On 9/28/20 12:45 PM, Nathan Chancellor wrote:
> After commit 01feba590cd6 ("ACPI: Do not create new NUMA domains from
> ACPI static tables that are not SRAT"):
>
> $ scripts/config --file arch/x86/configs/x86_64_defconfig -d NUMA -e ACPI_NFIT
>
> $ make -skj"$(nproc)" distclean defconfig drivers/acpi/nfit/
> drivers/acpi/nfit/core.c: In function ‘acpi_nfit_register_region’:
> drivers/acpi/nfit/core.c:3010:27: error: implicit declaration of
> function ‘pxm_to_node’; did you mean ‘xa_to_node’?
> [-Werror=implicit-function-declaration]
> 3010 | ndr_desc->target_node = pxm_to_node(spa->proximity_domain);
> | ^~~~~~~~~~~
> | xa_to_node
> cc1: some warnings being treated as errors
> ...
>
> Add a stub function like acpi_map_pxm_to_node had so that the build
> continues to work.
>
> Fixes: 01feba590cd6 ("ACPI: Do not create new NUMA domains from ACPI static tables that are not SRAT")
> Signed-off-by: Nathan Chancellor <natechancellor@xxxxxxxxx>
> ---
>
> I am not sure if this is the right place or value for this. It looks
> like there is going to be another stub function added here, which is
> going through -mm:
>
> https://lkml.kernel.org/r/159643094925.4062302.14979872973043772305.stgit@xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
>
> include/acpi/acpi_numa.h | 5 +++++
> 1 file changed, 5 insertions(+)
>
> diff --git a/include/acpi/acpi_numa.h b/include/acpi/acpi_numa.h
> index fdebcfc6c8df..09eb3bc20ff5 100644
> --- a/include/acpi/acpi_numa.h
> +++ b/include/acpi/acpi_numa.h
> @@ -22,5 +22,10 @@ extern int acpi_numa __initdata;
> extern void bad_srat(void);
> extern int srat_disabled(void);
>
> +#else /* CONFIG_ACPI_NUMA */
> +static inline int pxm_to_node(int pxm)
> +{
> + return 0;
> +}
> #endif /* CONFIG_ACPI_NUMA */
> #endif /* __ACP_NUMA_H */
>
> base-commit: eb6335b68ce3fc85a93c4c6cd3bb6bc5ac490efe
OK, that works/builds. It doesn't quite apply cleanly to linux-next-20200929
but that's a minor detail and easy to get around.
Thanks.
Acked-by: Randy Dunlap <rdunlap@xxxxxxxxxxxxx> # build-tested
--
~Randy
