Received-SPF: pass (google.com: domain of linux-nfs-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
From:   Benjamin Coddington <bcodding@redhat.com>
To:     Anna Schumaker <anna@kernel.org>
Cc:     linux-nfs@vger.kernel.org
Subject: Re: [RFC PATCH] NFS: add a sysfs file for enabling & disabling nfs
 features
Date:   Thu, 04 May 2023 08:56:37 -0400
Message-ID: <B1DC5ECE-ECA8-4282-92EB-7272D091AC87@redhat.com>
In-Reply-To: <20230421182738.901701-1-anna@kernel.org>
References: <20230421182738.901701-1-anna@kernel.org>
MIME-Version: 1.0
Content-Type: text/plain
Precedence: bulk

On 21 Apr 2023, at 14:27, Anna Schumaker wrote:

> From: Anna Schumaker <Anna.Schumaker@Netapp.com>
>
> And add some basic checking so we only enable features that are present
> in a given NFS version.
>
> Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
> ---

This is great, I like how you've kept the +/- notation similar to knfsd
supported versions.

Another way to do this would be an attribute file per capability, setting it
to 0 or 1 which is more inline with sysfs usage.

I think if we do use this, we ought to leave readdir plus out of it because
there's already a mount option for it.  Readdir plus can be turned on and
off with a remount already.  The issue for me would be how to work out what
the behavior should be when we have a mount that has "nordirplus" and then
someone tries to toggle it via sysfs.

Any other thoughts?

I'll add this patch to my future postings of sysfs work.

Ben