Received-SPF: pass (google.com: domain of linux-nfs-owner@vger.kernel.org designates 2620:137:e000::3:2 as permitted sender) client-ip=2620:137:e000::3:2;
From:   Benjamin Coddington <bcodding@redhat.com>
To:     Anna Schumaker <schumaker.anna@gmail.com>
Cc:     Cedric Blancher <cedric.blancher@gmail.com>,
        Linux NFS Mailing List <linux-nfs@vger.kernel.org>
Subject: Re: NFSv4.2: READ_PLUS - when safe to use?
Date:   Wed, 18 Oct 2023 15:41:47 -0400
Message-ID: <CCC3D7FC-BC0E-414D-A9CA-545EFC9AFC10@redhat.com>
In-Reply-To: <CAFX2JfkRpfRe6-T3MELWsAX_V5xyZoiv0Prq_1qq4-5Pi4PCag@mail.gmail.com>
References: <CALXu0Ufzu8FMdH=-_35tHNqu3c6ewf4d6a379=fUMwNvGq_rgQ@mail.gmail.com>
 <CAFX2JfkRpfRe6-T3MELWsAX_V5xyZoiv0Prq_1qq4-5Pi4PCag@mail.gmail.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
Precedence: bulk

On 18 Oct 2023, at 15:32, Anna Schumaker wrote:

> Hi Ced,
>
> On Sat, Oct 14, 2023 at 9:08 AM Cedric Blancher
> <cedric.blancher@gmail.com> wrote:
>>
>> Good afternoon!
>>
>> Since which kernel versions (NFS server, NFS client) is NFSv4.2
>> READ_PLUS safely usable?
>
> Linux 6.2 for the server side and 6.6 for the client.
>
>>
>> Also, could you make this a mount option, so people can turn this
>> on/off per mount, instead of using a kernel build option for this?
>
> The eventual plan is to remove the kernel build option entirely. We
> probably won't do a mount option, and just have it enabled by default
> when using NFS v4.2.

Anna, I'm still planning on either re-sending the patch you wrote to
enable/disable optional features per-mount, or re-work it to make it
file-per-feature as the sysfs creators intended.

We need the option to toggle client's default per-server or per-mount.

Ben