On 20.12.2023 16:42, Andrew Cooper wrote: > On 20/12/2023 3:10 pm, Roger Pau Monné wrote: >> On Wed, Dec 20, 2023 at 02:46:43PM +0000, Andrew Cooper wrote: >>> On 20/12/2023 2:22 pm, Roger Pau Monne wrote: >>>> @@ -978,24 +978,24 @@ void amd_check_zenbleed(void) >>>> val & chickenbit ? "chickenbit" : "microcode"); >>>> } >>>> >>>> -static void cf_check zen2_disable_c6(void *arg) >>>> +static void cf_check zen_disable_c6(void *arg) >>> fam17_disable_c6() ? I know Hygon is 0x18 but it's also reasonably well >>> know to be the same uarch. >>> >>> This particular algorithm is good for all Fam17 uarches, irrespective of >>> #1474, even if they happen to be the same set of CPUs in practice. >> Yeah, I was about to use fam17h prefix, but that wouldn't cover Hygon. >> I we are fine with it I can send an adjusted v2 using fam17h prefix. > > I think we're fine calling it fam17. Happy to do that consistently.
I agree. And it clearly cannot be just "zen", to avoid it also wrongly covering Zen3 / Zen4. Jan
