svn commit: r326383 - head/sys/x86/cpufreq

[Alexey Dokuchaev]

On Thu, Dec 07, 2017 at 01:26:48PM -0500, Jung-uk Kim wrote:
> On 12/07/2017 05:03, Alexey Dokuchaev wrote:
> > On Thu, Nov 30, 2017 at 03:08:49PM -0500, Jung-uk Kim wrote:
> >> ...
> >> Probably.  However, I am just trying to fix my FX-8350 and A10-6800 and
> >> I don't have Zen processors to verify the MSRs are actually working on
> >> those CPUs.
> > 
> > Ah, that's so lovely, thanks Jung-uk; I feel that our support for AMD
> > fam. 15h CPUs is lacking.  E.g. only four P-states are reported for my
> > A8-5550M, while it supports boosted P-states per BKDG, and reading MSRs
> > directly via `sysutils/amdmsrtweaker' reports eight of them (P0 .. P7),
> > with three turbo P-states P0 P1 P2.
> 
> You don't trust your BIOS? ;-)

Well, I kind of do, with a grain of salt.  BIOSes (or shall I say, their
CPU support packages) are often poorly written or being too conservative,
in order to simplify vendors' job making sure they work across all CPUs.

I also like to undervolt them; I'm doing this now with my Pentium M 780
CPU via patched DSDT.  I hope that I could do the same for A8-5550M, and
with having more P-states than currently provided via stock DSDT for a
more fine-grained control.

> Unless SSDT is unavailable or broken, we should use ACPI tables first.

I know that's correct and recommended procedure; in fact, under Ubuntu,
available frequency steps are: 2.10 GHz, 1.90 GHz, 1.70 GHz, 1.40 GHz;
also just four (with driver: acpi-cpufreq).  I'd have to boot into Win7
to see if it's doing more than that.

./danfe