----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: http://reviews.m5sim.org/r/909/#review1669 -----------------------------------------------------------
I thought about making this a warn_once, but I don't think it's a good idea since it may warn with one address and actually cause a fault (which was why Lisa added it) for a different address much later. In that case it would just be confusing the situation, not adding useful information. In my opinion we should get rid of it entirely and find a better way to help people discover what's going on with faults in SE, maybe by just changing what's printed when a memory fault happens. - Gabe On 2011-11-18 18:21:32, Ali Saidi wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > http://reviews.m5sim.org/r/909/ > ----------------------------------------------------------- > > (Updated 2011-11-18 18:21:32) > > > Review request for Default, Ali Saidi, Gabe Black, Steve Reinhardt, and > Nathan Binkert. > > > Summary > ------- > > SE: Change very noise warn message to warn_once > > This warn message gets called a lot when executing with O3 and there is > speculation. > > > Diffs > ----- > > src/sim/process.cc e66a566f2cfa > > Diff: http://reviews.m5sim.org/r/909/diff > > > Testing > ------- > > > Thanks, > > Ali > > _______________________________________________ gem5-dev mailing list [email protected] http://m5sim.org/mailman/listinfo/gem5-dev
