On Fri, 11 Feb 2005, Brad Boyer wrote:
> On Sat, Feb 12, 2005 at 04:04:49PM +1100, Finn Thain wrote:
> > Having finally looked at the definition of the __devexit_p macro, it seems
> > to me that the release method is supposed to be a null pointer. So
> > Noah's patch is not the problem at all, but it is not the entire solution
> > either. This patch is needed as well.
>
> I'm not sure this will get accepted in the mainline. In one sense, it's
> the right fix, because the release function being NULL only matters in
> the module case. However, it seems kind of messy. It might be a good
> idea to run a change like this one past the main scsi list. It looks
> like it is [EMAIL PROTECTED]
Indeed. __devexit_p() is used to set a function pointer for a routine that's in
the __exit section, i.e. a cleanup routine. If module support and hotplug are
disabled, these routines are thrown away at link time, and the pointer will be
replaced by NULL.
But I guess the real problem is in the NCR5380 driver, since there are no other
SCSI drivers that use __devexit_p() on the .release method. What's the problem
with always defining a .release method for NCR5380?
Gr{oetje,eeting}s,
Geert
--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- [EMAIL PROTECTED]
In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
-- Linus Torvalds
-
To unsubscribe from this list: send the line "unsubscribe linux-m68k" in
the body of a message to [EMAIL PROTECTED]
More majordomo info at http://vger.kernel.org/majordomo-info.html
