Re: [PATCH net v3 3/5] xsk: drain continuation descs after overflow in xsk_build_skb()

[sashiko-bot@kernel.org]

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [Medium] The commit message claims the function returns `-EOVERFLOW` after draining is complete, but the code merely uses `continue;`.
--

commit 5c0ce4e27ce234e8586a243e61aed44e90c34548
Author: Jason Xing <kerneljasonxing@gmail.com>

xsk: drain continuation descs after overflow in xsk_build_skb()

This patch adds a drain_cont flag to xdp_sock to properly drain leftover
continuation descriptors when a multi-buffer packet exceeds MAX_SKB_FRAGS.
The main TX loop handles these leftover descriptors individually by returning
them to the completion queue with standard backpressure mechanisms.

Closes: https://lore.kernel.org/all/20260425041726.85FB3C2BCB2@smtp.kernel.org/

> When the last fragment
> (without XDP_PKT_CONTD) is processed, the flag is cleared and the
> function returns -EOVERFLOW so the next call starts with a fresh
> budget for normal packets.

Is this description accurate?

> diff --git a/net/xdp/xsk.c b/net/xdp/xsk.c
> index 0a6203c425766..298194b7335e1 100644
> --- a/net/xdp/xsk.c
> +++ b/net/xdp/xsk.c

[ ... ]

> @@ -1062,6 +1064,23 @@ static int __xsk_generic_xmit(struct sock *sk)
>  			goto out;
>  		}
>  
> +		if (unlikely(xs->drain_cont)) {
> +			unsigned long flags;
> +			u32 idx;
> +
> +			spin_lock_irqsave(&xs->pool->cq_prod_lock, flags);
> +			idx = xskq_get_prod(xs->pool->cq);
> +			xskq_prod_write_addr(xs->pool->cq, idx, desc.addr);
> +			xskq_prod_submit_n(xs->pool->cq, 1);
> +			spin_unlock_irqrestore(&xs->pool->cq_prod_lock, flags);
> +
> +			xs->tx->invalid_descs++;
> +			xskq_cons_release(xs->tx);
> +			if (!xp_mb_desc(&desc))
> +				xs->drain_cont = false;
> +			continue;

Looking at the implementation here, the code clears the flag and executes
a continue statement, which proceeds to the next descriptor using the
remaining budget instead of returning -EOVERFLOW.

Could the code and the commit message be out of sync?

> +		}
> +
>  		skb = xsk_build_skb(xs, &desc);

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260517063311.28921-1-kerneljasonxing@gmail.com?part=3