[v3] net/iavf: fix VF startup coredump
Checks
Commit Message
When the vf starts to request multiple queues, the pf sends a reset
command to the vf. During the reset process, adminq sends an abnormal
message to pf for an unknown reason, and the resource request fails
resulting in a coredump.
This patch fixes the issue by checking the reset state before resetting.
v2: fix windows compilation errors
v3: add fix tag
Fixes: 3e6a5d2d310a ("net/iavf: add devargs to enable VF auto-reset")
Cc: stable@dpdk.org
Signed-off-by: Shiyang He <shiyangx.he@intel.com>
---
drivers/net/iavf/iavf.h | 3 ++-
drivers/net/iavf/iavf_ethdev.c | 48 +++++++++++++++++++++++++++-------
2 files changed, 40 insertions(+), 11 deletions(-)
Comments
> -----Original Message-----
> From: He, ShiyangX <shiyangx.he@intel.com>
> Sent: Wednesday, January 3, 2024 6:42 PM
> To: dev@dpdk.org
> Cc: Yang, Qiming <qiming.yang@intel.com>; He, ShiyangX
> <shiyangx.he@intel.com>; stable@dpdk.org; Wu, Jingjing
> <jingjing.wu@intel.com>; Xing, Beilei <beilei.xing@intel.com>; Wang, Liang-
> min <liang-min.wang@intel.com>; Zhang, Qi Z <qi.z.zhang@intel.com>
> Subject: [PATCH v3] net/iavf: fix VF startup coredump
>
> When the vf starts to request multiple queues, the pf sends a reset command
> to the vf. During the reset process, adminq sends an abnormal message to pf
> for an unknown reason, and the resource request fails resulting in a
> coredump.
>
> This patch fixes the issue by checking the reset state before resetting.
>
> v2: fix windows compilation errors
> v3: add fix tag
>
> Fixes: 3e6a5d2d310a ("net/iavf: add devargs to enable VF auto-reset")
> Cc: stable@dpdk.org
>
> Signed-off-by: Shiyang He <shiyangx.he@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
Applied to dpdk-next-net-intel.
Thanks
Qi
@@ -19,6 +19,7 @@
#define IAVF_AQ_LEN 32
#define IAVF_AQ_BUF_SZ 4096
#define IAVF_RESET_WAIT_CNT 500
+#define IAVF_RESET_DETECTED_CNT 500
#define IAVF_BUF_SIZE_MIN 1024
#define IAVF_FRAME_SIZE_MAX 9728
#define IAVF_QUEUE_BASE_ADDR_UNIT 128
@@ -511,5 +512,5 @@ int iavf_flow_sub_check(struct iavf_adapter *adapter,
struct iavf_fsub_conf *filter);
void iavf_dev_watchdog_enable(struct iavf_adapter *adapter);
void iavf_dev_watchdog_disable(struct iavf_adapter *adapter);
-int iavf_handle_hw_reset(struct rte_eth_dev *dev);
+void iavf_handle_hw_reset(struct rte_eth_dev *dev);
#endif /* _IAVF_ETHDEV_H_ */
@@ -1086,9 +1086,6 @@ iavf_dev_stop(struct rte_eth_dev *dev)
PMD_INIT_FUNC_TRACE();
- if (vf->vf_reset)
- return 0;
-
if (adapter->closed)
return -1;
@@ -2949,7 +2946,6 @@ iavf_dev_reset(struct rte_eth_dev *dev)
{
int ret;
struct iavf_hw *hw = IAVF_DEV_PRIVATE_TO_HW(dev->data->dev_private);
- struct iavf_info *vf = IAVF_DEV_PRIVATE_TO_VF(dev->data->dev_private);
/*
* Check whether the VF reset has been done and inform application,
@@ -2961,7 +2957,6 @@ iavf_dev_reset(struct rte_eth_dev *dev)
PMD_DRV_LOG(ERR, "Wait too long for reset done!\n");
return ret;
}
- vf->vf_reset = false;
PMD_DRV_LOG(DEBUG, "Start dev_reset ...\n");
ret = iavf_dev_uninit(dev);
@@ -2971,15 +2966,47 @@ iavf_dev_reset(struct rte_eth_dev *dev)
return iavf_dev_init(dev);
}
+static inline bool
+iavf_is_reset(struct iavf_hw *hw)
+{
+ return !(IAVF_READ_REG(hw, IAVF_VF_ARQLEN1) &
+ IAVF_VF_ARQLEN1_ARQENABLE_MASK);
+}
+
+static bool
+iavf_is_reset_detected(struct iavf_adapter *adapter)
+{
+ struct iavf_hw *hw = IAVF_DEV_PRIVATE_TO_HW(adapter);
+ int i;
+
+ /* poll until we see the reset actually happen */
+ for (i = 0; i < IAVF_RESET_DETECTED_CNT; i++) {
+ if (iavf_is_reset(hw))
+ return true;
+ rte_delay_ms(20);
+ }
+
+ return false;
+}
+
/*
* Handle hardware reset
*/
-int
+void
iavf_handle_hw_reset(struct rte_eth_dev *dev)
{
struct iavf_info *vf = IAVF_DEV_PRIVATE_TO_VF(dev->data->dev_private);
+ struct iavf_adapter *adapter = dev->data->dev_private;
int ret;
+ if (!dev->data->dev_started)
+ return;
+
+ if (!iavf_is_reset_detected(adapter)) {
+ PMD_DRV_LOG(DEBUG, "reset not start\n");
+ return;
+ }
+
vf->in_reset_recovery = true;
ret = iavf_dev_reset(dev);
@@ -2997,15 +3024,16 @@ iavf_handle_hw_reset(struct rte_eth_dev *dev)
ret = iavf_dev_start(dev);
if (ret)
goto error;
- dev->data->dev_started = 1;
+ dev->data->dev_started = 1;
vf->in_reset_recovery = false;
- return 0;
+
+ return;
error:
- PMD_DRV_LOG(DEBUG, "RESET recover with error code=%d\n", ret);
vf->in_reset_recovery = false;
- return ret;
+ iavf_uninit_vf(dev);
+ PMD_DRV_LOG(DEBUG, "failed to reset port %d\n", dev->data->port_id);
}
static int