From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <linux-kernel-owner@vger.kernel.org>
Date: Wed, 30 Aug 2017 16:28:55 +0100
From: Al Viro <viro@ZenIV.linux.org.uk>
To: Christoph Hellwig <hch@lst.de>
Cc: arve@android.com, riandrews@android.com,
        linux-fsdevel@vger.kernel.org, linux-kernel@vger.kernel.org
Subject: Re: [PATCH 01/16] ashmem: switch to ->read_iter
Message-ID: <20170830152855.GL5426@ZenIV.linux.org.uk>
References: <20170830150007.3953-1-hch@lst.de>
 <20170830150007.3953-2-hch@lst.de>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20170830150007.3953-2-hch@lst.de>
Sender: linux-kernel-owner@vger.kernel.org
List-ID: <linux-fsdevel.vger.kernel.org>

On Wed, Aug 30, 2017 at 04:59:52PM +0200, Christoph Hellwig wrote:
> And use the proper VFS helper for using the backing file.

Actually, I wonder if that update of ->f_pos of the underlying file is
correct.  That's unrelated to this patch, but might be worth sorting
out while we are at it.

How about
        mutex_unlock(&ashmem_mutex);
        /*
         * asma and asma->file are used outside the lock here.  We assume
         * once asma->file is set it will never be changed, and will not
         * be destroyed until all references to the file are dropped and
         * ashmem_release is called.
         */
	ret = vfs_iter_read(asma->file, iter, &iocb->ki_pos, 0);
        mutex_lock(&ashmem_mutex);
        if (ret >= 0)
                asma->file->f_pos = *pos;
out_unlock:
        mutex_unlock(&ashmem_mutex);
        return ret;

in there?