[PATCH v2] powerpc/powernv: Add pci_reset_phbs parameter to issue a PHB reset

Guilherme G. Piccoli gpiccoli at linux.vnet.ibm.com
Sat Dec 9 05:43:12 AEDT 2017


On 12/08/2017 10:03 AM, Michael Ellerman wrote:
> "Guilherme G. Piccoli" <gpiccoli at linux.vnet.ibm.com> writes:
> 
>> On 11/21/2017 12:35 AM, Balbir Singh wrote:
>>> On Thu, Nov 16, 2017 at 11:14 PM, Guilherme G. Piccoli
>>> <gpiccoli at linux.vnet.ibm.com> wrote:
>>>> On 11/16/2017 01:49 AM, Balbir Singh wrote:
>>>>> On Thu, Oct 26, 2017 at 2:27 AM, Guilherme G. Piccoli
>>>>> <gpiccoli at linux.vnet.ibm.com> wrote:
>>>>>> During a kdump kernel boot in PowerPC, we request a reset of the PHBs
>>>>>> to the FW. It makes sense, since if we are booting a kdump kernel it
>>>>>> means we had some trouble before and we cannot rely in the adapters'
>>>>>> health; they could be in a bad state, hence the reset is needed.
>>>>>>
>>>>>> But this reset is useful not only in kdump - there are situations,
>>>>>> specially when debugging drivers, that we could break an adapter in
>>>>>> a way it requires such reset. One can tell to just go ahead and
>>>>>> reboot the machine, but happens that many times doing kexec is much
>>>>>> faster, and so preferable than a full power cycle.
>>>>>>
>>>>>> This patch adds the pci_reset_phbs parameter to perform such reset
>>>>>> when desired by the user.
>>>>>>
>>>>>
>>>>> Do we care to reset specific phbs or all of them? I guess all based on
>>>>> your description.
>>>>
>>>> Exactly Balbir, it does reset all of them. We could add such
>>>> granularity, but I don't see much usability..
>>>> But if somebody feels it's useful, we can change...
>>>>
>>>
>>> OK.. makes sense, any reason why this can't be folded into reset_devices?
>>> I guess we want reset_phbs to be independent of reset_devices
>>
>> It was, in v1. But mpe asked it to be a powerpc specific parameter heheh
> 
> LOL. Turtles.
> 
> I still feel like we shouldn't be creating a generically named parameter
> like this, though we have done it many times in the past.
> 
> Can we call it "ppc_reset_phbs". And then I'll merge it without further
> quibbling, honest.

hahah sure! How about this one:  https://patchwork.ozlabs.org/patch/839123

Let me know your thoughts!
Cheers,


Guilherme
> 
> cheers
> 



More information about the Linuxppc-dev mailing list