On Fri, May 09, 2025 at 08:32:42 -0000, ray wang wrote: > Thanks a lot for the detailed review and suggestions.
Please for the next time include more context to your replies. I've already forgotten what I was refering to in most contexts that you are replying below so I'll have hard time following up. In general making review harder is detremental to your series because it will take longer. > > Reviewing this would likely be simpler if addition of the controller > > was split from the addition of the disk. > > > Okay, I will split the patch into two parts as suggested—one for introducing > the controller and another for the disk. > > > > > We ususally defer validation of minor infractions such as this to the > > schema rather than having code for it. > > > I will remove the validation logic and keep only the logic for retrieving the > 'serial' value. [...] > > According to the parser 'serial' seems to be optional. Use of the 's:' > > converter makes it mandatory. You likely need 'S:' if it's optional. > > > The 'serial' field is required—not optional—because the QEMU implementation > mandates its presence. Okay; you then need a validation check that mandates it.
