| Message ID | 20250205193751.297211-1-jdamato@fastly.com (mailing list archive) |
|---|---|
| State | Accepted |
| Delegated to: | Netdev Maintainers |
| Headers | show |
| Series | [net-next,v4] netdev-genl: Elide napi_id when not present | expand |
On 2/5/2025 11:37 AM, Joe Damato wrote: > There are at least two cases where napi_id may not present and the > napi_id should be elided: > > 1. Queues could be created, but napi_enable may not have been called > yet. In this case, there may be a NAPI but it may not have an ID and > output of a napi_id should be elided. > > 2. TX-only NAPIs currently do not have NAPI IDs. If a TX queue happens > to be linked with a TX-only NAPI, elide the NAPI ID from the netlink > output as a NAPI ID of 0 is not useful for users. > > Signed-off-by: Joe Damato <jdamato@fastly.com> Reviewed-by: Sridhar Samudrala <sridhar.samudrala@intel.com>
On Wed, Feb 5, 2025 at 8:38 PM Joe Damato <jdamato@fastly.com> wrote: > > There are at least two cases where napi_id may not present and the > napi_id should be elided: > > 1. Queues could be created, but napi_enable may not have been called > yet. In this case, there may be a NAPI but it may not have an ID and > output of a napi_id should be elided. > > 2. TX-only NAPIs currently do not have NAPI IDs. If a TX queue happens > to be linked with a TX-only NAPI, elide the NAPI ID from the netlink > output as a NAPI ID of 0 is not useful for users. > > Signed-off-by: Joe Damato <jdamato@fastly.com> Reviewed-by: Eric Dumazet <edumazet@google.com>
On Wed, 5 Feb 2025 19:37:47 +0000 Joe Damato wrote: > There are at least two cases where napi_id may not present and the > napi_id should be elided: > > 1. Queues could be created, but napi_enable may not have been called > yet. In this case, there may be a NAPI but it may not have an ID and > output of a napi_id should be elided. > > 2. TX-only NAPIs currently do not have NAPI IDs. If a TX queue happens > to be linked with a TX-only NAPI, elide the NAPI ID from the netlink > output as a NAPI ID of 0 is not useful for users. > > Signed-off-by: Joe Damato <jdamato@fastly.com> Applied, thanks!
diff --git a/include/net/busy_poll.h b/include/net/busy_poll.h index c39a426ebf52..741fa7754700 100644 --- a/include/net/busy_poll.h +++ b/include/net/busy_poll.h @@ -24,6 +24,11 @@ */ #define MIN_NAPI_ID ((unsigned int)(NR_CPUS + 1)) +static inline bool napi_id_valid(unsigned int napi_id) +{ + return napi_id >= MIN_NAPI_ID; +} + #define BUSY_POLL_BUDGET 8 #ifdef CONFIG_NET_RX_BUSY_POLL diff --git a/net/core/netdev-genl.c b/net/core/netdev-genl.c index 715f85c6b62e..b5c4e42351e6 100644 --- a/net/core/netdev-genl.c +++ b/net/core/netdev-genl.c @@ -364,6 +364,13 @@ int netdev_nl_napi_set_doit(struct sk_buff *skb, struct genl_info *info) return err; } +static int nla_put_napi_id(struct sk_buff *skb, const struct napi_struct *napi) +{ + if (napi && napi_id_valid(napi->napi_id)) + return nla_put_u32(skb, NETDEV_A_QUEUE_NAPI_ID, napi->napi_id); + return 0; +} + static int netdev_nl_queue_fill_one(struct sk_buff *rsp, struct net_device *netdev, u32 q_idx, u32 q_type, const struct genl_info *info) @@ -385,8 +392,7 @@ netdev_nl_queue_fill_one(struct sk_buff *rsp, struct net_device *netdev, switch (q_type) { case NETDEV_QUEUE_TYPE_RX: rxq = __netif_get_rx_queue(netdev, q_idx); - if (rxq->napi && nla_put_u32(rsp, NETDEV_A_QUEUE_NAPI_ID, - rxq->napi->napi_id)) + if (nla_put_napi_id(rsp, rxq->napi)) goto nla_put_failure; binding = rxq->mp_params.mp_priv; @@ -397,8 +403,7 @@ netdev_nl_queue_fill_one(struct sk_buff *rsp, struct net_device *netdev, break; case NETDEV_QUEUE_TYPE_TX: txq = netdev_get_tx_queue(netdev, q_idx); - if (txq->napi && nla_put_u32(rsp, NETDEV_A_QUEUE_NAPI_ID, - txq->napi->napi_id)) + if (nla_put_napi_id(rsp, txq->napi)) goto nla_put_failure; }
There are at least two cases where napi_id may not present and the napi_id should be elided: 1. Queues could be created, but napi_enable may not have been called yet. In this case, there may be a NAPI but it may not have an ID and output of a napi_id should be elided. 2. TX-only NAPIs currently do not have NAPI IDs. If a TX queue happens to be linked with a TX-only NAPI, elide the NAPI ID from the netlink output as a NAPI ID of 0 is not useful for users. Signed-off-by: Joe Damato <jdamato@fastly.com> --- v4: - Fixed a typo in the NETDEV_QUEUE_TYPE_TX case (rxq should've been txq). v3: https://lore.kernel.org/netdev/20250204192724.199209-1-jdamato@fastly.com/ - Took Eric's suggested patch to refactor the code to use helpers, which improves the code quality significantly. v2: https://lore.kernel.org/netdev/20250203191714.155526-1-jdamato@fastly.com/ - Updated to elide NAPI IDs for RX queues which may have not called napi_enable yet. rfc: https://lore.kernel.org/lkml/20250128163038.429864-1-jdamato@fastly.com/ include/net/busy_poll.h | 5 +++++ net/core/netdev-genl.c | 13 +++++++++---- 2 files changed, 14 insertions(+), 4 deletions(-) base-commit: 0bea93fdbaf8675b7e8124bdcaf51497dcc8bcfa