dev.dpdk.org archive mirror
 help / color / mirror / Atom feed
From: Bruce Richardson <bruce.richardson-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
To: Yerden Zhumabekov <e_zhumabekov-8EHiFRVJVgQ@public.gmane.org>
Cc: "dev-VfR2kkLFssw@public.gmane.org" <dev-VfR2kkLFssw@public.gmane.org>
Subject: Re: [PATCH 11/13] mbuf: move l2_len and l3_len to second cache line
Date: Thu, 4 Sep 2014 12:55:49 +0100	[thread overview]
Message-ID: <20140904115549.GA13208@sivswdev02.ir.intel.com> (raw)
In-Reply-To: <5408463C.8040805-8EHiFRVJVgQ@public.gmane.org>

On Thu, Sep 04, 2014 at 05:00:12PM +0600, Yerden Zhumabekov wrote:
> I get your point. I've also read throught the code of various PMDs and
> have found no indication of setting l2_len/l3_len fields as well.
> 
> As for testing, we'd be happy to test the patchset but we are now in
> process of building our testing facilities so we are not ready to
> provide enough workload for the hardware/software. I was also wondering
> if anyone has run some test and can provide some numbers on that matter.
> 
> Personally, I don't think frag/reassemly app is a perfect example for
> evaluating 2nd cache line performance penalty. The offsets to L3 and L4
> headers need to be calculated for all TCP/IP traffic and fragmented
> traffic is not representative in this case. Maybe it would be better to
> write an app which calculates these offsets for different set of mbufs
> and provides some stats. For example, l2fwd/l3fwd + additional l2_len
> and l3_len calculation.
> 
> And I'm also figuring out how to rewrite our app/libs (prefetch etc) to
> reflect the future changes in mbuf, hence my concerns :)
>
Just a final point on this. Note that the second cache line is always being
read by the TX leg of the code to free back mbufs to their mbuf pool post-
transmit. The overall fast-path RX+TX benchmarks show no performance
degradation due to that access.

For sample apps, you make a good point indeed about the existing app not being
very useful as they work on larger packets. I'll see what I can throw together
here to make a more realistic test.

/Bruce
 
> 
> 04.09.2014 16:27, Bruce Richardson ??????????:
> > Hi Yerden,
> >
> > I understand your concerns and it's good to have this discussion.
> >
> > There are a number of reasons why I've moved these particular fields
> > to the second cache line. Firstly, the main reason is that, obviously enough,
> > not all fields will fit in cache line 0, and we need to prioritize what does
> > get stored there. The guiding principle behind what fields get moved or not
> > that I've chosen to use for this patch set is to move fields that are not
> > used on the receive path (or the fastpath receive path, more specifically -
> > so that we can move fields only used by jumbo frames that span mbufs) to the
> > second cache line. From a search through the existing codebase, there are no
> > drivers which set the l2/l3 length fields on RX, this is only used in
> > reassembly libraries/apps and by the drivers on TX.
> >
> > The other reason for moving it to the second cache line is that it logically
> > belongs with all the other length fields that we need to add to enable
> > tunneling support. [To get an idea of the extra fields that I propose adding
> > to the mbuf, please see the RFC patchset I sent out previously as "[RFC 
> > PATCH 00/14] Extend the mbuf structure"]. While we probably can fit the 16-bits
> > needed for l2/l3 length on the mbuf line 0, there is not enough room for all
> > the lengths so we would end up splitting them with other fields in between.
> >
> > So, in terms of what do to about this particular issue. I would hope that for
> > applications that use these fields the impact should be small and/or possible
> > to work around e.g. maybe prefetch second cache line on RX in driver. If not,
> > then I'm happy to see about withdrawing this particular change and seeing if
> > we can keep l2/l3 lengths on cache line zero, with other length fields being
> > on cache line 1.
> >
> > Question: would you consider the ip fragmentation and reassembly example apps
> > in the Intel DPDK releases good examples to test to see the impacts of this
> > change, or is there some other test you would prefer that I look to do? 
> > Can you perhaps test out the patch sets for the mbuf that I've upstreamed so
> > far and let me know what regressions, if any, you see in your use-case
> > scenarios?
> >
> > Regards,
> > /Bruce
> >
> -- 
> Sincerely,
> 
> Yerden Zhumabekov
> STS, ACI
> Astana, KZ
> 

  parent reply	other threads:[~2014-09-04 11:55 UTC|newest]

Thread overview: 63+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-09-03 15:49 [PATCH 00/13] Mbuf Structure Rework, part 2 Bruce Richardson
     [not found] ` <1409759378-10113-1-git-send-email-bruce.richardson-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
2014-09-03 15:49   ` [PATCH 01/13] mbuf: replace data pointer by an offset Bruce Richardson
     [not found]     ` <1409759378-10113-2-git-send-email-bruce.richardson-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
2014-09-08  9:52       ` Olivier MATZ
     [not found]         ` <540D7C5F.8000406-pdR9zngts4EAvxtiuMwx3w@public.gmane.org>
2014-09-08  9:55           ` Olivier MATZ
2014-09-03 15:49   ` [PATCH 02/13] mbuf: reorder fields by time of use Bruce Richardson
     [not found]     ` <1409759378-10113-3-git-send-email-bruce.richardson-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
2014-09-08 10:17       ` Olivier MATZ
2014-09-03 15:49   ` [PATCH 03/13] mbuf: add packet_type field Bruce Richardson
     [not found]     ` <1409759378-10113-4-git-send-email-bruce.richardson-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
2014-09-08 10:17       ` Olivier MATZ
     [not found]         ` <540D8228.809-pdR9zngts4EAvxtiuMwx3w@public.gmane.org>
2014-09-08 10:33           ` Yerden Zhumabekov
     [not found]             ` <540D85E0.4030203-8EHiFRVJVgQ@public.gmane.org>
2014-09-08 11:17               ` Olivier MATZ
     [not found]                 ` <CAD16F236028A64DBBC0158B1636EA4510F3E4F8@SHSMSX104.ccr.corp.intel.com>
     [not found]                   ` <CAD16F236028A64DBBC0158B1636EA4510F3E4F8-0J0gbvR4kTg/UvCtAeCM4rfspsVTdybXVpNB7YpNyf8@public.gmane.org>
2014-09-09  3:57                     ` Liu, Jijiang
     [not found]                 ` <540D903E.1060206-pdR9zngts4EAvxtiuMwx3w@public.gmane.org>
2014-09-09  3:59                   ` Zhang, Helin
     [not found]                     ` <F35DEAC7BCE34641BA9FAC6BCA4A12E70A784E49-0J0gbvR4kTg/UvCtAeCM4rfspsVTdybXVpNB7YpNyf8@public.gmane.org>
2014-09-09  8:02                       ` Olivier MATZ
     [not found]                         ` <540EB428.9060706-pdR9zngts4EAvxtiuMwx3w@public.gmane.org>
2014-09-09  8:45                           ` Zhang, Helin
2014-09-09  9:47                           ` Richardson, Bruce
2014-09-09 15:05                   ` Jim Thompson
2014-09-03 15:49   ` [PATCH 04/13] mbuf: expand ol_flags field to 64-bits Bruce Richardson
     [not found]     ` <1409759378-10113-5-git-send-email-bruce.richardson-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
2014-09-08 10:25       ` Olivier MATZ
     [not found]         ` <540D8421.7070808-pdR9zngts4EAvxtiuMwx3w@public.gmane.org>
2014-09-09  9:00           ` Richardson, Bruce
2014-09-03 15:49   ` [PATCH 05/13] mbuf: introduce a flag to indicate a control mbuf Bruce Richardson
     [not found]     ` <1409759378-10113-6-git-send-email-bruce.richardson-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
2014-09-08 11:53       ` Olivier MATZ
2014-09-03 15:49   ` [PATCH 06/13] mbuf: minor changes for readability Bruce Richardson
     [not found]     ` <1409759378-10113-7-git-send-email-bruce.richardson-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
2014-09-08 12:03       ` Olivier MATZ
2014-09-03 15:49   ` [PATCH 07/13] mbuf: use macros only to access the mbuf metadata Bruce Richardson
     [not found]     ` <1409759378-10113-8-git-send-email-bruce.richardson-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
2014-09-08 12:05       ` Olivier MATZ
     [not found]         ` <540D9B95.3020504-pdR9zngts4EAvxtiuMwx3w@public.gmane.org>
2014-09-09  9:01           ` Richardson, Bruce
     [not found]             ` <59AF69C657FD0841A61C55336867B5B0343EFAA3-kPTMFJFq+rELt2AQoY/u9bfspsVTdybXVpNB7YpNyf8@public.gmane.org>
2014-09-12 16:56               ` Dumitrescu, Cristian
     [not found]                 ` <3EB4FA525960D640B5BDFFD6A3D891262E070D42-kPTMFJFq+rEMvF1YICWikbfspsVTdybXVpNB7YpNyf8@public.gmane.org>
2014-09-12 21:02                   ` Olivier MATZ
     [not found]                     ` <54135F63.2090401-pdR9zngts4EAvxtiuMwx3w@public.gmane.org>
2014-09-16 20:07                       ` Dumitrescu, Cristian
     [not found]                         ` <3EB4FA525960D640B5BDFFD6A3D891262E071FE6-kPTMFJFq+rEMvF1YICWikbfspsVTdybXVpNB7YpNyf8@public.gmane.org>
2014-09-16 22:06                           ` Ramia, Kannan Babu
     [not found]                             ` <682698A055A0F44AA47192B20141149711B1FFE6-yHIBzpp8AekFyVwBAnZdSLfspsVTdybXVpNB7YpNyf8@public.gmane.org>
2014-09-17 10:31                               ` Richardson, Bruce
     [not found]                                 ` <59AF69C657FD0841A61C55336867B5B0343F2BD2-kPTMFJFq+rELt2AQoY/u9bfspsVTdybXVpNB7YpNyf8@public.gmane.org>
2014-09-17 14:01                                   ` Thomas Monjalon
2014-09-10 15:09           ` Bruce Richardson
2014-09-10 15:31             ` Olivier MATZ
2014-09-03 15:49   ` [PATCH 08/13] mbuf: add named points inside the mbuf structure Bruce Richardson
     [not found]     ` <1409759378-10113-9-git-send-email-bruce.richardson-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
2014-09-08 12:08       ` Olivier MATZ
2014-09-03 15:49   ` [PATCH 09/13] ixgbe: rework vector pmd following mbuf changes Bruce Richardson
2014-09-03 15:49   ` [PATCH 10/13] mbuf: split mbuf across two cache lines Bruce Richardson
     [not found]     ` <1409759378-10113-11-git-send-email-bruce.richardson-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
2014-09-08 12:10       ` Olivier MATZ
2014-09-03 15:49   ` [PATCH 11/13] mbuf: move l2_len and l3_len to second cache line Bruce Richardson
     [not found]     ` <1409759378-10113-12-git-send-email-bruce.richardson-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
2014-09-04  5:08       ` Yerden Zhumabekov
     [not found]         ` <20140904102744.GA23231@sivswdev02.ir.intel.com>
     [not found]           ` <20140904102744.GA23231-IWE99D/oH1/+pXziaqXtF9h3ngVCH38I@public.gmane.org>
2014-09-04 11:00             ` Yerden Zhumabekov
     [not found]               ` <5408463C.8040805-8EHiFRVJVgQ@public.gmane.org>
2014-09-04 11:55                 ` Bruce Richardson [this message]
2014-09-03 15:49   ` [PATCH 12/13] ixgbe: Fix perf regression due to moved pool ptr Bruce Richardson
2014-09-03 15:49   ` [PATCH 13/13] ixgbe: Improve slow-path perf: vector scattered RX Bruce Richardson
2014-09-11 13:15   ` [PATCH v2 00/13] Mbuf Structure Rework, part 2 Bruce Richardson
     [not found]     ` <1410441347-22840-1-git-send-email-bruce.richardson-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
2014-09-11 13:15       ` [PATCH v2 01/13] mbuf: replace data pointer by an offset Bruce Richardson
2014-09-11 13:15       ` [PATCH v2 02/13] mbuf: reorder fields by time of use Bruce Richardson
     [not found]         ` <1410441347-22840-3-git-send-email-bruce.richardson-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
2014-09-15  7:11           ` Liu, Jijiang
     [not found]             ` <1ED644BD7E0A5F4091CF203DAFB8E4CC01D701BA-0J0gbvR4kThpB2pF5aRoyrfspsVTdybXVpNB7YpNyf8@public.gmane.org>
2014-09-15  8:19               ` Richardson, Bruce
2014-09-11 13:15       ` [PATCH v2 03/13] mbuf: expand ol_flags field to 64-bits Bruce Richardson
2014-09-11 13:15       ` [PATCH v2 04/13] mbuf: introduce a flag to indicate a control mbuf Bruce Richardson
2014-09-11 13:15       ` [PATCH v2 05/13] mbuf: minor changes for readability Bruce Richardson
2014-09-11 13:15       ` [PATCH v2 06/13] mbuf: use macros only to access the mbuf metadata Bruce Richardson
2014-09-11 13:15       ` [PATCH v2 07/13] mbuf: move metadata macros to rte_port library Bruce Richardson
2014-09-11 13:15       ` [PATCH v2 08/13] mbuf: add named points inside the mbuf structure Bruce Richardson
2014-09-11 13:15       ` [PATCH v2 09/13] ixgbe: rework vector pmd following mbuf changes Bruce Richardson
2014-09-11 13:15       ` [PATCH v2 10/13] mbuf: split mbuf across two cache lines Bruce Richardson
2014-09-11 13:15       ` [PATCH v2 11/13] mbuf: move l2_len and l3_len to second cache line Bruce Richardson
2014-09-11 13:15       ` [PATCH v2 12/13] ixgbe: Fix perf regression due to moved pool ptr Bruce Richardson
     [not found]         ` <1410441347-22840-13-git-send-email-bruce.richardson-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
2014-09-15 16:20           ` [PATCH v3 " Bruce Richardson
2014-09-11 13:15       ` [PATCH v2 13/13] ixgbe: Improve slow-path perf: vector scattered RX Bruce Richardson
2014-09-17 22:35       ` [PATCH v2 00/13] Mbuf Structure Rework, part 2 Thomas Monjalon

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=20140904115549.GA13208@sivswdev02.ir.intel.com \
    --to=bruce.richardson-ral2jqcrhueavxtiumwx3w@public.gmane.org \
    --cc=dev-VfR2kkLFssw@public.gmane.org \
    --cc=e_zhumabekov-8EHiFRVJVgQ@public.gmane.org \
    /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).