leandrof/linux
1
0
Fork 0
forked from Minki/linux
Code Issues Pull Requests Packages Projects Releases Wiki Activity

mm,madvise,hugetlb: fix unexpected data loss with MADV_DONTNEED on hugetlbfs

Browse Source 
A common use case for hugetlbfs is for the application to create
memory pools backed by huge pages, which then get handed over to
some malloc library (eg. jemalloc) for further management.

That malloc library may be doing MADV_DONTNEED calls on memory
that is no longer needed, expecting those calls to happen on
PAGE_SIZE boundaries.

However, currently the MADV_DONTNEED code rounds up any such
requests to HPAGE_PMD_SIZE boundaries. This leads to undesired
outcomes when jemalloc expects a 4kB MADV_DONTNEED, but 2MB of
memory get zeroed out, instead.

Use of pre-built shared libraries means that user code does not
always know the page size of every memory arena in use.

Avoid unexpected data loss with MADV_DONTNEED by rounding up
only to PAGE_SIZE (in do_madvise), and rounding down to huge
page granularity.

That way programs will only get as much memory zeroed out as
they requested.

Link: https://lkml.kernel.org/r/20221021192805.366ad573@imladris.surriel.com
Fixes: 90e7e7f5ef ("mm: enable MADV_DONTNEED for hugetlb mappings")
Signed-off-by: Rik van Riel <riel@surriel.com>
Reviewed-by: Mike Kravetz <mike.kravetz@oracle.com>
Cc: David Hildenbrand <david@redhat.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
This commit is contained in:
Rik van Riel 2022-10-21 19:28:05 -04:00 committed by Andrew Morton
parent fba4eaf931
commit 8ebe0a5eaa
1 changed files with 11 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

12
mm/madvise.c
View File

@ -813,7 +813,14 @@ static bool madvise_dontneed_free_valid_vma(struct vm_area_struct *vma,
if (start & ~huge_page_mask(hstate_vma(vma))) if (start & ~huge_page_mask(hstate_vma(vma)))
return false; return false;
*end = ALIGN(*end, huge_page_size(hstate_vma(vma))); /*
* Madvise callers expect the length to be rounded up to PAGE_SIZE
* boundaries, and may be unaware that this VMA uses huge pages.
* Avoid unexpected data loss by rounding down the number of
* huge pages freed.
*/
*end = ALIGN_DOWN(*end, huge_page_size(hstate_vma(vma)));
return true; return true;
} }
@ -828,6 +835,9 @@ static long madvise_dontneed_free(struct vm_area_struct *vma,
if (!madvise_dontneed_free_valid_vma(vma, start, &end, behavior)) if (!madvise_dontneed_free_valid_vma(vma, start, &end, behavior))
return -EINVAL; return -EINVAL;
if (start == end)
return 0;
if (!userfaultfd_remove(vma, start, end)) { if (!userfaultfd_remove(vma, start, end)) {
*prev = NULL; /* mmap_lock has been dropped, prev is stale */ *prev = NULL; /* mmap_lock has been dropped, prev is stale */