RE: [PATCH]: megaraid 2.20.4: Fixes a data corruption bug

[Bagalkote, Sreenivas]

Understood. Thank you. I verified that module loads fine even without
CONFIG_COMPAT.

Thanks,
Sreenivas

>-----Original Message-----
>From: James Bottomley [mailto:James.Bottomley@xxxxxxxxxxxx]
>Sent: Friday, October 01, 2004 4:58 PM
>To: Bagalkote, Sreenivas
>Cc: Mukker, Atul; 'linux-kernel@xxxxxxxxxxxxxxx';
>'linux-scsi@xxxxxxxxxxxxxxx'; 'bunk@xxxxxxxxx'; 'Andrew Morton';
>'Matt_Domsch@xxxxxxxx'; Ju, Seokmann
>Subject: RE: [PATCH]: megaraid 2.20.4: Fixes a data corruption bug
>
>
>On Fri, 2004-10-01 at 16:08, Bagalkote, Sreenivas wrote:
>> The submitted previous version of megaraid (2.20.3.1) had 
>> register_ioctl32_conversion & unregister_ioctl32_conversion 
>> defined to empty statements if CONFIG_COMPAT was _not_
>> defined.
>
>I know that.  However, when the empty statements were added to
>ioctl32.h, those had to be taken out of the megaraid driver.  That's
>this patch in the scsi-misc-2.6 tree:
>
> [PATCH] megaraid warning fix
>  
>  The ioctl32 conversion registration stubs are in ioctl32.h now.
>  
>  Signed-off-by: Andrew Morton <akpm@xxxxxxxx>
>
>> But I think the preferred way was to have the occurances of 
>> (un)register_ioctl32_conversion in the code surrounded by 
>> #ifdef CONFIG_COMPAT ... #endif directly. In the kernel source
>> only register_ioctl32_conversion has these #ifdef .. #endif. The
>> unregister_ioctl32_conversion doesn't.
>
>The current preferred way is to use the empty definitions in
>linux/ioctl32.h which means there's no necessity for adding the #ifdef
>CONFIG_COMPAT.  The correct thing is to remove the #ifdef CONFIG_COMPAT
>from around the register_ioctl32... part.
>
>James
>
>
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/