From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: by befuddled.reisers.ca (Postfix, from userid 65534) id C130A1EFD27; Sat, 19 Nov 2016 16:47:16 -0500 (EST) Received: from hera.aquilenet.fr (hera.aquilenet.fr [141.255.128.1]) by befuddled.reisers.ca (Postfix) with ESMTP id 86C661EFD0F for ; Sat, 19 Nov 2016 16:47:15 -0500 (EST) Received: from localhost (localhost [127.0.0.1]) by hera.aquilenet.fr (Postfix) with ESMTP id 783D39A6E for ; Sat, 19 Nov 2016 22:47:10 +0100 (CET) X-Virus-Scanned: Debian amavisd-new at aquilenet.fr Received: from hera.aquilenet.fr ([127.0.0.1]) by localhost (hera.aquilenet.fr [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id WNMe0h3XgvSy for ; Sat, 19 Nov 2016 22:47:10 +0100 (CET) Received: from var.youpi.perso.aquilenet.fr (unknown [IPv6:2a01:cb19:181:c200:3602:86ff:fe2c:6a19]) by hera.aquilenet.fr (Postfix) with ESMTPSA id F335DA83 for ; Sat, 19 Nov 2016 22:47:09 +0100 (CET) Received: from samy by var.youpi.perso.aquilenet.fr with local (Exim 4.88) (envelope-from ) id 1c8DTZ-0004Pw-IH for speakup@linux-speakup.org; Sat, 19 Nov 2016 22:47:09 +0100 Date: Sat, 19 Nov 2016 22:47:09 +0100 From: Samuel Thibault To: "Speakup is a screen review system for Linux." Subject: Re: using system serial drivers Message-ID: <20161119214709.GF3422@var.home> References: <20161119162834.GA2026@db> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20161119162834.GA2026@db> User-Agent: Mutt/1.5.21+34 (58baf7c9f32f) (2010-12-30) X-Bogosity: Ham, tests=bogofilter, spamicity=0.000000, version=1.2.4 X-BeenThere: speakup@linux-speakup.org X-Mailman-Version: 2.1.23 Precedence: list List-Id: "Speakup is a screen review system for Linux." List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sat, 19 Nov 2016 21:47:16 -0000 Hello, David, on Sat 19 Nov 2016 11:28:34 -0500, wrote: > Have any of you bothered to read the code in the driver you are calling > and see what overhead is imposed on entry/exit? It is quite abit. That's still extremely slow compared to the actual time spent by hardware to send the bytes over the wire. Really, nowadays we have billions of instructions per seconds with all kinds of hardware predictions to make things fast. The whole kernel has moved to making things object-oriented to make it more modular etc., speakup can really afford the same. > Some synths already have their own catch up handlers, so it would be quite > easy to write a new do_catchup that used the new serial > i/o and each synth could call the new one when appropriate. That'd bring more complexity that needed. Really, what is costly nowadays is human time to write & read code, not hardware time. Samuel