在 2026-06-06六的 11:09 -0500，Andrew Jones写道： > On Sat, Jun 06, 2026 at 07:21:32AM -0300, Daniel Henrique Barboza > wrote: > ... > > > I think we can just make RVSP a profile, and rvsp-ref CPU a BARE > > > CPU > > > based on that profile; then set the profile's min SATP to Sv48, > > > and the > > > BARE CPU's max SATP to Sv57. > > > > > > In this case if any other CPU complying to RVSP is added, it can > > > just > > > reuse the profile. > > > > Having riscv-server-ref being a profile CPU seems compelling indeed > > ... > > new CPUs could instance it with .profile = &rvserver-ref to have > > everything > > already sorted out, we would also have a flag to enable the profile > > for > > existing cpus, e.g. -cpu rv64,rvserver-ref=on. > > > > Drew, any thoughts? > > As long as the concept of 'profile' is internal to QEMU and we > comment its > [ab]use explaining why the riscv-server-ref is modeled that way, then > that > sounds good to me. If the concept is exposed to users then I fear > we'd > confuse things by inventing profiles that don't exist in the profiles > spec.
Well I think the problem here is that RVI doesn't restrict the definition of required processor feature sets to profiles spec. The best solution here is to make them create a proper profile describing the ISA extension requirement of RV Server Platform spec, instead of open-coding it in RVSP spec; otherwise they're effectively creating a profile, even if it's not in the profiles spec. Thanks, Icenowy > > Thanks, > drew
