request sockets are lockless, __tcp_oow_rate_limited() could be called
on the same object from different cpus. This is harmless.

Add READ_ONCE()/WRITE_ONCE() annotations to avoid a KCSAN report.

Fixes: 4ce7e93cb3fe ("tcp: rate limit ACK sent by SYN_RECV request sockets")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Remove ->sendpage() and ->sendpage_locked().  sendmsg() with
MSG_SPLICE_PAGES should be used instead.  This allows multiple pages and
multipage folios to be passed through.

Signed-off-by: David Howells <dhowells@redhat.com>
Acked-by: Marc Kleine-Budde <mkl@pengutronix.de> # for net/can
cc: Jens Axboe <axboe@kernel.dk>
cc: Matthew Wilcox <willy@infradead.org>
cc: linux-afs@lists.infradead.org
cc: mptcp@lists.linux.dev
cc: rds-devel@oss.oracle.com
cc: tipc-discussion@lists.sourceforge.net
cc: virtualization@lists.linux-foundation.org
Link: https://lore.kernel.org/r/20230623225513.2732256-16-dhowells@redhat.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

As MSG_SENDPAGE_NOTLAST is being phased out along with sendpage(), don't
use it further in than the sendpage methods, but rather translate it to
MSG_MORE and use that instead.

Signed-off-by: David Howells <dhowells@redhat.com>
cc: Willem de Bruijn <willemdebruijn.kernel@gmail.com>
cc: Bernard Metzler <bmt@zurich.ibm.com>
cc: Jason Gunthorpe <jgg@ziepe.ca>
cc: Leon Romanovsky <leon@kernel.org>
cc: John Fastabend <john.fastabend@gmail.com>
cc: Jakub Sitnicki <jakub@cloudflare.com>
cc: David Ahern <dsahern@kernel.org>
cc: Karsten Graul <kgraul@linux.ibm.com>
cc: Wenjia Zhang <wenjia@linux.ibm.com>
cc: Jan Karcher <jaka@linux.ibm.com>
cc: "D. Wythe" <alibuda@linux.alibaba.com>
cc: Tony Lu <tonylu@linux.alibaba.com>
cc: Wen Gu <guwen@linux.alibaba.com>
cc: Boris Pismenny <borisp@nvidia.com>
cc: Steffen Klassert <steffen.klassert@secunet.com>
cc: Herbert Xu <herbert@gondor.apana.org.au>
Link: https://lore.kernel.org/r/20230623225513.2732256-2-dhowells@redhat.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

Even when tcp_splice_read() reads all it was asked for, for blocking
sockets it'll release and immediately regrab the socket lock, loop
around and break on the while check.

Check tss.len right after we adjust it, and return if we're done.
That saves us one release_sock(); lock_sock(); pair per successful
blocking splice read.

Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Link: https://lore.kernel.org/r/80736a2cc6d478c383ea565ba825eaf4d1abd876.1687523671.git.asml.silence@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

Daniel Borkmann says:

====================
pull-request: bpf-next 2023-06-23

We've added 49 non-merge commits during the last 24 day(s) which contain
a total of 70 files changed, 1935 insertions(+), 442 deletions(-).

The main changes are:

1) Extend bpf_fib_lookup helper to allow passing the route table ID,
   from Louis DeLosSantos.

2) Fix regsafe() in verifier to call check_ids() for scalar registers,
   from Eduard Zingerman.

3) Extend the set of cpumask kfuncs with bpf_cpumask_first_and()
   and a rework of bpf_cpumask_any*() kfuncs. Additionally,
   add selftests, from David Vernet.

4) Fix socket lookup BPF helpers for tc/XDP to respect VRF bindings,
   from Gilad Sever.

5) Change bpf_link_put() to use workqueue unconditionally to fix it
   under PREEMPT_RT, from Sebastian Andrzej Siewior.

6) Follow-ups to address issues in the bpf_refcount shared ownership
   implementation, from Dave Marchevsky.

7) A few general refactorings to BPF map and program creation permissions
   checks which were part of the BPF token series, from Andrii Nakryiko.

8) Various fixes for benchmark framework and add a new benchmark
   for BPF memory allocator to BPF selftests, from Hou Tao.

9) Documentation improvements around iterators and trusted pointers,
   from Anton Protopopov.

10) Small cleanup in verifier to improve allocated object check,
    from Daniel T. Lee.

11) Improve performance of bpf_xdp_pointer() by avoiding access
    to shared_info when XDP packet does not have frags,
    from Jesper Dangaard Brouer.

12) Silence a harmless syzbot-reported warning in btf_type_id_size(),
    from Yonghong Song.

13) Remove duplicate bpfilter_umh_cleanup in favor of umd_cleanup_helper,
    from Jarkko Sakkinen.

14) Fix BPF selftests build for resolve_btfids under custom HOSTCFLAGS,
    from Viktor Malik.

* tag 'for-netdev' of https://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf-next: (49 commits)
  bpf, docs: Document existing macros instead of deprecated
  bpf, docs: BPF Iterator Document
  selftests/bpf: Fix compilation failure for prog vrf_socket_lookup
  selftests/bpf: Add vrf_socket_lookup tests
  bpf: Fix bpf socket lookup from tc/xdp to respect socket VRF bindings
  bpf: Call __bpf_sk_lookup()/__bpf_skc_lookup() directly via TC hookpoint
  bpf: Factor out socket lookup functions for the TC hookpoint.
  selftests/bpf: Set the default value of consumer_cnt as 0
  selftests/bpf: Ensure that next_cpu() returns a valid CPU number
  selftests/bpf: Output the correct error code for pthread APIs
  selftests/bpf: Use producer_cnt to allocate local counter array
  xsk: Remove unused inline function xsk_buff_discard()
  bpf: Keep BPF_PROG_LOAD permission checks clear of validations
  bpf: Centralize permissions checks for all BPF map types
  bpf: Inline map creation logic in map_create() function
  bpf: Move unprivileged checks into map_create() and bpf_prog_load()
  bpf: Remove in_atomic() from bpf_link_put().
  selftests/bpf: Verify that check_ids() is used for scalars in regsafe()
  bpf: Verify scalar ids mapping in regsafe() using check_ids()
  selftests/bpf: Check if mark_chain_precision() follows scalar ids
  ...
====================

Link: https://lore.kernel.org/r/20230623211256.8409-1-daniel@iogearbox.net
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

Spell "transmissions" properly.

Found by searching for keyword "tranm".

Signed-off-by: Yueh-Shun Li <shamrocklee@posteo.net>
Link: https://lore.kernel.org/r/20230622012627.15050-6-shamrocklee@posteo.net
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

Cross-merge networking fixes after downstream PR.

Conflicts:

tools/testing/selftests/net/fcnal-test.sh
  d7a2fc1437f7 ("selftests: net: fcnal-test: check if FIPS mode is enabled")
  dd017c72dde6 ("selftests: fcnal: Test SO_DONTROUTE on TCP sockets.")
https://lore.kernel.org/all/5007b52c-dd16-dbf6-8d64-b9701bfa498b@tessares.net/
https://lore.kernel.org/all/20230619105427.4a0df9b3@canb.auug.org.au/

No adjacent changes.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>

If there is no net-memcg associated with the sock, don't bother
calculating its memory usage for charge.

Signed-off-by: Abel Wu <wuyun.abel@bytedance.com>
Link: https://lore.kernel.org/r/20230620092712.16217-1-wuyun.abel@bytedance.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

ipsec-2023-06-20

Per-VMA locking allows us to lock a struct vm_area_struct without
taking the process-wide mmap lock in read mode.

Consider a process workload where the mmap lock is taken constantly in
write mode. In this scenario, all zerocopy receives are periodically
blocked during that period of time - though in principle, the memory
ranges being used by TCP are not touched by the operations that need
the mmap write lock. This results in performance degradation.

Now consider another workload where the mmap lock is never taken in
write mode, but there are many TCP connections using receive zerocopy
that are concurrently receiving. These connections all take the mmap
lock in read mode, but this does induce a lot of contention and atomic
ops for this process-wide lock. This results in additional CPU
overhead caused by contending on the cache line for this lock.

However, with per-vma locking, both of these problems can be avoided.

As a test, I ran an RPC-style request/response workload with 4KB
payloads and receive zerocopy enabled, with 100 simultaneous TCP
connections. I measured perf cycles within the
find_tcp_vma/mmap_read_lock/mmap_read_unlock codepath, with and
without per-vma locking enabled.

When using process-wide mmap semaphore read locking, about 1% of
measured perf cycles were within this path. With per-VMA locking, this
value dropped to about 0.45%.

Signed-off-by: Arjun Roy <arjunroy@google.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>