> On Jan 18, 2022, at 5:00 PM, Bruce Fields <bfields@xxxxxxxxxxxx> wrote:
>
> From: "J. Bruce Fields" <bfields@xxxxxxxxxx>
>
> I thought I was iterating over the array when actually the iteration is
> over the values contained in the array?
>
> Ugh, keep it simple.
>
> Symptoms were a null deference in vfs_lock_file() when an NFSv3 client
> that previously held a lock came back up and sent a notify.
>
> Reported-by: Jonathan Woithe <jwoithe@xxxxxxxxxx>
> Fixes: 7f024fcd5c97 ("Keep read and write fds with each nlm_file")
> Signed-off-by: J. Bruce Fields <bfields@xxxxxxxxxx>
> ---
> fs/lockd/svcsubs.c | 17 +++++++++--------
> 1 file changed, 9 insertions(+), 8 deletions(-)
>
> diff --git a/fs/lockd/svcsubs.c b/fs/lockd/svcsubs.c
> index cb3a7512c33e..54c2e42130ca 100644
> --- a/fs/lockd/svcsubs.c
> +++ b/fs/lockd/svcsubs.c
> @@ -179,19 +179,20 @@ nlm_delete_file(struct nlm_file *file)
> static int nlm_unlock_files(struct nlm_file *file)
> {
> struct file_lock lock;
> - struct file *f;
>
> lock.fl_type = F_UNLCK;
> lock.fl_start = 0;
> lock.fl_end = OFFSET_MAX;
> - for (f = file->f_file[0]; f <= file->f_file[1]; f++) {
> - if (f && vfs_lock_file(f, F_SETLK, &lock, NULL) < 0) {
> - pr_warn("lockd: unlock failure in %s:%d\n",
> - __FILE__, __LINE__);
> - return 1;
> - }
> - }
> + if (file->f_file[O_RDONLY] &&
> + vfs_lock_file(file->f_file[O_RDONLY], F_SETLK, &lock, NULL))
> + goto out_err;
> + if (file->f_file[O_WRONLY] &&
> + vfs_lock_file(file->f_file[O_WRONLY], F_SETLK, &lock, NULL))
> + goto out_err;
> return 0;
> +out_err:
> + pr_warn("lockd: unlock failure in %s:%d\n", __FILE__, __LINE__);
> + return 1;
> }
>
> /*
> --
> 2.34.1
>
Hi Bruce, thanks for the fixes. They've passed my basic smoke tests.
I've applied both patches for the very next nfsd PR. See:
git://git.kernel.org/pub/scm/linux/kernel/git/cel/linux.git for-next
--
Chuck Lever
