DPDK patches and discussions
 help / color / mirror / Atom feed
From: Bruce Richardson <bruce.richardson@intel.com>
To: "Walukiewicz, Miroslaw" <Miroslaw.Walukiewicz@intel.com>
Cc: "dev@dpdk.org" <dev@dpdk.org>
Subject: Re: [dpdk-dev] [RFC PATCH 0/7] support multi-phtread per lcore
Date: Tue, 23 Dec 2014 09:23:08 +0000	[thread overview]
Message-ID: <20141223092307.GA10244@bricha3-MOBL3> (raw)
In-Reply-To: <7C4248CAE043B144B1CD242D275626532FE29694@IRSMSX104.ger.corp.intel.com>

On Tue, Dec 23, 2014 at 09:19:54AM +0000, Walukiewicz, Miroslaw wrote:
> 
> 
> > -----Original Message-----
> > From: dev [mailto:dev-bounces@dpdk.org] On Behalf Of Stephen
> > Hemminger
> > Sent: Monday, December 22, 2014 7:29 PM
> > To: Richardson, Bruce
> > Cc: dev@dpdk.org
> > Subject: Re: [dpdk-dev] [RFC PATCH 0/7] support multi-phtread per lcore
> > 
> > On Mon, 22 Dec 2014 09:46:03 +0000
> > Bruce Richardson <bruce.richardson@intel.com> wrote:
> > 
> > > On Mon, Dec 22, 2014 at 01:51:27AM +0000, Liang, Cunming wrote:
> > > > ...
> > > > > I'm conflicted on this one. However, I think far more applications would
> > be
> > > > > broken
> > > > > to start having to use thread_id in place of an lcore_id than would be
> > broken
> > > > > by having the lcore_id no longer actually correspond to a core.
> > > > > I'm actually struggling to come up with a large number of scenarios
> > where it's
> > > > > important to an app to determine the cpu it's running on, compared to
> > the large
> > > > > number of cases where you need to have a data-structure per thread.
> > In DPDK
> > > > > libs
> > > > > alone, you see this assumption that lcore_id == thread_id a large
> > number of
> > > > > times.
> > > > >
> > > > > Despite the slight logical inconsistency, I think it's better to avoid
> > introducing
> > > > > a thread-id and continue having lcore_id representing a unique thread.
> > > > >
> > > > > /Bruce
> > > >
> > > > Ok, I understand it.
> > > > I list the implicit meaning if using lcore_id representing the unique thread.
> > > > 1). When lcore_id less than RTE_MAX_LCORE, it still represents the logical
> > core id.
> > > > 2). When lcore_id large equal than RTE_MAX_LCORE, it represents an
> > unique id for thread.
> > > > 3). Most of APIs(except rte_lcore_id()) in rte_lcore.h suggest to be used
> > only in CASE 1)
> > > > 4). rte_lcore_id() can be used in CASE 2), but the return value no matter
> > represent a logical core id.
> > > >
> > > > If most of us feel it's acceptable, I'll prepare for the RFC v2 base on this
> > conclusion.
> > > >
> > > > /Cunming
> > >
> > > Sorry, I don't like that suggestion either, as having lcore_id values greater
> > > than RTE_MAX_LCORE is terrible, as how will people know how to
> > dimension arrays
> > > to be indexes by lcore id? Given the choice, if we are not going to just use
> > > lcore_id as a generic thread id, which is always between 0 and
> > RTE_MAX_LCORE
> > > we can look to define a new thread_id variable to hold that. However, it
> > should
> > > have a bounded range.
> > > From an ease-of-porting perspective, I still think that the simplest option is
> > to
> > > use the existing lcore_id and accept the fact that it's now a thread id rather
> > > than an actual physical lcore. Question is, is would that cause us lots of
> > issues
> > > in the future?
> > >
> > > /Bruce
> > 
> > The current rte_lcore_id() has different meaning the thread. Your proposal
> > will
> > break code that uses lcore_id to do per-cpu statistics and the lcore_config
> > code in the samples.
> > q
> It depends on application context and how application treats rte_lcore_id() core. When number of the threads will not exceed the number of cores (let's say old-fashioned DPDK application) all stuff like per-cpu statistics will work correctly. 
> 
> When we treat threads on cores as ordinary threads as we introducing the special function rte_pthread_create() - the meaning of rte_lcore_id() changes to indicate 
>  thread number what is correct under new assumptions and new application model.
> 
> I do not  want to limit DPDK design  to only per-cpu application. There is much more application models that could be supported using DPDK. 
> Current per-cpu approach is only a subset of the possible applications.
> 
> Maybe we should indicate something like CONFIG_RTE_PTHREAD_ENABLE to change a meaning of rte_lcore_id() and introducing rte_pthread_create() family. 
> 
> Mirek
> 
>From the discussion it does look to me like we do need a separate thread id
value, separate from core id. Unfortunately that means that many(most?) places in libs
and examples where we use lcore_id right now, we probably need to use the new
thread id. :-(

/Bruce

  reply	other threads:[~2014-12-23  9:23 UTC|newest]

Thread overview: 37+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-12-11  2:04 Cunming Liang
2014-12-11  2:04 ` [dpdk-dev] [RFC PATCH 1/7] eal: add linear thread id as pthread-local variable Cunming Liang
2014-12-16  7:00   ` Qiu, Michael
2014-12-22 19:02   ` Ananyev, Konstantin
2014-12-23  9:56     ` Liang, Cunming
2014-12-11  2:04 ` [dpdk-dev] [RFC PATCH 2/7] mempool: use linear-tid as mempool cache index Cunming Liang
2014-12-11  2:04 ` [dpdk-dev] [RFC PATCH 3/7] ring: use linear-tid as ring debug stats index Cunming Liang
2014-12-11  2:04 ` [dpdk-dev] [RFC PATCH 4/7] eal: add simple API for multi-pthread Cunming Liang
2014-12-11  2:04 ` [dpdk-dev] [RFC PATCH 5/7] testpmd: support multi-pthread mode Cunming Liang
2014-12-11  2:04 ` [dpdk-dev] [RFC PATCH 6/7] sample: add new sample for multi-pthread Cunming Liang
2014-12-11  2:04 ` [dpdk-dev] [RFC PATCH 7/7] eal: macro for cpuset w/ or w/o CPU_ALLOC Cunming Liang
2014-12-11  2:54 ` [dpdk-dev] [RFC PATCH 0/7] support multi-phtread per lcore Jayakumar, Muthurajan
2014-12-11  9:56 ` Walukiewicz, Miroslaw
2014-12-12  5:44   ` Liang, Cunming
2014-12-15 11:10     ` Walukiewicz, Miroslaw
2014-12-15 11:53       ` Liang, Cunming
2014-12-18 12:20         ` Walukiewicz, Miroslaw
2014-12-18 14:32           ` Bruce Richardson
2014-12-18 15:11             ` Olivier MATZ
2014-12-18 16:04               ` Bruce Richardson
2014-12-18 16:15           ` Stephen Hemminger
2014-12-19  1:28           ` Liang, Cunming
2014-12-19 10:03             ` Bruce Richardson
2014-12-22  1:51               ` Liang, Cunming
2014-12-22  9:46                 ` Bruce Richardson
2014-12-22 10:01                   ` Walukiewicz, Miroslaw
2014-12-23  9:45                     ` Liang, Cunming
2014-12-22 18:28                   ` Stephen Hemminger
2014-12-23  9:19                     ` Walukiewicz, Miroslaw
2014-12-23  9:23                       ` Bruce Richardson [this message]
2014-12-23  9:51                     ` Liang, Cunming
2015-01-08 17:05                       ` Ananyev, Konstantin
2015-01-08 17:23                         ` Richardson, Bruce
2015-01-09  9:51                           ` Liang, Cunming
2015-01-09  9:40                         ` Liang, Cunming
2015-01-09 11:52                           ` Ananyev, Konstantin
2015-01-09  9:45                         ` Liang, Cunming

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=20141223092307.GA10244@bricha3-MOBL3 \
    --to=bruce.richardson@intel.com \
    --cc=Miroslaw.Walukiewicz@intel.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).