G5 Quad Fans full speed after 1 min

[Jason Bacon]

On 2020-01-19 05:13, Mark Millard via freebsd-ppc wrote:
> On 2020-Jan-19, at 00:38, Francis Little <oggy at farscape.co.uk> wrote:
>
>> Hi,
>>
>> My G5 Quad is running current from a few days ago, but this issue has been
>> happening for a long time.
>>
>> After about 1 min of uptime, the fans go full speed.
>>
>> As soon as I query anything like CPU temp or fan rpm with sysctl, the fans
>> return to a normal speed.
>>
>> 1 min later the fans go full speed again.
>>
>> I've been working round this for some time with a cron job that runs sysctl
>> with one of the cpu temp sensors to calm the system.
> QUOTING an old message:
>     The mftb figures on the various cores can be so far apart that
>     threads can end-up stuck sleeping, such as syncr, pmac_thermal,
>     and buf*deamon*  threads. (This can really mess things up by
>     not updating the storage correctly.) Such is still true of the
>     ELFv2 context.
>
>     (Most folks notice this via shutdown timeouts and the fans
>     going fast unnecessarily. But it is involved earlier as well.)
> END QUOTE
>
> Nothing in the boot sequence is forcing the CPUs/Cores to
> see any particular time relationship to each other and on
> the multi-socket PowerMacs it can vary widely (G4 and G5).
> Sometimes it will happen to end up okay, other times not.
>
> (I've no access to a single-socket, multi-core PowerMac,
> so I just do not know for that kind of context.)
>
> I run with patched boot-code that has cross-cpu/core time
> observations and adjustments to non-bsp time to see the
> bsp time as between the start and end of a round trip to
> the bsp from each non-bsp to get the bsp's time. It is
> based on the mid-point of the start and end times for
> the non-bsp's round trip vs. the bsp's returned time.
> With at most 4 cores, each non-bsp is done in sequence.
> The code only does this on PowerMacs, having no access
> to other types of PowerPC examples to test.
>
> I've not seen this type of problem since mid 2019-May on
> any of:
>
> 1 G5 with 2 sockets, 1 core each
> 2 G5's, 2 sockets, 2 cores each
> 2 G4's, 2 sockets, 1 core each
>
> (The mid-May time frame is when I adjusted the code to
> deal with the faster time increment on the slower
> 32-bit processors for the model that I have access to.
> I had to be more careful to avoid biasing the approximate
> symmetry to be less symmetric. On the G5's its been
> longer since I've seen this problem, based on earlier
> source code.)
>
> Unfortunately the "lab" the machines are in is powered
> down currently.
>
> FYI: Prior to this technique, I had a pure hack that
> was observed to avoid the problem. But it changed code
> used all the time --code that I did not want to have
> any hack's in if I could avoid it.
>
> FYI: I also run with other PowerMac related patches.
> Generally this mftb() time adjustment is one of the
> newest patches, possibly the newest. So my test
> context may be biased by the other patches.
>
>> If I boot to OS X 10.5 and load the system, the fans are stable.
> I've not done any investigation of the issue for the
> older contexts. But, if I remember right, I did see
> the problem on occasion back in that time frame.
>
>> Does anyone else get this?
> My understanding is everyone booting a fairly modern
> standard FreeBSD gets this sometimes for the kind of
> context that you specified. (I'm not sure of the
> variability if the frequency of the problem happening
> for that kind of context.)
>
> I certainly saw it before I investigated avoiding it.
>
> ===
> Mark Millard
> marklmi at yahoo.com
> ( dsl-only.net went
> away in early 2018-Mar)
>
> _______________________________________________
> freebsd-ppc at freebsd.org mailing list
> https://lists.freebsd.org/mailman/listinfo/freebsd-ppc
> To unsubscribe, send any mail to "freebsd-ppc-unsubscribe at freebsd.org"
On my dual CPU PowerMac G5, this issue happens for 80 - 90% of boots.

I'd love to test a patch if one is available.  Cutting the speed in half 
would be problematic for testing large ports.

Thanks,

     JB

-- 
Earth is a beta site.