On Fri, Aug 02, 2024 at 10:00:28AM +0100, Marc Zyngier wrote:
Also, the test predates the generated stuff by some margin.
Yeah, there were still defines in the main kernel source that were being retyped rather than shared previously which made me wonder.
Mark Brown broonie@kernel.org wrote:
I'd certainly be happy to convert, though that does seem a bit invasive for a fix.
Not for a point fix, for sure. And if you do, make sure it is entirely scripted.
When you say "entirely scripted" here I take it you're referring to the list of registers as well, and I guess also to the information about what is enumerated by which ID register values? I'd already been thinking about looking at the latter bit, and possibly also tracking wiring things up to traps (though that's only relevant inside the kernel). I agree that seems sensible, but I do think we can usefully do things in stages - even just replacing the magic numbers with use of the defines would be less error prone. It would be great if we just automatically covered every sysreg we know about in this test without any manual steps.