From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <linux-fsdevel-owner@vger.kernel.org>
Received: from zeniv.linux.org.uk ([195.92.253.2]:33074 "EHLO
	ZenIV.linux.org.uk" rhost-flags-OK-OK-OK-OK) by vger.kernel.org
	with ESMTP id S1751946AbcDVNX6 (ORCPT
	<rfc822;linux-fsdevel@vger.kernel.org>);
	Fri, 22 Apr 2016 09:23:58 -0400
Date: Fri, 22 Apr 2016 14:23:56 +0100
From: Al Viro <viro@ZenIV.linux.org.uk>
To: Emily Maier <emily@emilymaier.net>
Cc: linux-fsdevel@vger.kernel.org
Subject: Re: [PATCH] fuse: Expose read from directory to userspace
Message-ID: <20160422132356.GJ25498@ZenIV.linux.org.uk>
References: <1461329404-7967-1-git-send-email-emily@emilymaier.net>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <1461329404-7967-1-git-send-email-emily@emilymaier.net>
Sender: linux-fsdevel-owner@vger.kernel.org
List-ID: <linux-fsdevel.vger.kernel.org>

On Fri, Apr 22, 2016 at 08:50:04AM -0400, Emily Maier wrote:
> The VFS permits filesystems to implement read(2) on directories, but
> FUSE unconditionally returns -EISDIR for it. Expose this to userspace as
> FUSE_DIR_READ and continue returning the old behavior if the filesystem
> doesn't implement it.

What the hell for?  read() on directories had been an anachronism since
early 80s, has nothing resembling a sane semantics, etc.

Why would we want to implement that idiocy?