dispatched. It makes sense. Anyone have any thoughts about committing it? - Ali Saidi On May 13, 2011, 2:30 p.m., Yasuko Watanabe wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > http://reviews.gem5.org/r/693/ > ----------------------------------------------------------- > > (Updated May 13, 2011, 2:30 p.m.) > > > Review request for Default, Ali Saidi, Gabe Black, Nathan Binkert, and Steve > Reinhardt. > > > Repository: gem5 > > > Description > ------- > > O3: Fixed an LSQ full check condition at rename. > > > Diffs > ----- > > src/cpu/o3/rename.hh 54a65799e4c1 > src/cpu/o3/rename_impl.hh 54a65799e4c1 > > Diff: http://reviews.gem5.org/r/693/diff/ > > > Testing > ------- > > > Thanks, > > Yasuko Watanabe > > _______________________________________________ gem5-dev mailing list [email protected] http://m5sim.org/mailman/listinfo/gem5-dev
