summaryrefslogtreecommitdiff
path: root/tools/testing/selftests/net/lib/py/utils.py
diff options
context:
space:
mode:
authorEric Dumazet <edumazet@google.com>2024-12-15 20:56:29 +0300
committerJakub Kicinski <kuba@kernel.org>2024-12-18 06:37:48 +0300
commita853c609504e2d1d83e71285e3622fda1f1451d8 (patch)
treeb6d9cd4e2333cf6afc87beb0bc7675bc73777006 /tools/testing/selftests/net/lib/py/utils.py
parent50b362f21d6c10b0f7939c1482c6a1b43da82f1a (diff)
downloadlinux-a853c609504e2d1d83e71285e3622fda1f1451d8.tar.xz
inetpeer: do not get a refcount in inet_getpeer()
All inet_getpeer() callers except ip4_frag_init() don't need to acquire a permanent refcount on the inetpeer. They can switch to full RCU protection. Move the refcount_inc_not_zero() into ip4_frag_init(), so that all the other callers no longer have to perform a pair of expensive atomic operations on a possibly contended cache line. inet_putpeer() no longer needs to be exported. After this patch, my DUT can receive 8,400,000 UDP packets per second targeting closed ports, using 50% less cpu cycles than before. Also change two calls to l3mdev_master_ifindex() by l3mdev_master_ifindex_rcu() (Ido ideas) Fixes: 8c2bd38b95f7 ("icmp: change the order of rate limits") Signed-off-by: Eric Dumazet <edumazet@google.com> Link: https://patch.msgid.link/20241215175629.1248773-5-edumazet@google.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Diffstat (limited to 'tools/testing/selftests/net/lib/py/utils.py')
0 files changed, 0 insertions, 0 deletions