> On April 8, 2013, 8:12 a.m., Andreas Hansson wrote: > > src/mem/ruby/system/Sequencer.cc, line 568 > > <http://reviews.gem5.org/r/1814/diff/1/?file=34763#file34763line568> > > > > Should the same not apply here?
During cooldown, the CacheRecorder injects MemCmd::FlushReq packets, which do not require responses. Thus, the packet deconstructor *should* delete the request when this delete is called. I just ran a small test to verify, and that assessment appears correct. - Joel ----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: http://reviews.gem5.org/r/1814/#review4215 ----------------------------------------------------------- On April 7, 2013, 8:48 p.m., Joel Hestness wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > http://reviews.gem5.org/r/1814/ > ----------------------------------------------------------- > > (Updated April 7, 2013, 8:48 p.m.) > > > Review request for Default. > > > Description > ------- > > Changeset 9631:910e75681f78 > --------------------------- > Sequencer: Delete packet requests during warmup > > When warming up caches in Ruby, the CacheRecorder sends fetch requests into > Ruby Sequencers with packet types that require responses. Since responses are > never generated for these CacheRecorder requests, the requests are not deleted > in the packet destructor called from the Ruby hit callback. Free the request. > > > Diffs > ----- > > src/mem/ruby/system/Sequencer.cc fa31189e1fb5 > > Diff: http://reviews.gem5.org/r/1814/diff/ > > > Testing > ------- > > Tested with Valgrind to ensure that memory leaks are fixed. > > > Thanks, > > Joel Hestness > > _______________________________________________ gem5-dev mailing list [email protected] http://m5sim.org/mailman/listinfo/gem5-dev
