On Thu, Oct 05, 2023 at 08:07:35PM +0530, Gupta, Nipun wrote:
> Hi Greg,
>
> On 10/5/2023 7:30 PM, Greg KH wrote:
> > On Thu, Oct 05, 2023 at 03:46:34PM +0200, Thomas Gleixner wrote:
> > > On Thu, Oct 05 2023 at 12:24, Greg KH wrote:
> > > > > diff --git a/drivers/cdx/Kconfig b/drivers/cdx/Kconfig
> > > > > index a08958485e31..86df7ccb76bb 100644
> > > > > --- a/drivers/cdx/Kconfig
> > > > > +++ b/drivers/cdx/Kconfig
> > > > > @@ -8,6 +8,7 @@
> > > > > config CDX_BUS
> > > > > bool "CDX Bus driver"
> > > > > depends on OF && ARM64
> > > > > + select GENERIC_MSI_IRQ_DOMAIN
> > > >
> > > > This config option isn't in my tree anywhere, where did it come from?
> > > > What is it supposed to do?
> > >
> > > 13e7accb81d6 ("genirq: Get rid of GENERIC_MSI_IRQ_DOMAIN") :)
> >
> > Ok, so this hasn't been tested since the 6.2 release? Wow, I think
> > someone from AMD needs to take a deep look at this and verify that it
> > actually is doing what it is supposed to be doing...
>
> The patch Thomas mentioned renames "GENERIC_MSI_IRQ_DOMAIN" to
> "GENERIC_MSI_IRQ"; and in our testing "GENERIC_MSI_IRQ" is also selected as
> 'ARM64' is enabled which enables 'ARM_GIC_V3_ITS' which in-turn selects
> 'GENERIC_MSI_IRQ'.
Ok, but that's not what this patch "selects" at all :(
> The patch is tested for MSI functionality on 6.6-rc1. We will re-look into
> the config dependencies to avoid such issues, but please be assured that the
> patch has been validated.
How has the dependancy been validated as correct if there is no such
thing in this kernel?
confused,
greg k-h
On 10/5/2023 8:24 PM, Greg KH wrote:
> On Thu, Oct 05, 2023 at 08:07:35PM +0530, Gupta, Nipun wrote:
>> Hi Greg,
>>
>> On 10/5/2023 7:30 PM, Greg KH wrote:
>>> On Thu, Oct 05, 2023 at 03:46:34PM +0200, Thomas Gleixner wrote:
>>>> On Thu, Oct 05 2023 at 12:24, Greg KH wrote:
>>>>>> diff --git a/drivers/cdx/Kconfig b/drivers/cdx/Kconfig
>>>>>> index a08958485e31..86df7ccb76bb 100644
>>>>>> --- a/drivers/cdx/Kconfig
>>>>>> +++ b/drivers/cdx/Kconfig
>>>>>> @@ -8,6 +8,7 @@
>>>>>> config CDX_BUS
>>>>>> bool "CDX Bus driver"
>>>>>> depends on OF && ARM64
>>>>>> + select GENERIC_MSI_IRQ_DOMAIN
>>>>>
>>>>> This config option isn't in my tree anywhere, where did it come from?
>>>>> What is it supposed to do?
>>>>
>>>> 13e7accb81d6 ("genirq: Get rid of GENERIC_MSI_IRQ_DOMAIN") :)
>>>
>>> Ok, so this hasn't been tested since the 6.2 release? Wow, I think
>>> someone from AMD needs to take a deep look at this and verify that it
>>> actually is doing what it is supposed to be doing...
>>
>> The patch Thomas mentioned renames "GENERIC_MSI_IRQ_DOMAIN" to
>> "GENERIC_MSI_IRQ"; and in our testing "GENERIC_MSI_IRQ" is also selected as
>> 'ARM64' is enabled which enables 'ARM_GIC_V3_ITS' which in-turn selects
>> 'GENERIC_MSI_IRQ'.
>
> Ok, but that's not what this patch "selects" at all :(
"GENERIC_MSI_IRQ" gets enabled because of enabling "ARM64", and you are
right that what we are selecting here does not have any meaning now
(after 6.2).
>
>> The patch is tested for MSI functionality on 6.6-rc1. We will re-look into
>> the config dependencies to avoid such issues, but please be assured that the
>> patch has been validated.
>
> How has the dependancy been validated as correct if there is no such
> thing in this kernel?
Apology for the confusion. By validated I meant tested for MSI
functionality in CDX. Agree that "select GENERIC_MSI_IRQ_DOMAIN" needs
to be fixed here. Will update in next spin. :)
Thanks,
Nipun
>
> confused,
>
> greg k-h