[PATCH 0/5] riscv: Use READ_ONCE()/WRITE_ONCE() for pte accesses

Palmer Dabbelt palmer at dabbelt.com
Thu Dec 7 07:34:55 PST 2023


On Mon, 02 Oct 2023 08:10:26 PDT (-0700), alexghiti at rivosinc.com wrote:
> This series is a follow-up for riscv of a recent series from Ryan [1] which
> converts all direct dereferences of pte_t into a ptet_get() access.
>
> The goal here for riscv is to use READ_ONCE()/WRITE_ONCE() for all page
> table entries accesses to avoid any compiler transformation when the
> hardware can concurrently modify the page tables entries (A/D bits for
> example).
>
> I went a bit further and added pud/p4d/pgd_get() helpers as such concurrent
> modifications can happen too at those levels.
>
> [1] https://lore.kernel.org/all/20230612151545.3317766-1-ryan.roberts@arm.com/
>
> Alexandre Ghiti (5):
>   riscv: Use WRITE_ONCE() when setting page table entries
>   mm: Introduce pudp/p4dp/pgdp_get() functions
>   riscv: mm: Only compile pgtable.c if MMU
>   riscv: Suffix all page table entry pointers with 'p'
>   riscv: Use accessors to page table entries instead of direct
>     dereference
>
>  arch/riscv/include/asm/kfence.h     |   6 +-
>  arch/riscv/include/asm/kvm_host.h   |   2 +-
>  arch/riscv/include/asm/pgalloc.h    |  86 ++++++++++----------
>  arch/riscv/include/asm/pgtable-64.h |  26 +++---
>  arch/riscv/include/asm/pgtable.h    |  33 ++------
>  arch/riscv/kernel/efi.c             |   2 +-
>  arch/riscv/kvm/mmu.c                |  44 +++++-----
>  arch/riscv/mm/Makefile              |   3 +-
>  arch/riscv/mm/fault.c               |  38 ++++-----
>  arch/riscv/mm/hugetlbpage.c         |  80 +++++++++----------
>  arch/riscv/mm/init.c                |  30 +++----
>  arch/riscv/mm/kasan_init.c          | 119 ++++++++++++++--------------
>  arch/riscv/mm/pageattr.c            |  74 +++++++++--------
>  arch/riscv/mm/pgtable.c             |  71 +++++++++++------
>  include/linux/pgtable.h             |  21 +++++
>  15 files changed, 334 insertions(+), 301 deletions(-)

This has some build failures, I was just talking to Alex and he's going 
to fix them.



More information about the kvm-riscv mailing list