[RFCv2 1/9] efi/libstub: Ask efi_random_alloc() to skip unusable memory

Ard Biesheuvel ardb at kernel.org
Wed Aug 28 06:28:01 PDT 2024


On Mon, 19 Aug 2024 at 16:55, Pingfan Liu <piliu at redhat.com> wrote:
>
> efi_random_alloc() demands EFI_ALLOCATE_ADDRESS when allocate_pages(),
> but the current implement can not ensure the selected target locates
> inside free area, that is to exclude EFI_BOOT_SERVICES_*,
> EFI_RUNTIME_SERVICES_* etc.
>
> Fix the issue by checking md->type.
>
> Signed-off-by: Pingfan Liu <piliu at redhat.com>
> Cc: Ard Biesheuvel <ardb at kernel.org>
> To: linux-efi at vger.kernel.org
> ---
>  drivers/firmware/efi/libstub/randomalloc.c | 5 +++++
>  1 file changed, 5 insertions(+)
>
> diff --git a/drivers/firmware/efi/libstub/randomalloc.c b/drivers/firmware/efi/libstub/randomalloc.c
> index c41e7b2091cdd..7304e767688f2 100644
> --- a/drivers/firmware/efi/libstub/randomalloc.c
> +++ b/drivers/firmware/efi/libstub/randomalloc.c
> @@ -79,6 +79,8 @@ efi_status_t efi_random_alloc(unsigned long size,
>                 efi_memory_desc_t *md = (void *)map->map + map_offset;
>                 unsigned long slots;
>
> +               if (!(md->type & (EFI_CONVENTIONAL_MEMORY || EFI_PERSISTENT_MEMORY)))
> +                       continue;

This is wrong in 3 different ways:
- md->type is not a bitmask
- || is not bitwise but boolean
- get_entry_num_slots() ignores all memory types except
EFI_CONVENTIONAL_MEMORY anyway.

So what exactly are you trying to fix here?


>                 slots = get_entry_num_slots(md, size, ilog2(align), alloc_min,
>                                             alloc_max);
>                 MD_NUM_SLOTS(md) = slots;
> @@ -111,6 +113,9 @@ efi_status_t efi_random_alloc(unsigned long size,
>                 efi_physical_addr_t target;
>                 unsigned long pages;
>
> +               if (!(md->type & (EFI_CONVENTIONAL_MEMORY || EFI_PERSISTENT_MEMORY)))
> +                       continue;
> +
>                 if (total_mirrored_slots > 0 &&
>                     !(md->attribute & EFI_MEMORY_MORE_RELIABLE))
>                         continue;
> --
> 2.41.0
>



More information about the kexec mailing list