> On 2012-01-16 08:21:53, Steve Reinhardt wrote: > > I don't know all the pros and cons of doing this initialization in python > > vs. c++... would anyone ever want to change it? If so, it should be left > > in python. I'd also lean toward leaving it in python from a "if it ain't > > broke don't fix it" perspective. What are the benefits of moving it to c++?
This is, as far as I could tell back in the day, just a way of initializing certain registers like a particular MIPS CPU. I *seriously* doubt fiddling with those registers in python would be useful to anyone, and having it there is a significant irregularity compared to the other ISAs. I'd be happy to see setMipsOptions go away. - Gabe ----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: http://reviews.m5sim.org/r/994/#review1953 ----------------------------------------------------------- On 2012-01-16 08:45:46, Deyuan Guo wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > http://reviews.m5sim.org/r/994/ > ----------------------------------------------------------- > > (Updated 2012-01-16 08:45:46) > > > Review request for Default. > > > Summary > ------- > > [--splitting the mips-fs patch--] > We can do the CP0 initialization in configCP(), then make ISA::clear() > function call the configCP() in isa.cc. > After that, the setMipsOptions() function in fs.py will be no longer needed. > > > Diffs > ----- > > configs/example/fs.py f348cf78072c > src/arch/mips/isa.cc f348cf78072c > > Diff: http://reviews.m5sim.org/r/994/diff > > > Testing > ------- > > > Thanks, > > Deyuan > > _______________________________________________ gem5-dev mailing list [email protected] http://m5sim.org/mailman/listinfo/gem5-dev
