DPDK patches and discussions
 help / color / mirror / Atom feed
From: "Wei Hu (Xavier)" <huwei013@chinasoftinc.com>
To: <dev@dpdk.org>
Subject: [dpdk-dev] [PATCH 7/7] net/hns3: fix abnormal status after reset occurs repeatedly
Date: Tue, 17 Mar 2020 17:12:06 +0800	[thread overview]
Message-ID: <20200317091206.34928-8-huwei013@chinasoftinc.com> (raw)
In-Reply-To: <20200317091206.34928-1-huwei013@chinasoftinc.com>

From: "Wei Hu (Xavier)" <xavier.huwei@huawei.com>

Currently, when performing the following test case:
1. Run testpmd application based on hns3 PF device.
2. Inject reset(global/IMP reset) repeatedly.
After the reset, the network port can't link up.

In the RESET_STAGE_DEV_INIT stage of the reset process, the driver will
reinitialize the hardware. If global/IMP reset occurs at this time again,
the operation of reinitialize the hardware will fail because that firmware
don't respond to the configuration commands issued by driver. In current
driver, when failed to reinitialize the hardware, rollback operation is
done, such as clearing the relevant configuration of the command queue
registers.

If firmware detects that the function's command queue register is not
configured correctly, it will not complete the reset related hardware
configuration for this function, resulting in that driver can't detect
that the hardware reset has been completed. And then the reset process
of the driver exit abnormally, the hardware can not work normally after
reset.

This patch fixes it by avoid clearing the command queue related registers
when failed to reinitialize the hardware in the RESET_STAGE_DEV_INIT stage
of the reset process.

Fixes: 2790c6464725 ("net/hns3: support device reset")
Cc: stable@dpdk.org

Signed-off-by: Hongbo Zheng <zhenghongbo3@huawei.com>
Signed-off-by: Wei Hu (Xavier) <xavier.huwei@huawei.com>
Signed-off-by: Chunsong Feng <fengchunsong@huawei.com>
---
 drivers/net/hns3/hns3_cmd.c       |  2 +-
 drivers/net/hns3/hns3_ethdev.c    | 18 ++++--------------
 drivers/net/hns3/hns3_ethdev_vf.c | 16 ++++------------
 3 files changed, 9 insertions(+), 27 deletions(-)

diff --git a/drivers/net/hns3/hns3_cmd.c b/drivers/net/hns3/hns3_cmd.c
index c85168b31..c7993634e 100644
--- a/drivers/net/hns3/hns3_cmd.c
+++ b/drivers/net/hns3/hns3_cmd.c
@@ -529,7 +529,7 @@ hns3_cmd_init(struct hns3_hw *hw)
 	return 0;
 
 err_cmd_init:
-	hns3_cmd_uninit(hw);
+	rte_atomic16_set(&hw->reset.disable_cmd, 1);
 	return ret;
 }
 
diff --git a/drivers/net/hns3/hns3_ethdev.c b/drivers/net/hns3/hns3_ethdev.c
index 0d0f5ebb7..1d57de7f9 100644
--- a/drivers/net/hns3/hns3_ethdev.c
+++ b/drivers/net/hns3/hns3_ethdev.c
@@ -4126,13 +4126,10 @@ hns3_init_pf(struct rte_eth_dev *eth_dev)
 	rte_intr_disable(&pci_dev->intr_handle);
 	hns3_intr_unregister(&pci_dev->intr_handle, hns3_interrupt_handler,
 			     eth_dev);
-
 err_intr_callback_register:
-	hns3_cmd_uninit(hw);
-
 err_cmd_init:
+	hns3_cmd_uninit(hw);
 	hns3_cmd_destroy_queue(hw);
-
 err_cmd_init_queue:
 	hw->io_base = NULL;
 
@@ -4610,31 +4607,24 @@ hns3_reinit_dev(struct hns3_adapter *hns)
 	ret = hns3_reset_all_queues(hns);
 	if (ret) {
 		hns3_err(hw, "Failed to reset all queues: %d", ret);
-		goto err_init;
+		return ret;
 	}
 
 	ret = hns3_init_hardware(hns);
 	if (ret) {
 		hns3_err(hw, "Failed to init hardware: %d", ret);
-		goto err_init;
+		return ret;
 	}
 
 	ret = hns3_enable_hw_error_intr(hns, true);
 	if (ret) {
 		hns3_err(hw, "fail to enable hw error interrupts: %d",
 			     ret);
-		goto err_mac_init;
+		return ret;
 	}
 	hns3_info(hw, "Reset done, driver initialization finished.");
 
 	return 0;
-
-err_mac_init:
-	hns3_uninit_umv_space(hw);
-err_init:
-	hns3_cmd_uninit(hw);
-
-	return ret;
 }
 
 static bool
diff --git a/drivers/net/hns3/hns3_ethdev_vf.c b/drivers/net/hns3/hns3_ethdev_vf.c
index 1c508f06a..2e8acfee3 100644
--- a/drivers/net/hns3/hns3_ethdev_vf.c
+++ b/drivers/net/hns3/hns3_ethdev_vf.c
@@ -1486,11 +1486,9 @@ hns3vf_init_vf(struct rte_eth_dev *eth_dev)
 	hns3_intr_unregister(&pci_dev->intr_handle, hns3vf_interrupt_handler,
 			     eth_dev);
 err_intr_callback_register:
-	hns3_cmd_uninit(hw);
-
 err_cmd_init:
+	hns3_cmd_uninit(hw);
 	hns3_cmd_destroy_queue(hw);
-
 err_cmd_init_queue:
 	hw->io_base = NULL;
 
@@ -2104,7 +2102,7 @@ hns3vf_reinit_dev(struct hns3_adapter *hns)
 	ret = hns3_cmd_init(hw);
 	if (ret) {
 		hns3_err(hw, "Failed to init cmd: %d", ret);
-		goto err_cmd_init;
+		return ret;
 	}
 
 	if (hw->reset.level == HNS3_VF_FULL_RESET) {
@@ -2124,22 +2122,16 @@ hns3vf_reinit_dev(struct hns3_adapter *hns)
 	ret = hns3_reset_all_queues(hns);
 	if (ret) {
 		hns3_err(hw, "Failed to reset all queues: %d", ret);
-		goto err_init;
+		return ret;
 	}
 
 	ret = hns3vf_init_hardware(hns);
 	if (ret) {
 		hns3_err(hw, "Failed to init hardware: %d", ret);
-		goto err_init;
+		return ret;
 	}
 
 	return 0;
-
-err_cmd_init:
-	hns3vf_set_bus_master(pci_dev, false);
-err_init:
-	hns3_cmd_uninit(hw);
-	return ret;
 }
 
 static const struct eth_dev_ops hns3vf_eth_dev_ops = {
-- 
2.23.0


  parent reply	other threads:[~2020-03-17  9:25 UTC|newest]

Thread overview: 9+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-03-17  9:11 [dpdk-dev] [PATCH 0/7] misc updates for hns3 PMD driver Wei Hu (Xavier)
2020-03-17  9:12 ` [dpdk-dev] [PATCH 1/7] net/hns3: remove redundant MAC addr check when setting MAC Wei Hu (Xavier)
2020-03-17  9:12 ` [dpdk-dev] [PATCH 2/7] net/hns3: modify inappropriate names Wei Hu (Xavier)
2020-03-17  9:12 ` [dpdk-dev] [PATCH 3/7] net/hns3: fix packets's offload features flags in Rx Wei Hu (Xavier)
2020-03-17  9:12 ` [dpdk-dev] [PATCH 4/7] net/hns3: fix default error code of command interface Wei Hu (Xavier)
2020-03-17  9:12 ` [dpdk-dev] [PATCH 5/7] net/hns3: fix crash when flushing RSS flow rules with FLR Wei Hu (Xavier)
2020-03-17  9:12 ` [dpdk-dev] [PATCH 6/7] net/hns3: fix configuring illeagl VLAN pvid Wei Hu (Xavier)
2020-03-17  9:12 ` Wei Hu (Xavier) [this message]
2020-03-19  9:08 ` [dpdk-dev] [PATCH 0/7] misc updates for hns3 PMD driver Ferruh Yigit

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=20200317091206.34928-8-huwei013@chinasoftinc.com \
    --to=huwei013@chinasoftinc.com \
    --cc=dev@dpdk.org \
    /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).