On 30/04/2024 14:12, Conor Dooley wrote:
On Tue, Apr 30, 2024 at 01:58:11PM +0200, Clément Léger wrote:
Yeah, see what you mean. I think we also need to define if we want to expose all the ISA extensions in /proc/cpuinfo (ie no matter the config of the kernel) or not. If so, additional validate() callback would make sense. If we want to keep the full ISA string in /proc/info, then we will need another way of doing so.
If extensions aren't usable, they shouldn't be in /proc/cpuinfo either as there's programs that parse that to figure out what they can use, possibly even only checking a single cpu and using that as gospel. That's why there's that per-hart-isa thing that was added by one of your colleagues last year.
Acked. So indeed, validate() callback for F/V dependent extensions makes sense.
Clément