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>,
"Jakub Kicinski" <kuba@kernel.org>,
"Toke Høiland-Jørgensen" <toke@redhat.com>,
"Simon Horman" <simon.horman@netronome.com>,
oss-drivers@netronome.com
Subject: [PATCH bpf-next v3 11/16] nfp: remove rcu_read_lock() around XDP program invocation
Date: Thu, 17 Jun 2021 23:27:43 +0200 [thread overview]
Message-ID: <20210617212748.32456-12-toke@redhat.com> (raw)
In-Reply-To: <20210617212748.32456-1-toke@redhat.com>
The nfp 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.
While this is not actually an issue for the nfp driver because it doesn't
support XDP_REDIRECT (and thus doesn't call xdp_do_flush()), the
rcu_read_lock() is still unneeded. And 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: Simon Horman <simon.horman@netronome.com>
Cc: oss-drivers@netronome.com
Reviewed-by: Simon Horman <simon.horman@netronome.com>
Signed-off-by: Toke Høiland-Jørgensen <toke@redhat.com>
---
drivers/net/ethernet/netronome/nfp/nfp_net_common.c | 6 ++++--
1 file changed, 4 insertions(+), 2 deletions(-)
diff --git a/drivers/net/ethernet/netronome/nfp/nfp_net_common.c b/drivers/net/ethernet/netronome/nfp/nfp_net_common.c
index eeb30680b4dc..a3d59abed6ae 100644
--- a/drivers/net/ethernet/netronome/nfp/nfp_net_common.c
+++ b/drivers/net/ethernet/netronome/nfp/nfp_net_common.c
@@ -1819,7 +1819,6 @@ static int nfp_net_rx(struct nfp_net_rx_ring *rx_ring, int budget)
struct xdp_buff xdp;
int idx;
- rcu_read_lock();
xdp_prog = READ_ONCE(dp->xdp_prog);
true_bufsz = xdp_prog ? PAGE_SIZE : dp->fl_bufsz;
xdp_init_buff(&xdp, PAGE_SIZE - NFP_NET_RX_BUF_HEADROOM,
@@ -1919,6 +1918,10 @@ static int nfp_net_rx(struct nfp_net_rx_ring *rx_ring, int budget)
pkt_off - NFP_NET_RX_BUF_HEADROOM,
pkt_len, true);
+ /* This code is invoked within a single NAPI poll cycle
+ * and thus under local_bh_disable(), which provides the
+ * needed RCU protection.
+ */
act = bpf_prog_run_xdp(xdp_prog, &xdp);
pkt_len = xdp.data_end - xdp.data;
@@ -2036,7 +2039,6 @@ static int nfp_net_rx(struct nfp_net_rx_ring *rx_ring, int budget)
if (!nfp_net_xdp_complete(tx_ring))
pkts_polled = budget;
}
- rcu_read_unlock();
return pkts_polled;
}
--
2.32.0
next prev parent reply other threads:[~2021-06-17 21:28 UTC|newest]
Thread overview: 29+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-06-17 21:27 [PATCH bpf-next v3 00/16] Clean up and document RCU-based object protection for XDP_REDIRECT Toke Høiland-Jørgensen
2021-06-17 21:27 ` [PATCH bpf-next v3 01/16] rcu: Create an unrcu_pointer() to remove __rcu from a pointer Toke Høiland-Jørgensen
2021-06-17 21:27 ` [PATCH bpf-next v3 02/16] bpf: allow RCU-protected lookups to happen from bh context Toke Høiland-Jørgensen
2021-06-17 21:27 ` [PATCH bpf-next v3 03/16] xdp: add proper __rcu annotations to redirect map entries Toke Høiland-Jørgensen
2021-06-18 4:55 ` Martin KaFai Lau
2021-06-18 23:27 ` Daniel Borkmann
2021-06-21 21:39 ` Toke Høiland-Jørgensen
2021-06-21 22:15 ` Daniel Borkmann
2021-06-21 22:35 ` Toke Høiland-Jørgensen
2021-06-22 8:50 ` Daniel Borkmann
2021-06-22 13:55 ` Toke Høiland-Jørgensen
2021-06-22 20:26 ` Paul E. McKenney
2021-06-22 21:48 ` Toke Høiland-Jørgensen
2021-06-22 23:19 ` Paul E. McKenney
2021-06-23 10:55 ` Toke Høiland-Jørgensen
2021-06-17 21:27 ` [PATCH bpf-next v3 04/16] ena: remove rcu_read_lock() around XDP program invocation Toke Høiland-Jørgensen
2021-06-17 21:27 ` [PATCH bpf-next v3 05/16] bnxt: " Toke Høiland-Jørgensen
2021-06-17 21:27 ` [PATCH bpf-next v3 06/16] thunderx: " Toke Høiland-Jørgensen
2021-06-17 21:27 ` [PATCH bpf-next v3 07/16] freescale: " Toke Høiland-Jørgensen
2021-06-17 21:27 ` [PATCH bpf-next v3 08/16] net: intel: " Toke Høiland-Jørgensen
2021-06-17 21:27 ` [PATCH bpf-next v3 09/16] marvell: " Toke Høiland-Jørgensen
2021-06-17 21:27 ` [PATCH bpf-next v3 10/16] mlx4: " Toke Høiland-Jørgensen
2021-06-17 21:27 ` Toke Høiland-Jørgensen [this message]
2021-06-17 21:27 ` [PATCH bpf-next v3 12/16] qede: " Toke Høiland-Jørgensen
2021-06-17 21:27 ` [PATCH bpf-next v3 13/16] sfc: " Toke Høiland-Jørgensen
2021-06-17 21:27 ` [PATCH bpf-next v3 14/16] netsec: " Toke Høiland-Jørgensen
2021-06-17 21:27 ` [PATCH bpf-next v3 15/16] stmmac: " Toke Høiland-Jørgensen
2021-06-18 9:47 ` Wong Vee Khee
2021-06-17 21:27 ` [PATCH bpf-next v3 16/16] net: ti: " Toke Høiland-Jørgensen
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=20210617212748.32456-12-toke@redhat.com \
--to=toke@redhat.com \
--cc=bpf@vger.kernel.org \
--cc=brouer@redhat.com \
--cc=kafai@fb.com \
--cc=kuba@kernel.org \
--cc=liuhangbin@gmail.com \
--cc=magnus.karlsson@gmail.com \
--cc=netdev@vger.kernel.org \
--cc=oss-drivers@netronome.com \
--cc=paulmck@kernel.org \
--cc=simon.horman@netronome.com \
/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).