Re: [net-next v6 08/12] net: bnxt: Implement software USO

[Joe Damato]

On Sun, Mar 29, 2026 at 03:20:16PM -0700, Jakub Kicinski wrote:
> On Thu, 26 Mar 2026 16:52:27 -0700 Joe Damato wrote:
> > +	/* Upper bound on the number of descriptors needed.
> > +	 *
> > +	 * Each segment uses 1 long BD + 1 ext BD + payload BDs, which is
> > +	 * at most num_segs + nr_frags (each frag boundary crossing adds at
> > +	 * most 1 extra BD).
> > +	 */
> > +	bds_needed = 3 * num_segs + skb_shinfo(skb)->nr_frags + 1;
> > +
> > +	if (unlikely(bnxt_tx_avail(bp, txr) < bds_needed)) {
> > +		netif_txq_try_stop(txq, bnxt_tx_avail(bp, txr),
> > +				   bp->tx_wake_thresh);
> > +		return NETDEV_TX_BUSY;
> > +	}
> > +
> > +	slots = BNXT_SW_USO_MAX_SEGS - (txr->tx_inline_prod - txr->tx_inline_cons);
> > +
> > +	if (unlikely(slots < num_segs)) {
> > +		netif_txq_try_stop(txq, bnxt_tx_avail(bp, txr),
> 
> This looks sus, try_stop() will evaluate the bnxt_tx_avail(bp, txr)
> and leave the ring running.

Yea, I think the slot check can actually be removed entirely. Each segment
consumes 1 inline slot and at least 3 BDs. BNXT_SW_USO_MAX_SEGS is 64 and the
ring's minimum size is 2 * BNXT_SW_USO_MAX_DESCS (420).

Using 64 slots would consume at least 192 BDs, so the check above would fire
first.

I think as long as the ring size is constrained by the code in fix_features
and set_ringparam then this if block can be removed.

> > +				   bp->tx_wake_thresh);
> 
> Is tx_wake_thresh larger than the max USO even for smallest ring size?

Yes, it is.

Maybe its worth adding a comment in the code somewhere to make
this more clear? Not sure where would be an appropriate place, but maybe
bnxt_init_tx_rings?