Re: [PATCH net] net: bnxt: ring the doorbell when SW USO exits early

Joe Damato <[email protected]>
Newsgroups org.kernel.vger.netdev,org.kernel.vger.linux-kernel,org.kernel.vger.stable
Message-ID <[email protected]>
On Tue, Aug 18, 2026 at 03:36:42PM -0700, Michael Chan wrote:
> On Tue, Aug 18, 2026 at 2:15 PM Joe Damato <[email protected]> wrote:

[...]

> > diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_gso.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_gso.c
> > index f317f60414e8..3c58b1d94ac9 100644
> > --- a/drivers/net/ethernet/broadcom/bnxt/bnxt_gso.c
> > +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_gso.c
> > @@ -223,9 +223,7 @@ netdev_tx_t bnxt_sw_udp_gso_xmit(struct bnxt *bp,
> >         netdev_tx_sent_queue(txq, skb->len);
> >
> >         WRITE_ONCE(txr->tx_prod, prod);
> > -       /* Sync BDs before doorbell */
> > -       wmb();
> > -       bnxt_db_write(bp, &txr->tx_db, prod);
> > +       bnxt_txr_db_kick(bp, txr, prod);
> 
> The patch looks good to me.  But a slightly simpler approach is to
> just set txr->kick_pending here.  Upon return, the new logic above
> will always ring the doorbell for the success case.  By doing this, we
> don't need to export bnxt_txr_db_kick().

Ah, right. Sorry I missed that; I was rushing too quickly to get this
out :)

I'll send a v2 tomorrow that simplfies the patch as you suggested. Thanks.
lmpx.com only provides a reader for public news (NNTP) servers. It is not affiliated with the servers or forums shown here and is not responsible for the content of articles, which is written by their respective authors.