> On 2010-08-23 07:03:46, Nathan Binkert wrote: > > src/arch/x86/types.hh, line 48 > > <http://reviews.m5sim.org/r/209/diff/1/?file=1973#file1973line48> > > > > Do you really need to include this here, or can you just include it in > > the .cc file? Perhaps because of the template <>, you do need to include > > it here. In general, I'd like to support overloading paramIn and paramOut > > better and have this be the general mechanism for serializing sub-objects.
I have a patch that converts setInst and getInst and the backing storage to ExtMachInst which uses it, but I'm going to try passing the StaticInst to the fault's invoke like Steve suggested to get rid of setInst and getInst entirely. This still seemed useful, though, so I left it in. I think you'd get an error about paramIn and paramOut not being templates if you don't have that include. - Gabe ----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: http://reviews.m5sim.org/r/209/#review287 ----------------------------------------------------------- On 2010-08-22 18:53:37, Gabe Black wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > http://reviews.m5sim.org/r/209/ > ----------------------------------------------------------- > > (Updated 2010-08-22 18:53:37) > > > Review request for Default. > > > Summary > ------- > > X86: Make the x86 ExtMachInst serializable with (UN)SERIALIZE_SCALAR. > > > Diffs > ----- > > src/arch/x86/SConscript 405f840c4ae1 > src/arch/x86/types.hh 405f840c4ae1 > src/arch/x86/types.cc PRE-CREATION > > Diff: http://reviews.m5sim.org/r/209/diff > > > Testing > ------- > > > Thanks, > > Gabe > > _______________________________________________ m5-dev mailing list [email protected] http://m5sim.org/mailman/listinfo/m5-dev
