Re: Linux Sense buffer (was: sane 0.71 and microtek E3)

Rogier Wolff (
Tue, 31 Mar 1998 10:59:00 +0200 (MET DST)

David Mosberger-Tang wrote:
> But I agree it's bad taste to ship pre-compiled binaries assuming a
> 128KB kernel buffer. The real problem is that the Linux kernel
> doesn't allow to determine what the limit is. I'll see whether we
> could have that fixed for Linux 2.2.

David, I've been thinking about this. There is a possibility.

1) (Implemented for my standalone tamarack scan-driver). A -buf option
that allows the user to select the buffer size to be used at runtime.
(I've got a "decent" kernel right now, but the last half year I've been
running a kernel without the increased buffer size, and had to pass the
-buf 32000 every time... ;-)

2) (not implemented yet) Simply try with a large buffer, and if you
get "ENOMEM", reduce the buffer size automatically.


P.S. "large buffer" in my kernel means 1Mb.

If it's there and you can see it, it's REAL      |  |
If it's there and you can't see it, it's TRANSPARENT |  Tel: +31-15-2137555  |
If it's not there and you can see it, it's VIRTUAL   |__FAX:_+31-15-2138217  |
If it's not there and you can't see it, it's GONE! -- Roy Wilks, 1983  |_____|

Source code, list archive, and docs:
To unsubscribe: echo unsubscribe sane-devel | mail