Re: [PATCH v2] arm64: mm: Move KPTI helpers to mmu.c

[Will Deacon]

Hey Kevin,

On Fri, Sep 12, 2025 at 08:39:08AM +0100, Kevin Brodsky wrote:
> create_kpti_ng_temp_pgd() is currently defined (as an alias) in
> mmu.c without matching declaration in a header; instead cpufeature.c
> makes its own declaration. This is clearly not pretty, and as commit
> ceca927c86e6 ("arm64: mm: Fix CFI failure due to kpti_ng_pgd_alloc
> function signature") showed, it also makes it very easy for the
> prototypes to go out of sync.
> 
> All this would be much simpler if kpti_install_ng_mappings() and
> associated functions lived in mmu.c, where they logically belong.
> This is what this patch does:
> - Move kpti_install_ng_mappings() and associated functions from
>   cpufeature.c to mmu.c, add a declaration to <asm/mmu.h>
> - Remove create_kpti_ng_temp_pgd() and just call
>   __create_pgd_mapping_locked() directly instead
> - Mark all these functions __init
> - Move __initdata after kpti_ng_temp_alloc (as suggested by
>   checkpatch)
> 
> Reviewed-by: Ryan Roberts <ryan.roberts@xxxxxxx>
> Reviewed-by: Ard Biesheuvel <ardb@xxxxxxxxxx>
> Reviewed-by: Anshuman Khandual <anshuman.khandual@xxxxxxx>
> Signed-off-by: Kevin Brodsky <kevin.brodsky@xxxxxxx>
> ---
> v1..v2:
> * Removed create_kpti_ng_temp_pgd() instead of making it a wrapper
>   [Ryan's suggestion]
> * Added Reviewed-by's.
> ---
> Cc: Anshuman Khandual <anshuman.khandual@xxxxxxx>
> Cc: Ard Biesheuvel <ardb@xxxxxxxxxx>
> Cc: Catalin Marinas <catalin.marinas@xxxxxxx>
> Cc: Kees Cook <kees@xxxxxxxxxx>
> Cc: Mark Rutland <mark.rutland@xxxxxxx>
> Cc: Ryan Roberts <ryan.roberts@xxxxxxx>
> Cc: Suzuki K Poulose <suzuki.poulose@xxxxxxx>
> Cc: Will Deacon <will@xxxxxxxxxx>
> Cc: Yeoreum Yun <yeoreum.yun@xxxxxxx>
> ---
>  arch/arm64/include/asm/mmu.h   |  6 +++
>  arch/arm64/kernel/cpufeature.c | 97 ----------------------------------
>  arch/arm64/mm/mmu.c            | 96 +++++++++++++++++++++++++++++----
>  3 files changed, 93 insertions(+), 106 deletions(-)

Thanks, this looks fine to me. However, it conflicts with ceca927c86e6
("arm64: mm: Fix CFI failure due to kpti_ng_pgd_alloc function
signature") which landed after -rc1 so I'll either queue this late
(after merging in for-next/fixes to for-next/core) or we can defer it to
-rc1.

If we get to -rc2 and it's not in Linus' tree, then please repost
because it means I forgot about it :)

Will