DPDK patches and discussions
 help / color / mirror / Atom feed
From: "Qiu, Michael" <michael.qiu@intel.com>
To: "Iremonger, Bernard" <bernard.iremonger@intel.com>,
	"dev@dpdk.org" <dev@dpdk.org>
Cc: "He, Shaopeng" <shaopeng.he@intel.com>
Subject: Re: [dpdk-dev] [PATCH 2/2 v2] fm10k: Add hotplug support for fm10k
Date: Fri, 19 Jun 2015 07:41:17 +0000	[thread overview]
Message-ID: <533710CFB86FA344BFBF2D6802E602860469D92B@SHSMSX101.ccr.corp.intel.com> (raw)
In-Reply-To: <8CEF83825BEC744B83065625E567D7C204A37E29@IRSMSX108.ger.corp.intel.com>

On 6/19/2015 12:42 AM, Iremonger, Bernard wrote:
>> -----Original Message-----
>> From: Qiu, Michael
>> Sent: Wednesday, June 17, 2015 9:45 AM
>> To: dev@dpdk.org
>> Cc: Chen, Jing D; Iremonger, Bernard; He, Shaopeng; Qiu, Michael
>> Subject: [PATCH 2/2 v2] fm10k: Add hotplug support for fm10k
>>
>> Add hotplug support for fm10k.
>>
>> Signed-off-by: Michael Qiu <michael.qiu@intel.com>
> Hi Michael,
>
> There is one comment inline.
>
>> ---
>>  drivers/net/fm10k/fm10k_ethdev.c | 97
>> +++++++++++++++++++++++++++++++++++++++-
>>  1 file changed, 96 insertions(+), 1 deletion(-)
>>
>> diff --git a/drivers/net/fm10k/fm10k_ethdev.c
>> b/drivers/net/fm10k/fm10k_ethdev.c
>> index e310698..0d3eaf1 100644
>> --- a/drivers/net/fm10k/fm10k_ethdev.c
>> +++ b/drivers/net/fm10k/fm10k_ethdev.c
>> @@ -1412,6 +1412,36 @@ fm10k_dev_enable_intr_pf(struct rte_eth_dev
>> *dev)  }
>>
>>  static void
>> +fm10k_dev_disable_intr_pf(struct rte_eth_dev *dev) {
>> +	struct fm10k_hw *hw = FM10K_DEV_PRIVATE_TO_HW(dev->data-
>>> dev_private);
>> +	uint32_t int_map = FM10K_INT_MAP_DISABLE;
>> +
>> +	int_map |= 0;
>> +
>> +	FM10K_WRITE_REG(hw, FM10K_INT_MAP(fm10k_int_Mailbox),
>> int_map);
>> +	FM10K_WRITE_REG(hw, FM10K_INT_MAP(fm10k_int_PCIeFault),
>> int_map);
>> +	FM10K_WRITE_REG(hw,
>> FM10K_INT_MAP(fm10k_int_SwitchUpDown), int_map);
>> +	FM10K_WRITE_REG(hw, FM10K_INT_MAP(fm10k_int_SwitchEvent),
>> int_map);
>> +	FM10K_WRITE_REG(hw, FM10K_INT_MAP(fm10k_int_SRAM),
>> int_map);
>> +	FM10K_WRITE_REG(hw, FM10K_INT_MAP(fm10k_int_VFLR),
>> int_map);
>> +
>> +	/* Disable misc causes */
>> +	FM10K_WRITE_REG(hw, FM10K_EIMR,
>> FM10K_EIMR_DISABLE(PCA_FAULT) |
>> +				FM10K_EIMR_DISABLE(THI_FAULT) |
>> +				FM10K_EIMR_DISABLE(FUM_FAULT) |
>> +				FM10K_EIMR_DISABLE(MAILBOX) |
>> +				FM10K_EIMR_DISABLE(SWITCHREADY) |
>> +				FM10K_EIMR_DISABLE(SWITCHNOTREADY) |
>> +				FM10K_EIMR_DISABLE(SRAMERROR) |
>> +				FM10K_EIMR_DISABLE(VFLR));
>> +
>> +	/* Disable ITR 0 */
>> +	FM10K_WRITE_REG(hw, FM10K_ITR(0), FM10K_ITR_MASK_SET);
>> +	FM10K_WRITE_FLUSH(hw);
>> +}
>> +
>> +static void
>>  fm10k_dev_enable_intr_vf(struct rte_eth_dev *dev)  {
>>  	struct fm10k_hw *hw = FM10K_DEV_PRIVATE_TO_HW(dev->data-
>>> dev_private);
>> @@ -1429,6 +1459,22 @@ fm10k_dev_enable_intr_vf(struct rte_eth_dev
>> *dev)
>>  	FM10K_WRITE_FLUSH(hw);
>>  }
>>
>> +static void
>> +fm10k_dev_disable_intr_vf(struct rte_eth_dev *dev) {
>> +	struct fm10k_hw *hw = FM10K_DEV_PRIVATE_TO_HW(dev->data-
>>> dev_private);
>> +	uint32_t int_map = FM10K_INT_MAP_DISABLE;
>> +
>> +	int_map |= 0;
>> +
>> +	/* Only INT 0 available, other 15 are reserved. */
>> +	FM10K_WRITE_REG(hw, FM10K_VFINT_MAP, int_map);
>> +
>> +	/* Disable ITR 0 */
>> +	FM10K_WRITE_REG(hw, FM10K_VFITR(0), FM10K_ITR_MASK_SET);
>> +	FM10K_WRITE_FLUSH(hw);
>> +}
>> +
>>  static int
>>  fm10k_dev_handle_fault(struct fm10k_hw *hw, uint32_t eicr)  { @@ -
>> 1858,6 +1904,54 @@ eth_fm10k_dev_init(struct rte_eth_dev *dev)
>>  	return 0;
>>  }
>>
>> +static int
>> +eth_fm10k_dev_uninit(struct rte_eth_dev *dev) {
>> +	struct fm10k_hw *hw = FM10K_DEV_PRIVATE_TO_HW(dev->data-
>>> dev_private);
>> +
>> +	PMD_INIT_FUNC_TRACE();
>> +
>> +	/* only uninitialize in the primary process */
>> +	if (rte_eal_process_type() != RTE_PROC_PRIMARY)
>> +		return 0;
>> +
>> +	/* safe to close dev here */
> Should a flag be added so as not to call fm10k_dev_close() if it has been called already ?

It's OK for fm10k, for i40e for 82599, there has a flag in share code to
show whether has been closed, but for fm10k there doesn't.

Also function fm10k_dev_close() would check all pointer and resource,
that's why need first patch in fm10k_dev_stop().

Thanks,
Michael
> Regards,
>
> Bernard.
>
>> +	fm10k_dev_close(dev);
>> +
>> +	dev->dev_ops = NULL;
>> +	dev->rx_pkt_burst = NULL;
>> +	dev->tx_pkt_burst = NULL;
>> +
>> +	/* disable uio/vfio intr */
>> +	rte_intr_disable(&(dev->pci_dev->intr_handle));
>> +
>> +	/*PF/VF has different interrupt handling mechanism */
>> +	if (hw->mac.type == fm10k_mac_pf) {
>> +		/* disable interrupt */
>> +		fm10k_dev_disable_intr_pf(dev);
>> +
>> +		/* unregister callback func to eal lib */
>> +		rte_intr_callback_unregister(&(dev->pci_dev->intr_handle),
>> +			fm10k_dev_interrupt_handler_pf, (void *)dev);
>> +	} else {
>> +		/* disable interrupt */
>> +		fm10k_dev_disable_intr_vf(dev);
>> +
>> +		rte_intr_callback_unregister(&(dev->pci_dev->intr_handle),
>> +			fm10k_dev_interrupt_handler_vf, (void *)dev);
>> +	}
>> +
>> +	/* free mac memory */
>> +	if (dev->data->mac_addrs) {
>> +		rte_free(dev->data->mac_addrs);
>> +		dev->data->mac_addrs = NULL;
>> +	}
>> +
>> +	memset(hw, 0, sizeof(*hw));
>> +
>> +	return 0;
>> +}
>> +
>>  /*
>>   * The set of PCI devices this driver supports. This driver will enable both PF
>>   * and SRIOV-VF devices.
>> @@ -1873,9 +1967,10 @@ static struct eth_driver rte_pmd_fm10k = {
>>  	{
>>  		.name = "rte_pmd_fm10k",
>>  		.id_table = pci_id_fm10k_map,
>> -		.drv_flags = RTE_PCI_DRV_NEED_MAPPING,
>> +		.drv_flags = RTE_PCI_DRV_NEED_MAPPING |
>> RTE_PCI_DRV_DETACHABLE,
>>  	},
>>  	.eth_dev_init = eth_fm10k_dev_init,
>> +	.eth_dev_uninit = eth_fm10k_dev_uninit,
>>  	.dev_private_size = sizeof(struct fm10k_adapter),  };
>>
>> --
>> 1.9.3
>


  reply	other threads:[~2015-06-19  7:41 UTC|newest]

Thread overview: 35+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-05-31 14:37 [dpdk-dev] [RFC PATCH] librte_pmd_fm10k: " Michael Qiu
2015-06-02  8:26 ` Iremonger, Bernard
2015-06-02  9:36   ` Qiu, Michael
2015-06-10 12:21 ` [dpdk-dev] [PATCH 0/2 v2] Enable " Michael Qiu
2015-06-10 12:21   ` [dpdk-dev] [PATCH 1/2] fm10k: Free queues when close port Michael Qiu
2015-06-12  7:27     ` Chen, Jing D
2015-06-17  7:59     ` [dpdk-dev] [PATCH 1/2 v2] " Michael Qiu
2015-06-18  7:32       ` Chen, Jing D
2015-06-18 16:29       ` Iremonger, Bernard
2015-06-19  7:36         ` Qiu, Michael
2015-06-17  8:44     ` [dpdk-dev] [PATCH 2/2 v2] fm10k: Add hotplug support for fm10k Michael Qiu
2015-06-18 16:42       ` Iremonger, Bernard
2015-06-19  7:41         ` Qiu, Michael [this message]
2015-06-10 12:21   ` [dpdk-dev] [PATCH 2/2] " Michael Qiu
2015-06-10 12:24   ` [dpdk-dev] [PATCH 0/2 v2] Enable " Qiu, Michael
2015-06-19  8:31   ` [dpdk-dev] [PATCH 0/2 v3] " Michael Qiu
2015-06-19  8:31     ` [dpdk-dev] [PATCH 1/2 v3] fm10k: Free queues when close port Michael Qiu
2015-06-19  8:31     ` [dpdk-dev] [PATCH 2/2 v2] fm10k: Add hotplug support for fm10k Michael Qiu
2015-06-19  8:36       ` Qiu, Michael
2015-06-26  8:29     ` [dpdk-dev] [PATCH 0/2 v4] Enable " Michael Qiu
2015-06-26  8:29       ` [dpdk-dev] [PATCH 1/2 v4] fm10k: Free queues when close port Michael Qiu
2015-06-26 11:02         ` Iremonger, Bernard
2015-06-29  8:17           ` Qiu, Michael
2015-06-29  8:57             ` Iremonger, Bernard
2015-06-29  9:20               ` Qiu, Michael
2015-06-29  9:53                 ` Iremonger, Bernard
2015-06-29 14:58                   ` Qiu, Michael
2015-06-29 15:15                     ` Qiu, Michael
2015-06-29 11:08                 ` Ananyev, Konstantin
2015-06-29 15:14                   ` Qiu, Michael
2015-06-26  8:29       ` [dpdk-dev] [PATCH 2/2 v4] fm10k: Add hotplug support for fm10k Michael Qiu
2015-07-14 12:45       ` [dpdk-dev] [PATCH 0/2 v5] Enable " Michael Qiu
2015-07-14 12:45         ` [dpdk-dev] [PATCH 1/2 v5] fm10k: Free queues when close port Michael Qiu
2015-07-14 12:45         ` [dpdk-dev] [PATCH 2/2 v5] fm10k: Add hotplug support for fm10k Michael Qiu
2015-07-19 20:26         ` [dpdk-dev] [PATCH 0/2 v5] Enable " Thomas Monjalon

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=533710CFB86FA344BFBF2D6802E602860469D92B@SHSMSX101.ccr.corp.intel.com \
    --to=michael.qiu@intel.com \
    --cc=bernard.iremonger@intel.com \
    --cc=dev@dpdk.org \
    --cc=shaopeng.he@intel.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).