Re: IDE probing IDE_MAX_HWIFS

[Nick Warne <nick@linicks.net>]

> Now my question :-)  Is there a specific reason why this isn't included in
> other architectures?  I am asking as I guess one hell of a lot of people
> running on i386 have only two IDE interfaces anyway, and it could do with
> defining it as 2...

I have a patch here, works very well.  But I need to see if I am a bit 'Mickey 
Mouse' and need to ask all you proper coders if this would be an acceptable 
patch.

In drivers/ide/Kconfig

if IDE

config IDE_HWIFS_NUM
        bool "Specify the number of IDE Interfaces"
        depends on (ALPHA || SUPERH || X86)
        default n
        help

          ALPHA and SUPERH say 'y' here.

          X86 say 'y' to this if you wish to specify the number of IDE
          interfaces on your system.  If unsure, say 'n' to use
          the kernel default options (6 or 10).

config IDE_MAX_HWIFS
        int "Max IDE interfaces"
        depends on IDE_HWIFS_NUM
        default 4
        help
          This is the maximum number of IDE hardware interfaces that will
          be supported by the driver. Make sure it is at least as high as
          the number of IDE interfaces in your system.

          On X86 architecture default is (10 or 6) IDE interfaces if this
          is not used (IDE_HWIFS_NUM = n)



and in include/asm-i386/ide.h

#ifndef MAX_HWIFS
#ifndef CONFIG_IDE_HWIFS_NUM
# ifdef CONFIG_BLK_DEV_IDEPCI
#define MAX_HWIFS       10
# else
#define MAX_HWIFS       6
# endif
#else
#define MAX_HWIFS       CONFIG_IDE_MAX_HWIFS
#endif
#endif


I have just built and it works great - boot time seems to increase a lot (but 
I haven't measured as such).  It also eliminates me needing all the 
idex=noprobe also.

Nick
-- 
"When you're chewing on life's gristle,
Don't grumble, Give a whistle..."
-
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/