Hi Wenzhuo, I just noticed _rte_eth_dev_callback_process called by ixgbe_dev_interrupt_delayed_handler. This patch was incorrect, please ignore it. On 09/01/2015 04:09 PM, ??? wrote: > Hi Wenzhuo, > I bond 82599 ports with mode 802.3ad, the bonding port can't work > properly. But bonding with i350 ports works properly. I found the > registered callback function bond_ethdev_lsc_event_callback by > __eth_bond_slave_add_lock_free was not executed when link status > changed. I compared ixgbe driver with igb driver, ixgbe's LSC > interrupt action without _rte_eth_dev_callback_process. > On 09/01/2015 03:42 PM, Lu, Wenzhuo wrote: >> Hi Shelton, >> >>> -----Original Message----- >>> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Shelton Chia >>> Sent: Tuesday, September 1, 2015 2:22 PM >>> To: dev at dpdk.org >>> Cc: letaotor at gmail.com >>> Subject: [dpdk-dev] [PATCH] ixgbe: fix LSC callback >>> >>> add _rte_eth_dev_callback_process to call callback func >> Would you like to give more details about why we need this patch? >> Thanks. >>> Signed-off-by: Shelton Chia <jiaxt at sinogrid.com> >>> --- >>> drivers/net/ixgbe/ixgbe_ethdev.c | 2 ++ >>> 1 file changed, 2 insertions(+) >>> >>> diff --git a/drivers/net/ixgbe/ixgbe_ethdev.c >>> b/drivers/net/ixgbe/ixgbe_ethdev.c >>> index b8ee1e9..4e4c118 100644 >>> --- a/drivers/net/ixgbe/ixgbe_ethdev.c >>> +++ b/drivers/net/ixgbe/ixgbe_ethdev.c >>> @@ -2775,6 +2775,8 @@ ixgbe_dev_interrupt_action(struct rte_eth_dev >>> *dev) >>> >>> ixgbe_dev_link_status_print(dev); >>> >>> + __rte_eth_dev_callback_process(dev, >>> RTE_ETH_EVENT_INTR_LSC); >>> + >>> intr_enable_delay = true; >>> } >>> >>> -- >>> 2.5.0 >
-- ??? ??????????