If you go smaller than a single request the CPU generates things go pear shaped. So anything < 16 bytes won't end well. I strongly doubt non-power-of-two sizes work either.
Ali On 13.09.2012 17:48, Anthony Gutierrez wrote: >> On Sept. 13, 2012, 2:41 p.m., Andreas Hansson wrote: >> >>> Make it check if not 32 or 64? Would that be reasonable? If not I'm happy to discard the warning. >> Anthony Gutierrez wrote: Is there anything fundamentally limiting the space of block sizes in gem5? Otherwise, I don't see this error having a purpose. > > Sorry, warning, not error. > > - Anthony > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > http://reviews.gem5.org/r/1421/#review3471----------------------------------------------------------- > > On Sept. 13, 2012, 11:37 a.m., Anthony Gutierrez wrote: > >> ----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: http://reviews.gem5.org/r/1421/ [1] ----------------------------------------------------------- (Updated Sept. 13, 2012, 11:37 a.m.) Review request for Default. Description ------- Changeset 9224:a1a9c14e69b9 --------------------------- bus: removed outdated warn regarding 64 B block sizes this warn is outdated as 64 B blocks are very common, and even the default size for some CPU types. E.g., arm_detailed. Diffs ----- src/mem/bus.cc be1c1059438bbf7d181dc95a61ec685c2a52c696 Diff: http://reviews.gem5.org/r/1421/diff/ [2] Testing ------- Thanks, Anthony Gutierrez > > _______________________________________________ > gem5-dev mailing list > [email protected] > http://m5sim.org/mailman/listinfo/gem5-dev Links: ------ [1] http://reviews.gem5.org/r/1421/ [2] http://reviews.gem5.org/r/1421/diff/ _______________________________________________ gem5-dev mailing list [email protected] http://m5sim.org/mailman/listinfo/gem5-dev
