Re: [RFC][PATCH] cpqarray-1: Convert to modern module_init mechanism

Kai Germaschewski (kai@tp1.ruhr-uni-bochum.de)
Fri, 17 May 2002 09:35:31 -0500 (CDT)


On Thu, 16 May 2002, Adam Kropelin wrote:

> Below is a patch (against 2.5.15-dj1) to convert cpqarray over to the modern
> module_init mechanism. This eliminates the need to call cpqarray_init() from
> genhd.c and starts the process of simplifying the cpqarray init sequence.
> It lays the groundwork for converting over to the "new" PCI registration
> mechanism as well. Also included in the patch are some simple cleanups for
> a few obvious formatting flaws.
>
> Comments and critique are welcome. I'm also curious if this work is
> considered worthwhile. If so, I'll continue on and do the PCI init conversion
> (and any other fixups that may be warranted) as well.

Patch looks basically good to me (I basically have the same thing sitting
around here, as I was cleaning up drivers/block/genhd.c) If you want to I
can send you what I have, so you can base the further changes (e.g. PCI)
on it.

--Kai

> /*
> * This is it. Find all the controllers and register them. I really hate
> * stealing all these major device numbers.
> - * returns the number of block devices registered.
> + * returns 0 on success, -EIO on failure

I'd suggest to go ahead and make it return sensible values, i.e. -EBUSY if
it can't get the major, -ENODEV if there's no hardware, -ENOMEM if the
allocations fail.

> int __init cpqarray_init(void)

This should be static now. Also, you need to remove the explicit call to
cpqarray_init() from drivers/block/genhd.c, otherwise it'll get called
twice. (That part's already in my patch).

--Kai

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/