DPDK patches and discussions
 help / color / mirror / Atom feed
From: "Min Hu (Connor)" <humin29@huawei.com>
To: <dev@dpdk.org>
Cc: <ferruh.yigit@intel.com>, <thomas@monjalon.net>
Subject: [dpdk-dev] [PATCH 2/4] net/hns3: fix MP action register and unregister
Date: Tue, 2 Nov 2021 09:38:27 +0800	[thread overview]
Message-ID: <20211102013829.42345-3-humin29@huawei.com> (raw)
In-Reply-To: <20211102013829.42345-1-humin29@huawei.com>

From: Huisong Li <lihuisong@huawei.com>

The multi-process has the following problems:
1) After a port in primary process is closed, the mp action of the process
   is unregistered. which will cause that other device in the primary
   process cannot respond to requests from secondary processes.
2) Because variable "hns3_inited" is set to true without returning an
   initial value, the mp action cannot be registered again after it is
   unregistered.
3) The mp action of primary and secondary process need to be registered
   only once regardless of port numbers in the process. That's what
   variable "hns3_inited" does. But the variable is difficult to
   understand.

This patch adds a hns3_process_local_data structure to resolve above
problems.

Fixes: 9570b1fdbdad ("net/hns3: check multi-process action register result")
Fixes: 23d4b61fee5d ("net/hns3: support multiple process")

Signed-off-by: Huisong Li <lihuisong@huawei.com>
Signed-off-by: Min Hu (Connor) <humin29@huawei.com>
---
 drivers/net/hns3/hns3_ethdev.c    |  2 ++
 drivers/net/hns3/hns3_ethdev_vf.c |  2 ++
 drivers/net/hns3/hns3_mp.c        | 37 ++++++++++++++++++-------------
 drivers/net/hns3/hns3_mp.h        |  7 ++++++
 4 files changed, 33 insertions(+), 15 deletions(-)

diff --git a/drivers/net/hns3/hns3_ethdev.c b/drivers/net/hns3/hns3_ethdev.c
index 2dd9881040..658baf9ec3 100644
--- a/drivers/net/hns3/hns3_ethdev.c
+++ b/drivers/net/hns3/hns3_ethdev.c
@@ -7379,6 +7379,7 @@ hns3_dev_init(struct rte_eth_dev *eth_dev)
 			goto err_mp_init_secondary;
 		}
 		__atomic_fetch_add(&hw->secondary_cnt, 1, __ATOMIC_RELAXED);
+		process_data.eth_dev_cnt++;
 		hns3_tx_push_init(eth_dev);
 		return 0;
 	}
@@ -7390,6 +7391,7 @@ hns3_dev_init(struct rte_eth_dev *eth_dev)
 			     ret);
 		goto err_mp_init_primary;
 	}
+	process_data.eth_dev_cnt++;
 
 	hw->adapter_state = HNS3_NIC_UNINITIALIZED;
 	hns->is_vf = false;
diff --git a/drivers/net/hns3/hns3_ethdev_vf.c b/drivers/net/hns3/hns3_ethdev_vf.c
index 0f7eeb1d7e..7286a858a9 100644
--- a/drivers/net/hns3/hns3_ethdev_vf.c
+++ b/drivers/net/hns3/hns3_ethdev_vf.c
@@ -2687,6 +2687,7 @@ hns3vf_dev_init(struct rte_eth_dev *eth_dev)
 			goto err_mp_init_secondary;
 		}
 		__atomic_fetch_add(&hw->secondary_cnt, 1, __ATOMIC_RELAXED);
+		process_data.eth_dev_cnt++;
 		hns3_tx_push_init(eth_dev);
 		return 0;
 	}
@@ -2698,6 +2699,7 @@ hns3vf_dev_init(struct rte_eth_dev *eth_dev)
 			     ret);
 		goto err_mp_init_primary;
 	}
+	process_data.eth_dev_cnt++;
 
 	hw->adapter_state = HNS3_NIC_UNINITIALIZED;
 	hns->is_vf = true;
diff --git a/drivers/net/hns3/hns3_mp.c b/drivers/net/hns3/hns3_mp.c
index c28598a53a..1a79d249b8 100644
--- a/drivers/net/hns3/hns3_mp.c
+++ b/drivers/net/hns3/hns3_mp.c
@@ -12,7 +12,8 @@
 #include "hns3_rxtx.h"
 #include "hns3_mp.h"
 
-static bool hns3_inited;
+/* local data for primary or secondary process. */
+struct hns3_process_local_data process_data;
 
 /*
  * Initialize IPC message.
@@ -230,14 +231,15 @@ int hns3_mp_init_primary(void)
 {
 	int ret;
 
-	if (!hns3_inited) {
-		/* primary is allowed to not support IPC */
-		ret = rte_mp_action_register(HNS3_MP_NAME, mp_primary_handle);
-		if (ret && rte_errno != ENOTSUP)
-			return ret;
+	if (process_data.init_done)
+		return 0;
 
-		hns3_inited = true;
-	}
+	/* primary is allowed to not support IPC */
+	ret = rte_mp_action_register(HNS3_MP_NAME, mp_primary_handle);
+	if (ret && rte_errno != ENOTSUP)
+		return ret;
+
+	process_data.init_done = true;
 
 	return 0;
 }
@@ -247,8 +249,12 @@ int hns3_mp_init_primary(void)
  */
 void hns3_mp_uninit_primary(void)
 {
-	if (hns3_inited)
+	process_data.eth_dev_cnt--;
+
+	if (process_data.eth_dev_cnt == 0) {
 		rte_mp_action_unregister(HNS3_MP_NAME);
+		process_data.init_done = false;
+	}
 }
 
 /*
@@ -258,13 +264,14 @@ int hns3_mp_init_secondary(void)
 {
 	int ret;
 
-	if (!hns3_inited) {
-		ret = rte_mp_action_register(HNS3_MP_NAME, mp_secondary_handle);
-		if (ret)
-			return ret;
+	if (process_data.init_done)
+		return 0;
 
-		hns3_inited = true;
-	}
+	ret = rte_mp_action_register(HNS3_MP_NAME, mp_secondary_handle);
+	if (ret)
+		return ret;
+
+	process_data.init_done = true;
 
 	return 0;
 }
diff --git a/drivers/net/hns3/hns3_mp.h b/drivers/net/hns3/hns3_mp.h
index e0e4aeaf6c..b49532f985 100644
--- a/drivers/net/hns3/hns3_mp.h
+++ b/drivers/net/hns3/hns3_mp.h
@@ -5,6 +5,13 @@
 #ifndef _HNS3_MP_H_
 #define _HNS3_MP_H_
 
+/* Local data for primary or secondary process. */
+struct hns3_process_local_data {
+	bool init_done; /* Process action register completed flag. */
+	int eth_dev_cnt; /* Ethdev count under the current process. */
+};
+extern struct hns3_process_local_data process_data;
+
 void hns3_mp_req_start_rxtx(struct rte_eth_dev *dev);
 void hns3_mp_req_stop_rxtx(struct rte_eth_dev *dev);
 void hns3_mp_req_start_tx(struct rte_eth_dev *dev);
-- 
2.33.0


  parent reply	other threads:[~2021-11-02  1:40 UTC|newest]

Thread overview: 7+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-11-02  1:38 [dpdk-dev] [PATCH 0/4] bugfix for multi process of hns3 PMD Min Hu (Connor)
2021-11-02  1:38 ` [dpdk-dev] [PATCH 1/4] net/hns3: decrease the count when secondary process exits Min Hu (Connor)
2021-11-02  1:38 ` Min Hu (Connor) [this message]
2021-11-04 14:22   ` [dpdk-dev] [PATCH 2/4] net/hns3: fix MP action register and unregister Ferruh Yigit
2021-11-02  1:38 ` [dpdk-dev] [PATCH 3/4] net/hns3: fix lack of unregistering MP action for secondary Min Hu (Connor)
2021-11-02  1:38 ` [dpdk-dev] [PATCH 4/4] net/hns3: refactor multi-process initialization Min Hu (Connor)
2021-11-04 14:22 ` [dpdk-dev] [PATCH 0/4] bugfix for multi process of hns3 PMD 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=20211102013829.42345-3-humin29@huawei.com \
    --to=humin29@huawei.com \
    --cc=dev@dpdk.org \
    --cc=ferruh.yigit@intel.com \
    --cc=thomas@monjalon.net \
    /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).