From: Ferruh Yigit <ferruh.yigit@intel.com>
To: Pascal Mazon <pascal.mazon@6wind.com>
Cc: dev@dpdk.org, Keith Wiles <keith.wiles@intel.com>,
Vipin Varghese <vipin.varghese@intel.com>,
stable@dpdk.org
Subject: Re: [dpdk-dev] [PATCH 1/2] net/tap: fix setting speed by argument
Date: Tue, 24 Oct 2017 18:24:52 -0700 [thread overview]
Message-ID: <9e4a5217-ef03-861f-e17c-7eed22018b62@intel.com> (raw)
In-Reply-To: <8cf71f00-ca73-86b4-b6dd-fa24aac3167f@6wind.com>
On 9/19/2017 5:45 AM, Pascal Mazon wrote:
> Hi,
>
> The patch looks mainly ok to me.
>
> I'll put some comments inline.
>
> On 18/09/2017 20:47, Ferruh Yigit wrote:
>> From: Vipin Varghese <vipin.varghese@intel.com>
>>
>> tap speed argument is not working without generating any error.
> Can you describe the error, paste the log?
>>
>> This patch sets the configured speed during device start.
>>
>> Fixes: 02f96a0a82d1 ("net/tap: add TUN/TAP device PMD")
>> Cc: stable@dpdk.org
>>
>> Signed-off-by: Vipin Varghese <vipin.varghese@intel.com>
>> ---
>> drivers/net/tap/rte_eth_tap.c | 27 +++++++++++++++++++++++++++
>> drivers/net/tap/rte_eth_tap.h | 2 ++
>> 2 files changed, 29 insertions(+)
>>
>> diff --git a/drivers/net/tap/rte_eth_tap.c b/drivers/net/tap/rte_eth_tap.c
>> index 01cba0fa1..00dad167f 100644
>> --- a/drivers/net/tap/rte_eth_tap.c
>> +++ b/drivers/net/tap/rte_eth_tap.c
>> @@ -545,6 +545,7 @@ tap_ioctl(struct pmd_internals *pmd, unsigned long request,
>> case SIOCGIFHWADDR:
>> case SIOCSIFHWADDR:
>> case SIOCSIFMTU:
>> + case SIOCETHTOOL:
>> break;
>> default:
>> RTE_ASSERT(!"unsupported request type: must not happen");
>> @@ -585,6 +586,32 @@ static int
>> tap_dev_start(struct rte_eth_dev *dev)
>> {
>> int err;
>> + struct ifreq ifr;
>> + struct ethtool_cmd edata = {0};
>> + struct pmd_internals *pmd = dev->data->dev_private;
>> +
>> + /*set & get speed to device*/
>> + edata.speed = pmd_link.link_speed;
>> + edata.cmd = ETHTOOL_SSET;
>> + ifr.ifr_data = (caddr_t)&edata;
>> + if (tap_ioctl(pmd, SIOCETHTOOL, &ifr, 0, LOCAL_ONLY) < 0) {
> I think it would be best to also try setting the speed on the remote
> (tap_ioctl(REMOTE_ONLY)), but continue execution even if that fails.
Hi Pascal,
It looks like this ioctl fails to set tap speed. And as far as I can see speed
is hardcoded in kernel [1], do you know if it is possible to set speed of tap
interface?
Thanks,
ferruh
[1]
http://elixir.free-electrons.com/linux/v4.13.9/source/drivers/net/tun.c#L2460
>> + RTE_LOG(WARNING, PMD,
>> + "Could not set param speed %d for ifindex for %s.\n",
>> + pmd_link.link_speed,
>> + pmd->name);
>> +
>> + /* fetch current speed of created device */
>> + edata.cmd = ETHTOOL_GSET;
>> + ifr.ifr_data = (caddr_t)&edata;
>> + if (tap_ioctl(pmd, SIOCETHTOOL, &ifr, 0, LOCAL_ONLY) < 0)
>> + return 0;
>> +
>> + pmd_link.link_speed = edata.speed;
>> + RTE_LOG(INFO, PMD, "get speed %d for ifindex for %s.\n",
> I would say "got".
>
>> + pmd_link.link_speed,
>> + pmd->name);
>> + }
>> + dev->data->dev_link = pmd_link;
>>
>> err = tap_intr_handle_set(dev, 1);
>> if (err)
>> diff --git a/drivers/net/tap/rte_eth_tap.h b/drivers/net/tap/rte_eth_tap.h
>> index 928a0454e..3e91db4ae 100644
>> --- a/drivers/net/tap/rte_eth_tap.h
>> +++ b/drivers/net/tap/rte_eth_tap.h
>> @@ -40,6 +40,8 @@
>> #include <net/if.h>
>>
>> #include <linux/if_tun.h>
>> +#include <linux/ethtool.h>
>> +#include <linux/sockios.h>
>>
>> #include <rte_ethdev.h>
>> #include <rte_ether.h>
> These includes are only useful inside rte_eth_tap.c, I would put them
> there only.
>
next prev parent reply other threads:[~2017-10-25 1:24 UTC|newest]
Thread overview: 7+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-09-18 18:47 Ferruh Yigit
2017-09-18 18:47 ` [dpdk-dev] [PATCH 2/2] net/tap: fix unregistering callback with invalid fd Ferruh Yigit
2017-09-19 12:46 ` Pascal Mazon
2017-09-20 16:05 ` Ferruh Yigit
2017-09-19 12:45 ` [dpdk-dev] [PATCH 1/2] net/tap: fix setting speed by argument Pascal Mazon
2017-10-25 1:24 ` Ferruh Yigit [this message]
2017-10-25 6:27 ` Pascal Mazon
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=9e4a5217-ef03-861f-e17c-7eed22018b62@intel.com \
--to=ferruh.yigit@intel.com \
--cc=dev@dpdk.org \
--cc=keith.wiles@intel.com \
--cc=pascal.mazon@6wind.com \
--cc=stable@dpdk.org \
--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).