| lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC | |
|
Open Source and information security mailing list archives
| ||
|
Message-ID: <788b4f16-2627-45d5-a23c-b69cbdc7bbac@linux.microsoft.com>
Date: Mon, 27 Oct 2025 18:58:09 +0530
From: Aditya Garg <gargaditya@...ux.microsoft.com>
To: kys@...rosoft.com, haiyangz@...rosoft.com, wei.liu@...nel.org,
decui@...rosoft.com, andrew+netdev@...n.ch, davem@...emloft.net,
edumazet@...gle.com, kuba@...nel.org, pabeni@...hat.com,
longli@...rosoft.com, kotaranov@...rosoft.com, horms@...nel.org,
shradhagupta@...ux.microsoft.com, ssengar@...ux.microsoft.com,
ernis@...ux.microsoft.com, dipayanroy@...ux.microsoft.com,
shirazsaleem@...rosoft.com, linux-hyperv@...r.kernel.org,
netdev@...r.kernel.org, linux-kernel@...r.kernel.org,
linux-rdma@...r.kernel.org, gargaditya@...rosoft.com
Subject: Re: [RFC PATCH net-next v2] net: mana: Handle SKB if TX SGEs exceed
hardware limit
On 23-10-2025 02:13, Aditya Garg wrote:
> The MANA hardware supports a maximum of 30 scatter-gather entries (SGEs)
> per TX WQE. Exceeding this limit can cause TX failures.
> Add ndo_features_check() callback to validate SKB layout before
> transmission. For GSO SKBs that would exceed the hardware SGE limit, clear
> NETIF_F_GSO_MASK to enforce software segmentation in the stack.
> Add a fallback in mana_start_xmit() to linearize non-GSO SKBs that still
> exceed the SGE limit.
>
> Return NETDEV_TX_BUSY only for -ENOSPC from mana_gd_post_work_request(),
> send other errors to free_sgl_ptr to free resources and record the tx
> drop.
>
> Co-developed-by: Dipayaan Roy <dipayanroy@...ux.microsoft.com>
> Signed-off-by: Dipayaan Roy <dipayanroy@...ux.microsoft.com>
> Signed-off-by: Aditya Garg <gargaditya@...ux.microsoft.com>
> ---
> drivers/net/ethernet/microsoft/mana/mana_en.c | 48 +++++++++++++++++--
> include/net/mana/gdma.h | 6 ++-
> include/net/mana/mana.h | 1 +
> 3 files changed, 49 insertions(+), 6 deletions(-)
>
> diff --git a/drivers/net/ethernet/microsoft/mana/mana_en.c b/drivers/net/ethernet/microsoft/mana/mana_en.c
> index 0142fd98392c..8e23a87a779d 100644
> --- a/drivers/net/ethernet/microsoft/mana/mana_en.c
> +++ b/drivers/net/ethernet/microsoft/mana/mana_en.c
> @@ -11,6 +11,7 @@
> #include <linux/mm.h>
> #include <linux/pci.h>
> #include <linux/export.h>
> +#include <linux/skbuff.h>
>
> #include <net/checksum.h>
> #include <net/ip6_checksum.h>
> @@ -289,6 +290,21 @@ netdev_tx_t mana_start_xmit(struct sk_buff *skb, struct net_device *ndev)
> cq = &apc->tx_qp[txq_idx].tx_cq;
> tx_stats = &txq->stats;
>
> + if (MAX_SKB_FRAGS + 2 > MAX_TX_WQE_SGL_ENTRIES &&
> + skb_shinfo(skb)->nr_frags + 2 > MAX_TX_WQE_SGL_ENTRIES) {
> + /* GSO skb with Hardware SGE limit exceeded is not expected here
> + * as they are handled in mana_features_check() callback
> + */
> + if (skb_is_gso(skb))
> + netdev_warn_once(ndev, "GSO enabled skb exceeds max SGE limit\n");
> + if (skb_linearize(skb)) {
> + netdev_warn_once(ndev, "Failed to linearize skb with nr_frags=%d and is_gso=%d\n",
> + skb_shinfo(skb)->nr_frags,
> + skb_is_gso(skb));
> + goto tx_drop_count;
> + }
> + }
> +
> pkg.tx_oob.s_oob.vcq_num = cq->gdma_id;
> pkg.tx_oob.s_oob.vsq_frame = txq->vsq_frame;
>
> @@ -402,8 +418,6 @@ netdev_tx_t mana_start_xmit(struct sk_buff *skb, struct net_device *ndev)
> }
> }
>
> - WARN_ON_ONCE(pkg.wqe_req.num_sge > MAX_TX_WQE_SGL_ENTRIES);
> -
> if (pkg.wqe_req.num_sge <= ARRAY_SIZE(pkg.sgl_array)) {
> pkg.wqe_req.sgl = pkg.sgl_array;
> } else {
> @@ -438,9 +452,13 @@ netdev_tx_t mana_start_xmit(struct sk_buff *skb, struct net_device *ndev)
>
> if (err) {
> (void)skb_dequeue_tail(&txq->pending_skbs);
> + mana_unmap_skb(skb, apc);
> netdev_warn(ndev, "Failed to post TX OOB: %d\n", err);
> - err = NETDEV_TX_BUSY;
> - goto tx_busy;
> + if (err == -ENOSPC) {
> + err = NETDEV_TX_BUSY;
> + goto tx_busy;
> + }
> + goto free_sgl_ptr;
> }
>
> err = NETDEV_TX_OK;
> @@ -478,6 +496,25 @@ netdev_tx_t mana_start_xmit(struct sk_buff *skb, struct net_device *ndev)
> return NETDEV_TX_OK;
> }
>
> +static netdev_features_t mana_features_check(struct sk_buff *skb,
> + struct net_device *ndev,
> + netdev_features_t features)
> +{
> + if (MAX_SKB_FRAGS + 2 > MAX_TX_WQE_SGL_ENTRIES &&
> + skb_shinfo(skb)->nr_frags + 2 > MAX_TX_WQE_SGL_ENTRIES) {
> + /* Exceeds HW SGE limit.
> + * GSO case:
> + * Disable GSO so the stack will software-segment the skb
> + * into smaller skbs that fit the SGE budget.
> + * Non-GSO case:
> + * The xmit path will attempt skb_linearize() as a fallback.
> + */
> + if (skb_is_gso(skb))
> + features &= ~NETIF_F_GSO_MASK;
> + }
> + return features;
> +}
> +
> static void mana_get_stats64(struct net_device *ndev,
> struct rtnl_link_stats64 *st)
> {
> @@ -838,6 +875,7 @@ static const struct net_device_ops mana_devops = {
> .ndo_open = mana_open,
> .ndo_stop = mana_close,
> .ndo_select_queue = mana_select_queue,
> + .ndo_features_check = mana_features_check,
> .ndo_start_xmit = mana_start_xmit,
> .ndo_validate_addr = eth_validate_addr,
> .ndo_get_stats64 = mana_get_stats64,
> @@ -1606,7 +1644,7 @@ static int mana_move_wq_tail(struct gdma_queue *wq, u32 num_units)
> return 0;
> }
>
> -static void mana_unmap_skb(struct sk_buff *skb, struct mana_port_context *apc)
> +void mana_unmap_skb(struct sk_buff *skb, struct mana_port_context *apc)
> {
> struct mana_skb_head *ash = (struct mana_skb_head *)skb->head;
> struct gdma_context *gc = apc->ac->gdma_dev->gdma_context;
> diff --git a/include/net/mana/gdma.h b/include/net/mana/gdma.h
> index 57df78cfbf82..b35ecc58fbab 100644
> --- a/include/net/mana/gdma.h
> +++ b/include/net/mana/gdma.h
> @@ -591,6 +591,9 @@ enum {
> /* Driver can self reset on FPGA Reconfig EQE notification */
> #define GDMA_DRV_CAP_FLAG_1_HANDLE_RECONFIG_EQE BIT(17)
>
> +/* Driver supports linearizing the skb when num_sge exceeds hardware limit */
> +#define GDMA_DRV_CAP_FLAG_1_SKB_LINEARIZE BIT(20)
> +
> #define GDMA_DRV_CAP_FLAGS1 \
> (GDMA_DRV_CAP_FLAG_1_EQ_SHARING_MULTI_VPORT | \
> GDMA_DRV_CAP_FLAG_1_NAPI_WKDONE_FIX | \
> @@ -599,7 +602,8 @@ enum {
> GDMA_DRV_CAP_FLAG_1_DEV_LIST_HOLES_SUP | \
> GDMA_DRV_CAP_FLAG_1_DYNAMIC_IRQ_ALLOC_SUPPORT | \
> GDMA_DRV_CAP_FLAG_1_SELF_RESET_ON_EQE | \
> - GDMA_DRV_CAP_FLAG_1_HANDLE_RECONFIG_EQE)
> + GDMA_DRV_CAP_FLAG_1_HANDLE_RECONFIG_EQE | \
> + GDMA_DRV_CAP_FLAG_1_SKB_LINEARIZE)
>
> #define GDMA_DRV_CAP_FLAGS2 0
>
> diff --git a/include/net/mana/mana.h b/include/net/mana/mana.h
> index 0921485565c0..330e1bb088bb 100644
> --- a/include/net/mana/mana.h
> +++ b/include/net/mana/mana.h
> @@ -580,6 +580,7 @@ int mana_set_bw_clamp(struct mana_port_context *apc, u32 speed,
> void mana_query_phy_stats(struct mana_port_context *apc);
> int mana_pre_alloc_rxbufs(struct mana_port_context *apc, int mtu, int num_queues);
> void mana_pre_dealloc_rxbufs(struct mana_port_context *apc);
> +void mana_unmap_skb(struct sk_buff *skb, struct mana_port_context *apc);
>
> extern const struct ethtool_ops mana_ethtool_ops;
> extern struct dentry *mana_debugfs_root;
Hi Eric,
I hope this approach is okay. If there are no issues, I'll go ahead and
send the patch for final review.
Let me know if you need more time for this RFC.
Regards,
Aditya
Powered by blists - more mailing lists