DPDK patches and discussions
 help / color / mirror / Atom feed
From: Bruce Richardson <bruce.richardson@intel.com>
To: Olivier MATZ <olivier.matz@6wind.com>
Cc: "dev@dpdk.org" <dev@dpdk.org>,
	"Damjan Marion \(damarion\)" <damarion@cisco.com>
Subject: Re: [dpdk-dev] rte_mbuf.next in 2nd cacheline
Date: Mon, 15 Jun 2015 15:46:28 +0100	[thread overview]
Message-ID: <20150615144628.GB580@bricha3-MOBL3> (raw)
In-Reply-To: <557EE1A0.609@6wind.com>

On Mon, Jun 15, 2015 at 04:30:56PM +0200, Olivier MATZ wrote:
> 
> 
> On 06/15/2015 04:12 PM, Bruce Richardson wrote:
> > On Mon, Jun 15, 2015 at 04:05:05PM +0200, Olivier MATZ wrote:
> >> Hi,
> >>
> >> On 06/15/2015 03:54 PM, Ananyev, Konstantin wrote:
> >>>
> >>>
> >>>> -----Original Message-----
> >>>> From: dev [mailto:dev-bounces@dpdk.org] On Behalf Of Bruce Richardson
> >>>> Sent: Monday, June 15, 2015 2:44 PM
> >>>> To: Olivier MATZ
> >>>> Cc: dev@dpdk.org; Damjan Marion (damarion)
> >>>> Subject: Re: [dpdk-dev] rte_mbuf.next in 2nd cacheline
> >>>>
> >>>> On Mon, Jun 15, 2015 at 03:20:22PM +0200, Olivier MATZ wrote:
> >>>>> Hi Damjan,
> >>>>>
> >>>>> On 06/10/2015 11:47 PM, Damjan Marion (damarion) wrote:
> >>>>>>
> >>>>>> Hi,
> >>>>>>
> >>>>>> We noticed 7% performance improvement by simply moving rte_mbuf.next field to the 1st cache line.
> >>>>>>
> >>>>>> Currently, it falls under /* second cache line - fields only used in slow path or on TX */
> >>>>>> but it is actually used at several places in rx fast path. (e.g.: i40e_rx_alloc_bufs() is setting that field to NULL).
> >>>>>>
> >>>>>> Is there anything we can do here (stop using next field, or move it to 1st cache line)?
> >>>>>
> >>>>> Agree, this is also something I noticed, see:
> >>>>> http://dpdk.org/ml/archives/dev/2015-February/014400.html
> >>>>>
> >>>>> I did not have the time to do performance testing, but it's something
> >>>>> I'd like to do as soon as I can. I don't see any obvious reason not to
> >>>>> do it.
> >>>>>
> >>>>> It seems we currently just have enough room to do it (8 bytes are
> >>>>> remaining in the first cache line when compiled in 64 bits).
> >>>>
> >>>> This, to me, is the obvious reason not to do it! It prevents us from taking in
> >>>> any other offload fields in the RX fast-path into the mbuf.
> >>>>
> >>>> That being said, I can see why we might want to look to move it - but from the
> >>>> work done in the ixgbe driver, I'd be hopeful we can get as much performance with
> >>>> it on the second cache line for most cases, through judicious use of prefetching,
> >>>> or otherwise.
> >>>>
> >>>> It took a lot of work and investigation to get free space in the mbuf - especially
> >>>> in cache line 0, and I'd like to avoid just filling the cache line up again as
> >>>> long as we possibly can!
> >>>
> >>> Yep, agree with Bruce here.
> >>> Plus, with packet_type going to be 4B and vlan_tci_outer,
> >>> we just don't have 8 free bytes at the first cache line any more.
> >>
> >> I don't understand why m->next would not be a better candidate than
> >> rx offload fields to be in the first cache line. For instance, m->next
> >> is mandatory and must be initialized when allocating a mbuf (to be
> >> compared with m->seqn for instance, which is also in the first cache
> >> line). So if we want to do some room in the first cache line, I
> >> think we can.
> >>
> >> To me, the only reason for not doing it now is because we don't
> >> have a full performance test report (several use-cases, drivers, ...)
> >> that shows it's better.
> >>
> > Because the "next" field is not mandatory to be set on initialization. It can
> > instead be set only when needed, and cleared on free if it is used.
> > 
> > The next pointers always start out as NULL when the mbuf pool is created. The
> > only time it is set to non-NULL is when we have chained mbufs. If we never have
> > any chained mbufs, we never need to touch the next field, or even read it - since
> > we have the num-segments count in the first cache line. If we do have a multi-segment
> > mbuf, it's likely to be a big packet, so we have more processing time available
> > and we can then take the hit of setting the next pointer. Whenever we go to
> > free that mbuf for that packet, the code to do the freeing obviously needs to
> > read the next pointer so as to free all the buffers in the chain, and so it can
> > also reset the next pointer to NULL when doing so.
> > 
> > In this way, we can ensure that the next pointer on cache line 1 is not a problem
> > in our fast path.
> 
> This is a good idea, but looking at the drivers, it seems that today
> they all set m->next to NULL in the rx function. What you are suggesting
> is to remove all of them, and document somewhere that all mbufs in a
> pool are supposed to have their m->next set to NULL, correct?

Yes. However, this restriction should not be visible to the applications, as
it's generally necessary to call pktmbuf_free (or a function based off it) to
free an mbuf - whether in user code or a library.

> 
> I think what you are describing could also apply to reference counter
> (set to 1 by default), right?

Yes, except that on free, the refcnt is actually set to zero, as we need to
do so because of the atomic nature of a refcnt e.g. it's 2 and then two
threads decrement it simultaneously. In a case like this, it doesn't matter
whether its set to 1 again on free or on allocation, as it's on cacheline zero.

/Bruce

> 
> 
> Olivier
> 
> 
> > 
> > /Bruce
> > 

  reply	other threads:[~2015-06-15 14:46 UTC|newest]

Thread overview: 26+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-06-10 21:47 Damjan Marion (damarion)
2015-06-15 13:20 ` Olivier MATZ
2015-06-15 13:44   ` Bruce Richardson
2015-06-15 13:54     ` Ananyev, Konstantin
2015-06-15 14:05       ` Olivier MATZ
2015-06-15 14:12         ` Bruce Richardson
2015-06-15 14:30           ` Olivier MATZ
2015-06-15 14:46             ` Bruce Richardson [this message]
2015-06-15 14:52             ` Ananyev, Konstantin
2015-06-15 15:19               ` Olivier MATZ
2015-06-15 15:23                 ` Bruce Richardson
2015-06-15 15:28                   ` Ananyev, Konstantin
2015-06-15 15:39                     ` Bruce Richardson
2015-06-15 15:59                       ` Ananyev, Konstantin
2015-06-15 16:02                         ` Bruce Richardson
2015-06-15 16:10                           ` Ananyev, Konstantin
2015-06-15 16:23                             ` Bruce Richardson
2015-06-15 18:34                               ` Ananyev, Konstantin
2015-06-15 20:47                                 ` Stephen Hemminger
2015-06-16  8:20                                   ` Bruce Richardson
2015-06-17 13:55           ` Damjan Marion (damarion)
2015-06-17 14:04             ` Thomas Monjalon
2015-06-17 14:06             ` Bruce Richardson
2015-06-17 14:23               ` Damjan Marion (damarion)
2015-06-17 16:32                 ` Thomas Monjalon
     [not found]               ` <0DE313B5-C9F0-4879-9D92-838ED088202C@cisco.com>
     [not found]                 ` <27EA8870B328F74E88180827A0F816396BD43720@xmb-aln-x10.cisco.com>
     [not found]                   ` <59AF69C657FD0841A61C55336867B5B0345592CD@IRSMSX103.ger.corp.intel.com>
     [not found]                     ` <1FD9B82B8BF2CF418D9A1000154491D97450B186@ORSMSX102.amr.corp.intel.com>
     [not found]                       ` <27EA8870B328F74E88180827A0F816396BD43891@xmb-aln-x10.cisco.com>
     [not found]                         ` <2601191342CEEE43887BDE71AB97725836A1237C@irsmsx105.ger.corp.intel.com>
2015-06-17 18:50                           ` Ananyev, Konstantin

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=20150615144628.GB580@bricha3-MOBL3 \
    --to=bruce.richardson@intel.com \
    --cc=damarion@cisco.com \
    --cc=dev@dpdk.org \
    --cc=olivier.matz@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).