On Wed, Jul 25, 2018 at 12:11:26AM +0900, Tetsuo Handa wrote:
> On 2018/07/19 7:58, Omar Sandoval wrote:
> > From: Omar Sandoval <osandov@xxxxxx>
> >
> > Now that we're using an rwsem, we can hold it during the entirety of
> > read_kcore() and have a common return path. This is preparation for the
> > next change.
> >
> > Signed-off-by: Omar Sandoval <osandov@xxxxxx>
> > ---
> > fs/proc/kcore.c | 70 ++++++++++++++++++++++++++++---------------------
> > 1 file changed, 40 insertions(+), 30 deletions(-)
> >
> > diff --git a/fs/proc/kcore.c b/fs/proc/kcore.c
> > index 95aa988c5b5d..e317ac890871 100644
> > --- a/fs/proc/kcore.c
> > +++ b/fs/proc/kcore.c
> > @@ -440,19 +440,18 @@ static ssize_t
> > read_kcore(struct file *file, char __user *buffer, size_t buflen, loff_t *fpos)
> > {
> > char *buf = file->private_data;
> > - ssize_t acc = 0;
> > size_t size, tsz;
> > size_t elf_buflen;
> > int nphdr;
> > unsigned long start;
> > + size_t orig_buflen = buflen;
> > + int ret = 0;
> >
> > down_read(&kclist_lock);
>
> (...snipped...)
>
> > +out:
> > + up_write(&kclist_lock);
>
> Oops. This needs to be up_read().
Yeah, thanks, I'll fix this and rerun my tests with lockdep.
