This is correct. SMT hasnt been compiled, tested, or anything for FS mode.
In theory, it should work out the box but there is most definitely glue code that usually can only be debugged by trial/error and maybe comparison to a known working model (like SimpleCPU). We're currently working on adding more rigorous regression testing for SE mode SMT as well. On Wed, May 27, 2009 at 12:48 AM, ef <[email protected]> wrote: > Here is a quick answer: > I dont think SMT and FS mode are supported together, their might be patches. > You should wait for someone more knowledgable to comment. > Thanks > > On Tue, May 26, 2009 at 8:52 PM, Alexander Neckar > <[email protected]> wrote: >> >> Hi, >> >> We're simulating an SMT with ALPHA_FS. When we change the smtFetchPolicy >> parameter in o3_cpu.py we don't observe any change whatsoever in simulator >> behavior. Even with a bogus input (like 'smtFetchPolicy = >> param.string('xyz',"smt fetch policy"') we don't see any warnings or errors. >> Any insight would be appreciated. >> >> Thanks, >> Alex Neckar >> >> _______________________________________________ >> m5-users mailing list >> [email protected] >> http://m5sim.org/cgi-bin/mailman/listinfo/m5-users > > > _______________________________________________ > m5-users mailing list > [email protected] > http://m5sim.org/cgi-bin/mailman/listinfo/m5-users > -- - Korey _______________________________________________ m5-users mailing list [email protected] http://m5sim.org/cgi-bin/mailman/listinfo/m5-users
