Re: [PATCH net-next 1/2] net/smc: transition to RDMA core CQ pooling

[D. Wythe]

On Tue, May 12, 2026 at 10:31:09AM +0200, Paolo Abeni wrote:
> On 5/8/26 8:37 AM, D. Wythe wrote:
> > -void smc_wr_rx_cq_handler(struct ib_cq *ib_cq, void *cq_context)
> > -{
> > -	struct smc_ib_device *dev = (struct smc_ib_device *)cq_context;
> > +/***************************** init, exit, misc ******************************/
> >  
> > -	tasklet_schedule(&dev->recv_tasklet);
> > +static inline void smc_wr_reg_init_cqe(struct ib_cqe *cqe)
> 

Thanks for the review, Paolo.

> This and the next 3 helpers are used at init time, hopefully not
> critical/fast path; the `inline` annotation should really be avoided.
> 

Agreed, will drop the inline from these init-time helpers in v2.

> Note that the sashiko gemini instance has more concerns on patch 1,
> please have a look:
> 
> https://sashiko.dev/#/patchset/20260508063718.101622-1-alibuda%40linux.alibaba.com
> 
> /P

I've reviewed the sashiko feedback and will address those concerns in v2
as well. It did make sense.

Best wishes,
D. Wythe