From mboxrd@z Thu Jan 1 00:00:00 1970 From: Benno Senoner Subject: Re: [linux-audio-dev] Re: new latency report Date: Sun, 9 Jul 2000 20:13:47 +0200 Content-Type: text/plain References: In-Reply-To: MIME-Version: 1.0 Message-Id: <00070920393600.02245@smp> Content-Transfer-Encoding: 8bit Sender: owner-linux-mm@kvack.org Return-Path: To: Alan Cox , Roger Larsson Cc: "linux-kernel@vger.rutgers.edu" , "linux-mm@kvack.org" , "linux-audio-dev@ginette.musique.umontreal.ca" List-ID: On Sun, 09 Jul 2000, Alan Cox wrote: > > and the 704ms used to busy loop in modprobe... > > (SB16 non PnP) > > I take patches for the sb16 if it bugs you enough to fix it. The point is that modprobe will be a general problem: many modules will freeze your box for dozen if not hundreds of msecs. ( eg aic7xxx ) We can live with this if we require that the user insmods all the modules at boottime. The problem could be the audiomatic module loading / cleaning (kmod). For example how do we know in advance that the user wants to use pppd ? (ppp.o , slhc.o ) If he is offline while doing low-latency audio , and suddenly needs something from the net, as soon as he fires up pppd, a latency-peak may occur. So a way to avoid latency peaks would be to inform the user, that if (during his audio recording sessions) he wants to do some stuff which requires module loading , he has to preload the modules at boottime, and disable automatic module cleanup. Anyone better ideas ? Benno. -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@kvack.org. For more info on Linux MM, see: http://www.linux.eu.org/Linux-MM/