diff options
| author | Jakub Kicinski <kuba@kernel.org> | 2024-05-13 14:44:13 -0700 |
|---|---|---|
| committer | Jakub Kicinski <kuba@kernel.org> | 2024-05-13 14:44:13 -0700 |
| commit | e6e43570fd98ac609c903bc91d6db163ba2e82d0 (patch) | |
| tree | 1a1e029a9f9feb22836769c4ff8a4b715fd39fd6 /net/ipv4/udp_offload.c | |
| parent | 9af9b891fc6b44bb336933d63be526ca5cc6ee25 (diff) | |
| parent | bc21faefbe58feff0ae7cae8b52c4145073e2208 (diff) | |
| download | linux-e6e43570fd98ac609c903bc91d6db163ba2e82d0.tar.gz linux-e6e43570fd98ac609c903bc91d6db163ba2e82d0.tar.bz2 linux-e6e43570fd98ac609c903bc91d6db163ba2e82d0.zip | |
Merge branch 'net-gro-remove-network_header-use-move-p-flush-flush_id-calculations-to-l4'
Richard Gobert says:
====================
net: gro: remove network_header use, move p->{flush/flush_id} calculations to L4
The cb fields network_offset and inner_network_offset are used instead of
skb->network_header throughout GRO.
These fields are then leveraged in the next commit to remove flush_id state
from napi_gro_cb, and stateful code in {ipv6,inet}_gro_receive which may be
unnecessarily complicated due to encapsulation support in GRO. These fields
are checked in L4 instead.
3rd patch adds tests for different flush_id flows in GRO.
====================
Link: https://lore.kernel.org/r/20240509190819.2985-1-richardbgobert@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Diffstat (limited to 'net/ipv4/udp_offload.c')
| -rw-r--r-- | net/ipv4/udp_offload.c | 9 |
1 files changed, 1 insertions, 8 deletions
diff --git a/net/ipv4/udp_offload.c b/net/ipv4/udp_offload.c index f6d3c442e260..59448a2dbf2c 100644 --- a/net/ipv4/udp_offload.c +++ b/net/ipv4/udp_offload.c @@ -478,14 +478,7 @@ static struct sk_buff *udp_gro_receive_segment(struct list_head *head, return p; } - flush = NAPI_GRO_CB(p)->flush; - - if (NAPI_GRO_CB(p)->flush_id != 1 || - NAPI_GRO_CB(p)->count != 1 || - !NAPI_GRO_CB(p)->is_atomic) - flush |= NAPI_GRO_CB(p)->flush_id; - else - NAPI_GRO_CB(p)->is_atomic = false; + flush = gro_receive_network_flush(uh, uh2, p); /* Terminate the flow on len mismatch or if it grow "too much". * Under small packet flood GRO count could elsewhere grow a lot |