On Wed, Dec 4, 2024 at 11:27 AM Jann Horn <jannh@xxxxxxxxxx> wrote: > > The current check_memfd_seals() is racy: Since we first do > check_memfd_seals() and then udmabuf_pin_folios() without holding any > relevant lock across both, F_SEAL_WRITE can be set in between. > This is problematic because we can end up holding pins to pages in a > write-sealed memfd. > > Fix it using the inode lock, that's probably the easiest way. > In the future, we might want to consider moving this logic into memfd, > especially if anyone else wants to use memfd_pin_folios(). I am curious, why is it not possible to have a reproducer for this issue, is it not reproducible and is theoretical? thanks, - Joel > > Reported-by: Julian Orth <ju.orth@xxxxxxxxx> > Closes: https://bugzilla.kernel.org/show_bug.cgi?id=219106 > Closes: https://lore.kernel.org/r/CAG48ez0w8HrFEZtJkfmkVKFDhE5aP7nz=obrimeTgpD+StkV9w@xxxxxxxxxxxxxx > Fixes: fbb0de795078 ("Add udmabuf misc device") > Cc: stable@xxxxxxxxxxxxxxx > Signed-off-by: Jann Horn <jannh@xxxxxxxxxx> > --- > drivers/dma-buf/udmabuf.c | 13 +++++++++---- > 1 file changed, 9 insertions(+), 4 deletions(-) > > diff --git a/drivers/dma-buf/udmabuf.c b/drivers/dma-buf/udmabuf.c > index 8ce1f074c2d32a0a9f59ff7184359e37d56548c6..c1d8c2766d6d36fc5fe1b3d73057f6e01ec6678f 100644 > --- a/drivers/dma-buf/udmabuf.c > +++ b/drivers/dma-buf/udmabuf.c > @@ -436,14 +436,19 @@ static long udmabuf_create(struct miscdevice *device, > goto err; > } > > + /* > + * Take the inode lock to protect against concurrent > + * memfd_add_seals(), which takes this lock in write mode. > + */ > + inode_lock_shared(file_inode(memfd)); > ret = check_memfd_seals(memfd); > - if (ret < 0) { > - fput(memfd); > - goto err; > - } > + if (ret) > + goto out_unlock; > > ret = udmabuf_pin_folios(ubuf, memfd, list[i].offset, > list[i].size, folios); > +out_unlock: > + inode_unlock_shared(file_inode(memfd)); > fput(memfd); > if (ret) > goto err; > > -- > 2.47.0.338.g60cca15819-goog >
