[dpdk-dev,v3] net/ixgbe: fix interrupt block issue
Checks
Commit Message
When handle link status change interrupt, interrupt
will be blocked until delayed handler finish, the
duration is at least 1 second, this may cause following
VF to PF mailbox traffic be blocked and sometimes PF
can't ack to VF in time before VF think it's time out.
This patch remove this limitation, during the period
after interrupt handler finish and before delayed handler
only lsc interrupt will be disabled.
Fixes: 0a45657a6794 ("pci: rework interrupt handling")
Signed-off-by: Qi Zhang <qi.z.zhang@intel.com>
---
v3:
- during the period after interrupt handler finish and before delayed handler only lsc interrupt will be disabled.
v2:
- rebase to dpdk-next-net
drivers/net/ixgbe/ixgbe_ethdev.c | 24 ++++++++++++++----------
drivers/net/ixgbe/ixgbe_ethdev.h | 2 ++
2 files changed, 16 insertions(+), 10 deletions(-)
Comments
2017-01-16 14:23, Qi Zhang:
> v3:
> - during the period after interrupt handler finish and before delayed handler only lsc interrupt will be disabled.
>
> v2:
> - rebase to dpdk-next-net
Please, use --in-reply-to to keep revisions in the same thread.
It will help to compare (and will sort things in archives).
On 1/16/2017 7:23 PM, Qi Zhang wrote:
> When handle link status change interrupt, interrupt
> will be blocked until delayed handler finish, the
> duration is at least 1 second, this may cause following
> VF to PF mailbox traffic be blocked and sometimes PF
> can't ack to VF in time before VF think it's time out.
> This patch remove this limitation, during the period
> after interrupt handler finish and before delayed handler
> only lsc interrupt will be disabled.
>
> Fixes: 0a45657a6794 ("pci: rework interrupt handling")
>
> Signed-off-by: Qi Zhang <qi.z.zhang@intel.com>
Applied to dpdk-next-net/master, thanks.
@@ -3762,7 +3762,6 @@ ixgbe_dev_interrupt_action(struct rte_eth_dev *dev,
IXGBE_DEV_PRIVATE_TO_INTR(dev->data->dev_private);
int64_t timeout;
struct rte_eth_link link;
- int intr_enable_delay = false;
struct ixgbe_hw *hw =
IXGBE_DEV_PRIVATE_TO_HW(dev->data->dev_private);
@@ -3795,20 +3794,19 @@ ixgbe_dev_interrupt_action(struct rte_eth_dev *dev,
timeout = IXGBE_LINK_DOWN_CHECK_TIMEOUT;
ixgbe_dev_link_status_print(dev);
-
- intr_enable_delay = true;
- }
-
- if (intr_enable_delay) {
+ intr->mask_original = intr->mask;
+ /* only disable lsc interrupt */
+ intr->mask &= ~IXGBE_EIMS_LSC;
if (rte_eal_alarm_set(timeout * 1000,
ixgbe_dev_interrupt_delayed_handler, (void *)dev) < 0)
PMD_DRV_LOG(ERR, "Error setting alarm");
- } else {
- PMD_DRV_LOG(DEBUG, "enable intr immediately");
- ixgbe_enable_intr(dev);
- rte_intr_enable(intr_handle);
+ else
+ intr->mask = intr->mask_original;
}
+ PMD_DRV_LOG(DEBUG, "enable intr immediately");
+ ixgbe_enable_intr(dev);
+ rte_intr_enable(intr_handle);
return 0;
}
@@ -3839,6 +3837,8 @@ ixgbe_dev_interrupt_delayed_handler(void *param)
IXGBE_DEV_PRIVATE_TO_HW(dev->data->dev_private);
uint32_t eicr;
+ ixgbe_disable_intr(hw);
+
eicr = IXGBE_READ_REG(hw, IXGBE_EICR);
if (eicr & IXGBE_EICR_MAILBOX)
ixgbe_pf_mbx_process(dev);
@@ -3861,6 +3861,10 @@ ixgbe_dev_interrupt_delayed_handler(void *param)
intr->flags &= ~IXGBE_FLAG_MACSEC;
}
+ /* restore original mask */
+ intr->mask = intr->mask_original;
+ intr->mask_original = 0;
+
PMD_DRV_LOG(DEBUG, "enable intr in delayed handler S[%08x]", eicr);
ixgbe_enable_intr(dev);
rte_intr_enable(intr_handle);
@@ -200,6 +200,8 @@ struct ixgbe_hw_fdir_info {
struct ixgbe_interrupt {
uint32_t flags;
uint32_t mask;
+ /*to save original mask during delayed handler */
+ uint32_t mask_original;
};
struct ixgbe_stat_mapping_registers {