Re: [PATCH 1/3] hugetlbfs: extend hugetlb_vma_lock to private VMAs

[Mike Kravetz]

On 09/25/23 16:28, riel@xxxxxxxxxxx wrote:
> From: Rik van Riel <riel@xxxxxxxxxxx>
> 
> Extend the locking scheme used to protect shared hugetlb mappings
> from truncate vs page fault races, in order to protect private
> hugetlb mappings (with resv_map) against MADV_DONTNEED.

The only time an anon hugetlb mapping would not have a resv_map would
be in the case where MAP_NORESERVE was specified.  In this case the user
should be prepared for allocation failures.  I am good with this approach
not adding locking for the MAP_NORESERVE case.

> Add a read-write semaphore to the resv_map data structure, and
> use that from the hugetlb_vma_(un)lock_* functions, in preparation
> for closing the race between MADV_DONTNEED and page faults.
> 
> Signed-off-by: Rik van Riel <riel@xxxxxxxxxxx>
> ---
>  include/linux/hugetlb.h |  6 ++++++
>  mm/hugetlb.c            | 41 +++++++++++++++++++++++++++++++++++++----
>  2 files changed, 43 insertions(+), 4 deletions(-)

Reviewed-by: Mike Kravetz <mike.kravetz@xxxxxxxxxx>

-- 
Mike Kravetz