Re: [PATCH V1 1/2] bnxt_en: Add TPH support in BNXT driver

[Wei Huang]

On 1/8/25 05:33, Jiri Slaby wrote:
> On 15. 11. 24, 23:04, Jakub Kicinski wrote:
>> On Fri, 15 Nov 2024 14:04:11 -0600 Wei Huang wrote:
>>> +static void bnxt_irq_affinity_release(struct kref __always_unused *ref)
>>
>> unused? you're using it now
>>
>>> +{
>>> +    struct irq_affinity_notify *notify =
>>> +        (struct irq_affinity_notify *)
>>> +        container_of(ref, struct irq_affinity_notify, kref);
>>
>> this is ugly, and cast is unnecessary.
>>
>>> +    struct bnxt_irq *irq;
>>> +
>>> +    irq = container_of(notify, struct bnxt_irq, affinity_notify);
>>
>> since you init irq out of line you can as well init notify here
>>
>>> +    if (pcie_tph_set_st_entry(irq->bp->pdev, irq->msix_nr, 0)) {
>>
>> You checked this function can sleep, right? Because rtnl_lock()
>> will sleep.
> 
> Based on the above, I assume a new version was expected, but I cannot
> find any. So, Wei Huang, what's the status of this?

Currently the Broadcom team is driving the changes for upstream bnxt. I
will leave this question to them (Andy, Somnath, Michael).

> 
> thanks,