Commit 05dc895b authored Jun 30, 2021 by Ziye Yang Committed by Jim Harris Jul 01, 2021

sock/posix: Fix the coredump when removing the sock from socks_has_data list



When entering the if case to order the list, there is bug should be
fixed. The original code does not address this.

The way this happens is when there is a connection left in the socks_with_data list
between polls and there are enough new events detected that it would exceed the
maximal number of events. A connection is left on this list between polls if it isn't
fully drained via reads by the upper layer on each poll loop.

Currently, the maximal socket event num is 32. Then we did not hit this issue
in our normal test cases. But when you use NVMe-oF tcp target to test which is
described in #2105, there are more than 32 active sockets, and it exceeeds
the maximal num of events of polling (32), so we will trigger this issue.

Fixes issue #2015

Change-Id: I9384476fdba8826f5fe55a5d2594e3f4ed3832ba
Signed-off-by: Ziye Yang <ziye.yang@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/8541


Community-CI: Mellanox Build Bot
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>

parent 358b84b4

module/sock/posix/posix.c

+3 −1

Original line number	Diff line number	Diff line
		@@ -1499,7 +1499,6 @@ posix_sock_group_impl_poll(struct spdk_sock_group_impl *_group, int max_events,

		/* Break the link between C and D */
		pc->link.tqe_next = NULL;
		pd->link.tqe_prev = NULL;

		/* Connect F to A */
		pf->link.tqe_next = pa;
		@@ -1508,6 +1507,9 @@ posix_sock_group_impl_poll(struct spdk_sock_group_impl *_group, int max_events,
		/* Fix up the list first/last pointers */
		group->socks_with_data.tqh_first = pd;
		group->socks_with_data.tqh_last = &pc->link.tqe_next;

		/* D is in front of the list, make tqe prev pointer point to the head of list */
		pd->link.tqe_prev = &group->socks_with_data.tqh_first;
		}

		return num_events;