DPDK patches and discussions
 help / color / mirror / Atom feed
From: Bruce Richardson <bruce.richardson@intel.com>
To: Ruifeng Wang <Ruifeng.Wang@arm.com>
Cc: Honnappa Nagarahalli <Honnappa.Nagarahalli@arm.com>,
	"thomas@monjalon.net" <thomas@monjalon.net>,
	David Marchand <david.marchand@redhat.com>, dev <dev@dpdk.org>,
	"jerinj@marvell.com" <jerinj@marvell.com>, nd <nd@arm.com>
Subject: Re: [dpdk-dev] Use WFE for spinlock and ring
Date: Fri, 7 May 2021 11:24:46 +0100	[thread overview]
Message-ID: <YJUVbrQ1a2V24N0j@bricha3-MOBL.ger.corp.intel.com> (raw)
In-Reply-To: <AM5PR0802MB24652885F87CD9EC56837EB79E579@AM5PR0802MB2465.eurprd08.prod.outlook.com>

On Fri, May 07, 2021 at 10:18:52AM +0000, Ruifeng Wang wrote:
> > -----Original Message-----
> > From: Bruce Richardson <bruce.richardson@intel.com>
> > Sent: Friday, April 30, 2021 10:19 PM
> > To: Honnappa Nagarahalli <Honnappa.Nagarahalli@arm.com>
> > Cc: thomas@monjalon.net; Ruifeng Wang <Ruifeng.Wang@arm.com>; David
> > Marchand <david.marchand@redhat.com>; dev <dev@dpdk.org>;
> > jerinj@marvell.com; nd <nd@arm.com>
> > Subject: Re: Use WFE for spinlock and ring
> > 
> > On Fri, Apr 30, 2021 at 01:41:22PM +0000, Honnappa Nagarahalli wrote:
> > > <snip>
> > >
> > > > > > > > > >
> > > > > > > > > > The rte_wait_until_equal_xxx APIs abstract the
> > > > > > > > > > functionality of 'polling for a memory location to
> > > > > > > > > > become equal to a given
> > > > value'[1].
> > > > > > > > > >
> > > > > > > > > > Use the API for the rte spinlock and ring implementations.
> > > > > > > > > > With the wait until equal APIs being stable, changes
> > > > > > > > > > will not impact
> > > > ABI.
> > > > > > > > >
> > > > > > > > > Afaics, there is no ARM target with WFE enabled and we
> > > > > > > > > lost ability to enable WFE support with removal of the
> > > > > > > > > make build
> > > > system.
> > > > > > > >
> > > > > > > > WFE can be enabled with direct meson file change.
> > > > > > > > WFE is not intended to be enabled by default. It can be
> > > > > > > > enabled based on benchmarking result on hardware.
> > > > > > > > >
> > > > > > > > > $ git grep RTE_ARM_USE_WFE
> > > > > > > > > config/arm/meson.build:        ['RTE_ARM_USE_WFE', false],
> > > > > > > > > lib/eal/arm/include/rte_pause_64.h:#ifdef RTE_ARM_USE_WFE
> > > > > > > > >
> > > > > > > > > How did you enable WFE to test this series?
> > > > > > > >
> > > > > > > > I modified meson file to test.
> > > > > > > > Tests were also done with WFE disabled to make sure no
> > > > > > > > degradation with
> > > > > > > generic implementation.
> > > > > > >
> > > > > > > I don't understand the usage.
> > > > > > > Which platform should use it?
> > > > > >
> > > > > > Platforms that implement WFE semantic (e.g. N1) can use.
> > > > > > The user can enable this feature for power efficiency purpose.
> > > > > > But there is something to note as described in commit message
> > > > > > 1be7855d77
> > > > when the API was introduced.
> > > > > >
> > > > > > > Should it be a compile-time option?
> > > > > >
> > > > > > Yes, it should be a compile-time option.
> > > > > > It can be configured via c_args meson option?
> > > > >
> > > > > +Cc Bruce for discussing how to enable such feature.
> > > > >
> > > > > The problem with c_args is that the application has no way to know.
> > > > >
> > > > Agree about c_args not being a great choice. Why does this need to
> > > > be a compile-time option? Can runtime support not be detected in
> > > > some manner?
> > > The problem is inconsistency in performance on different Arm platforms.
> > We had decided that each platform needs to enable it after some testing.
> > 
> > Then it sounds like it does indeed need to be a build option. Does it need to
> > be added to the meson_options.txt, or can it just be specified in cross-files
> > and optionally via c_args?
> 
> Add it to the meson_options.txt is good as the option will be clearly exposed.
> My concern is more options (e.g. RTE_ARCH_ARM64_MEMCPY) need to be added in.
> Will the options bloat meson_options.txt?

That bloat would indeed become a concern. We may need to look at more use
of auto-detection and cross files for such options.

  reply	other threads:[~2021-05-07 10:24 UTC|newest]

Thread overview: 60+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-04-24  7:07 [dpdk-dev] [PATCH v1 0/2] " Gavin Hu
2020-04-24  7:07 ` [dpdk-dev] [PATCH v1 1/2] spinlock: use wfe to reduce contention on aarch64 Gavin Hu
2020-04-24  7:07 ` [dpdk-dev] [PATCH v1 2/2] ring: use wfe to wait for ring tail update " Gavin Hu
2020-04-24 13:23   ` Ananyev, Konstantin
2020-04-26  8:41     ` Gavin Hu
2020-04-24 14:46   ` Honnappa Nagarahalli
2020-04-26  8:33     ` Gavin Hu
2020-04-26  8:39 ` [dpdk-dev] [PATCH v2 0/2] Use WFE for spinlock and ring Gavin Hu
2020-05-01  9:47   ` Ananyev, Konstantin
2020-05-03 14:54   ` David Marchand
2020-05-03 21:31     ` Honnappa Nagarahalli
2020-09-25 14:30       ` David Marchand
2021-03-25 14:58         ` David Marchand
2021-04-25  6:06           ` Ruifeng Wang
2021-04-25 15:56             ` Thomas Monjalon
2020-04-26  8:39 ` [dpdk-dev] [PATCH v2 1/2] spinlock: use wfe to reduce contention on aarch64 Gavin Hu
2020-04-28 12:22   ` Jerin Jacob
2020-04-26  8:39 ` [dpdk-dev] [PATCH v2 2/2] ring: use wfe to wait for ring tail update " Gavin Hu
2021-04-25  5:56 ` [dpdk-dev] Use WFE for spinlock and ring Ruifeng Wang
2021-04-25  5:56   ` [dpdk-dev] [PATCH v3 1/2] spinlock: use wfe to reduce contention on aarch64 Ruifeng Wang
2021-04-25  5:56   ` [dpdk-dev] [PATCH v3 2/2] ring: use wfe to wait for ring tail update " Ruifeng Wang
2021-04-26  5:38     ` Jerin Jacob
2021-04-28 17:17     ` Stephen Hemminger
2021-04-29 14:35       ` Ruifeng Wang
2021-04-29 15:05         ` Stephen Hemminger
2021-05-07  8:25           ` Ruifeng Wang
2021-04-28  7:42   ` [dpdk-dev] Use WFE for spinlock and ring David Marchand
2021-04-28  9:30     ` Ruifeng Wang
2021-04-28 11:13       ` Thomas Monjalon
2021-04-29 14:28         ` Ruifeng Wang
2021-04-29 15:20           ` Thomas Monjalon
2021-04-30  9:16             ` Bruce Richardson
2021-04-30 13:41               ` Honnappa Nagarahalli
2021-04-30 14:19                 ` Bruce Richardson
2021-05-07 10:18                   ` Ruifeng Wang
2021-05-07 10:24                     ` Bruce Richardson [this message]
2021-07-05  8:51                       ` David Marchand
2021-07-05 10:21                         ` Ruifeng Wang
2021-07-07 14:47   ` Stephen Hemminger
2021-07-08  9:41     ` Ruifeng Wang
2021-07-08 16:58       ` Honnappa Nagarahalli
2021-07-07  5:43 ` [dpdk-dev] [PATCH v4 0/3] " Ruifeng Wang
2021-07-07  5:43   ` [dpdk-dev] [PATCH v4 1/3] spinlock: use wfe to reduce contention on aarch64 Ruifeng Wang
2021-07-07  5:43   ` [dpdk-dev] [PATCH v4 2/3] ring: use wfe to wait for ring tail update " Ruifeng Wang
2021-07-07  5:43   ` [dpdk-dev] [PATCH v4 3/3] build: add option to enable wait until equal Ruifeng Wang
2021-07-07 12:15     ` David Marchand
2021-07-07  5:48 ` [dpdk-dev] [PATCH v4 0/3] Use WFE for spinlock and ring Ruifeng Wang
2021-07-07  5:48   ` [dpdk-dev] [PATCH v4 1/3] spinlock: use wfe to reduce contention on aarch64 Ruifeng Wang
2021-07-07  5:48   ` [dpdk-dev] [PATCH v4 2/3] ring: use wfe to wait for ring tail update " Ruifeng Wang
2021-07-07  5:48   ` [dpdk-dev] [PATCH v4 3/3] build: add option to enable wait until equal Ruifeng Wang
2021-07-07  6:32     ` Thomas Monjalon
2021-07-07  6:46       ` Ruifeng Wang
2021-07-07 12:27         ` Bruce Richardson
2021-07-07 12:36           ` Jerin Jacob
2021-07-08  6:25             ` Ruifeng Wang
2021-07-08  6:32               ` Jerin Jacob
2021-07-08  7:32                 ` Thomas Monjalon
2021-07-08  9:21                   ` Ruifeng Wang
2021-07-08 10:28                     ` Thomas Monjalon
2021-07-09 18:39   ` [dpdk-dev] [PATCH v4 0/3] Use WFE for spinlock and ring 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=YJUVbrQ1a2V24N0j@bricha3-MOBL.ger.corp.intel.com \
    --to=bruce.richardson@intel.com \
    --cc=Honnappa.Nagarahalli@arm.com \
    --cc=Ruifeng.Wang@arm.com \
    --cc=david.marchand@redhat.com \
    --cc=dev@dpdk.org \
    --cc=jerinj@marvell.com \
    --cc=nd@arm.com \
    --cc=thomas@monjalon.net \
    /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).