* [dpdk-dev] [Bug 278] bond_ethdev_rx_burst multithread access bond's different queue will crash
@ 2019-05-15 6:18 bugzilla
2019-05-15 6:18 ` bugzilla
0 siblings, 1 reply; 2+ messages in thread
From: bugzilla @ 2019-05-15 6:18 UTC (permalink / raw)
To: dev
https://bugs.dpdk.org/show_bug.cgi?id=278
Bug ID: 278
Summary: bond_ethdev_rx_burst multithread access bond's
different queue will crash
Product: DPDK
Version: 18.11
Hardware: All
OS: All
Status: CONFIRMED
Severity: normal
Priority: Normal
Component: ethdev
Assignee: dev@dpdk.org
Reporter: wavespider@sina.com
Target Milestone: ---
when multi-thread access net-bonding interface's different queue, calling
function
below will result in crash because of "out of array boundary":
rte_eth_rx_burst
-> bond_ethdev_rx_burst
one thread loop over and update internals->active_slave:
if (++internals->active_slave >= slave_count)
internals->active_slave = 0;
another thread just read internals->active_slave, maybe equal slave_count:
active_slave = internals->active_slave;
num_rx_slave =
rte_eth_rx_burst(internals->active_slaves[active_slave], /*Here out of
boundary*/
bd_rx_q->queue_id,
bufs + num_rx_total, nb_pkts);
default, internals->active_slaves[active_slave] maybe the value is 0, then call
rte_eth_rx_burst function:
port-id == 0 and queue_id = 20 (example):
when port-id == 0's interface have not enough queue-number(<20), then will
result in crash ( dev->data->rx_queues[queue_id] == NULL pointer access ).
--
You are receiving this mail because:
You are the assignee for the bug.
^ permalink raw reply [flat|nested] 2+ messages in thread* [dpdk-dev] [Bug 278] bond_ethdev_rx_burst multithread access bond's different queue will crash
2019-05-15 6:18 [dpdk-dev] [Bug 278] bond_ethdev_rx_burst multithread access bond's different queue will crash bugzilla
@ 2019-05-15 6:18 ` bugzilla
0 siblings, 0 replies; 2+ messages in thread
From: bugzilla @ 2019-05-15 6:18 UTC (permalink / raw)
To: dev
https://bugs.dpdk.org/show_bug.cgi?id=278
Bug ID: 278
Summary: bond_ethdev_rx_burst multithread access bond's
different queue will crash
Product: DPDK
Version: 18.11
Hardware: All
OS: All
Status: CONFIRMED
Severity: normal
Priority: Normal
Component: ethdev
Assignee: dev@dpdk.org
Reporter: wavespider@sina.com
Target Milestone: ---
when multi-thread access net-bonding interface's different queue, calling
function
below will result in crash because of "out of array boundary":
rte_eth_rx_burst
-> bond_ethdev_rx_burst
one thread loop over and update internals->active_slave:
if (++internals->active_slave >= slave_count)
internals->active_slave = 0;
another thread just read internals->active_slave, maybe equal slave_count:
active_slave = internals->active_slave;
num_rx_slave =
rte_eth_rx_burst(internals->active_slaves[active_slave], /*Here out of
boundary*/
bd_rx_q->queue_id,
bufs + num_rx_total, nb_pkts);
default, internals->active_slaves[active_slave] maybe the value is 0, then call
rte_eth_rx_burst function:
port-id == 0 and queue_id = 20 (example):
when port-id == 0's interface have not enough queue-number(<20), then will
result in crash ( dev->data->rx_queues[queue_id] == NULL pointer access ).
--
You are receiving this mail because:
You are the assignee for the bug.
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2019-05-15 6:18 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2019-05-15 6:18 [dpdk-dev] [Bug 278] bond_ethdev_rx_burst multithread access bond's different queue will crash bugzilla
2019-05-15 6:18 ` bugzilla
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).