diff options
| author | Felix Fietkau <nbd@nbd.name> | 2022-10-10 11:43:38 +0200 |
|---|---|---|
| committer | Johannes Berg <johannes.berg@intel.com> | 2022-12-01 15:09:10 +0100 |
| commit | 7d360f6061db01830adfdb1eaa3977b19db0c30b (patch) | |
| tree | a5ab0938c568e495556356c813633693e01cb80a /include/net/fq_impl.h | |
| parent | 209d70d34a7fd32e6fb6784f107913b0d18fe501 (diff) | |
wifi: mac80211: add support for restricting netdev features per vif
This can be used to selectively disable feature flags for checksum offload,
scatter/gather or GSO by changing vif->netdev_features.
Removing features from vif->netdev_features does not affect the netdev
features themselves, but instead fixes up skbs in the tx path so that the
offloads are not needed in the driver.
Aside from making it easier to deal with vif type based hardware limitations,
this also makes it possible to optimize performance on hardware without native
GSO support by declaring GSO support in hw->netdev_features and removing it
from vif->netdev_features. This allows mac80211 to handle GSO segmentation
after the sta lookup, but before itxq enqueue, thus reducing the number of
unnecessary sta lookups, as well as some other per-packet processing.
Signed-off-by: Felix Fietkau <nbd@nbd.name>
Link: https://lore.kernel.org/r/20221010094338.78070-1-nbd@nbd.name
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Diffstat (limited to 'include/net/fq_impl.h')
| -rw-r--r-- | include/net/fq_impl.h | 16 |
1 files changed, 10 insertions, 6 deletions
diff --git a/include/net/fq_impl.h b/include/net/fq_impl.h index 524b510f1c68..9467e33dfb36 100644 --- a/include/net/fq_impl.h +++ b/include/net/fq_impl.h @@ -200,6 +200,7 @@ static void fq_tin_enqueue(struct fq *fq, fq_skb_free_t free_func) { struct fq_flow *flow; + struct sk_buff *next; bool oom; lockdep_assert_held(&fq->lock); @@ -214,11 +215,15 @@ static void fq_tin_enqueue(struct fq *fq, } flow->tin = tin; - flow->backlog += skb->len; - tin->backlog_bytes += skb->len; - tin->backlog_packets++; - fq->memory_usage += skb->truesize; - fq->backlog++; + skb_list_walk_safe(skb, skb, next) { + skb_mark_not_on_list(skb); + flow->backlog += skb->len; + tin->backlog_bytes += skb->len; + tin->backlog_packets++; + fq->memory_usage += skb->truesize; + fq->backlog++; + __skb_queue_tail(&flow->queue, skb); + } if (list_empty(&flow->flowchain)) { flow->deficit = fq->quantum; @@ -226,7 +231,6 @@ static void fq_tin_enqueue(struct fq *fq, &tin->new_flows); } - __skb_queue_tail(&flow->queue, skb); oom = (fq->memory_usage > fq->memory_limit); while (fq->backlog > fq->limit || oom) { flow = fq_find_fattest_flow(fq); |