On 2026-04-07 at 12:52:07 -0700, Sohil Mehta wrote:
>On 4/7/2026 10:45 AM, Maciej Wieczor-Retman wrote:
>
>> diff --git a/arch/x86/kernel/process_64.c b/arch/x86/kernel/process_64.c
>> index 08e72f429870..1a0e96835bbc 100644
>> --- a/arch/x86/kernel/process_64.c
>> +++ b/arch/x86/kernel/process_64.c
>> @@ -797,7 +797,7 @@ static long prctl_map_vdso(const struct vdso_image
>> *image, unsigned long addr)
>>
>> #ifdef CONFIG_ADDRESS_MASKING
>>
>> -#define LAM_U57_BITS 6
>> +#define LAM_DEFAULT_BITS 4
>>
>
>As Dave mentioned earlier, the default wording makes it seem like this
>can be configured. But that isn't true anymore, right?
>
>How about naming this as LAM_TAG_BITS to match ARCH_GET_MAX_TAG_BITS?
Right, I'll rename it, thanks!
>> static void enable_lam_func(void *__mm)
>> {
>> @@ -814,7 +814,7 @@ static void enable_lam_func(void *__mm)
>> static void mm_enable_lam(struct mm_struct *mm)
>> {
>> mm->context.lam_cr3_mask = X86_CR3_LAM_U57;
>> - mm->context.untag_mask = ~GENMASK(62, 57);
>> + mm->context.untag_mask = ~GENMASK(57 + LAM_DEFAULT_BITS - 1, 57);
>>
>
>Also, would it be useful to calculate the LAM mask as a #define because
>it might need to be reused later or copied over to the selftest (as in
>patch 3)?
I think as it's only used during this initialization here it's not very useful
to give it a separate #define. And if it's only for selftest purposes it's
probably not worth it to export it to userspace.
--
Kind regards
Maciej Wieczór-Retman
