On 3/9/21 1:40 PM, Oscar Salvador wrote:
> +static void __meminit vmemmap_use_new_sub_pmd(unsigned long start, unsigned long end)
> +{
> + /*
> + * Could be our memmap page is filled with PAGE_UNUSED already from a
> + * previous remove. Make sure to reset it.
> + */
> + vmemmap_use_sub_pmd(start);
> +
> + /*
> + * Mark with PAGE_UNUSED the unused parts of the new memmap range
> + */
> + if (!IS_ALIGNED(start, PMD_SIZE))
> + memset((void *)start, PAGE_UNUSED,
> + start - ALIGN_DOWN(start, PMD_SIZE));
> + if (!IS_ALIGNED(end, PMD_SIZE))
> + memset((void *)end, PAGE_UNUSED,
> + ALIGN(end, PMD_SIZE) - end);
> +}
> +#endif
This is apparently under both CONFIG_SPARSEMEM_VMEMMAP and
CONFIG_MEMORY_HOTPLUG #ifdefs. It errors out at compile-time with this
config: https://sr71.net/~dave/intel/config-mmotm-20210311
> linux.git/arch/x86/mm/init_64.c: In function 'vmemmap_populate_hugepages':
> linux.git/arch/x86/mm/init_64.c:1585:6: error: implicit declaration of function 'vmemmap_use_new_sub_pmd' [-Werror=implicit-function-declaration]
> vmemmap_use_new_sub_pmd(addr, next);
> ^~~~~~~~~~~~~~~~~~~~~~~
> /home/davehans/linux.git/arch/x86/mm/init_64.c:1591:4: error: implicit declaration of function 'vmemmap_use_sub_pmd' [-Werror=implicit-function-declaration]
> vmemmap_use_sub_pmd(addr, next);
> ^~~~~~~~~~~~~~~~~~~
I didn't see a quick fix other than #ifdef'ing the call sites, which is
pretty ugly.
