Subject: Re: USB CF reader problems with 2.0beta
To: Jukka Marin <jmarin@embedtronics.fi>
From: Steven M. Bellovin <smb@research.att.com>
List: port-i386
Date: 05/02/2004 13:31:39
In message <20040502170156.GA4540@kyyhky.embedtronics.fi>, Jukka Marin writes:
>On Sun, May 02, 2004 at 12:49:52PM -0400, Lennart Augustsson wrote:
>> Yes, it's a 64K buffer.  The controller can deal with non-contigous
>> pages and it's just laziness on my part that it doesn't.
>> BUT, it used to work just fine allocating 64K.  I don't understand
>> how memory can get so fragmented.
>
>On my laptop, it works just fine.  It's running
>
>NetBSD 1.6ZE (NAUKU) #5: Sun Nov 16 21:47:25 EET 2003
>
>Uptime is 29 days and the laptop has been used for many things in that
>time.  On the other hand, the 2.0beta system was rebooted yesterday and
>the memory is too fragmented for umass already (even though the 2.0beta
>machine has 256 MB more RAM than the laptop).
>

I find that things tend to work better if I have a USB mass storage 
device plugged in at boot time.  No guarantees -- but it preserves one 
contiguous area, at least.

		--Steve Bellovin, http://www.research.att.com/~smb