All of lore.kernel.org
 help / color / mirror / Atom feed
From: Luca Boccassi <bluca@debian.org>
To: Louis Luo <llouis@vmware.com>, "dev@dpdk.org" <dev@dpdk.org>
Cc: "maxime.coquelin@redhat.com" <maxime.coquelin@redhat.com>,
	"tiwei.bie@intel.com" <tiwei.bie@intel.com>,
	Yong Wang <yongwang@vmware.com>,
	"3chas3@gmail.com" <3chas3@gmail.com>,
	"bruce.richardson@intel.com" <bruce.richardson@intel.com>,
	"jianfeng.tan@intel.com" <jianfeng.tan@intel.com>,
	"anatoly.burakov@intel.com" <anatoly.burakov@intel.com>,
	"stable@dpdk.org" <stable@dpdk.org>,
	Brian Russell <brussell@brocade.com>
Subject: Re: [PATCH 2/3] net/vmxnet3: fix vmxnet3 dev_uninit() hot-unplug
Date: Tue, 18 Sep 2018 14:14:54 +0100	[thread overview]
Message-ID: <1537276494.10481.7.camel@debian.org> (raw)
In-Reply-To: <FF963A5D-98DC-4EE2-B429-161CCCCC102D@vmware.com>

Hi,

The application must already stop and close before detaching (which
will call uninit). Quoting from the documentation:

"*  Before detaching, they must be stopped and closed.

    DPDK applications must call "rte_eth_dev_stop()" and
    "rte_eth_dev_close()" APIs before detaching ports. These functions will
    start finalization sequence of the PMDs."

http://doc.dpdk.org/guides/prog_guide/port_hotplug_framework.html

On Mon, 2018-09-17 at 19:06 +0000, Louis Luo wrote:
> Hi Luca,
> 
> When eth_vmxnet3_dev_uninit() is called, is it guaranteed that
> vmxnet3_dev_close/ vmxnet3_dev_stop must have been called? I'm not
> familiar with the hot-plug procedure, so just wonder if there is any
> chance that eth_vmxnet3_dev_uninit() is called without calling
> vmxnet3_dev_close/ vmxnet3_dev_stop.
> 
> Thanks,
> Louis
> 
> On 8/16/18, 6:51 AM, "dev on behalf of Luca Boccassi" <dev-bounces@d
> pdk.org on behalf of bluca@debian.org> wrote:
> 
>     The vmxnet3 driver can't call back into dev_close(), and possibly
>     dev_stop(), in dev_uninit().  When dev_uninit() is called,
> anything
>     that those routines would want to clean up has already been
> released.
>     Further, for complete cleanup, it is necessary to release any of
> the
>     queue resources during dev_close().
>     This allows a vmxnet3 device to be hot-unplugged without leaking
>     queues.
>     
>     Fixes: dfaff37fc46d ("vmxnet3: import new vmxnet3 poll mode
> driver implementation")
>     Cc: stable@dpdk.org
>     
>     Signed-off-by: Brian Russell <brussell@brocade.com>
>     Signed-off-by: Luca Boccassi <bluca@debian.org>
>     ---
>      drivers/net/vmxnet3/vmxnet3_ethdev.c | 36 ++++++++++++++++++++
> --------
>      1 file changed, 26 insertions(+), 10 deletions(-)
>     
>     diff --git a/drivers/net/vmxnet3/vmxnet3_ethdev.c
> b/drivers/net/vmxnet3/vmxnet3_ethdev.c
>     index 2613cd1358..b5d4be5e24 100644
>     --- a/drivers/net/vmxnet3/vmxnet3_ethdev.c
>     +++ b/drivers/net/vmxnet3/vmxnet3_ethdev.c
>     @@ -348,16 +348,11 @@ eth_vmxnet3_dev_init(struct rte_eth_dev
> *eth_dev)
>      static int
>      eth_vmxnet3_dev_uninit(struct rte_eth_dev *eth_dev)
>      {
>     -	struct vmxnet3_hw *hw = eth_dev->data->dev_private;
>     -
>      	PMD_INIT_FUNC_TRACE();
>      
>      	if (rte_eal_process_type() != RTE_PROC_PRIMARY)
>      		return 0;
>      
>     -	if (hw->adapter_stopped == 0)
>     -		vmxnet3_dev_close(eth_dev);
>     -
>      	eth_dev->dev_ops = NULL;
>      	eth_dev->rx_pkt_burst = NULL;
>      	eth_dev->tx_pkt_burst = NULL;
>     @@ -803,7 +798,7 @@ vmxnet3_dev_stop(struct rte_eth_dev *dev)
>      	PMD_INIT_FUNC_TRACE();
>      
>      	if (hw->adapter_stopped == 1) {
>     -		PMD_INIT_LOG(DEBUG, "Device already closed.");
>     +		PMD_INIT_LOG(DEBUG, "Device already stopped.");
>      		return;
>      	}
>      
>     @@ -827,7 +822,6 @@ vmxnet3_dev_stop(struct rte_eth_dev *dev)
>      	/* reset the device */
>      	VMXNET3_WRITE_BAR1_REG(hw, VMXNET3_REG_CMD,
> VMXNET3_CMD_RESET_DEV);
>      	PMD_INIT_LOG(DEBUG, "Device reset.");
>     -	hw->adapter_stopped = 0;
>      
>      	vmxnet3_dev_clear_queues(dev);
>      
>     @@ -837,6 +831,30 @@ vmxnet3_dev_stop(struct rte_eth_dev *dev)
>      	link.link_speed = ETH_SPEED_NUM_10G;
>      	link.link_autoneg = ETH_LINK_FIXED;
>      	rte_eth_linkstatus_set(dev, &link);
>     +
>     +	hw->adapter_stopped = 1;
>     +}
>     +
>     +static void
>     +vmxnet3_free_queues(struct rte_eth_dev *dev)
>     +{
>     +	int i;
>     +
>     +	PMD_INIT_FUNC_TRACE();
>     +
>     +	for (i = 0; i < dev->data->nb_rx_queues; i++) {
>     +		void *rxq = dev->data->rx_queues[i];
>     +
>     +		vmxnet3_dev_rx_queue_release(rxq);
>     +	}
>     +	dev->data->nb_rx_queues = 0;
>     +
>     +	for (i = 0; i < dev->data->nb_tx_queues; i++) {
>     +		void *txq = dev->data->tx_queues[i];
>     +
>     +		vmxnet3_dev_tx_queue_release(txq);
>     +	}
>     +	dev->data->nb_tx_queues = 0;
>      }
>      
>      /*
>     @@ -845,12 +863,10 @@ vmxnet3_dev_stop(struct rte_eth_dev *dev)
>      static void
>      vmxnet3_dev_close(struct rte_eth_dev *dev)
>      {
>     -	struct vmxnet3_hw *hw = dev->data->dev_private;
>     -
>      	PMD_INIT_FUNC_TRACE();
>      
>      	vmxnet3_dev_stop(dev);
>     -	hw->adapter_stopped = 1;
>     +	vmxnet3_free_queues(dev);
>      }
>      
>      static void
>     -- 
>     2.18.0
>     
>     
> 

-- 
Kind regards,
Luca Boccassi

  reply	other threads:[~2018-09-18 13:14 UTC|newest]

Thread overview: 30+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-08-16 13:50 [PATCH 0/3] Fix hot plug/unplug of virtual devices Luca Boccassi
2018-08-16 13:50 ` [PATCH 1/3] net/virtio: register/unregister intr handler on start/stop Luca Boccassi
2018-08-16 13:50 ` [PATCH 2/3] net/vmxnet3: fix vmxnet3 dev_uninit() hot-unplug Luca Boccassi
2018-09-17 19:06   ` Louis Luo
2018-09-18 13:14     ` Luca Boccassi [this message]
2018-09-18 18:14       ` Louis Luo
2018-09-18 18:29         ` Luca Boccassi
2018-09-18 18:48           ` Louis Luo
2018-09-19 12:58             ` Luca Boccassi
2018-08-16 13:50 ` [PATCH 3/3] eal/linux: handle uio read failure in interrupt handler Luca Boccassi
2018-09-19 12:57 ` [PATCH v2 1/3] net/virtio: register/unregister intr handler on start/stop Luca Boccassi
2018-09-19 12:57   ` [PATCH v2 2/3] net/vmxnet3: fix vmxnet3 dev_uninit() hot-unplug Luca Boccassi
2018-09-19 15:47     ` Chas Williams
2018-09-19 16:08       ` Luca Boccassi
2018-10-27 15:09       ` Thomas Monjalon
2018-10-31 17:27         ` Thomas Monjalon
2018-10-31 17:46           ` Luca Boccassi
2018-10-31 18:02             ` Thomas Monjalon
2018-10-31 18:54             ` Louis Luo
2018-09-27  8:39     ` Luca Boccassi
2018-09-19 12:57   ` [PATCH v2 3/3] eal/linux: handle uio read failure in interrupt handler Luca Boccassi
2018-10-11 10:32     ` Thomas Monjalon
2018-09-27  8:40   ` [PATCH v2 1/3] net/virtio: register/unregister intr handler on start/stop Luca Boccassi
2018-09-27 10:51     ` Maxime Coquelin
2018-09-27 11:14   ` Maxime Coquelin
2018-10-31 18:39 ` [PATCH v3 " Luca Boccassi
2018-10-31 18:39   ` [PATCH v3 2/3] net/vmxnet3: fix vmxnet3 dev_uninit() hot-unplug Luca Boccassi
2018-10-31 18:39   ` [PATCH v3 3/3] eal/linux: handle uio read failure in interrupt handler Luca Boccassi
2018-11-02  9:49   ` [dpdk-stable] [PATCH v3 1/3] net/virtio: register/unregister intr handler on start/stop Thomas Monjalon
2018-11-02 11:14     ` Luca Boccassi

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=1537276494.10481.7.camel@debian.org \
    --to=bluca@debian.org \
    --cc=3chas3@gmail.com \
    --cc=anatoly.burakov@intel.com \
    --cc=bruce.richardson@intel.com \
    --cc=brussell@brocade.com \
    --cc=dev@dpdk.org \
    --cc=jianfeng.tan@intel.com \
    --cc=llouis@vmware.com \
    --cc=maxime.coquelin@redhat.com \
    --cc=stable@dpdk.org \
    --cc=tiwei.bie@intel.com \
    --cc=yongwang@vmware.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.