From: "Toke Høiland-Jørgensen" <toke@redhat.com>
To: bpf@vger.kernel.org, netdev@vger.kernel.org
Cc: "Martin KaFai Lau" <kafai@fb.com>,
"Hangbin Liu" <liuhangbin@gmail.com>,
"Jesper Dangaard Brouer" <brouer@redhat.com>,
"Magnus Karlsson" <magnus.karlsson@gmail.com>,
"Paul E . McKenney" <paulmck@kernel.org>,
"Toke Høiland-Jørgensen" <toke@redhat.com>,
"Edward Cree" <ecree.xilinx@gmail.com>,
"Martin Habets" <habetsm.xilinx@gmail.com>
Subject: [PATCH bpf-next 14/17] sfc: remove rcu_read_lock() around XDP program invocation
Date: Wed, 9 Jun 2021 12:33:23 +0200 [thread overview]
Message-ID: <20210609103326.278782-15-toke@redhat.com> (raw)
In-Reply-To: <20210609103326.278782-1-toke@redhat.com>
The sfc driver has rcu_read_lock()/rcu_read_unlock() pairs around XDP
program invocations. However, the actual lifetime of the objects referred
by the XDP program invocation is longer, all the way through to the call to
xdp_do_flush(), making the scope of the rcu_read_lock() too small. This
turns out to be harmless because it all happens in a single NAPI poll
cycle (and thus under local_bh_disable()), but it makes the rcu_read_lock()
misleading.
Rather than extend the scope of the rcu_read_lock(), just get rid of it
entirely. With the addition of RCU annotations to the XDP_REDIRECT map
types that take bh execution into account, lockdep even understands this to
be safe, so there's really no reason to keep it around.
Cc: Edward Cree <ecree.xilinx@gmail.com>
Cc: Martin Habets <habetsm.xilinx@gmail.com>
Signed-off-by: Toke Høiland-Jørgensen <toke@redhat.com>
---
drivers/net/ethernet/sfc/rx.c | 9 ++-------
1 file changed, 2 insertions(+), 7 deletions(-)
diff --git a/drivers/net/ethernet/sfc/rx.c b/drivers/net/ethernet/sfc/rx.c
index 17b8119c48e5..606750938b89 100644
--- a/drivers/net/ethernet/sfc/rx.c
+++ b/drivers/net/ethernet/sfc/rx.c
@@ -260,18 +260,14 @@ static bool efx_do_xdp(struct efx_nic *efx, struct efx_channel *channel,
s16 offset;
int err;
- rcu_read_lock();
- xdp_prog = rcu_dereference(efx->xdp_prog);
- if (!xdp_prog) {
- rcu_read_unlock();
+ xdp_prog = rcu_dereference_bh(efx->xdp_prog);
+ if (!xdp_prog)
return true;
- }
rx_queue = efx_channel_get_rx_queue(channel);
if (unlikely(channel->rx_pkt_n_frags > 1)) {
/* We can't do XDP on fragmented packets - drop. */
- rcu_read_unlock();
efx_free_rx_buffers(rx_queue, rx_buf,
channel->rx_pkt_n_frags);
if (net_ratelimit())
@@ -296,7 +292,6 @@ static bool efx_do_xdp(struct efx_nic *efx, struct efx_channel *channel,
rx_buf->len, false);
xdp_act = bpf_prog_run_xdp(xdp_prog, &xdp);
- rcu_read_unlock();
offset = (u8 *)xdp.data - *ehp;
--
2.31.1
next prev parent reply other threads:[~2021-06-09 10:38 UTC|newest]
Thread overview: 37+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-06-09 10:33 [PATCH bpf-next 00/17] Clean up and document RCU-based object protection for XDP_REDIRECT Toke Høiland-Jørgensen
2021-06-09 10:33 ` [PATCH bpf-next 01/17] rcu: Create an unrcu_pointer() to remove __rcu from a pointer Toke Høiland-Jørgensen
2021-06-09 10:33 ` [PATCH bpf-next 02/17] bpf: allow RCU-protected lookups to happen from bh context Toke Høiland-Jørgensen
2021-06-10 18:38 ` Alexei Starovoitov
2021-06-10 21:24 ` Daniel Borkmann
2021-06-10 22:27 ` Toke Høiland-Jørgensen
2021-06-10 19:33 ` Martin KaFai Lau
2021-06-09 10:33 ` [PATCH bpf-next 03/17] dev: add rcu_read_lock_bh_held() as a valid check when getting a RCU dev ref Toke Høiland-Jørgensen
2021-06-10 19:37 ` Martin KaFai Lau
2021-06-10 23:05 ` Toke Høiland-Jørgensen
2021-06-09 10:33 ` [PATCH bpf-next 04/17] xdp: add proper __rcu annotations to redirect map entries Toke Høiland-Jørgensen
2021-06-10 21:09 ` Martin KaFai Lau
2021-06-10 23:19 ` Toke Høiland-Jørgensen
2021-06-10 23:32 ` Martin KaFai Lau
2021-06-10 23:41 ` Toke Høiland-Jørgensen
2021-06-09 10:33 ` [PATCH bpf-next 05/17] ena: remove rcu_read_lock() around XDP program invocation Toke Høiland-Jørgensen
2021-06-09 13:57 ` Paul E. McKenney
2021-06-09 10:33 ` [PATCH bpf-next 06/17] bnxt: " Toke Høiland-Jørgensen
2021-06-09 13:58 ` Paul E. McKenney
2021-06-10 8:47 ` Toke Høiland-Jørgensen
2021-06-09 10:33 ` [PATCH bpf-next 07/17] thunderx: " Toke Høiland-Jørgensen
2021-06-09 10:33 ` [PATCH bpf-next 08/17] freescale: " Toke Høiland-Jørgensen
2021-06-09 10:33 ` [PATCH bpf-next 09/17] net: intel: " Toke Høiland-Jørgensen
2021-06-09 10:33 ` [PATCH bpf-next 10/17] marvell: " Toke Høiland-Jørgensen
2021-06-09 10:33 ` [PATCH bpf-next 11/17] mlx4: " Toke Høiland-Jørgensen
2021-06-10 7:10 ` Tariq Toukan
2021-06-09 10:33 ` [PATCH bpf-next 12/17] nfp: " Toke Høiland-Jørgensen
2021-06-11 16:30 ` Simon Horman
2021-06-09 10:33 ` [PATCH bpf-next 13/17] qede: " Toke Høiland-Jørgensen
2021-06-09 10:33 ` Toke Høiland-Jørgensen [this message]
2021-06-09 12:15 ` [PATCH bpf-next 14/17] sfc: " Edward Cree
2021-06-09 10:33 ` [PATCH bpf-next 15/17] netsec: " Toke Høiland-Jørgensen
2021-06-10 5:30 ` Ilias Apalodimas
2021-06-09 10:33 ` [PATCH bpf-next 16/17] stmmac: " Toke Høiland-Jørgensen
2021-06-09 10:33 ` [PATCH bpf-next 17/17] net: ti: " Toke Høiland-Jørgensen
2021-06-09 17:04 ` Grygorii Strashko
2021-06-10 0:18 ` [PATCH bpf-next 00/17] Clean up and document RCU-based object protection for XDP_REDIRECT Yonghong Song
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20210609103326.278782-15-toke@redhat.com \
--to=toke@redhat.com \
--cc=bpf@vger.kernel.org \
--cc=brouer@redhat.com \
--cc=ecree.xilinx@gmail.com \
--cc=habetsm.xilinx@gmail.com \
--cc=kafai@fb.com \
--cc=liuhangbin@gmail.com \
--cc=magnus.karlsson@gmail.com \
--cc=netdev@vger.kernel.org \
--cc=paulmck@kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).