lkml.org 
[lkml]   [2018]   [Jul]   [15]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
Patch in this message
/
From
Subject[PATCH v2] datagram: return from __skb_recv_datagram() as soon as possible
Date
We got a soft lockup in a heavy busy cloud server where RIP is
at _raw_spin_unlock_irqrestore+0x1b/0x40:
[] finish_wait+0x56/0x70
[] __skb_recv_datagram+0x3fb/0x5a0
[] ? datagram_poll+0x100/0x100
[] skb_recv_datagram+0x41/0x60
[] netlink_recvmsg+0x62/0x450
[] sock_recvmsg+0xbf/0x100
[] ? futex_wait+0x193/0x280
[] ? finish_task_switch+0x108/0x170
[] SYSC_recvfrom+0xe8/0x160
[] ? __schedule+0x3c8/0x990
[] SyS_recvfrom+0xe/0x10
[] system_call_fastpath+0x16/0x1b

In fact, a mistake exists in __skb_recv_datagram(). For example,
if a datagram come in persistently after go through the socket
queue, then __skb_wait_for_more_packets() will find out that the
last peeked skb is not the real last one, so it return 0. this
results in long time outer loop, and can trigger soft lockup.

So this patch changes the loop condition to prevent soft lockup.

Signed-off-by: Baoyou Xie <baoyou.xie@gmail.com>
---
net/core/datagram.c | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/net/core/datagram.c b/net/core/datagram.c
index 9938952..76c1001 100644
--- a/net/core/datagram.c
+++ b/net/core/datagram.c
@@ -295,9 +295,11 @@ struct sk_buff *__skb_recv_datagram(struct sock *sk, unsigned int flags,
int *peeked, int *off, int *err)
{
struct sk_buff *skb, *last;
+ unsigned long expire;
long timeo;

timeo = sock_rcvtimeo(sk, flags & MSG_DONTWAIT);
+ expire = jiffies + timeo;

do {
skb = __skb_try_recv_datagram(sk, flags, destructor, peeked,
@@ -307,7 +309,7 @@ struct sk_buff *__skb_recv_datagram(struct sock *sk, unsigned int flags,

if (*err != -EAGAIN)
break;
- } while (timeo &&
+ } while (time_before(jiffies, expire) &&
!__skb_wait_for_more_packets(sk, err, &timeo, last));

return NULL;
--
2.7.4
\
 
 \ /
  Last update: 2018-07-15 22:07    [W:0.053 / U:3.552 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site