On Sat, Jun 8, 2019 at 9:35 PM Matt Coffin <mcoffin13@xxxxxxxxx> wrote: > > [Why] The why and how are very important, but please loose the annotations. > Currently, if metacopy is enabled via a module param/kernel config, then > the setting for redirect_dir is overriden, unless passed as a mount > option specifically. Despite this overriding, the redirect_dir option is > not printed when displaying the mount options (see /proc/mounts), so it > is hard for a user to know that it has happened. > > [How] > A bit of code that does the overriding based on the mount options was > setting redirect_dir, but not updating the redirect_mode string. By > adding that update, the information is now printed (and the config > struct has the correct information in all places). You need to Sign-off your patch. Please run ./scripts/checkpatch.pl before posting It will tell you that and so will: https://github.com/torvalds/linux/blob/master/Documentation/process/submitting-patches.rst > --- > fs/overlayfs/super.c | 3 +++ > 1 file changed, 3 insertions(+) > > diff --git a/fs/overlayfs/super.c b/fs/overlayfs/super.c > index 5ec4fc2f5d7e..46fab4660bfe 100644 > --- a/fs/overlayfs/super.c > +++ b/fs/overlayfs/super.c > @@ -598,6 +598,9 @@ static int ovl_parse_opt(char *opt, struct ovl_config *config) > config->metacopy = false; > } else { > /* Automatically enable redirect otherwise. */ > + pr_info("overlayfs: enabling redirect_dir due to implicit metacopy=%d\n", ...metacopy=on\n" We know the value and better display to user in mount option format. > + config->metacopy); > + config->redirect_mode = "on"; I'm afraid that string is allocated: kfree(config->redirect_mode); config->redirect_mode = kstrdup("on", GFP_KERNEL); if (!config->redirect_mode) return -ENOMEM; Better wrap that up in a helper ovl_set_redirect_mode("on"). Thanks, Amir.
