From mboxrd@z Thu Jan 1 00:00:00 1970 From: Michael Tokarev Subject: Re: [PATCH v2] Introduce a version6 of autofs interface, to fix design error. Date: Sun, 29 Apr 2012 23:53:48 +0400 Message-ID: <4F9D9C4C.7000904@msgid.tls.msk.ru> References: <20120427.151433.1007849975903946491.davem@davemloft.net> <20120427.151651.2274764174217164454.davem@davemloft.net> <20120427.152404.2292425516870981391.davem@davemloft.net> <20120427234215.079a1565@pyramind.ukuu.org.uk> <4F9CE18C.2060806@msgid.tls.msk.ru> <4F9CF191.5020607@msgid.tls.msk.ru> Mime-Version: 1.0 Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: Sender: linux-kernel-owner@vger.kernel.org List-ID: Content-Type: text/plain; charset="us-ascii" To: Linus Torvalds Cc: Alan Cox , Thomas Meyer , David Miller , linux-kernel@vger.kernel.org, autofs@vger.kernel.org, raven@themaw.net, stable@kernel.org On 29.04.2012 23:09, Linus Torvalds wrote: > > This keeps Alan's idea of using O_DIRECT, but only makes it matter for > the writer - because now the "packet" thing is a per-pipe-buffer state > (we already had per-pipe-buffer flags, so this is not anything new). > So when you *write* using a O_DIRECT pipe, it will create packetized > buffers, and the reader just sees that directly. > > So this should "just work", and doesn't need any extra flags in the > pipe_inode_info. And the user space interface remains the same: you > can create these packet pipes with "pipe2(fd, O_DIRECT)" if you want > to, and my test-program gives exactly the same output. Ok. I verified this on 3.3 kernel (with the original fix reverted), and now both autofs5 and systemd works. I also verified it on 3.0 kernel (3.0.30), the patch also applies there just fine (with 1..2 lines offsets) and works too -- at least on 32/64 bits. So, the result appears to be an excellent solution to a bad problem... Thank you all for the efforts and support! You can add my Tested-off-by: Michael Tokarev (I'd add a Signed-off-by, but it is not my patch ;) /mjt