DPDK patches and discussions
 help / color / mirror / Atom feed
From: Thomas Monjalon <thomas@monjalon.net>
To: Bruce Richardson <bruce.richardson@intel.com>,
	David Hunt <david.hunt@intel.com>,
	David Marchand <david.marchand@redhat.com>
Cc: dev@dpdk.org
Subject: Re: [dpdk-dev] [PATCH v1 1/6] build: increase default of max lcores to 512
Date: Fri, 10 Sep 2021 10:24:59 +0200	[thread overview]
Message-ID: <5148108.61zlV0aQVf@thomas> (raw)
In-Reply-To: <CAJFAV8wTCpp907i3=-ZLjJJbcvE4_Rk0KPcMfL1JrhC1qL-T_Q@mail.gmail.com>

10/09/2021 10:06, David Marchand:
> On Fri, Sep 10, 2021 at 9:54 AM Bruce Richardson
> <bruce.richardson@intel.com> wrote:
> >
> > On Fri, Sep 10, 2021 at 08:51:04AM +0200, David Marchand wrote:
> > > On Thu, Sep 9, 2021 at 4:38 PM Bruce Richardson
> > > <bruce.richardson@intel.com> wrote:
> > > >
> > > > On Thu, Sep 09, 2021 at 02:45:06PM +0100, David Hunt wrote:
> > > > > Modern processors are coming with an ever increasing number of cores,
> > > > > and 128 does not seem like a sensible max limit any more, especially
> > > > > when you consider multi-socket systems with Hyper-Threading enabled.
> > > > >
> > > > > This patch increases max_lcores default from 128 to 512.
> > > > >
> > > > > Signed-off-by: David Hunt <david.hunt@intel.com>
> > >
> > > Why should we need this?
> > >
> > > --lcores makes it possible to pin 128 lcores to any physical core on
> > > your system.
> > > And for applications that have their own thread management, they can
> > > pin thread, then use rte_thread_register.
> > >
> > > Do you have applications that require more than 128 lcores?
> > >
> > The trouble is that using the --lcores syntax for mapping high core numbers
> > to low lcore ids is much more awkward to use. Every case of DPDK use I've
> > seen uses -c with a coremask, or -l with just giving a few core numbers on
> > it. This simple scheme won't work with core numbers greater than 128, and
> > there are already systems available with more than that number of cores.
> >
> > Apart from the memory footprint issues - which this patch is already making
> > a good start in addressing, why would we not increase the default
> > max_lcores to that seen on real systems?
> 
> The memory footprint is a major issue to me, and reserving all those
> lcores won't be needed in any system.
> We will also have to decide on a "640k ought to be enough" value to
> avoid ABI issue with the next processor that comes out and has more
> than 512 cores.
> 
> Could we wire the -c / -l options to --lcores behavior ?
> It breaks the 1:1 lcore/physical core assumption, but it solves your
> usability issue.

Why would we change existing options while we already have an option
(--lcores) which solves the issue above?
I think the only issue is to educate users.
Is there something to improve in the documentation?



  reply	other threads:[~2021-09-10  8:25 UTC|newest]

Thread overview: 46+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-09-09 13:45 [dpdk-dev] build: Increase the default value of RTE_MAX_LCORE David Hunt
2021-09-09 13:45 ` [dpdk-dev] [PATCH v1 1/6] build: increase default of max lcores to 512 David Hunt
2021-09-09 14:37   ` Bruce Richardson
2021-09-10  6:51     ` David Marchand
2021-09-10  7:54       ` Bruce Richardson
2021-09-10  8:06         ` David Marchand
2021-09-10  8:24           ` Thomas Monjalon [this message]
2021-09-14  9:34             ` David Hunt
2021-09-14 10:00               ` David Marchand
2021-09-14 11:07                 ` David Hunt
2021-09-14 11:29                   ` David Marchand
2021-09-15 12:13                     ` David Hunt
2021-11-17 15:55                   ` Morten Brørup
2021-11-17 19:01                     ` David Hunt
2021-09-15 12:11   ` [dpdk-dev] [PATCH v2] eal: add additional info if lcore exceeds max cores David Hunt
2021-09-16 12:34     ` David Marchand
2021-09-20  9:30       ` David Hunt
2021-09-21 11:50     ` [dpdk-dev] [PATCH v3 1/2] eal: add additional info if core list too long David Hunt
2021-09-21 11:50       ` [dpdk-dev] [PATCH v3 2/2] eal: add additional info if core mask " David Hunt
2021-09-21 12:00         ` Bruce Richardson
2021-09-21 11:57       ` [dpdk-dev] [PATCH v3 1/2] eal: add additional info if core list " Bruce Richardson
2021-09-21 12:04         ` David Hunt
2021-09-21 13:16           ` David Hunt
2021-09-21 13:20             ` Bruce Richardson
2021-09-21 13:51       ` David Marchand
2021-09-21 15:10         ` David Hunt
2021-09-22 12:29       ` [dpdk-dev] [PATCH v4 " David Hunt
2021-09-22 12:29         ` [dpdk-dev] [PATCH v4 2/2] eal: add additional info if core mask " David Hunt
2021-09-23  8:12           ` David Marchand
2021-09-23 10:21             ` David Hunt
2021-09-23  8:11         ` [dpdk-dev] [PATCH v4 1/2] eal: add additional info if core list " David Marchand
2021-09-23  9:47           ` David Hunt
2021-09-23 11:02         ` [dpdk-dev] [PATCH v5 " David Hunt
2021-09-23 11:02           ` [dpdk-dev] [PATCH v5 2/2] eal: add additional info if core mask " David Hunt
2021-11-02 17:45             ` David Marchand
2021-11-03 10:27               ` David Hunt
2021-11-03 10:29                 ` David Marchand
2021-11-03 13:30                 ` David Hunt
2021-11-03 14:32           ` [dpdk-dev] [PATCH v6 1/2] eal: add additional info if core list " David Hunt
2021-11-03 14:32             ` [dpdk-dev] [PATCH v6 2/2] eal: add additional info if core mask " David Hunt
2021-11-05 10:50               ` David Marchand
2021-09-09 13:45 ` [dpdk-dev] [PATCH v1 2/6] lib/power: reduce memory footprint of acpi lib David Hunt
2021-09-09 13:45 ` [dpdk-dev] [PATCH v1 3/6] lib/power: reduce memory footprint of pstate lib David Hunt
2021-09-09 13:45 ` [dpdk-dev] [PATCH v1 4/6] lib/power: reduce memory footprint of cppc lib David Hunt
2021-09-09 13:45 ` [dpdk-dev] [PATCH v1 5/6] lib/power: reduce memory footprint of channels David Hunt
2021-09-09 13:45 ` [dpdk-dev] [PATCH v1 6/6] lib/power: switch empty poll to max cores config David Hunt

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=5148108.61zlV0aQVf@thomas \
    --to=thomas@monjalon.net \
    --cc=bruce.richardson@intel.com \
    --cc=david.hunt@intel.com \
    --cc=david.marchand@redhat.com \
    --cc=dev@dpdk.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).