DPDK patches and discussions
 help / color / mirror / Atom feed
From: "Wiles, Keith" <keith.wiles@intel.com>
To: "Varghese, Vipin" <vipin.varghese@intel.com>
Cc: Ophir Munk <ophirmu@mellanox.com>, "dev@dpdk.org" <dev@dpdk.org>,
	"Yigit,  Ferruh" <ferruh.yigit@intel.com>,
	Thomas Monjalon <thomas@monjalon.net>,
	"Olga Shern" <olgas@mellanox.com>
Subject: Re: [dpdk-dev] [PATCH v2] net/tap: perform proto field update for tun only
Date: Tue, 22 May 2018 13:55:09 +0000	[thread overview]
Message-ID: <23DB1EDD-4358-419E-B1CE-65CFB6CFB3A9@intel.com> (raw)
In-Reply-To: <4C9E0AB70F954A408CC4ADDBF0F8FA7D4D1F84BF@BGSMSX101.gar.corp.intel.com>



> On May 22, 2018, at 6:05 AM, Varghese, Vipin <vipin.varghese@intel.com> wrote:
> 
> Hi Ophir, Keith and Ferruh,
> 
> I have shared v3 with all the suggested changes. 

I just sent another change, sorry I did not catch if before, which I obviously was not paying attention.

> 
> Thanks
> Vipin Varghese
> 
>> -----Original Message-----
>> From: Varghese, Vipin
>> Sent: Tuesday, May 22, 2018 12:04 PM
>> To: Ophir Munk <ophirmu@mellanox.com>; dev@dpdk.org; Wiles, Keith
>> <keith.wiles@intel.com>; Yigit, Ferruh <ferruh.yigit@intel.com>
>> Cc: Thomas Monjalon <thomas@monjalon.net>; Olga Shern
>> <olgas@mellanox.com>
>> Subject: RE: [PATCH v2] net/tap: perform proto field update for tun only
>> 
>> Hi Ophir,
>> 
>> Thanks for the suggestion, once these are merged to dpdk or dpdk-net mainline,
>> I can start the modifications.
>> 
>> Thanks
>> Vipin Varghese
>> 
>>> -----Original Message-----
>>> From: Ophir Munk [mailto:ophirmu@mellanox.com]
>>> Sent: Wednesday, May 16, 2018 3:59 AM
>>> To: Varghese, Vipin <vipin.varghese@intel.com>; dev@dpdk.org; Wiles,
>>> Keith <keith.wiles@intel.com>; Yigit, Ferruh <ferruh.yigit@intel.com>
>>> Cc: Thomas Monjalon <thomas@monjalon.net>; Olga Shern
>>> <olgas@mellanox.com>
>>> Subject: RE: [PATCH v2] net/tap: perform proto field update for tun
>>> only
>>> 
>>> Hi,
>>> I am adding my review although this patch has already been acked.
>>> Please find some comments inline. Otherwise it looks OK.
>>> 
>>> Regards,
>>> Ophir
>>> 
>>>> -----Original Message-----
>>>> From: Vipin Varghese [mailto:vipin.varghese@intel.com]
>>>> Sent: Tuesday, May 15, 2018 5:50 PM
>>>> To: dev@dpdk.org; keith.wiles@intel.com; ferruh.yigit@intel.com;
>>>> Ophir Munk <ophirmu@mellanox.com>
>>>> Cc: Vipin Varghese <vipin.varghese@intel.com>
>>>> Subject: [PATCH v2] net/tap: perform proto field update for tun only
>>>> 
>>>> The TX function is shared between TAP and TUN PMD. Checking TUN-TAP
>>>> type field will ensure the TAP PMD will always have protocol field as 0.
>>>> 
>>>> Signed-off-by: Vipin Varghese <vipin.varghese@intel.com>
>>>> Suggested-by: Ferruh Yigit <ferruh.yigit@intel.com>
>>>> ---
>>>> 
>>>> Changes in V2:
>>>> updated in comment wording - Keith Wiles remove debug print from tx
>>>> code
>>>> - Keith Wiles
>>>> ---
>>>> drivers/net/tap/rte_eth_tap.c | 48
>>>> ++++++++++++++++++++++++++-----------
>>>> ------
>>>> drivers/net/tap/rte_eth_tap.h | 10 +++++++++
>>>> 2 files changed, 39 insertions(+), 19 deletions(-)
>>>> 
>>>> diff --git a/drivers/net/tap/rte_eth_tap.c
>>>> b/drivers/net/tap/rte_eth_tap.c index ea6d899..fafd557 100644
>>>> --- a/drivers/net/tap/rte_eth_tap.c
>>>> +++ b/drivers/net/tap/rte_eth_tap.c
>>>> @@ -115,7 +115,8 @@ tun_alloc(struct pmd_internals *pmd)
>>>> 	 * Do not set IFF_NO_PI as packet information header will be needed
>>>> 	 * to check if a received packet has been truncated.
>>>> 	 */
>>>> -	ifr.ifr_flags = (tap_type) ? IFF_TAP : IFF_TUN | IFF_POINTOPOINT;
>>>> +	ifr.ifr_flags = (pmd->type == ETH_TUNTAP_TYPE_TAP) ?
>>>> +			IFF_TAP : IFF_TUN | IFF_POINTOPOINT;
>>>> 	snprintf(ifr.ifr_name, IFNAMSIZ, "%s", pmd->name);
>>>> 
>>>> 	TAP_LOG(DEBUG, "ifr_name '%s'", ifr.ifr_name); @@ -502,20
>>>> +503,22 @@ pmd_tx_burst(void *queue, struct rte_mbuf **bufs,
>>>> +uint16_t
>>>> nb_pkts)
>>>> 		if (rte_pktmbuf_pkt_len(mbuf) > max_size)
>>>> 			break;
>>>> 
>>>> -		/*
>>>> -		 * TUN and TAP are created with IFF_NO_PI disabled.
>>>> -		 * For TUN PMD this mandatory as fields are used by
>>>> -		 * Kernel tun.c to determine whether its IP or non IP
>>>> -		 * packets.
>>>> -		 *
>>>> -		 * The logic fetches the first byte of data from mbuf.
>>>> -		 * compares whether its v4 or v6. If none matches default
>>>> -		 * value 0x00 is taken for protocol field.
>>>> -		 */
>>>> -		char *buff_data = rte_pktmbuf_mtod(seg, void *);
>>>> -		j = (*buff_data & 0xf0);
>>>> -		pi.proto = (j == 0x40) ? 0x0008 :
>>>> -				(j == 0x60) ? 0xdd86 : 0x00;
>>>> +		if (txq->type == ETH_TUNTAP_TYPE_TUN) {
>>>> +			/*
>>>> +			 * TUN and TAP are created with IFF_NO_PI disabled.
>>>> +			 * For TUN PMD this mandatory as fields are used by
>>>> +			 * Kernel tun.c to determine whether its IP or non IP
>>>> +			 * packets.
>>>> +			 *
>>>> +			 * The logic fetches the first byte of data from mbuf
>>>> +			 * then compares whether its v4 or v6. If first byte
>>>> +			 * is 4 or 6, then protocol field is updated.
>>>> +			 */
>>>> +			char *buff_data = rte_pktmbuf_mtod(seg, void *);
>>>> +			j = (*buff_data & 0xf0);
>>>> +			pi.proto = (j == 0x40) ? 0x0008 :
>>>> +					(j == 0x60) ? 0xdd86 : 0x00;
>>>> +		}
>>>> 
>>>> 		iovecs[0].iov_base = &pi;
>>>> 		iovecs[0].iov_len = sizeof(pi);
>>>> @@ -1052,6 +1055,9 @@ tap_setup_queue(struct rte_eth_dev *dev,
>>>> 	tx->mtu = &dev->data->mtu;
>>>> 	rx->rxmode = &dev->data->dev_conf.rxmode;
>>>> 
>>>> +	tx->type = pmd->type;
>>>> +	rx->type = pmd->type;
>>> 
>>> It seems there is no usage for rx->type in TAP code. Therefore - can
>>> you please remove rx->type assignment?
>>>> +
>>>> 	return *fd;
>>>> }
>>>> 
>>>> @@ -1386,6 +1392,7 @@ eth_dev_tap_create(struct rte_vdev_device
>>>> *vdev, char *tap_name,
>>>> 	pmd = dev->data->dev_private;
>>>> 	pmd->dev = dev;
>>>> 	snprintf(pmd->name, sizeof(pmd->name), "%s", tap_name);
>>>> +	pmd->type = ETH_TUNTAP_TYPE_UNKNOWN;
>>>> 
>>>> 	pmd->ioctl_sock = socket(AF_INET, SOCK_DGRAM, 0);
>>>> 	if (pmd->ioctl_sock == -1) {
>>>> @@ -1421,7 +1428,9 @@ eth_dev_tap_create(struct rte_vdev_device
>>>> *vdev, char *tap_name,
>>>> 		pmd->txq[i].fd = -1;
>>>> 	}
>>>> 
>>>> -	if (tap_type) {
>>>> +	pmd->type = (tap_type) ? ETH_TUNTAP_TYPE_TAP :
>>>> ETH_TUNTAP_TYPE_TUN;
>>> 
>>> 1. It seem you are using 'tap_type' as a global static variable to
>>> pass a value to eth_dev_tap_create().
>>> Can you please pass this value by adding a parameter to
>>> eth_dev_tap_create() call and remove the global static variable?
>>> 2. You are using tap_type with values 0 and 1 which are converted to
>>> enums ETH_TUNTAP_TYPE_TUN and ETH_TUNTAP_TYPE_TAP respectively.
>> Can
>>> you please always just use the enums (avoiding the conversion)?
>>> 
>>>> +
>>>> +	if (pmd->type == ETH_TUNTAP_TYPE_TAP) {
>>>> 		if (is_zero_ether_addr(mac_addr))
>>>> 			eth_random_addr((uint8_t *)&pmd->eth_addr);
>>>> 		else
>>>> @@ -1440,7 +1449,7 @@ eth_dev_tap_create(struct rte_vdev_device
>>>> *vdev, char *tap_name,
>>>> 	if (tap_ioctl(pmd, SIOCSIFMTU, &ifr, 1, LOCAL_AND_REMOTE) < 0)
>>>> 		goto error_exit;
>>>> 
>>>> -	if (tap_type) {
>>>> +	if (pmd->type == ETH_TUNTAP_TYPE_TAP) {
>>>> 		memset(&ifr, 0, sizeof(struct ifreq));
>>>> 		ifr.ifr_hwaddr.sa_family = AF_LOCAL;
>>>> 		rte_memcpy(ifr.ifr_hwaddr.sa_data, &pmd->eth_addr, @@ -
>>>> 1812,7 +1821,9 @@ rte_pmd_tap_remove(struct rte_vdev_device *dev)
>>>> 	struct pmd_internals *internals;
>>>> 	int i;
>>>> 
>>>> -	TAP_LOG(DEBUG, "Closing TUN/TAP Ethernet device on numa %u",
>>>> +	internals = eth_dev->data->dev_private;
>>>> +	TAP_LOG(DEBUG, "Closing %s Ethernet device on numa %u",
>>>> +		(internals->type == ETH_TUNTAP_TYPE_TAP) ? "TAP" :
>>>> "TUN",
>>>> 		rte_socket_id());
>>>> 
>>>> 	/* find the ethdev entry */
>>>> @@ -1820,7 +1831,6 @@ rte_pmd_tap_remove(struct rte_vdev_device
>>>> *dev)
>>>> 	if (!eth_dev)
>>>> 		return 0;
>>>> 
>>>> -	internals = eth_dev->data->dev_private;
>>>> 	if (internals->nlsk_fd) {
>>>> 		tap_flow_flush(eth_dev, NULL);
>>>> 		tap_flow_implicit_flush(internals, NULL); diff --git
>>>> a/drivers/net/tap/rte_eth_tap.h b/drivers/net/tap/rte_eth_tap.h
>>>> index 67c9d4b..8b0da5a 100644
>>>> --- a/drivers/net/tap/rte_eth_tap.h
>>>> +++ b/drivers/net/tap/rte_eth_tap.h
>>>> @@ -23,6 +23,13 @@
>>>> #define RTE_PMD_TAP_MAX_QUEUES	1
>>>> #endif
>>>> 
>>>> +enum rte_tuntap_type {
>>>> +	ETH_TUNTAP_TYPE_UNKNOWN,
>>>> +	ETH_TUNTAP_TYPE_TUN,
>>>> +	ETH_TUNTAP_TYPE_TAP,
>>>> +	ETH_TUNTAP_TYPE_MAX,
>>>> +};
>>>> +
>>>> struct pkt_stats {
>>>> 	uint64_t opackets;              /* Number of output packets */
>>>> 	uint64_t ipackets;              /* Number of input packets */
>>>> @@ -38,6 +45,7 @@ struct rx_queue {
>>>> 	uint32_t trigger_seen;          /* Last seen Rx trigger value */
>>>> 	uint16_t in_port;               /* Port ID */
>>>> 	int fd;
>>>> +	int type;			  /* Type field - TUN|TAP */
>>> 
>>> In seems there is no usage for 'type' field from rx_queue struct in
>>> the code. If so
>>> - can you please remove it?
>>> 
>>>> 	struct pkt_stats stats;         /* Stats for this RX queue */
>>>> 	uint16_t nb_rx_desc;            /* max number of mbufs available */
>>>> 	struct rte_eth_rxmode *rxmode;  /* RX features */ @@ -48,6 +56,7
>>> @@
>>>> struct rx_queue {
>>>> 
>>>> struct tx_queue {
>>>> 	int fd;
>>>> +	int type;			  /* Type field - TUN|TAP */
>>> 
>>> Please use spaces (like all other fields in this struct) instead of
>>> tabs, so all fields look aligned.
>>> 
>>>> 	uint16_t *mtu;                  /* Pointer to MTU from dev_data */
>>>> 	uint16_t csum:1;                /* Enable checksum offloading */
>>>> 	struct pkt_stats stats;         /* Stats for this TX queue */
>>>> @@ -57,6 +66,7 @@ struct pmd_internals {
>>>> 	struct rte_eth_dev *dev;          /* Ethernet device. */
>>>> 	char remote_iface[RTE_ETH_NAME_MAX_LEN]; /* Remote netdevice
>>> name */
>>>> 	char name[RTE_ETH_NAME_MAX_LEN];  /* Internal Tap device name */
>>>> +	int type;			  /* Type field - TUN|TAP */
>>> 
>>> Same comment here about replacing tabs with spaces
>>> 
>>>> 	struct ether_addr eth_addr;       /* Mac address of the device port */
>>>> 	struct ifreq remote_initial_flags;   /* Remote netdevice flags on init
>>>> */
>>>> 	int remote_if_index;              /* remote netdevice IF_INDEX */
>>>> --
>>>> 2.7.4
> 

Regards,
Keith

  reply	other threads:[~2018-05-22 13:55 UTC|newest]

Thread overview: 19+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-05-14  4:53 [dpdk-dev] [PATCH] " Vipin Varghese
2018-05-14 22:27 ` Wiles, Keith
2018-05-15  9:08   ` Varghese, Vipin
2018-05-15 14:49 ` [dpdk-dev] [PATCH v2] " Vipin Varghese
2018-05-15 12:31   ` Wiles, Keith
2018-05-15 22:28   ` Ophir Munk
2018-05-22  6:33     ` Varghese, Vipin
2018-05-22 11:05       ` Varghese, Vipin
2018-05-22 13:55         ` Wiles, Keith [this message]
2018-05-22 10:47   ` [dpdk-dev] [PATCH v3] " Vipin Varghese
2018-05-22 13:53     ` Wiles, Keith
2018-05-23  4:33       ` Varghese, Vipin
2018-05-22 15:04     ` Ophir Munk
2018-05-23  4:37       ` Varghese, Vipin
2018-05-22 22:04     ` [dpdk-dev] [PATCH v4] " Ferruh Yigit
2018-05-22 22:06       ` Ferruh Yigit
2018-05-22 22:09       ` [dpdk-dev] [PATCH v5] net/tap: fix proto field for tun Ferruh Yigit
2018-05-22 22:19         ` Ferruh Yigit
2018-05-22 22:25           ` 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=23DB1EDD-4358-419E-B1CE-65CFB6CFB3A9@intel.com \
    --to=keith.wiles@intel.com \
    --cc=dev@dpdk.org \
    --cc=ferruh.yigit@intel.com \
    --cc=olgas@mellanox.com \
    --cc=ophirmu@mellanox.com \
    --cc=thomas@monjalon.net \
    --cc=vipin.varghese@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).