> On March 14, 2013, 10:29 a.m., Nilay Vaish wrote: > > src/proto/packet.proto, line 61 > > <http://reviews.gem5.org/r/1768/diff/1/?file=34501#file34501line61> > > > > Just for curiosity, why optional?
I was on the fence regarding this one, but I can see cases where we simply wouldn't store the flags if they are all zero (which is the common case). Similarly, if the trace comes from something that is not a CPU it will most likely not use the flags in the Request. - Andreas ----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: http://reviews.gem5.org/r/1768/#review4114 ----------------------------------------------------------- On March 14, 2013, 7:25 a.m., Andreas Hansson wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > http://reviews.gem5.org/r/1768/ > ----------------------------------------------------------- > > (Updated March 14, 2013, 7:25 a.m.) > > > Review request for Default. > > > Description > ------- > > Changeset 9586:cb51609ddbe8 > --------------------------- > mem: Add optional request flags to the packet trace > > This patch adds an optional flags field to the packet trace to encode > the request flags that contain information about whether the request > is (un)cacheable, instruction fetch, preftech etc. > > > Diffs > ----- > > src/mem/comm_monitor.cc 1a21964b7227 > src/proto/packet.proto 1a21964b7227 > > Diff: http://reviews.gem5.org/r/1768/diff/ > > > Testing > ------- > > All regressions pass > > > Thanks, > > Andreas Hansson > > _______________________________________________ gem5-dev mailing list [email protected] http://m5sim.org/mailman/listinfo/gem5-dev
