From alsa-devel-owner@alsa.jcu.cz  Fri Feb 19 02:13:53 1999
Received: from rumms.uni-mannheim.de (rumms.uni-mannheim.de [134.155.50.52])
	by marvin.jcu.cz (8.9.1a/8.9.1) with ESMTP id CAA31988
	for <alsa-devel@alsa.jcu.cz>; Fri, 19 Feb 1999 02:13:09 +0100
Received: from rumms.uni-mannheim.de (rhb.swm.uni-mannheim.de [134.155.20.115])
	by rumms.uni-mannheim.de (8.8.7/8.8.7) with ESMTP id CAA18980;
	Fri, 19 Feb 1999 02:13:07 +0100 (MET)
Message-ID: <36CCBA79.A9A2A7D6@rumms.uni-mannheim.de>
Date: Fri, 19 Feb 1999 02:12:25 +0100
From: Raphael Becker <beckerra@rumms.uni-mannheim.de>
MIME-Version: 1.0
To: Ulf Carlsson <ulfc@bun.falkenberg.se>
CC: alsa-devel@alsa.jcu.cz,
        Linux-Kernel Mailinglist <linux-kernel@vger.rutgers.edu>,
        support@4front-tech.com
Subject: Re: 2.2.x memory allocation (was Re: TEAC CD-R55S doesn't work)
References: <36CB62ED.28F5E547@rumms.uni-mannheim.de> <Pine.OSF.3.96.990218225200.10998A-100000@alpha.hut.fi> <19990219000754.A15878@bun.falkenberg.se>
Content-Type: text/plain; charset=iso-8859-1
Content-Transfer-Encoding: 8bit
Reply-To: alsa-devel@alsa.jcu.cz
Sender: alsa-devel-owner@alsa.jcu.cz
Precedence: list

Ulf Carlsson wrote:
> 
> > Hmm... I think there's something wrong with the way 2.2.1 allocates memory...
> > ... as in there seems to be no memory for some parts of the kernel... like
> > DMA:ble memory for example
> > let me show you what trying to get sound working w/ 2.2.1 gave me:
> 
> [snip]
> 
> When you check if you can allacate a DMA buffer you cannot just check how much
> free RAM you have. You have to check if those pages are in blocks which are
> large enough. If your memory is very fragmented you might very well be unable to
> allocate a big DMA buffer. Check this with alt-sysrq-m.

Ok, for this time i tried the easiest way (like in good ol Windows): 
"rebooting the system"!!!

--> after that "top" said, I have 44M free RAM-Space. I think a fresh
rebooted system doesnt have too much fragmentation in the free memory
blocks (somewhere in this 44M free RAM should be a block which is big
enough for cdrecord, or am I thinking wrong?)

Why doesnt it make some problems on other systems? 
Why does it work if I just boot my old 2.0.36, with all the options I
have in 2.2.1 too, all SCSI-reated options are static compiled into
the kernel in both cases. 

I and a good frind from Heidelberg (greetings to him) think, that it
is a problem of the Adaptec-related things in the new Kernel, he has
2.2.1 too and the same CD-R, my CD-R in his system works fine, he
doesnt have a Adaptec 2940AU (PCI), I have ...

Greetings 
Raphael
-- 
 _    _ Powered by SuSE___   ___   _  
| |  (_)_ _ _  ___ __ |_  ) |_  ) / | 
| |__| | ' \ || \ \ /  / / _ / / _| | 
|____|_|_||_\_,_/_\_\ /___(_)___(_)_|

