From mboxrd@z Thu Jan 1 00:00:00 1970 From: Grant Edwards Subject: Re: locking changes in tty broke low latency feature Date: Mon, 24 Feb 2014 15:44:59 +0000 (UTC) Message-ID: References: <20140219230623.736E8406062@ip-64-139-1-69.sjc.megapath.net> <53056E99.9070900@hurleysoftware.com> <53064672.3000807@hurleysoftware.com> Cc: linux-kernel@vger.kernel.org, linux-rt-users@vger.kernel.org To: linux-serial@vger.kernel.org Return-path: Sender: linux-serial-owner@vger.kernel.org List-Id: linux-rt-users.vger.kernel.org On 2014-02-23, Thomas Gleixner wrote: > On Thu, 20 Feb 2014, Peter Hurley wrote: >> Using Alan's idea to mock up a latency test, I threw together a test >> jig using two computers running 3.14-rc1 and my fwserial driver >> (modified to not aggregrate writes) in raw mode where the target does >> this: > > This is a complete pointless test. No, it isn't. It tested exactly what it was supposed to test: latency between the driver pushing bytes to the line discipline and user-space read waking up. > Use a bog standard 8250 UART on the PC and connect a microcontroller > on the other end which serves you an continous stream of data at > 115200 Baud. > > There is no way you can keep up with that without the low latency > option neither on old and nor on new machines if you have enough > other stuff going on in the system. That makes no sense. Enabling the low-latency option historically made it _harder_ to keep up since it caused more overhead in the form buffer processing and context switches. -- Grant Edwards grant.b.edwards Yow! Look into my eyes and at try to forget that you have gmail.com a Macy's charge card!