On Thu, Dec 7, 2017 at 1:05 PM, Greg Kroah-Hartman gregkh@linuxfoundation.org wrote:
On Thu, Dec 07, 2017 at 11:22:50AM +0100, Jack Wang wrote:
From: "Kirill A. Shutemov" kirill.shutemov@linux.intel.com
commit 0a85e51d37645e9ce57e5e1a30859e07810ed07c upstream.
Patch series "thp: fix few MADV_DONTNEED races"
For MADV_DONTNEED to work properly with huge pages, it's critical to not clear pmd intermittently unless you hold down_write(mmap_sem).
Otherwise MADV_DONTNEED can miss the THP which can lead to userspace breakage.
See example of such race in commit message of patch 2/4.
All these races are found by code inspection. I haven't seen them triggered. I don't think it's worth to apply them to stable@.
This patch (of 4):
What about all 4 of these?
The 4 patches were from here: http://lkml.iu.edu/hypermail/linux/kernel/1703.0/01274.html
thp: reduce indentation level in change_huge_pmd() thp: fix MADV_DONTNEED vs. numa balancing race thp: fix MADV_DONTNEED vs. MADV_FREE race (the function exists only since 4.5+ ) thp: fix MADV_DONTNEED vs clear soft dirty race (this one already included since 4.4.63)
And should this also go into 4.9? I don't want to include fixes into 4.4-stable without them also being in 4.9-stable, as that would be a regression for people upgrading.
In 4.9, the last 2 were already applied, I will try to backport these 3 commits also to 4.9 stable.
thanks,
greg k-h
Thanks,