On Mon, Sep 07, 2026 at 03:55:43PM +0100, Mark Brown wrote: > On Mon, Sep 07, 2026 at 03:14:34PM +0100, Lorenzo Stoakes (ARM) wrote: > > On Fri, Sep 04, 2026 at 10:07:49PM +0100, Mark Brown wrote: > > > On Fri, Sep 04, 2026 at 09:54:51AM +0100, Lorenzo Stoakes (ARM) wrote: > > > > I think the best thing overall would be to leave the runtime paths as > > > they are and refuse to run with an architecturally invalid setup, that > > > would avoid bloating the fast path. > > > Ack, it'd be nice to not have to worry about that. > > > Presumably we definitely want an oops on a GCS violation in general? > > Yes, the guest is going to want GCS exceptions delivering if they've > enabled them and something triggers them. It might also be relying on > the GCS for stack traces even if nothing triggers any GCS related fault.
OK so the answer seems definitely to be disallowing the invalid config at runtime then I think! :) -- Cheers, Lorenzo
