> On 2010-08-23 11:01:56, Ali Saidi wrote: > > src/cpu/simple/atomic.cc, line 692 > > <http://reviews.m5sim.org/r/233/diff/1/?file=2123#file2123line692> > > > > Well, the code needs to delete trace data and set it to null either > > way. Only if the execfaulting flag is set does it need to dump() > > Gabe Black wrote: > It will be later like it would have been with no fault in postExecute. I > added that blob of code is to suppress dumping later if there was a fault, so > everything will work fine if you suppress the suppressor and make it work > like it temporarily work like it did before I added that. By making it dump > here, you're making it do the same thing in two different places depending on > the circumstances which might be confusing later on.
ok. - Ali ----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: http://reviews.m5sim.org/r/233/#review309 ----------------------------------------------------------- On 2010-08-23 09:34:20, Ali Saidi wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > http://reviews.m5sim.org/r/233/ > ----------------------------------------------------------- > > (Updated 2010-08-23 09:34:20) > > > Review request for Default. > > > Summary > ------- > > CPU: Print out traces for faluting instructions when the flag for it > (ExecFaulting) is set > > > Diffs > ----- > > src/cpu/simple/atomic.cc 47d9409b2b7f > src/cpu/simple/timing.cc 47d9409b2b7f > > Diff: http://reviews.m5sim.org/r/233/diff > > > Testing > ------- > > > Thanks, > > Ali > > _______________________________________________ m5-dev mailing list [email protected] http://m5sim.org/mailman/listinfo/m5-dev
