| Message ID | 20220310132350.78729-1-sean@mess.org (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | media: lirc: suppress false positive smatch warning | expand |
On Thu, Mar 10, 2022 at 01:23:50PM +0000, Sean Young wrote: > The latest smatch says: > > drivers/media/rc/lirc_dev.c:632 lirc_read_mode2() error: uninitialized symbol 'copied'. > drivers/media/rc/lirc_dev.c:671 lirc_read_scancode() error: uninitialized symbol 'copied'. > > This is a false positive since in all reaching code paths, copied will > be set. Work around this by providing a zero initializer for copied. > > Cc: Dan Carpenter <dan.carpenter@oracle.com> > Cc: smatch@vger.kernel.org > Signed-off-by: Sean Young <sean@mess.org> I'm sorry for this. I can't reproduce this on my system. I just did a push for something unrelated and I don't know what the previous HEAD was... regards, dan carpenter
On Thu, Mar 10, 2022 at 05:42:30PM +0300, Dan Carpenter wrote: > On Thu, Mar 10, 2022 at 01:23:50PM +0000, Sean Young wrote: > > The latest smatch says: > > > > drivers/media/rc/lirc_dev.c:632 lirc_read_mode2() error: uninitialized symbol 'copied'. > > drivers/media/rc/lirc_dev.c:671 lirc_read_scancode() error: uninitialized symbol 'copied'. > > > > This is a false positive since in all reaching code paths, copied will > > be set. Work around this by providing a zero initializer for copied. > > > > Cc: Dan Carpenter <dan.carpenter@oracle.com> > > Cc: smatch@vger.kernel.org > > Signed-off-by: Sean Young <sean@mess.org> > > I'm sorry for this. I can't reproduce this on my system. I just did a > push for something unrelated and I don't know what the previous HEAD > was... You're right, with the latest HEAD the warning goes away. Great, let's drop this patch then. Thanks for a great tool Sean
diff --git a/drivers/media/rc/lirc_dev.c b/drivers/media/rc/lirc_dev.c index 765375bda0c6..efa09beae6a7 100644 --- a/drivers/media/rc/lirc_dev.c +++ b/drivers/media/rc/lirc_dev.c @@ -601,7 +601,7 @@ static ssize_t lirc_read_mode2(struct file *file, char __user *buffer, { struct lirc_fh *fh = file->private_data; struct rc_dev *rcdev = fh->rc; - unsigned int copied; + unsigned int copied = 0; int ret; if (length < sizeof(unsigned int) || length % sizeof(unsigned int)) @@ -639,7 +639,7 @@ static ssize_t lirc_read_scancode(struct file *file, char __user *buffer, { struct lirc_fh *fh = file->private_data; struct rc_dev *rcdev = fh->rc; - unsigned int copied; + unsigned int copied = 0; int ret; if (length < sizeof(struct lirc_scancode) ||
The latest smatch says: drivers/media/rc/lirc_dev.c:632 lirc_read_mode2() error: uninitialized symbol 'copied'. drivers/media/rc/lirc_dev.c:671 lirc_read_scancode() error: uninitialized symbol 'copied'. This is a false positive since in all reaching code paths, copied will be set. Work around this by providing a zero initializer for copied. Cc: Dan Carpenter <dan.carpenter@oracle.com> Cc: smatch@vger.kernel.org Signed-off-by: Sean Young <sean@mess.org> --- drivers/media/rc/lirc_dev.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)