RE: [PATCH] Basic support for checking NFSv4 ACLs in Linux

From: Ondrej Valousek <ondrej.valousek.xm@renesas.com>
To: "Andreas Grünbacher" <andreas.gruenbacher@gmail.com>
Cc: "bug-gnulib@gnu.org" <bug-gnulib@gnu.org>
Subject: RE: [PATCH] Basic support for checking NFSv4 ACLs in Linux
Date: Wed, 16 Nov 2022 09:51:46 +0000	[thread overview]
Message-ID: <TYXPR01MB1854BAFA66C76EAAE952ECC7D9079@TYXPR01MB1854.jpnprd01.prod.outlook.com> (raw)
In-Reply-To: <CAHpGcMLD3UMSxf-0m3Nt9AK7EKJpAfdmT93cKTnq4r4Rzu_W0A@mail.gmail.com>

> I'm not really sure what to do about ACE4_READ_NAMED_ATTRS and ACE4_WRITE_NAMED_ATTRS; those are not the same as Linux extended attributes. This will need a bit of testing against various NFSv4 servers to give reasonable results.

Can we just commit the code as-is without the extra mask checks for now?
Frankly speaking:
- I still do not understand why we should be doing it, I perceive it as an unnecessary extra potentially fragile code
- the existing code just "works" well to me. Tried various scenarios on Netapp and Linux NFS server. It just works.
- I understand it might not be perfect, but I think even you will agree that it's far better than what we have now in there (i.e. nothing)
- since most of the heavy lifting has already been done, we can patch the masks later on (if needed) as it should be fairly easy job once we know exactly what to do there.

I am happy to incorporate the other suggestions you had (thanks for them), but I have reached the point that I can't invest more time into it.
Thanks,

Ondrej