From: Avi Kivity <avi@cloudius-systems.com>
To: "Richardson, Bruce" <bruce.richardson@intel.com>,
Vladislav Zolotarov <vladz@cloudius-systems.com>
Cc: "dev@dpdk.org" <dev@dpdk.org>
Subject: Re: [dpdk-dev] [PATCH v1] ixgbe_pmd: forbid tx_rs_thresh above 1 for all NICs but 82598
Date: Fri, 11 Sep 2015 20:44:54 +0300 [thread overview]
Message-ID: <55F31316.2090807@cloudius-systems.com> (raw)
In-Reply-To: <59AF69C657FD0841A61C55336867B5B0359263BC@IRSMSX103.ger.corp.intel.com>
On 09/11/2015 07:07 PM, Richardson, Bruce wrote:
>
>> -----Original Message-----
>> From: dev [mailto:dev-bounces@dpdk.org] On Behalf Of Vladislav Zolotarov
>> Sent: Friday, September 11, 2015 5:04 PM
>> To: Avi Kivity
>> Cc: dev@dpdk.org
>> Subject: Re: [dpdk-dev] [PATCH v1] ixgbe_pmd: forbid tx_rs_thresh above 1
>> for all NICs but 82598
>>
>> On Sep 11, 2015 6:43 PM, "Avi Kivity" <avi@cloudius-systems.com> wrote:
>>> On 09/11/2015 06:12 PM, Vladislav Zolotarov wrote:
>>>>
>>>> On Sep 11, 2015 5:55 PM, "Thomas Monjalon"
>>>> <thomas.monjalon@6wind.com>
>> wrote:
>>>>> 2015-09-11 17:47, Avi Kivity:
>>>>>> On 09/11/2015 05:25 PM, didier.pallard wrote:
>>>>>>> On 08/25/2015 08:52 PM, Vlad Zolotarov wrote:
>>>>>>>> Helin, the issue has been seen on x540 devices. Pls., see a
>> chapter
>>>>>>>> 7.2.1.1 of x540 devices spec:
>>>>>>>>
>>>>>>>> A packet (or multiple packets in transmit segmentation) can
>>>>>>>> span
>> any
>>>>>>>> number of
>>>>>>>> buffers (and their descriptors) up to a limit of 40 minus
>>>>>>>> WTHRESH minus 2 (see Section 7.2.3.3 for Tx Ring details and
>>>>>>>> section Section 7.2.3.5.1
>> for
>>>>>>>> WTHRESH
>>>>>>>> details). For best performance it is recommended to minimize
>>>>>>>> the number of buffers as possible.
>>>>>>>>
>>>>>>>> Could u, pls., clarify why do u think that the maximum number
>>>>>>>> of
>> data
>>>>>>>> buffers is limited by 8?
>>>>>>>>
>>>>>>>> thanks,
>>>>>>>> vlad
>>>>>>> Hi vlad,
>>>>>>>
>>>>>>> Documentation states that a packet (or multiple packets in
>>>>>>> transmit
>>>>>>> segmentation) can span any number of buffers (and their
>>>>>>> descriptors) up to a limit of 40 minus WTHRESH minus 2.
>>>>>>>
>>>>>>> Shouldn't there be a test in transmit function that drops
>>>>>>> properly
>> the
>>>>>>> mbufs with a too large number of segments, while incrementing a
>>>>>>> statistic; otherwise transmit
>> function
>>>>>>> may be locked by the faulty packet without notification.
>>>>>>>
>>>>>> What we proposed is that the pmd expose to dpdk, and dpdk expose
>>>>>> to
>> the
>>>>>> application, an mbuf check function. This way applications that
>>>>>> can generate complex packets can verify that the device will be
>>>>>> able to process them, and applications that only generate simple
>>>>>> mbufs can
>> avoid
>>>>>> the overhead by not calling the function.
>>>>> More than a check, it should be exposed as a capability of the port.
>>>>> Anyway, if the application sends too much segments, the driver must
>>>>> drop it to avoid hang, and maintain a dedicated statistic counter
>>>>> to
>> allow
>>>>> easy debugging.
>>>> I agree with Thomas - this should not be optional. Malformed packets
>> should be dropped. In the icgbe case it's a very simple test - it's a
>> single branch per packet so i doubt that it could impose any measurable
>> performance degradation.allows
>>>>
>>> A drop allows the application no chance to recover. The driver must
>> either provide the ability for the application to know that it cannot
>> accept the packet, or it must fix it up itself.
>>
>> An appropriate statistics counter would be a perfect tool to detect such
>> issues. Knowingly sending a packet that will cause a HW to hang is not
>> acceptable.
> I would agree. Drivers should provide a function to query the max number of
> segments they can accept and the driver should be able to discard any packets
> exceeding that number, and just track it via a stat.
>
There is no such max number of segments. The i40e card, as an extreme
example, allows 8 fragments per packet, but that is after TSO
segmentation. So if the header is in three fragments, that leaves 5
data fragments per packet. Another card (ixgbe) has a 38-fragment
pre-TSO limit. With such a variety of limitations, the only generic way
to expose them is via a function.
next prev parent reply other threads:[~2015-09-11 17:44 UTC|newest]
Thread overview: 48+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-08-13 18:06 Vlad Zolotarov
2015-08-13 20:28 ` Zhang, Helin
2015-08-14 5:37 ` Vlad Zolotarov
2015-08-19 0:42 ` Lu, Wenzhuo
2015-08-19 4:55 ` Vladislav Zolotarov
2015-08-19 7:43 ` Ananyev, Konstantin
2015-08-19 10:02 ` Vlad Zolotarov
2015-08-20 8:41 ` Ananyev, Konstantin
2015-08-20 8:56 ` Vlad Zolotarov
2015-08-20 9:05 ` Vlad Zolotarov
2015-08-20 9:06 ` Vlad Zolotarov
2015-08-25 17:33 ` Ananyev, Konstantin
2015-08-25 17:39 ` Avi Kivity
2015-08-19 17:29 ` Zhang, Helin
2015-08-25 18:13 ` Zhang, Helin
2015-08-25 18:33 ` Vladislav Zolotarov
2015-08-25 18:43 ` Zhang, Helin
2015-08-25 18:52 ` Vlad Zolotarov
2015-08-25 19:16 ` Zhang, Helin
2015-08-25 19:23 ` Avi Kivity
2015-08-25 19:30 ` Vladislav Zolotarov
2015-08-25 20:07 ` Vlad Zolotarov
2015-08-25 20:13 ` Zhang, Helin
2015-09-09 12:18 ` Thomas Monjalon
2015-09-09 13:19 ` Ananyev, Konstantin
2015-09-11 15:17 ` Vladislav Zolotarov
2015-09-11 14:25 ` didier.pallard
2015-09-11 14:47 ` Avi Kivity
2015-09-11 14:55 ` Thomas Monjalon
2015-09-11 15:12 ` Vladislav Zolotarov
2015-09-11 15:43 ` Avi Kivity
2015-09-11 16:04 ` Vladislav Zolotarov
2015-09-11 16:07 ` Richardson, Bruce
2015-09-11 16:14 ` Vladislav Zolotarov
2015-09-11 17:44 ` Avi Kivity [this message]
2015-09-11 16:08 ` Thomas Monjalon
2015-09-11 16:18 ` Vladislav Zolotarov
2015-09-11 17:17 ` Matthew Hall
2015-09-11 17:42 ` Ananyev, Konstantin
2015-09-11 17:58 ` Matthew Hall
2015-09-11 17:48 ` Avi Kivity
2015-09-13 11:47 ` Ananyev, Konstantin
2015-09-13 12:24 ` Vlad Zolotarov
2015-09-13 12:32 ` Avi Kivity
2015-09-13 15:54 ` Ananyev, Konstantin
2015-09-13 16:01 ` Avi Kivity
2015-09-11 16:00 ` Richardson, Bruce
2015-09-11 16:13 ` Vladislav Zolotarov
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=55F31316.2090807@cloudius-systems.com \
--to=avi@cloudius-systems.com \
--cc=bruce.richardson@intel.com \
--cc=dev@dpdk.org \
--cc=vladz@cloudius-systems.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).