[PATCH] armv8: mmu: don't panic when unmapping a page which isn't mapped

[Ilias Apalodimas]

On Fri, 12 Jun 2026 at 14:32, Casey Connolly <casey.connolly at linaro.org> wrote:
>
> If a caller tries to unmap a single page which is already part of an
> unmapped block, set_one_region() will panic since there isn't a valid
> block or table to contain the page. Detect this case explicitly and
> return since no action needs to be taken.
>
> Signed-off-by: Casey Connolly <casey.connolly at linaro.org>
> ---
> Depends on my previous series which fixes unmapping regions
>
> https://lore.kernel.org/u-boot/20260608-b4-mmu-unmap-fixes-v6-0-ac0764cccf40@linaro.org/
> ---

Acked-by: Ilias Apalodimas <ilias.apalodimas at linaro.org>

>  arch/arm/cpu/armv8/cache_v8.c | 5 +++++
>  1 file changed, 5 insertions(+)
>
> diff --git a/arch/arm/cpu/armv8/cache_v8.c b/arch/arm/cpu/armv8/cache_v8.c
> index 6c85022556ad..472887b5d99b 100644
> --- a/arch/arm/cpu/armv8/cache_v8.c
> +++ b/arch/arm/cpu/armv8/cache_v8.c
> @@ -1003,8 +1003,13 @@ static u64 set_one_region(u64 start, u64 size, u64 attrs, bool flag, int level)
>         int levelshift = level2shift(level);
>         u64 levelsize = 1ULL << levelshift;
>         u64 *pte = find_pte(start, level);
>
> +       /* Trying to unmap a region which is part of an already unmapped block, nothing to do */
> +       if (pte_type(pte) == PTE_TYPE_FAULT && attrs == PTE_TYPE_FAULT && size < levelsize) {
> +               return size;
> +       }
> +
>         /* Can we can just modify the current level block/page? */
>         if (is_aligned(start, size, levelsize)) {
>                 if (attrs == PTE_TYPE_FAULT) {
>                         if (pte_type(pte) == PTE_TYPE_TABLE && level < 3)
> --
> 2.53.0
>