From: Kevin Traynor <ktraynor@redhat.com>
To: stable@dpdk.org, xuemingl@nvidia.com
Cc: Kevin Traynor <ktraynor@redhat.com>
Subject: [PATCH 22.11] Revert "net/iavf: fix abnormal disable HW interrupt"
Date: Fri, 12 Jan 2024 11:20:46 +0000 [thread overview]
Message-ID: <20240112112046.224346-1-ktraynor@redhat.com> (raw)
This reverts commit 2bdeeb53da0dda4e36a022553d3403c1217d677e.
The change in behaviour [0] for handling some admin queue messages
from polling to interrupt mode has changed the behaviour and is leading
to a deadlock with OVS.
[0]
commit 2bdeeb53da0dda4e36a022553d3403c1217d677e
Author: Mingjin Ye <mingjinx.ye@intel.com>
Date: Wed Jun 14 09:53:03 2023 +0000
net/iavf: fix abnormal disable HW interrupt
[ upstream commit 675a104e2e940ec476e8b469725e8465d01c0098 ]
Bugzilla ID: 1337
Reported-by: Kevin Traynor <ktraynor@redhat.com>
Signed-off-by: Kevin Traynor <ktraynor@redhat.com>
---
drivers/net/iavf/iavf_ethdev.c | 25 +++++++++---------
drivers/net/iavf/iavf_vchnl.c | 48 +++++++++++++++++++++++++++-------
2 files changed, 50 insertions(+), 23 deletions(-)
diff --git a/drivers/net/iavf/iavf_ethdev.c b/drivers/net/iavf/iavf_ethdev.c
index 62d72a0155..a12ea39444 100644
--- a/drivers/net/iavf/iavf_ethdev.c
+++ b/drivers/net/iavf/iavf_ethdev.c
@@ -2760,17 +2760,4 @@ iavf_dev_close(struct rte_eth_dev *dev)
ret = iavf_dev_stop(dev);
-
- /*
- * Release redundant queue resource when close the dev
- * so that other vfs can re-use the queues.
- */
- if (vf->lv_enabled) {
- ret = iavf_request_queues(dev, IAVF_MAX_NUM_QUEUES_DFLT);
- if (ret)
- PMD_DRV_LOG(ERR, "Reset the num of queues failed");
-
- vf->max_rss_qregion = IAVF_MAX_NUM_QUEUES_DFLT;
- }
-
adapter->closed = true;
@@ -2789,4 +2776,16 @@ iavf_dev_close(struct rte_eth_dev *dev)
iavf_config_promisc(adapter, false, false);
+ /*
+ * Release redundant queue resource when close the dev
+ * so that other vfs can re-use the queues.
+ */
+ if (vf->lv_enabled) {
+ ret = iavf_request_queues(dev, IAVF_MAX_NUM_QUEUES_DFLT);
+ if (ret)
+ PMD_DRV_LOG(ERR, "Reset the num of queues failed");
+
+ vf->max_rss_qregion = IAVF_MAX_NUM_QUEUES_DFLT;
+ }
+
iavf_shutdown_adminq(hw);
if (vf->vf_res->vf_cap_flags & VIRTCHNL_VF_OFFLOAD_WB_ON_ITR) {
diff --git a/drivers/net/iavf/iavf_vchnl.c b/drivers/net/iavf/iavf_vchnl.c
index c2f6b617d6..aeffb07cca 100644
--- a/drivers/net/iavf/iavf_vchnl.c
+++ b/drivers/net/iavf/iavf_vchnl.c
@@ -319,5 +319,4 @@ iavf_execute_vf_cmd(struct iavf_adapter *adapter, struct iavf_cmd_info *args,
switch (args->ops) {
case VIRTCHNL_OP_RESET_VF:
- case VIRTCHNL_OP_REQUEST_QUEUES:
/*no need to wait for response */
_clear_cmd(vf);
@@ -343,4 +342,31 @@ iavf_execute_vf_cmd(struct iavf_adapter *adapter, struct iavf_cmd_info *args,
_clear_cmd(vf);
break;
+ case VIRTCHNL_OP_REQUEST_QUEUES:
+ /*
+ * ignore async reply, only wait for system message,
+ * vf_reset = true if get VIRTCHNL_EVENT_RESET_IMPENDING,
+ * if not, means request queues failed.
+ */
+ do {
+ result = iavf_read_msg_from_pf(adapter, args->out_size,
+ args->out_buffer);
+ if (result == IAVF_MSG_SYS && vf->vf_reset) {
+ break;
+ } else if (result == IAVF_MSG_CMD ||
+ result == IAVF_MSG_ERR) {
+ err = -1;
+ break;
+ }
+ iavf_msec_delay(ASQ_DELAY_MS);
+ /* If don't read msg or read sys event, continue */
+ } while (i++ < MAX_TRY_TIMES);
+ if (i >= MAX_TRY_TIMES ||
+ vf->cmd_retval != VIRTCHNL_STATUS_SUCCESS) {
+ err = -1;
+ PMD_DRV_LOG(ERR, "No response or return failure (%d)"
+ " for cmd %d", vf->cmd_retval, args->ops);
+ }
+ _clear_cmd(vf);
+ break;
default:
if (rte_thread_is_intr()) {
@@ -2042,9 +2068,9 @@ iavf_request_queues(struct rte_eth_dev *dev, uint16_t num)
IAVF_DEV_PRIVATE_TO_ADAPTER(dev->data->dev_private);
struct iavf_info *vf = IAVF_DEV_PRIVATE_TO_VF(adapter);
+ struct rte_pci_device *pci_dev = RTE_ETH_DEV_TO_PCI(dev);
struct virtchnl_vf_res_request vfres;
struct iavf_cmd_info args;
uint16_t num_queue_pairs;
int err;
- int i = 0;
if (!(vf->vf_res->vf_cap_flags &
@@ -2067,5 +2093,14 @@ iavf_request_queues(struct rte_eth_dev *dev, uint16_t num)
if (vf->vf_res->vf_cap_flags & VIRTCHNL_VF_OFFLOAD_WB_ON_ITR) {
- err = iavf_execute_vf_cmd_safe(adapter, &args, 0);
+ /* disable interrupt to avoid the admin queue message to be read
+ * before iavf_read_msg_from_pf.
+ *
+ * don't disable interrupt handler until ready to execute vf cmd.
+ */
+ rte_spinlock_lock(&vf->aq_lock);
+ rte_intr_disable(pci_dev->intr_handle);
+ err = iavf_execute_vf_cmd(adapter, &args, 0);
+ rte_intr_enable(pci_dev->intr_handle);
+ rte_spinlock_unlock(&vf->aq_lock);
} else {
rte_eal_alarm_cancel(iavf_dev_alarm_handler, dev);
@@ -2080,11 +2115,4 @@ iavf_request_queues(struct rte_eth_dev *dev, uint16_t num)
}
- /* wait for interrupt notification vf is resetting */
- while (i++ < MAX_TRY_TIMES) {
- if (vf->vf_reset)
- break;
- iavf_msec_delay(ASQ_DELAY_MS);
- }
-
/* request queues succeeded, vf is resetting */
if (vf->vf_reset) {
--
2.43.0
next reply other threads:[~2024-01-12 11:20 UTC|newest]
Thread overview: 7+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-01-12 11:20 Kevin Traynor [this message]
2024-01-14 23:47 ` Chris Brezovec (cbrezove)
2024-01-15 10:02 ` Kevin Traynor
2024-01-18 11:41 ` Kevin Traynor
2024-01-18 12:03 ` Xueming(Steven) Li
2024-01-18 12:23 ` Kevin Traynor
2024-01-19 13:56 ` Xueming(Steven) Li
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20240112112046.224346-1-ktraynor@redhat.com \
--to=ktraynor@redhat.com \
--cc=stable@dpdk.org \
--cc=xuemingl@nvidia.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
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).