Thomas Monjalon <tho...@monjalon.net> wrote on 03/22/2019 10:51:17 AM:
> From: Thomas Monjalon <tho...@monjalon.net>
> To: Pradeep Satyanarayana <prad...@us.ibm.com>
> Cc: bruce.richard...@intel.com, Chao Zhu
> <chao...@linux.vnet.ibm.com>, Dekel Peled <dek...@mellanox.com>,
> dev@dpdk.org, David Christensen <d...@ibm.com>,
> honnappa.nagaraha...@arm.com, konstantin.anan...@intel.com,
> ola.liljed...@arm.com, Ori Kam <or...@mellanox.com>, Shahaf Shuler
> <shah...@mellanox.com>, David Wilder <wil...@us.ibm.com>, Yongseok
> Koh <ys...@mellanox.com>
> Date: 03/22/2019 10:51 AM
> Subject: Re: [PATCH] eal/ppc: remove fix of memory barrier for IBM POWER
>
> 22/03/2019 16:30, Pradeep Satyanarayana:
> > Thomas Monjalon <tho...@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
Thanks
Pradeep
prad...@us.ibm.com
