mm/memfd: add an F_SEAL_FUTURE_WRITE seal to memfd
authorJoel Fernandes (Google) <joel@joelfernandes.org>
Tue, 5 Mar 2019 23:47:54 +0000 (15:47 -0800)
committerLinus Torvalds <torvalds@linux-foundation.org>
Wed, 6 Mar 2019 05:07:19 +0000 (21:07 -0800)
commitab3948f58ff841e51feb845720624665ef5b7ef3
tree7258ed463b4e48b91add44c6ee09217ef8611679
parent7f18825174203526a47c127c12a50f897ee0b511
mm/memfd: add an F_SEAL_FUTURE_WRITE seal to memfd

Android uses ashmem for sharing memory regions.  We are looking forward
to migrating all usecases of ashmem to memfd so that we can possibly
remove the ashmem driver in the future from staging while also
benefiting from using memfd and contributing to it.  Note staging
drivers are also not ABI and generally can be removed at anytime.

One of the main usecases Android has is the ability to create a region
and mmap it as writeable, then add protection against making any
"future" writes while keeping the existing already mmap'ed
writeable-region active.  This allows us to implement a usecase where
receivers of the shared memory buffer can get a read-only view, while
the sender continues to write to the buffer.  See CursorWindow
documentation in Android for more details:

  https://developer.android.com/reference/android/database/CursorWindow

This usecase cannot be implemented with the existing F_SEAL_WRITE seal.
To support the usecase, this patch adds a new F_SEAL_FUTURE_WRITE seal
which prevents any future mmap and write syscalls from succeeding while
keeping the existing mmap active.

A better way to do F_SEAL_FUTURE_WRITE seal was discussed [1] last week
where we don't need to modify core VFS structures to get the same
behavior of the seal.  This solves several side-effects pointed by Andy.
self-tests are provided in later patch to verify the expected semantics.

[1] https://lore.kernel.org/lkml/20181111173650.GA256781@google.com/

Thanks a lot to Andy for suggestions to improve code.

Link: http://lkml.kernel.org/r/20190112203816.85534-2-joel@joelfernandes.org
Signed-off-by: Joel Fernandes (Google) <joel@joelfernandes.org>
Acked-by: John Stultz <john.stultz@linaro.org>
Cc: Andy Lutomirski <luto@kernel.org>
Cc: Minchan Kim <minchan@kernel.org>
Cc: Jann Horn <jannh@google.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Andy Lutomirski <luto@kernel.org>
Cc: Hugh Dickins <hughd@google.com>
Cc: J. Bruce Fields <bfields@fieldses.org>
Cc: Jeff Layton <jlayton@kernel.org>
Cc: Marc-Andr Lureau <marcandre.lureau@redhat.com>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: Mike Kravetz <mike.kravetz@oracle.com>
Cc: Shuah Khan <shuah@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
fs/hugetlbfs/inode.c
include/uapi/linux/fcntl.h
mm/memfd.c
mm/shmem.c