From: Manjunath Patil <manjunath.b.patil@oracle.com>
To: santosh.shilimkar@oracle.com
Cc: netdev@vger.kernel.org, linux-rdma@vger.kernel.org,
aruna.ramakrishna@oracle.com, rama.nichanamatlu@oracle.com
Subject: [PATCH 1/1] net/rds: suppress page allocation failure error in recv buffer refill
Date: Wed, 16 Sep 2020 12:08:46 -0700 [thread overview]
Message-ID: <1600283326-30323-1-git-send-email-manjunath.b.patil@oracle.com> (raw)
RDS/IB tries to refill the recv buffer in softirq context using
GFP_NOWAIT flag. However alloc failure is handled by queueing a work to
refill the recv buffer with GFP_KERNEL flag. This means failure to
allocate with GFP_NOWAIT isn't fatal. Do not print the PAF warnings if
softirq context fails to refill the recv buffer, instead print a one
line warning once a day.
Signed-off-by: Manjunath Patil <manjunath.b.patil@oracle.com>
Reviewed-by: Aruna Ramakrishna <aruna.ramakrishna@oracle.com>
---
net/rds/ib_recv.c | 16 +++++++++++++---
1 file changed, 13 insertions(+), 3 deletions(-)
diff --git a/net/rds/ib_recv.c b/net/rds/ib_recv.c
index 694d411dc72f..38d2894f6bb2 100644
--- a/net/rds/ib_recv.c
+++ b/net/rds/ib_recv.c
@@ -310,8 +310,8 @@ static int rds_ib_recv_refill_one(struct rds_connection *conn,
struct rds_ib_connection *ic = conn->c_transport_data;
struct ib_sge *sge;
int ret = -ENOMEM;
- gfp_t slab_mask = GFP_NOWAIT;
- gfp_t page_mask = GFP_NOWAIT;
+ gfp_t slab_mask = gfp;
+ gfp_t page_mask = gfp;
if (gfp & __GFP_DIRECT_RECLAIM) {
slab_mask = GFP_KERNEL;
@@ -406,6 +406,16 @@ void rds_ib_recv_refill(struct rds_connection *conn, int prefill, gfp_t gfp)
recv = &ic->i_recvs[pos];
ret = rds_ib_recv_refill_one(conn, recv, gfp);
if (ret) {
+ static unsigned long warn_time;
+ /* warn max once per day. This should be enough to
+ * warn users about low mem situation.
+ */
+ if (printk_timed_ratelimit(&warn_time,
+ 24 * 60 * 60 * 1000))
+ pr_warn("RDS/IB: failed to refill recv buffer for <%pI6c,%pI6c,%d>, waking worker\n",
+ &conn->c_laddr, &conn->c_faddr,
+ conn->c_tos);
+
must_wake = true;
break;
}
@@ -1020,7 +1030,7 @@ void rds_ib_recv_cqe_handler(struct rds_ib_connection *ic,
rds_ib_stats_inc(s_ib_rx_ring_empty);
if (rds_ib_ring_low(&ic->i_recv_ring)) {
- rds_ib_recv_refill(conn, 0, GFP_NOWAIT);
+ rds_ib_recv_refill(conn, 0, GFP_NOWAIT | __GFP_NOWARN);
rds_ib_stats_inc(s_ib_rx_refill_from_cq);
}
}
--
2.27.0.112.g101b320
next reply other threads:[~2020-09-16 19:10 UTC|newest]
Thread overview: 10+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-09-16 19:08 Manjunath Patil [this message]
2020-09-16 19:27 ` [PATCH 1/1] net/rds: suppress page allocation failure error in recv buffer refill santosh.shilimkar
2020-09-16 21:15 ` Manjunath Patil
2020-09-16 21:25 ` santosh.shilimkar
2020-09-16 21:35 ` Manjunath Patil
-- strict thread matches above, loose matches on Subject: below --
2020-10-02 20:05 Manjunath Patil
2020-10-02 20:10 ` santosh.shilimkar
2020-10-02 20:23 ` Manjunath Patil
2020-10-04 0:26 ` David Miller
2020-10-05 21:39 ` Manjunath Patil
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=1600283326-30323-1-git-send-email-manjunath.b.patil@oracle.com \
--to=manjunath.b.patil@oracle.com \
--cc=aruna.ramakrishna@oracle.com \
--cc=linux-rdma@vger.kernel.org \
--cc=netdev@vger.kernel.org \
--cc=rama.nichanamatlu@oracle.com \
--cc=santosh.shilimkar@oracle.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).