Re: [PATCH] allow i8042 register location override

[Dmitry Torokhov <dtor_core@ameritech.net>]

On Monday 06 September 2004 07:22 am, Vojtech Pavlik wrote:
> On Fri, Sep 03, 2004 at 09:41:24AM -0600, Bjorn Helgaas wrote:
> > On Friday 03 September 2004 12:57 am, Dmitry Torokhov wrote:
> > > What do you think about the patch below? I renamed some function/variable
> > > names to be more in line with the rest of i8042 code, other than that
> > > its pretty much your code.
> > 
> > That looks great to me, and it works fine on my DL360.
> > 
> > My only comment is that in an ideal world, we would not have to
> > change any drivers if a new architecture started supporting ACPI.
> > With the current patch, we'd have to twiddle some of the i8042-XXX.h
> > files a bit.  But I don't think it's worth the trouble of restructuring
> > them to fix that.
> > 
> > Thanks for all your help!
> 
> One bug that I could spot immediately is that the patch sets i8042_reset
> on i386. This doesn't seem intentional, and is quite wrong, too, since
> on some older machines it confuses the BIOS.

Nope it does not. It is only on IA64:

> +#if defined(__ia64__)
> +        i8042_reset = 1;
> +#endif
which is in i8042-x86ia64io.h

This fragment is in i8042-io.h which is not included in by Intel-originated
arches anymore.

> -#if !defined(__i386__) && !defined(__x86_64__)
>          i8042_reset = 1;
> -#endif

Did I miss something?

-- 
Dmitry