Re: [PATCH stable] mm/khugepaged: write all dirty file folios when collapsing

[Pedro Falcato]

On Fri, Jul 03, 2026 at 01:11:29PM +0800, Lance Yang wrote:
> 
> On Thu, Jul 02, 2026 at 05:54:09PM +0100, Pedro Falcato wrote:
> >As-is, khugepaged and writable-file opening exclude each other. A file
> >cannot be open writeable and have THPs (because the filesystem is not aware
> >of them). khugepaged will never collapse file pages for files that are
> >opened writeable. On an open(O_RDWR/O_WRONLY), the page cache for that
> >particular file is dropped. This is fine because nothing could've been
> >dirtied.
> >
> >However, there is an edge-case: collapse_file() might not be able to
> >coexist with concurrent writers, but it can coexist with dirty folios
> >(from previous writers). Therefore, the following can happen:
> >
> >open(file, O_RDWR)
> >write(file)
> >close(file)
> >madvise(file_mapping, MADV_COLLAPSE, some non-dirty range)
> >open(file, O_RDWR)
> > nr_thps > 0
> >  truncate_inode_pages()
> >    /* THPs are cleared out, but so are the dirty folios */
> >
> >When this edge-case happens, there is data loss, as the dirty folios are
> >fully discarded.
> 
> Well spotted, thanks!

Well, Gregg deserves a lot of the credit :)

> 
> >
> >Fix it by fully writing back the page cache (and waiting) when collapsing
> >file THPs. Doing so provides the guarantee that no dirty folio will be
> >observed while there are active THPs. To fully ensure this is safe, the
> >invalidate_lock needs to be held while doing the writeout, so that
> >do_dentry_open()'s page cache truncation excludes this write-and-wait.
> >
> >Cc: stable@xxxxxxxxxxxxxxx
> >Cc: Alexander Viro <viro@xxxxxxxxxxxxxxxxxx>
> >Cc: Christian Brauner <brauner@xxxxxxxxxx>
> >Cc: Jan Kara <jack@xxxxxxx>
> >Cc: Matthew Wilcox <willy@xxxxxxxxxxxxx>
> >Cc: Song Liu <song@xxxxxxxxxx>
> >Cc: Eric Hagberg <ehagberg@xxxxxxxxxxxxxx>
> >Cc: Zi Yan <ziy@xxxxxxxxxx>
> >Fixes: 99cb0dbd47a1 ("mm,thp: add read-only THP support for (non-shmem) FS")
> >Reported-by: Gregg Leventhal <gleventhal@xxxxxxxxxxxxxx>
> >Closes: https://lore.kernel.org/linux-mm/CAFN_u7H_0ECF3jixP=T=U7AH5=Q3wQNvJMo8an3VqUDMerQfUw@xxxxxxxxxxxxxx/
> >Tested-by: Zi Yan <ziy@xxxxxxxxxx>
> >Signed-off-by: Pedro Falcato <pfalcato@xxxxxxx>
> >---
> 
> Tested on v7.1.2. I no longer see the data loss with this patch applied.
> 
> Tested-by: Lance Yang <lance.yang@xxxxxxxxx>

Thanks!

-- 
Pedro