[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20080718174929.GA981@elte.hu>
Date: Fri, 18 Jul 2008 19:49:29 +0200
From: Ingo Molnar <mingo@...e.hu>
To: Jack Steiner <steiner@....com>
Cc: tglx@...utronix.de, linux-kernel@...r.kernel.org,
"H. Peter Anvin" <hpa@...or.com>
Subject: Re: [PATCH] - Earlier detection of system type (UV or non-UV)
* Ingo Molnar <mingo@...e.hu> wrote:
>
> * Jack Steiner <steiner@....com> wrote:
>
> > A patch is coming that requires earlier detection of system type (UV
> > or non-UV) Move the logic that detects whether running on UV to
> > acpi_boot_table_init().
>
> > ===================================================================
> > --- linux.orig/arch/x86/kernel/acpi/boot.c 2008-07-09 11:14:03.000000000 -0500
> > +++ linux/arch/x86/kernel/acpi/boot.c 2008-07-09 11:14:33.000000000 -0500
> > @@ -1712,6 +1712,10 @@ int __init acpi_boot_table_init(void)
> > return error;
> > }
> >
> > +#ifdef CONFIG_X86_64
> > + acpi_table_parse(ACPI_SIG_MADT, acpi_early_madt_oem_check);
> > +#endif
> > +
> > acpi_table_parse(ACPI_SIG_BOOT, acpi_parse_sbf);
>
> why the assymetry? We should parse it regardless of 32-bit or 64-bit.
> (even if UV is 64-bit only, obviously)
also, it doesnt build with the attached config:
arch/x86/kernel/genapic_64.c: In function ‘acpi_early_madt_oem_check':
arch/x86/kernel/genapic_64.c:71: error: dereferencing pointer to
incomplete type
Ingo
View attachment "config" of type "text/plain" (31107 bytes)
Powered by blists - more mailing lists