ThunderX Panic after r368370

[Andrew Turner]

> On 7 Dec 2020, at 16:42, Michal Meloun <meloun.michal at gmail.com> wrote:
> 
> 
> 
> On 07.12.2020 17:33, Marcel Flores wrote:
>>> On Dec 7, 2020, at 1:13 AM, Michal Meloun <meloun.michal at gmail.com> wrote:
>>> 
>>> On 07.12.2020 9:21, Mark Millard wrote:
>>>> On 2020-Dec-6, at 13:30, Mark Millard <marklmi at yahoo.com> wrote:
>>>>> On 2020-Dec-6, at 03:51, Michal Meloun <meloun.michal at gmail.com> wrote:
>>>>> 
>>>>> On 06.12.2020 10:47, Mark Millard wrote:
>>>>>>> On 2020-Dec-6, at 00:17, Michal Meloun <meloun.michal at gmail.com> wrote:
>>>>>>>> On 06.12.2020 3:21, Marcel Flores wrote:
>>>>>>>>> Hi All,
>>>>>>>>> Looks like the ThunderX started panicking at boot after r368370:
>>>>>>>>> https://reviews.freebsd.org/rS368370
>>>>>>>>> From a verbose boot, it looks like it bails in gic0 redistributor setup(?):
>>>>>>>>> gic0: CPU29 Re-Distributor woke up
>>>>>>>>> gic0: CPU24 enabled CPU interface via system registers
>>>>>>>>> gic0: CPU17 enabled CPU interface via system registers
>>>>>>>>> gic0: CPU29 enabled CPU interface via system registers
>>>>>>>>> done
>>>> Looks to me like fdt_cpuid's use in cpu_init_fdt is one of the issues
>>>> with what is added to each cpuset_domain[domain] :
>>>> fdt_cpuid's initial value can not be added by this code: it is
>>>> incremented first.
>>> 
>>> Good catch. This ThunderX system is *FDT* based. And I spend half of night by digging in ACPI sources in attempt to understand who screw cpuset_domain :P
>>> 
>>> Marcel, can you please try following patch? I hope that it fix this issue.
>>> 
>>> https://github.com/strejda/freebsd/commit/1404960be6a7a6c75af86e86331acf951cfdf9b1
>>> 
>> That did the trick — no more panic.
>> Working debug-patch output just in case:
>> https://gist.github.com/mesflores/be330f23df2faa93a082825104b986a1
>> Appreciate the quick response!
> 
> In meantime, Andrew made expanded version of this fix
> https://reviews.freebsd.org/D27497 <https://reviews.freebsd.org/D27497>
> 
> I sure that this will be committed ASAP.
> Many thanks for cooperation and help. Sorry for troubles.

I’ve committed it in r368416.

Andrew