DPDK patches and discussions
 help / color / mirror / Atom feed
From: Ferruh Yigit <ferruh.yigit@intel.com>
To: Nalla Pradeep <pnalla@marvell.com>,
	Radha Mohan Chintakuntla <radhac@marvell.com>,
	Veerasenareddy Burru <vburru@marvell.com>,
	Anatoly Burakov <anatoly.burakov@intel.com>
Cc: jerinj@marvell.com, dev@dpdk.org, sburla@marvell.com
Subject: Re: [dpdk-dev] [PATCH v2 03/11] net/octeontx_ep: add device init and uninit
Date: Tue, 26 Jan 2021 15:25:43 +0000	[thread overview]
Message-ID: <c18c7267-be1d-967e-5e13-6bebd9c72079@intel.com> (raw)
In-Reply-To: <20210118093602.5449-3-pnalla@marvell.com>

On 1/18/2021 9:35 AM, Nalla Pradeep wrote:
> Add basic init and uninit function which includes
> initializing fields of ethdev private structure.
> 
> Signed-off-by: Nalla Pradeep <pnalla@marvell.com>

<...>

ep/otx_ep_common.h
> +++ b/drivers/net/octeontx_ep/otx_ep_common.h
> @@ -4,11 +4,28 @@
>   #ifndef _OTX_EP_COMMON_H_
>   #define _OTX_EP_COMMON_H_
>   
> +#define otx_ep_printf(level, fmt, args...)		\
> +	rte_log(RTE_LOG_ ## level, RTE_LOGTYPE_PMD,		\
> +		 fmt, ##args)

Can you please register a new type specific to PMD, instead of using the 
'RTE_LOGTYPE_PMD'?

<...>

> +static int
> +otx_ep_chip_specific_setup(struct otx_ep_device *otx_epvf)
> +{
> +	struct rte_pci_device *pdev = otx_epvf->pdev;
> +	uint32_t dev_id = pdev->id.device_id;
> +	int ret;
> +
> +	switch (dev_id) {
> +	case PCI_DEVID_OCTEONTX_EP_VF:
> +		otx_epvf->chip_id = dev_id;
> +		break;
> +	case PCI_DEVID_OCTEONTX2_EP_NET_VF:
> +	case PCI_DEVID_CN98XX_EP_NET_VF:
> +		otx_epvf->chip_id = dev_id;
> +		break;
> +	default:
> +		otx_ep_err("Unsupported device\n");
> +		ret = -EINVAL;
> +	}
> +
> +	if (!ret)
> +		otx_ep_info("OTX_EP dev_id[%d]\n", dev_id);

'ret' may be used uninitialized here. (I see it is fixed in later patches, but 
please fix here too)

<...>

> +
> +	return ret;
> +}
> +
> +/* OTX_EP VF device initialization */
> +static int
> +otx_epdev_init(struct otx_ep_device *otx_epvf)
> +{
> +	if (otx_ep_chip_specific_setup(otx_epvf)) {
> +		otx_ep_err("Chip specific setup failed\n");
> +		goto setup_fail;
> +	}
> +
> +	return 0;
> +
> +setup_fail:
> +	return -ENOMEM;

Is 'NOMEM' correct return type, there seems nothing related to the memory.

> +}
> +
>   static int
>   otx_ep_eth_dev_uninit(struct rte_eth_dev *eth_dev)
>   {
> -	RTE_SET_USED(eth_dev);
> +	struct otx_ep_device *otx_epvf = OTX_EP_DEV(eth_dev);
> +
> +	if (rte_eal_process_type() != RTE_PROC_PRIMARY)
> +		return 0;
> +	otx_epvf->port_configured = 0;
> +
> +	if (eth_dev->data->mac_addrs != NULL)
> +		rte_free(eth_dev->data->mac_addrs);
>   

After 'otx_ep_eth_dev_uninit()' the 'rte_eth_dev_release_port()' is called:
rte_eth_dev_pci_generic_remove
   dev_uninit()
   rte_eth_dev_release_port()

'rte_eth_dev_pci_generic_remove()' also frees the 'eth_dev->data->mac_addrs' 
leading double free.

You can either free yourself and explicitly set the pointer to NULL, or relay 
the freeing of it to the 'rte_eth_dev_pci_generic_remove()'.

> -	return -ENODEV;
> +	return 0;
>   }
>   
>   static int
>   otx_ep_eth_dev_init(struct rte_eth_dev *eth_dev)
>   {
> -	RTE_SET_USED(eth_dev);
> +	struct rte_pci_device *pdev = RTE_ETH_DEV_TO_PCI(eth_dev);
> +	struct otx_ep_device *otx_epvf = OTX_EP_DEV(eth_dev);
> +	unsigned char vf_mac_addr[RTE_ETHER_ADDR_LEN];
>   
> -	return -ENODEV;
> +	/* Single process support */
> +	if (rte_eal_process_type() != RTE_PROC_PRIMARY)
> +		return 0;
> +
> +	rte_eth_copy_pci_info(eth_dev, pdev);
> +

'rte_eth_copy_pci_info()' already done before this point, as:
rte_eth_dev_pci_generic_probe
   rte_eth_dev_pci_allocate
     rte_eth_copy_pci_info

Can you please double check is the copying again required?

> +	if (pdev->mem_resource[0].addr) {
> +		otx_ep_info("OTX_EP BAR0 is mapped:\n");
> +	} else {
> +		otx_ep_err("OTX_EP: Failed to map device BARs\n");
> +		otx_ep_err("BAR0 %p\n", pdev->mem_resource[0].addr);
> +		return -ENODEV;
> +	}

Why do you need to check if BAR0 is mapped or not, when it can be unmapped?

> +	otx_epvf->eth_dev = eth_dev;
> +	otx_epvf->port_id = eth_dev->data->port_id;
> +	eth_dev->data->mac_addrs = rte_zmalloc("otx_ep", RTE_ETHER_ADDR_LEN, 0);
> +	if (eth_dev->data->mac_addrs == NULL) {
> +		otx_ep_err("MAC addresses memory allocation failed\n");
> +		return -ENOMEM;
> +	}
> +	rte_eth_random_addr(vf_mac_addr);
> +	memcpy(eth_dev->data->mac_addrs, vf_mac_addr, RTE_ETHER_ADDR_LEN);

You can use 'rte_ether_addr_copy()' to copy mac.

> +	otx_epvf->hw_addr = pdev->mem_resource[0].addr;
> +	otx_epvf->pdev = pdev;
> +
> +	otx_epdev_init(otx_epvf);
> +	otx_epvf->port_configured = 0;

Is 'port_configured' used? It seems not checked at all.

> +
> +	return 0;
>   }
>   
>   static int
> @@ -42,7 +121,6 @@ otx_ep_eth_dev_pci_remove(struct rte_pci_device *pci_dev)
>   					      otx_ep_eth_dev_uninit);
>   }
>   
> -

Please fix these kind of things in the original patch that introduces it.

>   /* Set of PCI devices this driver supports */
>   static const struct rte_pci_id pci_id_otx_ep_map[] = {
>   	{ RTE_PCI_DEVICE(PCI_VENDOR_ID_CAVIUM, PCI_DEVID_OCTEONTX_EP_VF) },
> 


  parent reply	other threads:[~2021-01-26 15:25 UTC|newest]

Thread overview: 76+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-01-18  9:35 [dpdk-dev] [PATCH v2 01/11] net/octeontx_ep: add build and doc infrastructure Nalla Pradeep
2021-01-18  9:35 ` [dpdk-dev] [PATCH v2 02/11] net/octeontx_ep: add ethdev probe and remove Nalla Pradeep
2021-01-19 11:55   ` Jerin Jacob
2021-01-26 15:27   ` Ferruh Yigit
2021-01-27 11:45     ` [dpdk-dev] [EXT] " Pradeep Kumar Nalla
2021-01-18  9:35 ` [dpdk-dev] [PATCH v2 03/11] net/octeontx_ep: add device init and uninit Nalla Pradeep
2021-01-19 12:06   ` Jerin Jacob
2021-01-26 15:25   ` Ferruh Yigit [this message]
2021-01-18  9:35 ` [dpdk-dev] [PATCH v2 04/11] net/octeontx_ep: Added basic device setup Nalla Pradeep
2021-01-19 12:09   ` Jerin Jacob
2021-01-26 15:29   ` Ferruh Yigit
2021-01-18  9:35 ` [dpdk-dev] [PATCH v2 05/11] net/octeontx_ep: Add dev info get and configure Nalla Pradeep
2021-01-26 15:31   ` Ferruh Yigit
2021-01-28  7:03     ` [dpdk-dev] [EXT] " Pradeep Kumar Nalla
2021-01-18  9:35 ` [dpdk-dev] [PATCH v2 06/11] net/octeontx_ep: Added rxq setup and release Nalla Pradeep
2021-01-18  9:35 ` [dpdk-dev] [PATCH v2 07/11] net/octeontx_ep: Added tx queue " Nalla Pradeep
2021-01-18  9:35 ` [dpdk-dev] [PATCH v2 08/11] net/octeontx_ep: Setting up iq and oq registers Nalla Pradeep
2021-01-26 15:32   ` Ferruh Yigit
2021-01-18  9:36 ` [dpdk-dev] [PATCH v2 09/11] net/octeontx_ep: Added dev start and stop Nalla Pradeep
2021-01-26 15:32   ` Ferruh Yigit
2021-01-18  9:36 ` [dpdk-dev] [PATCH v2 10/11] net/octeontx_ep: Receive data path function added Nalla Pradeep
2021-01-26 15:33   ` Ferruh Yigit
2021-01-18  9:36 ` [dpdk-dev] [PATCH v2 11/11] net/octeontx_ep: Transmit " Nalla Pradeep
2021-01-26 15:35   ` Ferruh Yigit
2021-01-19 11:50 ` [dpdk-dev] [PATCH v2 01/11] net/octeontx_ep: add build and doc infrastructure Jerin Jacob
2021-01-26 15:09 ` Ferruh Yigit
2021-01-26 15:41 ` Ferruh Yigit
2021-01-28 15:22 ` [dpdk-dev] [PATCH v5 00/11] Octeon Tx/Tx2 Endpoint pmd Nalla Pradeep
2021-01-29  0:16   ` [dpdk-dev] [PATCH v6 00/12] " Nalla Pradeep
2021-01-29 12:44     ` [dpdk-dev] [PATCH v7 " Nalla Pradeep
2021-01-29 14:08       ` Ferruh Yigit
2021-01-29 12:44     ` [dpdk-dev] [PATCH v7 01/12] raw/octeontx_ep: changed device id Nalla Pradeep
2021-02-11  0:58       ` Radha Mohan
2021-01-29 12:45     ` [dpdk-dev] [PATCH v7 02/12] net/octeontx_ep: add build and doc infrastructure Nalla Pradeep
2021-01-29 12:45     ` [dpdk-dev] [PATCH v7 03/12] net/octeontx_ep: add ethdev probe and remove Nalla Pradeep
2021-01-29 12:45     ` [dpdk-dev] [PATCH v7 04/12] net/octeontx_ep: add device init and uninit Nalla Pradeep
2021-01-29 12:45     ` [dpdk-dev] [PATCH v7 05/12] net/octeontx_ep: added basic device setup Nalla Pradeep
2021-01-29 12:45     ` [dpdk-dev] [PATCH v7 06/12] net/octeontx_ep: add dev info get and configure Nalla Pradeep
2021-01-29 12:45     ` [dpdk-dev] [PATCH v7 07/12] net/octeontx_ep: added rxq setup and release Nalla Pradeep
2021-01-29 14:04       ` Ferruh Yigit
2021-01-29 12:45     ` [dpdk-dev] [PATCH v7 08/12] net/octeontx_ep: added tx queue " Nalla Pradeep
2021-01-29 14:04       ` Ferruh Yigit
2021-01-29 12:45     ` [dpdk-dev] [PATCH v7 09/12] net/octeontx_ep: setting up iq and oq registers Nalla Pradeep
2021-01-29 12:45     ` [dpdk-dev] [PATCH v7 10/12] net/octeontx_ep: added dev start and stop Nalla Pradeep
2021-01-29 12:45     ` [dpdk-dev] [PATCH v7 11/12] net/octeontx_ep: receive data path function added Nalla Pradeep
2021-01-29 12:45     ` [dpdk-dev] [PATCH v7 12/12] net/octeontx_ep: transmit " Nalla Pradeep
2021-01-29  0:16   ` [dpdk-dev] [PATCH v6 01/12] raw/octeontx_ep: changed device id Nalla Pradeep
2021-01-29  0:16   ` [dpdk-dev] [PATCH v6 02/12] net/octeontx_ep: add build and doc infrastructure Nalla Pradeep
2021-01-29  0:16   ` [dpdk-dev] [PATCH v6 03/12] net/octeontx_ep: add ethdev probe and remove Nalla Pradeep
2021-01-29  0:16   ` [dpdk-dev] [PATCH v6 04/12] net/octeontx_ep: add device init and uninit Nalla Pradeep
2021-01-29  9:19     ` Ferruh Yigit
2021-01-29  0:16   ` [dpdk-dev] [PATCH v6 05/12] net/octeontx_ep: added basic device setup Nalla Pradeep
2021-01-29  0:16   ` [dpdk-dev] [PATCH v6 06/12] net/octeontx_ep: add dev info get and configure Nalla Pradeep
2021-01-29  0:16   ` [dpdk-dev] [PATCH v6 07/12] net/octeontx_ep: added rxq setup and release Nalla Pradeep
2021-01-29  0:16   ` [dpdk-dev] [PATCH v6 08/12] net/octeontx_ep: added tx queue " Nalla Pradeep
2021-01-29  0:16   ` [dpdk-dev] [PATCH v6 09/12] net/octeontx_ep: setting up iq and oq registers Nalla Pradeep
2021-01-29  0:16   ` [dpdk-dev] [PATCH v6 10/12] net/octeontx_ep: added dev start and stop Nalla Pradeep
2021-01-29  0:16   ` [dpdk-dev] [PATCH v6 11/12] net/octeontx_ep: receive data path function added Nalla Pradeep
2021-01-29  9:24     ` Ferruh Yigit
2021-01-29  0:16   ` [dpdk-dev] [PATCH v6 12/12] net/octeontx_ep: transmit " Nalla Pradeep
2021-01-28 15:22 ` [dpdk-dev] [PATCH v5 01/11] net/octeontx_ep: add build and doc infrastructure Nalla Pradeep
2021-01-28 15:22 ` [dpdk-dev] [PATCH v5 02/11] net/octeontx_ep: add ethdev probe and remove Nalla Pradeep
2021-01-28 15:22 ` [dpdk-dev] [PATCH v5 03/11] net/octeontx_ep: add device init and uninit Nalla Pradeep
2021-01-28 16:55   ` Ferruh Yigit
2021-01-28 15:22 ` [dpdk-dev] [PATCH v5 04/11] net/octeontx_ep: added basic device setup Nalla Pradeep
2021-01-28 15:22 ` [dpdk-dev] [PATCH v5 05/11] net/octeontx_ep: add dev info get and configure Nalla Pradeep
2021-01-28 16:56   ` Ferruh Yigit
2021-01-28 15:22 ` [dpdk-dev] [PATCH v5 06/11] net/octeontx_ep: added rxq setup and release Nalla Pradeep
2021-01-28 15:22 ` [dpdk-dev] [PATCH v5 07/11] net/octeontx_ep: added tx queue " Nalla Pradeep
2021-01-28 16:57   ` Ferruh Yigit
2021-01-28 15:22 ` [dpdk-dev] [PATCH v5 08/11] net/octeontx_ep: setting up iq and oq registers Nalla Pradeep
2021-01-28 15:22 ` [dpdk-dev] [PATCH v5 09/11] net/octeontx_ep: added dev start and stop Nalla Pradeep
2021-01-28 15:22 ` [dpdk-dev] [PATCH v5 10/11] net/octeontx_ep: receive data path function added Nalla Pradeep
2021-01-28 16:58   ` Ferruh Yigit
2021-01-28 15:22 ` [dpdk-dev] [PATCH v5 11/11] net/octeontx_ep: transmit " Nalla Pradeep
2021-01-28 16:59   ` 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=c18c7267-be1d-967e-5e13-6bebd9c72079@intel.com \
    --to=ferruh.yigit@intel.com \
    --cc=anatoly.burakov@intel.com \
    --cc=dev@dpdk.org \
    --cc=jerinj@marvell.com \
    --cc=pnalla@marvell.com \
    --cc=radhac@marvell.com \
    --cc=sburla@marvell.com \
    --cc=vburru@marvell.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).