[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20220119144920.GL8034@ziepe.ca>
Date: Wed, 19 Jan 2022 10:49:20 -0400
From: Jason Gunthorpe <jgg@...pe.ca>
To: Praveen Kannoju <praveen.kannoju@...cle.com>
Cc: Leon Romanovsky <leon@...nel.org>,
Santosh Shilimkar <santosh.shilimkar@...cle.com>,
"David S . Miller" <davem@...emloft.net>,
"kuba@...nel.org" <kuba@...nel.org>,
"netdev@...r.kernel.org" <netdev@...r.kernel.org>,
"linux-rdma@...r.kernel.org" <linux-rdma@...r.kernel.org>,
"rds-devel@....oracle.com" <rds-devel@....oracle.com>,
"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
Rama Nichanamatlu <rama.nichanamatlu@...cle.com>,
Rajesh Sivaramasubramaniom
<rajesh.sivaramasubramaniom@...cle.com>
Subject: Re: [PATCH RFC] rds: ib: Reduce the contention caused by the
asynchronous workers to flush the mr pool
On Wed, Jan 19, 2022 at 02:08:48PM +0000, Praveen Kannoju wrote:
> From: Jason Gunthorpe [mailto:jgg@...pe.ca]
> Sent: 19 January 2022 06:47 PM
> To: Praveen Kannoju <praveen.kannoju@...cle.com>
> Cc: Leon Romanovsky <leon@...nel.org>; Santosh Shilimkar <santosh.shilimkar@...cle.com>; David S . Miller <davem@...emloft.net>; kuba@...nel.org; netdev@...r.kernel.org; linux-rdma@...r.kernel.org; rds-devel@....oracle.com; linux-kernel@...r.kernel.org; Rama Nichanamatlu <rama.nichanamatlu@...cle.com>; Rajesh Sivaramasubramaniom <rajesh.sivaramasubramaniom@...cle.com>
> Subject: Re: [PATCH RFC] rds: ib: Reduce the contention caused by the asynchronous workers to flush the mr pool
>
> On Wed, Jan 19, 2022 at 01:12:29PM +0000, Praveen Kannoju wrote:
>
> > Yes, we are using the barriers. I was justifying the usage of
> > smp_rmb() and smp_wmb() over smp_load_acquire() and
> > smp_store_release() in the patch.
>
> You failed to justify it.
>
> Jason
>
> Apologies, if my earlier point is not clear, Jason.
> Let me reframe:
>
> 1. The introduced bool variable "flush_ongoing", is being accessed only in the function "rds_ib_free_mr" while spawning asynchronous workers.
>
> 2. The ordering guaranteed by smp_rmb() and smp_wmb() would be
> sufficient for such simple usage and hence we did not use
> smp_load_acquire() and smp_store_release().
Again you haven't defined why these barriers are any differnet from
acquire/release or even *what they are doing*
Jason
Powered by blists - more mailing lists