DPDK patches and discussions
 help / color / mirror / Atom feed
* ETH_RSS_IP only does not seem to balance traffic
@ 2022-01-12 15:52 Yasuhiro Ohara
  2022-01-13 15:05 ` Bruce Richardson
  0 siblings, 1 reply; 5+ messages in thread
From: Yasuhiro Ohara @ 2022-01-12 15:52 UTC (permalink / raw)
  To: dev


Hi,

My system developper friend recently ran into a problem where
l3fwd does not appear to receive balanced traffic on Intel XL710,
but it is resolved when the attached patch is applied.

-            .rss_hf = ETH_RSS_IP,
+            .rss_hf = ETH_RSS_IP | ETH_RSS_TCP | ETH_RSS_UDP,

IIRC I ran into a similar problem 3 or 4 years back,
but didn't report then because I believed I was doing something silly.
But since my friend is an experienced engineer, I feel like
it may be better (for the community) to ask this in the list.

We are using dpdk-stable-18.11.6 and igb_uio.

What are we doing wrong?

If it is not a FAQ, I can test it again with more recent stable,
and will report the details.

Thanks.

Best,
Yasu


^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: ETH_RSS_IP only does not seem to balance traffic
  2022-01-12 15:52 ETH_RSS_IP only does not seem to balance traffic Yasuhiro Ohara
@ 2022-01-13 15:05 ` Bruce Richardson
  2022-01-14  1:41   ` Yasuhiro Ohara
  2022-01-14  1:44   ` Xing, Beilei
  0 siblings, 2 replies; 5+ messages in thread
From: Bruce Richardson @ 2022-01-13 15:05 UTC (permalink / raw)
  To: Yasuhiro Ohara; +Cc: dev, Beilei Xing

On Thu, Jan 13, 2022 at 12:52:04AM +0900, Yasuhiro Ohara wrote:
> 
> Hi,
> 
> My system developper friend recently ran into a problem where
> l3fwd does not appear to receive balanced traffic on Intel XL710,
> but it is resolved when the attached patch is applied.
> 
> -            .rss_hf = ETH_RSS_IP,
> +            .rss_hf = ETH_RSS_IP | ETH_RSS_TCP | ETH_RSS_UDP,
> 
> IIRC I ran into a similar problem 3 or 4 years back,
> but didn't report then because I believed I was doing something silly.
> But since my friend is an experienced engineer, I feel like
> it may be better (for the community) to ask this in the list.
> 
> We are using dpdk-stable-18.11.6 and igb_uio.
> 
> What are we doing wrong?
> 
> If it is not a FAQ, I can test it again with more recent stable,
> and will report the details.
> 
For XL710 NICs, I believe that ETH_RSS_IP load balances only IP frames that
do not have TCP or UDP headers also. Adding i40e driver maintainer on CC
to comment further.

/Bruce

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: ETH_RSS_IP only does not seem to balance traffic
  2022-01-13 15:05 ` Bruce Richardson
@ 2022-01-14  1:41   ` Yasuhiro Ohara
  2022-01-14  1:44   ` Xing, Beilei
  1 sibling, 0 replies; 5+ messages in thread
From: Yasuhiro Ohara @ 2022-01-14  1:41 UTC (permalink / raw)
  To: bruce.richardson; +Cc: dev, beilei.xing


That makes sense. Thank you.
It would be great to have further comments from the maintener.

For the RTE framework integrity, would it be better for us
to have a consistent meaning for the ETH_RSS_XXX flags?
Do the other drivers act differently?

Best regards,
Yasu

From: Bruce Richardson <bruce.richardson@intel.com>
Subject: Re: ETH_RSS_IP only does not seem to balance traffic
Date: Thu, 13 Jan 2022 15:05:53 +0000
Message-ID: <YeA/0SQjcKJpCBu9@bricha3-MOBL.ger.corp.intel.com>

> On Thu, Jan 13, 2022 at 12:52:04AM +0900, Yasuhiro Ohara wrote:
>> 
>> Hi,
>> 
>> My system developper friend recently ran into a problem where
>> l3fwd does not appear to receive balanced traffic on Intel XL710,
>> but it is resolved when the attached patch is applied.
>> 
>> -            .rss_hf = ETH_RSS_IP,
>> +            .rss_hf = ETH_RSS_IP | ETH_RSS_TCP | ETH_RSS_UDP,
>> 
>> IIRC I ran into a similar problem 3 or 4 years back,
>> but didn't report then because I believed I was doing something silly.
>> But since my friend is an experienced engineer, I feel like
>> it may be better (for the community) to ask this in the list.
>> 
>> We are using dpdk-stable-18.11.6 and igb_uio.
>> 
>> What are we doing wrong?
>> 
>> If it is not a FAQ, I can test it again with more recent stable,
>> and will report the details.
>> 
> For XL710 NICs, I believe that ETH_RSS_IP load balances only IP frames that
> do not have TCP or UDP headers also. Adding i40e driver maintainer on CC
> to comment further.
> 
> /Bruce
> 

^ permalink raw reply	[flat|nested] 5+ messages in thread

* RE: ETH_RSS_IP only does not seem to balance traffic
  2022-01-13 15:05 ` Bruce Richardson
  2022-01-14  1:41   ` Yasuhiro Ohara
@ 2022-01-14  1:44   ` Xing, Beilei
  2022-01-14  1:55     ` Yasuhiro Ohara
  1 sibling, 1 reply; 5+ messages in thread
From: Xing, Beilei @ 2022-01-14  1:44 UTC (permalink / raw)
  To: Richardson, Bruce, Yasuhiro Ohara; +Cc: dev



> -----Original Message-----
> From: Richardson, Bruce <bruce.richardson@intel.com>
> Sent: Thursday, January 13, 2022 11:06 PM
> To: Yasuhiro Ohara <yasu@nttv6.jp>
> Cc: dev@dpdk.org; Xing, Beilei <beilei.xing@intel.com>
> Subject: Re: ETH_RSS_IP only does not seem to balance traffic
> 
> On Thu, Jan 13, 2022 at 12:52:04AM +0900, Yasuhiro Ohara wrote:
> >
> > Hi,
> >
> > My system developper friend recently ran into a problem where l3fwd
> > does not appear to receive balanced traffic on Intel XL710, but it is
> > resolved when the attached patch is applied.
> >
> > -            .rss_hf = ETH_RSS_IP,
> > +            .rss_hf = ETH_RSS_IP | ETH_RSS_TCP | ETH_RSS_UDP,
> >
> > IIRC I ran into a similar problem 3 or 4 years back, but didn't report
> > then because I believed I was doing something silly.
> > But since my friend is an experienced engineer, I feel like it may be
> > better (for the community) to ask this in the list.
> >
> > We are using dpdk-stable-18.11.6 and igb_uio.
> >
> > What are we doing wrong?
> >
> > If it is not a FAQ, I can test it again with more recent stable, and
> > will report the details.
> >
> For XL710 NICs, I believe that ETH_RSS_IP load balances only IP frames that do
> not have TCP or UDP headers also. Adding i40e driver maintainer on CC to
> comment further.

Yes, Bruce is right. For XL710, ETH_RSS_IP doesn't cover TCP and UDP packets.

> 
> /Bruce

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: ETH_RSS_IP only does not seem to balance traffic
  2022-01-14  1:44   ` Xing, Beilei
@ 2022-01-14  1:55     ` Yasuhiro Ohara
  0 siblings, 0 replies; 5+ messages in thread
From: Yasuhiro Ohara @ 2022-01-14  1:55 UTC (permalink / raw)
  To: beilei.xing; +Cc: bruce.richardson, dev


Thank you for the confirmation!

Best regards,
Yasu

From: "Xing, Beilei" <beilei.xing@intel.com>
Subject: RE: ETH_RSS_IP only does not seem to balance traffic
Date: Fri, 14 Jan 2022 01:44:45 +0000
Message-ID: <DM6PR11MB3098E698BC0A31AE3ABA9099F7549@DM6PR11MB3098.namprd11.prod.outlook.com>

> 
> 
>> -----Original Message-----
>> From: Richardson, Bruce <bruce.richardson@intel.com>
>> Sent: Thursday, January 13, 2022 11:06 PM
>> To: Yasuhiro Ohara <yasu@nttv6.jp>
>> Cc: dev@dpdk.org; Xing, Beilei <beilei.xing@intel.com>
>> Subject: Re: ETH_RSS_IP only does not seem to balance traffic
>> 
>> On Thu, Jan 13, 2022 at 12:52:04AM +0900, Yasuhiro Ohara wrote:
>> >
>> > Hi,
>> >
>> > My system developper friend recently ran into a problem where l3fwd
>> > does not appear to receive balanced traffic on Intel XL710, but it is
>> > resolved when the attached patch is applied.
>> >
>> > -            .rss_hf = ETH_RSS_IP,
>> > +            .rss_hf = ETH_RSS_IP | ETH_RSS_TCP | ETH_RSS_UDP,
>> >
>> > IIRC I ran into a similar problem 3 or 4 years back, but didn't report
>> > then because I believed I was doing something silly.
>> > But since my friend is an experienced engineer, I feel like it may be
>> > better (for the community) to ask this in the list.
>> >
>> > We are using dpdk-stable-18.11.6 and igb_uio.
>> >
>> > What are we doing wrong?
>> >
>> > If it is not a FAQ, I can test it again with more recent stable, and
>> > will report the details.
>> >
>> For XL710 NICs, I believe that ETH_RSS_IP load balances only IP frames that do
>> not have TCP or UDP headers also. Adding i40e driver maintainer on CC to
>> comment further.
> 
> Yes, Bruce is right. For XL710, ETH_RSS_IP doesn't cover TCP and UDP packets.
> 
>> 
>> /Bruce
> 

^ permalink raw reply	[flat|nested] 5+ messages in thread

end of thread, other threads:[~2022-01-14  1:55 UTC | newest]

Thread overview: 5+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2022-01-12 15:52 ETH_RSS_IP only does not seem to balance traffic Yasuhiro Ohara
2022-01-13 15:05 ` Bruce Richardson
2022-01-14  1:41   ` Yasuhiro Ohara
2022-01-14  1:44   ` Xing, Beilei
2022-01-14  1:55     ` Yasuhiro Ohara

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).