From: Thomas Monjalon <thomas@monjalon.net>
To: "pradeep@us.ibm.com" <pradeep@us.ibm.com>,
Chao Zhu <chaozhu@linux.vnet.ibm.com>
Cc: dev@dpdk.org, Dekel Peled <dekelp@mellanox.com>,
Shahaf Shuler <shahafs@mellanox.com>,
"bruce.richardson@intel.com" <bruce.richardson@intel.com>,
David Christensen <drc@ibm.com>,
"honnappa.nagarahalli@arm.com" <honnappa.nagarahalli@arm.com>,
"konstantin.ananyev@intel.com" <konstantin.ananyev@intel.com>,
"ola.liljedahl@arm.com" <ola.liljedahl@arm.com>,
Ori Kam <orika@mellanox.com>, David Wilder <wilder@us.ibm.com>,
Yongseok Koh <yskoh@mellanox.com>,
Idan Werpoler <Idanw@mellanox.com>,
Olga Shern <olgas@mellanox.com>
Subject: Re: [dpdk-dev] [PATCH] eal/ppc: remove fix of memory barrier for IBM POWER
Date: Wed, 27 Mar 2019 10:19:03 +0100 [thread overview]
Message-ID: <1793899.zbSyOp9Bm4@xps> (raw)
Message-ID: <20190327091903.wkqjF1W115IEJViTrcGV3yuHqzMM55itnG3OWUHfmiA@z> (raw)
In-Reply-To: <VI1PR05MB4224613FA0B8DB3737F8F23DB65F0@VI1PR05MB4224.eurprd05.prod.outlook.com>
Pradeep, Chao,
Do we have more news?
We must merge this patch for DPDK 19.05-rc1.
I understand you want to try improving performance
by using lightweight sync for SMP barrier,
and this change can be done later.
First priority is to fix the bug of the general barrier.
That's why I should push this patch before the end of the week.
> Shahaf Shuler <shahafs@mellanox.com<mailto:shahafs@mellanox.com>> wrote on 03/23/2019 11:37:42 PM:
> > From: Shahaf Shuler <shahafs@mellanox.com<mailto:shahafs@mellanox.com>>
> > Pradeep Satyanarayana wrote on Saturday, March 23, 2019 12:58 AM
> > >Thomas Monjalon <thomas@monjalon.net<mailto:thomas@monjalon.net>> wrote on 03/22/2019 10:51:17 AM:
> > >> 22/03/2019 16:30, Pradeep Satyanarayana:
> > >> > Thomas Monjalon <thomas@monjalon.net<mailto:thomas@monjalon.net>> wrote on 03/22/2019 01:49:03 AM:
> > >> > > 22/03/2019 02:40, Pradeep Satyanarayana:
> > >> > > > - rte_[rw]mb (general memory barrier) --> should be lwsync
> > >> > >
> > >> > > This is what may be discussed.
> > >> > > The assumption is that the general memory barrier should cover
> > >> > > all cases (CPU caches, SMP and I/O).
> > >> > > That's why we think it should "sync" for Power.
> > >> >
> > >> > In that case, at a minimum we must de-link rte_smp_[rw]mb from rte_[rw]mb
> > >> > and retain it as lwsync. Agreed?
> > >>
> > >> I have no clue about what is needed for SMP barrier in Power.
> > >> As long as it works as expected, no problem.
> > >>
> > >
> > >We will try that out and report back here, later next week
> >
> > Till then, i think there are 2 orthogonal issues:
> > 1. ppc rte_wmb is incorrect
> > 2. ppc rte_smp_[rw]mb may be improved.
> >
> > for #1 the current patch from Dekel seems to be OK. do you agree?
> > for #2 i guess you will check and come back w/ patch/answer?
>
> That has been the line of thinking. However, we need to do some extensive testing
> to confirm that it all holds up.
next prev parent reply other threads:[~2019-03-27 9:19 UTC|newest]
Thread overview: 42+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-03-18 12:58 Dekel Peled
2019-03-18 12:58 ` Dekel Peled
2019-03-19 3:24 ` Chao Zhu
2019-03-19 3:24 ` Chao Zhu
2019-03-19 10:05 ` Dekel Peled
2019-03-19 10:05 ` Dekel Peled
2019-03-19 11:14 ` Thomas Monjalon
2019-03-19 11:14 ` Thomas Monjalon
2019-03-19 19:42 ` Shahaf Shuler
2019-03-19 19:42 ` Shahaf Shuler
2019-03-19 20:45 ` Thomas Monjalon
2019-03-19 20:45 ` Thomas Monjalon
2019-03-20 22:40 ` Pradeep Satyanarayana
2019-03-20 22:40 ` Pradeep Satyanarayana
2019-03-21 8:49 ` Shahaf Shuler
2019-03-21 8:49 ` Shahaf Shuler
2019-03-22 1:40 ` Pradeep Satyanarayana
2019-03-22 1:40 ` Pradeep Satyanarayana
2019-03-22 8:49 ` Thomas Monjalon
2019-03-22 8:49 ` Thomas Monjalon
2019-03-22 15:30 ` Pradeep Satyanarayana
2019-03-22 15:30 ` Pradeep Satyanarayana
2019-03-22 17:51 ` Thomas Monjalon
2019-03-22 17:51 ` Thomas Monjalon
2019-03-22 22:57 ` Pradeep Satyanarayana
2019-03-22 22:57 ` Pradeep Satyanarayana
2019-03-24 6:37 ` Shahaf Shuler
2019-03-24 6:37 ` Shahaf Shuler
2019-03-24 17:37 ` Pradeep Satyanarayana
2019-03-24 17:37 ` Pradeep Satyanarayana
2019-03-26 9:15 ` Dekel Peled
2019-03-26 9:15 ` Dekel Peled
2019-03-27 9:19 ` Thomas Monjalon [this message]
2019-03-27 9:19 ` Thomas Monjalon
2019-03-27 23:50 ` Pradeep Satyanarayana
2019-03-27 23:50 ` Pradeep Satyanarayana
[not found] ` <OF456B0ECC.006EF7E7-ON882583CA.00827A75-882583CA.0082F7BE@LocalDomain>
2019-03-28 17:51 ` Pradeep Satyanarayana
2019-03-28 17:51 ` Pradeep Satyanarayana
2019-03-28 17:56 ` Thomas Monjalon
2019-03-28 17:56 ` Thomas Monjalon
2019-03-28 22:50 ` [dpdk-dev] [dpdk-stable] " Thomas Monjalon
2019-03-28 22:50 ` 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=1793899.zbSyOp9Bm4@xps \
--to=thomas@monjalon.net \
--cc=Idanw@mellanox.com \
--cc=bruce.richardson@intel.com \
--cc=chaozhu@linux.vnet.ibm.com \
--cc=dekelp@mellanox.com \
--cc=dev@dpdk.org \
--cc=drc@ibm.com \
--cc=honnappa.nagarahalli@arm.com \
--cc=konstantin.ananyev@intel.com \
--cc=ola.liljedahl@arm.com \
--cc=olgas@mellanox.com \
--cc=orika@mellanox.com \
--cc=pradeep@us.ibm.com \
--cc=shahafs@mellanox.com \
--cc=wilder@us.ibm.com \
--cc=yskoh@mellanox.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).