在 2013-02-04一的 11:31 -0800，David Rientjes写道：
> On Mon, 4 Feb 2013, liguang wrote:
>
> > diff --git a/arch/x86/include/asm/acpi.h b/arch/x86/include/asm/acpi.h
> > index 0c44630..c4e0875 100644
> > --- a/arch/x86/include/asm/acpi.h
> > +++ b/arch/x86/include/asm/acpi.h
> > @@ -181,7 +181,8 @@ static inline void disable_acpi(void) { }
> > #define ARCH_HAS_POWER_INIT 1
> >
> > #ifdef CONFIG_ACPI_NUMA
> > -extern int acpi_numa;
> > +extern void bad_srat(void);
> > +extern bool srat_disabled(void);
>
> So anything that wants to disable acpi_numa, such as xen, must call a
> function named bad_srat()? That seems a little misnamed itself.
I would assume every caller have the sense of what SRAT is.
(may be a little bold assumption)
>
> > extern int x86_acpi_numa_init(void);
> > #endif /* CONFIG_ACPI_NUMA */
> >
> > diff --git a/arch/x86/mm/numa.c b/arch/x86/mm/numa.c
> > index 2d125be..92986d8 100644
> > --- a/arch/x86/mm/numa.c
> > +++ b/arch/x86/mm/numa.c
> > @@ -47,7 +47,7 @@ static __init int numa_setup(char *opt)
> > #endif
> > #ifdef CONFIG_ACPI_NUMA
> > if (!strncmp(opt, "noacpi", 6))
> > - acpi_numa = -1;
> > + bad_srat();
>
> This will print an error line saying "SRAT: SRAT not used." which wasn't
> emitted before. I don't think that's the desire.
yes, you're right.
>
> > #endif
> > return 0;
> > }
> > diff --git a/arch/x86/mm/srat.c b/arch/x86/mm/srat.c
> > index 4ddf497..a837c95 100644
> > --- a/arch/x86/mm/srat.c
> > +++ b/arch/x86/mm/srat.c
> > @@ -24,22 +24,22 @@
> > #include <asm/apic.h>
> > #include <asm/uv/uv.h>
> >
> > -int acpi_numa __initdata;
> > +static bool acpi_numa __initdata;
> >
>
> You've changed this to be type bool but didn't update the existing code
> that still does "acpi_numa = 1".
good catch!
>
> > static __init int setup_node(int pxm)
> > {
> > return acpi_map_pxm_to_node(pxm);
> > }
> >
> > -static __init void bad_srat(void)
> > +void __init bad_srat(void)
> > {
> > printk(KERN_ERR "SRAT: SRAT not used.\n");
> > - acpi_numa = -1;
> > + acpi_numa = false;
> > }
> >
> > -static __init inline int srat_disabled(void)
> > +bool __init srat_disabled(void)
> > {
> > - return acpi_numa < 0;
> > + return acpi_numa;
> > }
> >
> > /* Callback for SLIT parsing */
> > diff --git a/arch/x86/xen/enlighten.c b/arch/x86/xen/enlighten.c
> > index 138e566..94c8c70 100644
> > --- a/arch/x86/xen/enlighten.c
> > +++ b/arch/x86/xen/enlighten.c
> > @@ -1415,7 +1415,7 @@ asmlinkage void __init xen_start_kernel(void)
> > * any NUMA information the kernel tries to get from ACPI will
> > * be meaningless. Prevent it from trying.
> > */
> > - acpi_numa = -1;
> > + bad_srat();
>
> Same as before. It seems like you want to introduce a new function
> specifically to disable it:
>
> void acpi_numa_disable(void)
> {
> acpi_numa = false;
> }
>
> and then bad_srat() should call into it?
good idea, Thanks!
--
To unsubscribe from this list: send the line "unsubscribe linux-acpi" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
