From: Jakub Kicinski <kuba@kernel.org>
To: Shannon Nelson <shannon.nelson@amd.com>
Cc: <netdev@vger.kernel.org>, <davem@davemloft.net>,
<edumazet@google.com>, <pabeni@redhat.com>,
<brett.creeley@amd.com>, <drivers@pensando.io>
Subject: Re: [PATCH net-next 6/9] ionic: Add XDP_TX support
Date: Wed, 31 Jan 2024 18:25:05 -0800 [thread overview]
Message-ID: <20240131182505.67eeedd9@kernel.org> (raw)
In-Reply-To: <20240130013042.11586-7-shannon.nelson@amd.com>
On Mon, 29 Jan 2024 17:30:39 -0800 Shannon Nelson wrote:
> case XDP_TX:
> + xdpf = xdp_convert_buff_to_frame(&xdp_buf);
> + if (!xdpf)
> + goto out_xdp_abort;
> +
> + txq = rxq->partner;
> + nq = netdev_get_tx_queue(netdev, txq->index);
> + __netif_tx_lock(nq, smp_processor_id());
> +
> + if (netif_tx_queue_stopped(nq) ||
> + unlikely(ionic_maybe_stop_tx(txq, 1))) {
> + __netif_tx_unlock(nq);
> + goto out_xdp_abort;
> + }
> +
> + dma_unmap_page(rxq->dev, buf_info->dma_addr,
> + IONIC_PAGE_SIZE, DMA_FROM_DEVICE);
> +
> + err = ionic_xdp_post_frame(netdev, txq, xdpf, XDP_TX,
> + buf_info->page,
> + buf_info->page_offset,
> + true);
I think that you need txq_trans_cond_update() somewhere, otherwise
if XDP starves stack Tx the stack will think the queue is stalled.
> + __netif_tx_unlock(nq);
> + if (err) {
> + netdev_dbg(netdev, "tx ionic_xdp_post_frame err %d\n", err);
> + goto out_xdp_abort;
> + }
> + stats->xdp_tx++;
> +
> + /* the Tx completion will free the buffers */
> + break;
> +
> case XDP_ABORTED:
> default:
> - trace_xdp_exception(netdev, xdp_prog, xdp_action);
> - ionic_rx_page_free(rxq, buf_info);
> - stats->xdp_aborted++;
> + goto out_xdp_abort;
> }
>
> + return true;
> +
> +out_xdp_abort:
> + trace_xdp_exception(netdev, xdp_prog, xdp_action);
> + ionic_rx_page_free(rxq, buf_info);
> + stats->xdp_aborted++;
> +
> return true;
> }
>
> @@ -880,6 +1001,16 @@ static void ionic_tx_clean(struct ionic_queue *q,
> struct sk_buff *skb = cb_arg;
> u16 qi;
>
> + if (desc_info->xdpf) {
> + ionic_xdp_tx_desc_clean(q->partner, desc_info);
> + stats->clean++;
> +
> + if (unlikely(__netif_subqueue_stopped(q->lif->netdev, q->index)))
> + netif_wake_subqueue(q->lif->netdev, q->index);
> +
> + return;
> + }
You can't complete XDP if NAPI budget is 0, you may be in hard IRQ
context if its netpoll calling :(
next prev parent reply other threads:[~2024-02-01 2:25 UTC|newest]
Thread overview: 12+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-01-30 1:30 [PATCH net-next 0/9] ionic: add XDP support Shannon Nelson
2024-01-30 1:30 ` [PATCH net-next 1/9] ionic: set adminq irq affinity Shannon Nelson
2024-01-30 1:30 ` [PATCH net-next 2/9] ionic: add helpers for accessing buffer info Shannon Nelson
2024-01-30 1:30 ` [PATCH net-next 3/9] ionic: use dma range APIs Shannon Nelson
2024-01-30 1:30 ` [PATCH net-next 4/9] ionic: add initial framework for XDP support Shannon Nelson
2024-01-30 1:30 ` [PATCH net-next 5/9] ionic: Add XDP packet headroom Shannon Nelson
2024-01-30 1:30 ` [PATCH net-next 6/9] ionic: Add XDP_TX support Shannon Nelson
2024-02-01 2:25 ` Jakub Kicinski [this message]
2024-02-01 17:41 ` Nelson, Shannon
2024-01-30 1:30 ` [PATCH net-next 7/9] ionic: Add XDP_REDIRECT support Shannon Nelson
2024-01-30 1:30 ` [PATCH net-next 8/9] ionic: add ndo_xdp_xmit Shannon Nelson
2024-01-30 1:30 ` [PATCH net-next 9/9] ionic: implement xdp frags support Shannon Nelson
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20240131182505.67eeedd9@kernel.org \
--to=kuba@kernel.org \
--cc=brett.creeley@amd.com \
--cc=davem@davemloft.net \
--cc=drivers@pensando.io \
--cc=edumazet@google.com \
--cc=netdev@vger.kernel.org \
--cc=pabeni@redhat.com \
--cc=shannon.nelson@amd.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).