From: "Wiles, Keith" <keith.wiles@intel.com>
To: Pascal Mazon <pascal.mazon@6wind.com>
Cc: "Yigit, Ferruh" <ferruh.yigit@intel.com>, "dev@dpdk.org" <dev@dpdk.org>
Subject: Re: [dpdk-dev] [PATCH 1/6] net/tap: use correct tap name
Date: Tue, 31 Jan 2017 16:06:40 +0000 [thread overview]
Message-ID: <74B7FC24-638D-4A48-A02D-C80B5B4E7E2B@intel.com> (raw)
In-Reply-To: <e1b5c4ec-7609-5a40-311a-61087ad2cad9@6wind.com>
> On Jan 31, 2017, at 9:44 AM, Pascal Mazon <pascal.mazon@6wind.com> wrote:
>
> On 01/31/2017 04:44 PM, Wiles, Keith wrote:
>>
>>> On Jan 31, 2017, at 9:38 AM, Yigit, Ferruh <ferruh.yigit@intel.com> wrote:
>>>
>>> On 1/31/2017 3:30 PM, Pascal Mazon wrote:
>>>> On 01/31/2017 04:28 PM, Ferruh Yigit wrote:
>>>>> On 1/31/2017 2:23 PM, Pascal Mazon wrote:
>>>>>> On 01/31/2017 02:06 PM, Ferruh Yigit wrote:> On 1/31/2017 9:42 AM,
>>>>>> Pascal Mazon wrote:
>>>>>>>> dev->data->name contains "net_tap", the device driver name.
>>>>>>>
>>>>>>> I see what patch does, just as a note to commit log:
>>>>>>>
>>>>>>> AFAIK, "dev->data->name" is device name, and for this case it is
>>>>>>> "net_tap#", like "net_tap0", "net_tap1" ...
>>>>>>>
>>>>>>> "dev->data_drv_name" is the driver name which is "net_tap"
>>>>>>
>>>>>> Indeed, dev->data->name is the device name, looking like "net_tap#",
>>>>>> with number increasing for each vdev.
>>>>>> I'll put the following commit log line if that's ok:
>>>>>>
>>>>>> dev->data->name contains the device name, e.g. "net_tap0".
>>>>>>
>>>>>>>
>>>>>>>> dev->data->dev_private->name contains the actual iface name,
>>>>>>>> e.g. "dtap0".
>>>>>>>
>>>>>>> Right, I agree this is correct comparing "dev->data->name"
>>>>>>>
>>>>>>> But the problem is pmd->name is per eth_dev.
>>>>>>>
>>>>>>> If I read code correct, for multiple queue support, each queue pair will
>>>>>>> create a tap device, so each needs a different name.
>>>>>>>
>>>>>>> So can't just use pmd->name. Need to create a name per queue pair, it
>>>>>>> can be combination of pmd->name + "_" + queue_id? Or can keep a name per
>>>>>>> queue pair, instead of eth_dev.
>>>>>>>
>>>>>>> What do you think?
>>>>>>
>>>>>> Actually that's not exactly how it goes.
>>>>>> Adding a queue to a netdevice requires to open("/dev/net/tun") and setting
>>>>>> TUNSETIFF (through ioctl) on the resulting fd.
>>>>>> That's the important part: queues for a given tap device must set TUNSETIFF
>>>>>> with a common ifreq.ifr_name (in our case, pmd->name).
Looking at the changes to set the link up/down and the adding the two functions. I noticed in the stop/start routines I was set the link in DPDK and not adjusting the interface link. Should the stop/start routine also do the same thing?
>>>>>>
>>>>>> This is best explained in the kernel doc, there:
>>>>>
>>>>> Thank you for the clarification.
>>>>> If so, why PMD keeps a fd per queue?
>>>>>
>>>>> Overall, patch looks good except mentioned detail in commit log.
>>>>>
>>>>> I suggest waiting Keith's patch, and rebase this set on top of it.
>>>>>
>>>>> Thanks,
>>>>> ferruh
>>>>>
>>>>>>
>>>>>> [1]
>>>>>> https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/tree/Documentation/networking/tuntap.txt#n108
>>>>>>
>>>>> <...>
>>>>>
>>>>
>>>> I would say it is because dev->{r/t}x_pkt_burst() is done per queue.
>>>> In these functions, we need to make our read() and write() accesses on the
>>>> right fd considering the queue we're required to process.
>>>
>>> If fd is same for all queues, it is possible to keep one instance in pmd
>>> private_data, and access it from queues. I think it is confusing to have
>>> multiple copy of it, but I see your point.
>>
>> In my changes I removed the fds[] array as it was not required only the rx/tx_queue has an fd variable.
>
> Sounds good to me.
>
> Ferruh, just to make sure we're on the same page, in multiqueue, there is definitely a different fd for each queue.
>
>>
>>>
>>>>
>>>> I'll wait for Keith's patch, then.
>>>
>>> Thanks again.
>>>
>>>>
>>>> Best regards,
>>
>> Regards,
>> Keith
>>
>
>
> --
> Pascal Mazon
> www.6wind.com
Regards,
Keith
next prev parent reply other threads:[~2017-01-31 16:06 UTC|newest]
Thread overview: 45+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-01-31 9:42 Pascal Mazon
2017-01-31 9:42 ` [dpdk-dev] [PATCH 2/6] net/tap: use correct channel for error logs Pascal Mazon
2017-01-31 13:07 ` Ferruh Yigit
2017-01-31 16:58 ` Stephen Hemminger
2017-01-31 17:04 ` Wiles, Keith
2017-01-31 9:42 ` [dpdk-dev] [PATCH 3/6] net/tap: don't set fd value overwritten just below Pascal Mazon
2017-01-31 9:42 ` [dpdk-dev] [PATCH 4/6] net/tap: keep kernel-assigned MAC address Pascal Mazon
2017-01-31 13:13 ` Ferruh Yigit
2017-01-31 9:42 ` [dpdk-dev] [PATCH 5/6] net/tap: display tap name after parsing Pascal Mazon
2017-01-31 13:16 ` Ferruh Yigit
2017-01-31 9:42 ` [dpdk-dev] [PATCH 6/6] net/tap: implement link up and down callbacks Pascal Mazon
2017-01-31 13:21 ` Ferruh Yigit
2017-01-31 14:31 ` Pascal Mazon
2017-01-31 13:06 ` [dpdk-dev] [PATCH 1/6] net/tap: use correct tap name Ferruh Yigit
2017-01-31 14:23 ` Pascal Mazon
2017-01-31 15:28 ` Ferruh Yigit
2017-01-31 15:30 ` Pascal Mazon
2017-01-31 15:38 ` Ferruh Yigit
2017-01-31 15:44 ` Wiles, Keith
2017-01-31 15:44 ` Pascal Mazon
2017-01-31 16:06 ` Wiles, Keith [this message]
2017-01-31 16:39 ` Pascal Mazon
2017-01-31 23:29 ` Wiles, Keith
2017-02-01 8:11 ` Pascal Mazon
2017-02-01 15:25 ` Wiles, Keith
2017-02-01 15:40 ` Pascal Mazon
2017-02-01 15:55 ` Wiles, Keith
2017-02-01 17:50 ` Ferruh Yigit
2017-02-02 8:05 ` Pascal Mazon
2017-02-02 8:25 ` Pascal Mazon
2017-02-02 10:23 ` Ferruh Yigit
2017-01-31 14:52 ` Wiles, Keith
2017-01-31 15:14 ` Ferruh Yigit
2017-01-31 15:19 ` Wiles, Keith
2017-02-02 13:46 ` Wiles, Keith
2017-02-02 16:17 ` [dpdk-dev] [PATCH v2 1/7] " Pascal Mazon
2017-02-02 16:17 ` [dpdk-dev] [PATCH v2 2/7] net/tap: use correct channel for error logs Pascal Mazon
2017-02-02 16:18 ` [dpdk-dev] [PATCH v2 3/7] net/tap: don't set fd value overwritten just below Pascal Mazon
2017-02-02 16:18 ` [dpdk-dev] [PATCH v2 4/7] net/tap: keep kernel-assigned MAC address Pascal Mazon
2017-02-02 16:18 ` [dpdk-dev] [PATCH v2 5/7] net/tap: display tap name after parsing Pascal Mazon
2017-02-02 16:18 ` [dpdk-dev] [PATCH v2 6/7] net/tap: implement link up and down callbacks Pascal Mazon
2017-02-02 16:18 ` [dpdk-dev] [PATCH v2 7/7] net/tap: support promiscuous and allmulti setting Pascal Mazon
2017-02-02 16:23 ` [dpdk-dev] [PATCH v2 1/7] net/tap: use correct tap name Wiles, Keith
2017-02-02 16:24 ` Wiles, Keith
2017-02-02 21:55 ` 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=74B7FC24-638D-4A48-A02D-C80B5B4E7E2B@intel.com \
--to=keith.wiles@intel.com \
--cc=dev@dpdk.org \
--cc=ferruh.yigit@intel.com \
--cc=pascal.mazon@6wind.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).