Re: [mm] 9b12e49e9b: BUG:Bad_page_state_in_process

[Matthew Wilcox]

On Thu, Jun 09, 2022 at 12:42:16PM +0800, Baolin Wang wrote:
> diff --git a/arch/x86/mm/pgtable.c b/arch/x86/mm/pgtable.c
> index 6cccf52e156a..cae74e972426 100644
> --- a/arch/x86/mm/pgtable.c
> +++ b/arch/x86/mm/pgtable.c
> @@ -858,6 +858,7 @@ int pmd_free_pte_page(pmd_t *pmd, unsigned long addr)
>         /* INVLPG to clear all paging-structure caches */
>         flush_tlb_kernel_range(addr, addr + PAGE_SIZE-1);
> 
> +       pgtable_clear_and_dec(virt_to_page(pte));
>         free_page((unsigned long)pte);
> 
>         return 1;

If you're going to call virt_to_page() here, you may as well cache the
result and call __free_page(page) to avoid calling virt_to_page() twice.