Re: [PATCH 1/1] x86/mm/pae: Align up pteval_t, pmdval_t and pudval_t to avoid split locks

[Dave Hansen <dave.hansen@intel.com>]

On 4/1/24 09:57, Javier Pello wrote:
> -typedef u64	pteval_t;
> -typedef u64	pmdval_t;
> -typedef u64	pudval_t;
> -typedef u64	p4dval_t;
> -typedef u64	pgdval_t;
> -typedef u64	pgprotval_t;
> +/*
> + * Variables of these types are subject to atomic compare-and-exchange
> + * operations, so they have to be properly aligned to avoid split locks.
> + */
> +typedef u64	pteval_t	__aligned(8);
> +typedef u64	pmdval_t	__aligned(8);
> +typedef u64	pudval_t	__aligned(8);
> +typedef u64	p4dval_t	__aligned(8);
> +typedef u64	pgdval_t	__aligned(8);
> +typedef u64	pgprotval_t	__aligned(8);

First of all, how is it that you're running a PAE kernel on new, 64-bit
hardware?  That's rather odd.

The case that you're hitting is actually an on-stack pmd_t.  The fun
part is that it's not shared and doesn't even _need_ atomics.  I think
it's just using pmd_populate() because it's convenient.

I'd honestly much rather just disable split lock support in 32-bit
builds than mess with this stuff.  You really shouldn't be running
32-but kernels on this hardware.