Ryzen public erratas

[Gary Jennejohn]

On Tue, 26 Jun 2018 07:05:22 -0700
Eitan Adler <lists at eitanadler.com> wrote:

> On 19 June 2018 at 02:50, Gary Jennejohn <gljennjohn at gmail.com> wrote:
> > On Mon, 18 Jun 2018 22:44:13 -0700
> > Eitan Adler <lists at eitanadler.com> wrote:
> >  
> >> On 13 June 2018 at 04:16, Eitan Adler <lists at eitanadler.com> wrote:  
> >> > On 13 June 2018 at 03:35, Konstantin Belousov <kostikbel at gmail.com> wrote:  
> >> >> Today I noted that AMD published the public errata document for Ryzens,
> >> >> https://developer.amd.com/wp-content/resources/55449_1.12.pdf
> >> >>
> >> >> Some of the issues listed there looks quite relevant to the potential
> >> >> hangs that some people still experience with the machines.  I wrote
> >> >> a script which should apply the recommended workarounds to the erratas
> >> >> that I find interesting.
> >> >>
> >> >> To run it, kldload cpuctl, then apply the latest firmware update to your
> >> >> CPU, then run the following shell script.  Comments indicate the errata
> >> >> number for the workarounds.
> >> >>
> >> >> Please report the results.  If the script helps, I will code the kernel
> >> >> change to apply the workarounds.
> >> >>
> >> >> #!/bin/sh
> >> >>
> >> >> # Enable workarounds for erratas listed in
> >> >> # https://developer.amd.com/wp-content/resources/55449_1.12.pdf
> >> >>
> >> >> # 1057, 1109
> >> >> sysctl machdep.idle_mwait=0
> >> >> sysctl machdep.idle=hlt  
> >> >
> >> >
> >> > Is this needed if it was previously machdep.idle: acpi ?  
> >>
> >> This might explain why I've never seen the lockup issues mentioned by
> >> other people. What would cause my machine to differ from others?
> >>  
> >
> > I had sysctl machdep.idle_mwait=1 and machdep.idle=acpi before
> > applying the shell script.  I had multiple lockups every week,
> > sometimes multiple lockups per day.  
> 
> This makes me curious about why I didn't experience lockups.  Perhaps my
> BIOS defaulted to something else?
> 
> With these settings:
> 
> machdep.idle: acpi
> machdep.idle_mwait: 1
> 

I can only say that after updating the processor's microcde and
applying the errata script my system runs much more stabily.  No
lockups for days.

I suspect that updating the microcode helped quite a bit.

I have a first-generation Ryzen 5 1600 with all the errata.

-- 
Gary Jennejohn