Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Date:   Fri, 28 Apr 2023 20:27:53 +0900
From:   Dominique Martinet <asmadeus@codewreck.org>
To:     Dave Chinner <david@fromorbit.com>
Cc:     Alexander Viro <viro@zeniv.linux.org.uk>,
        Christian Brauner <brauner@kernel.org>,
        Jens Axboe <axboe@kernel.dk>,
        Pavel Begunkov <asml.silence@gmail.com>,
        Stefan Roesch <shr@fb.com>, linux-fsdevel@vger.kernel.org,
        linux-kernel@vger.kernel.org, io-uring@vger.kernel.org
Subject: Re: [PATCH RFC 2/2] io_uring: add support for getdents
Message-ID: <ZEutuerMIcKpWAfP@codewreck.org>
References: <20230422-uring-getdents-v1-0-14c1db36e98c@codewreck.org>
 <20230422-uring-getdents-v1-2-14c1db36e98c@codewreck.org>
 <20230423224045.GS447837@dread.disaster.area>
 <ZEXChAJfCRPv9vbs@codewreck.org>
 <20230428050640.GA1969623@dread.disaster.area>
 <ZEtkXJ1vMsFR3tkN@codewreck.org>
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Disposition: inline
In-Reply-To: <ZEtkXJ1vMsFR3tkN@codewreck.org>
Precedence: bulk

Dominique Martinet wrote on Fri, Apr 28, 2023 at 03:14:52PM +0900:
> > We already pass a struct dir_context to ->iterate_shared(), so we
> > have a simple way to add context specific flags down the filesystem
> > from iterate_dir(). This is similar to the iocb for file data IO
> > that contains the flags field that holds the IOCB_NOWAIT context for
> > io_uring based IO. So the infrastructure to plumb it all the way
> > down the fs implementation of ->iterate_shared is already there.
> 
> Sure, that sounds like a good approach that isn't breaking the API (not
> breaking iterate/iterate_shared implementations that don't look at the
> flags and allowing the fs that want to look at it to do so)

Hmm actually I said that, but io_getdents() needs to know if the flag
will be honored or not (if it will be honored, we can call this when
issue_flags & IO_URING_F_NONBLOCK but if we're not sure the fs handles
it then we risk blocking)

I'm not familiar with this part of the VFS, but I do not see any kind of
flags for the filesystem to signal if it'll handle it or not -- this is
actually similar to iterate vs. iterate_shared so that'll mean adding
iterate_shared_hasnonblock or something, which is getting silly.

I'm sure you understand this better than me and I'm missing something
obvious here, but I don't think I'll be able to make something I'm happy
with here (in a reasonable timeframe anyway).


Thanks,
-- 
Dominique Martinet | Asmadeus