Hi Oliver,
I'll take care of these at my next opportunity. I know I removed the Minolta
entry - at least in one of the sources - just not this one. ;)
I don't quite understand about the races - but that's because I don't
understand the underlying mechanics. Can you elaborate or point me to some
material that better explains what's going on?
Thanks,
/\/elson
On Fri, 28 Sep 2001, Oliver Neukum wrote:
> Am Freitag, 28. September 2001 04:34 schrieb David Nelson:
> > Hi All,
> >
> > Well, here it is! And it even has devfs support! And best of all, it
> > seems to work. ;)
> >
> > This is not a patch. I've still yet to update all the documentation and
> > what not. I've also got some open issues that I'd like to resolve/cleanup
> > before submitting a patch for inclusion into the kernel. So, in the
> > meantime, give it a run and let me know if you have any problems.
> >
> > If I've missed something (a patch, ID's, recognition, etc), just let me
> > know.
>
> Please remove the Minolta Scan Dual. It needs a special driver as it uses
> SCSI over USB.
>
> Your disconnect_scanner() is racing with read, write and open.
> You need to take the BKL not only in open, but in disconnect and close, too.
> Then you make the freeing of memory in disconnect conditional on the flag you
> set at open time. If it is not set you set a new flag to be added which would
> cause close to do what disconnect would do. You may also check this flag at
> the beginning of read and write to quickly return ENODEV.
>
> Regards
> Oliver
>
--
~~ ** ~~ UNIX remains, to this very day, a dark endless maze of ~~ ** ~~
catwalks and mantraps, an eternal hard-hat area that
kills the foolish and shelters the brave. (UNIX Manifesto)
_______________________________________________
[EMAIL PROTECTED]
To unsubscribe, use the last form field at:
https://lists.sourceforge.net/lists/listinfo/linux-usb-devel
