From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from www62.your-server.de (www62.your-server.de [213.133.104.62]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id C16CB3F1667; Thu, 2 Apr 2026 23:10:59 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=213.133.104.62 ARC-Seal:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1775171463; cv=none; b=VblKnLln5uFVaQRa8YGAOx2jV+OI2ZaZ6fKG7VHjxO2Eb7QtGzTpadvDhsnKW0a41DqHbeGzOlQo+CNwm03gkVor+XPiQgGIgGXJyzsUuQlncdkV4L6aKek7e2j9KiBO9wYcEYDu2YWja5RWFyvxEmUJ2nsL+euo1K592BQEguo= ARC-Message-Signature:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1775171463; c=relaxed/simple; bh=i+g0vC4eFoI6yao3/LrtW0tAIHtLsyRVDczUsOokHZI=; h=From:To:Cc:Subject:Date:Message-ID:In-Reply-To:References: MIME-Version; b=JU30tiJgoNXjMA0ei6xapgkBFIo3EEZfIY0EGXVEZ8R96UVUjTyxIABDGwzdgcb0aQPuWIpiQJxXWhkYeYY4dLjtYhKFXTB/LMd0khRSoL2oKDRtlbIqH2a4WLAZroTkwWALHlH6XufeH8XvGxZxdv6AjoKj2lkq6N8qzY2j1so= ARC-Authentication-Results:i=1; smtp.subspace.kernel.org; dmarc=pass (p=reject dis=none) header.from=iogearbox.net; spf=pass smtp.mailfrom=iogearbox.net; dkim=pass (2048-bit key) header.d=iogearbox.net header.i=@iogearbox.net header.b=OkGS8oqz; arc=none smtp.client-ip=213.133.104.62 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=reject dis=none) header.from=iogearbox.net Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=iogearbox.net Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=iogearbox.net header.i=@iogearbox.net header.b="OkGS8oqz" DKIM-Signature: v=1; a=rsa-sha256; q=dns/txt; c=relaxed/relaxed; d=iogearbox.net; s=default2302; h=Content-Transfer-Encoding:MIME-Version: References:In-Reply-To:Message-ID:Date:Subject:Cc:To:From:Sender:Reply-To: Content-Type:Content-ID:Content-Description:Resent-Date:Resent-From: Resent-Sender:Resent-To:Resent-Cc:Resent-Message-ID; bh=GnY28wpFkMbn93yfj9OXJLREKu4h8R7relqWa3gv2cU=; b=OkGS8oqzxqIvSiNy66F+2odBUd eScvssMHn/Tkc6mDhwiCzTq3Ava2R1HlBoKk+xkr+dv3E+hlyBKPd/vbFgSE8EYIKeaiftvbYgf7c cB5u9h510CzXcd2heH6VUZqhfnvwoL5mrydnAbtwtOHwC1SEGKyB7nl3OhttNhKQ6xjf8eadqAWTu /ntpKm2jou1P40/hkXnHqic7ZHAqUVrar0/Ftpatxt0jOoS2b7PE5GyMKwFe9OTGtfTZjjiTAQ+Tw y2hLjPveEbbJ8cz1WrWUOgInhb9CuuELy3SD8gr1vT7EQZAEHp933PkzQ/SgEJgxsWpmiWi0NZkTg tt/N7jIg==; Received: from localhost ([127.0.0.1]) by www62.your-server.de with esmtpsa (TLS1.3) tls TLS_AES_256_GCM_SHA384 (Exim 4.96.2) (envelope-from ) id 1w8RBR-0000mw-1o; Fri, 03 Apr 2026 01:10:41 +0200 From: Daniel Borkmann To: netdev@vger.kernel.org Cc: bpf@vger.kernel.org, kuba@kernel.org, davem@davemloft.net, razor@blackwall.org, pabeni@redhat.com, willemb@google.com, sdf@fomichev.me, john.fastabend@gmail.com, martin.lau@kernel.org, jordan@jrife.io, maciej.fijalkowski@intel.com, magnus.karlsson@intel.com, dw@davidwei.uk, toke@redhat.com, yangzhenze@bytedance.com, wangdongdong.6@bytedance.com Subject: [PATCH net-next v11 07/14] net: Proxy netdev_queue_get_dma_dev for leased queues Date: Fri, 3 Apr 2026 01:10:24 +0200 Message-ID: <20260402231031.447597-8-daniel@iogearbox.net> X-Mailer: git-send-email 2.43.0 In-Reply-To: <20260402231031.447597-1-daniel@iogearbox.net> References: <20260402231031.447597-1-daniel@iogearbox.net> Precedence: bulk X-Mailing-List: netdev@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: 8bit X-Virus-Scanned: Clear (ClamAV 1.4.3/27959/Thu Apr 2 08:25:45 2026) From: David Wei Extend netdev_queue_get_dma_dev to return the physical device of the real rxq for DMA in case the queue was leased. This allows memory providers like io_uring zero-copy or devmem to bind to the physically leased rxq via virtual devices such as netkit. Signed-off-by: David Wei Co-developed-by: Daniel Borkmann Signed-off-by: Daniel Borkmann Reviewed-by: Nikolay Aleksandrov --- include/net/netdev_queues.h | 4 +++- io_uring/zcrx.c | 3 ++- net/core/netdev-genl.c | 5 +++-- net/core/netdev_queues.c | 41 +++++++++++++++++++++++++++++-------- 4 files changed, 40 insertions(+), 13 deletions(-) diff --git a/include/net/netdev_queues.h b/include/net/netdev_queues.h index 748b70552ed1..70c9fe9e83cc 100644 --- a/include/net/netdev_queues.h +++ b/include/net/netdev_queues.h @@ -380,7 +380,9 @@ static inline unsigned int netif_xmit_timeout_ms(struct netdev_queue *txq) get_desc, start_thrs); \ }) -struct device *netdev_queue_get_dma_dev(struct net_device *dev, int idx); +struct device *netdev_queue_get_dma_dev(struct net_device *dev, + unsigned int idx, + enum netdev_queue_type type); bool netdev_can_create_queue(const struct net_device *dev, struct netlink_ext_ack *extack); bool netdev_can_lease_queue(const struct net_device *dev, diff --git a/io_uring/zcrx.c b/io_uring/zcrx.c index d3ec63c83d0c..f4a7809ba0c2 100644 --- a/io_uring/zcrx.c +++ b/io_uring/zcrx.c @@ -829,7 +829,8 @@ int io_register_zcrx_ifq(struct io_ring_ctx *ctx, } netdev_hold(ifq->netdev, &ifq->netdev_tracker, GFP_KERNEL); - ifq->dev = netdev_queue_get_dma_dev(ifq->netdev, reg.if_rxq); + ifq->dev = netdev_queue_get_dma_dev(ifq->netdev, reg.if_rxq, + NETDEV_QUEUE_TYPE_RX); if (!ifq->dev) { ret = -EOPNOTSUPP; goto netdev_put_unlock; diff --git a/net/core/netdev-genl.c b/net/core/netdev-genl.c index 515832854251..056460d01940 100644 --- a/net/core/netdev-genl.c +++ b/net/core/netdev-genl.c @@ -976,7 +976,8 @@ netdev_nl_get_dma_dev(struct net_device *netdev, unsigned long *rxq_bitmap, for_each_set_bit(rxq_idx, rxq_bitmap, netdev->real_num_rx_queues) { struct device *rxq_dma_dev; - rxq_dma_dev = netdev_queue_get_dma_dev(netdev, rxq_idx); + rxq_dma_dev = netdev_queue_get_dma_dev(netdev, rxq_idx, + NETDEV_QUEUE_TYPE_RX); if (dma_dev && rxq_dma_dev != dma_dev) { NL_SET_ERR_MSG_FMT(extack, "DMA device mismatch between queue %u and %u (multi-PF device?)", rxq_idx, prev_rxq_idx); @@ -1153,7 +1154,7 @@ int netdev_nl_bind_tx_doit(struct sk_buff *skb, struct genl_info *info) goto err_unlock_netdev; } - dma_dev = netdev_queue_get_dma_dev(netdev, 0); + dma_dev = netdev_queue_get_dma_dev(netdev, 0, NETDEV_QUEUE_TYPE_TX); binding = net_devmem_bind_dmabuf(netdev, dma_dev, DMA_TO_DEVICE, dmabuf_fd, priv, info->extack); if (IS_ERR(binding)) { diff --git a/net/core/netdev_queues.c b/net/core/netdev_queues.c index 177401828e79..265161e12a9c 100644 --- a/net/core/netdev_queues.c +++ b/net/core/netdev_queues.c @@ -6,27 +6,50 @@ #include "dev.h" +static struct device * +__netdev_queue_get_dma_dev(struct net_device *dev, unsigned int idx) +{ + const struct netdev_queue_mgmt_ops *queue_ops = dev->queue_mgmt_ops; + struct device *dma_dev; + + if (queue_ops && queue_ops->ndo_queue_get_dma_dev) + dma_dev = queue_ops->ndo_queue_get_dma_dev(dev, idx); + else + dma_dev = dev->dev.parent; + + return dma_dev && dma_dev->dma_mask ? dma_dev : NULL; +} + /** * netdev_queue_get_dma_dev() - get dma device for zero-copy operations * @dev: net_device * @idx: queue index + * @type: queue type (RX or TX) * - * Get dma device for zero-copy operations to be used for this queue. - * When such device is not available or valid, the function will return NULL. + * Get dma device for zero-copy operations to be used for this queue. If + * the queue is an RX queue leased from a physical queue, we retrieve the + * physical queue's dma device. When the dma device is not available or + * valid, the function will return NULL. * * Return: Device or NULL on error */ -struct device *netdev_queue_get_dma_dev(struct net_device *dev, int idx) +struct device *netdev_queue_get_dma_dev(struct net_device *dev, + unsigned int idx, + enum netdev_queue_type type) { - const struct netdev_queue_mgmt_ops *queue_ops = dev->queue_mgmt_ops; + struct net_device *orig_dev = dev; struct device *dma_dev; - if (queue_ops && queue_ops->ndo_queue_get_dma_dev) - dma_dev = queue_ops->ndo_queue_get_dma_dev(dev, idx); - else - dma_dev = dev->dev.parent; + /* Only RX side supports queue leasing today. */ + if (type != NETDEV_QUEUE_TYPE_RX || !netif_rxq_is_leased(dev, idx)) + return __netdev_queue_get_dma_dev(dev, idx); - return dma_dev && dma_dev->dma_mask ? dma_dev : NULL; + if (!netif_get_rx_queue_lease_locked(&dev, &idx)) + return NULL; + + dma_dev = __netdev_queue_get_dma_dev(dev, idx); + netif_put_rx_queue_lease_locked(orig_dev, dev); + return dma_dev; } bool netdev_can_create_queue(const struct net_device *dev, -- 2.43.0