mm: more ptep_get() conversion
authorRyan Roberts <ryan.roberts@arm.com>
Tue, 14 Nov 2023 15:49:45 +0000 (15:49 +0000)
committerAndrew Morton <akpm@linux-foundation.org>
Wed, 15 Nov 2023 23:30:09 +0000 (15:30 -0800)
commitafccb0804fc74ac2f6737af6a139632606cb461d
treef40e6663e27cc7be0f49900d08d08cfd7c54c397
parent5f74f820f6fc844b95f9e5e406e0a07d97510420
mm: more ptep_get() conversion

Commit c33c794828f2 ("mm: ptep_get() conversion") converted all (non-arch)
call sites to use ptep_get() instead of doing a direct dereference of the
pte.  Full rationale can be found in that commit's log.

Since then, three new call sites have snuck in, which directly dereference
the pte, so let's fix those up.

Unfortunately there is no reliable automated mechanism to catch these; I'm
relying on a combination of Coccinelle (which throws up a lot of false
positives) and some compiler magic to force a compiler error on
dereference (While this approach finds dereferences, it also yields a
non-booting kernel so can't be committed).

Link: https://lkml.kernel.org/r/20231114154945.490401-1-ryan.roberts@arm.com
Signed-off-by: Ryan Roberts <ryan.roberts@arm.com>
Cc: Matthew Wilcox (Oracle) <willy@infradead.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
mm/filemap.c
mm/ksm.c
mm/userfaultfd.c