Re: [kbuild] drivers/firmware/arm_scmi/clock.c:242:40: warning: Variable 'msg' is not assigned a value. [unassignedVariable]

[Cristian Marussi <cristian.marussi@arm.com>]

On Mon, May 30, 2022 at 10:04:55AM +0100, Cristian Marussi wrote:
> On Fri, May 27, 2022 at 11:57:29AM +0300, Dan Carpenter wrote:
> > tree:   https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git  master
> > head:   7e284070abe53d448517b80493863595af4ab5f0
> > commit: 7bc7caafe6b1e5b882255a42bc1bf112fa87b69b firmware: arm_scmi: Use common iterators in the clock protocol
> > compiler: arm-linux-gnueabi-gcc (GCC) 11.3.0
> > reproduce (cppcheck warning):
> >         # apt-get install cppcheck
> >         git checkout 7bc7caafe6b1e5b882255a42bc1bf112fa87b69b
> >         cppcheck --quiet --enable=style,performance,portability --template=gcc FILE
> > 
> > If you fix the issue, kindly add following tag where applicable
> > Reported-by: kernel test robot <lkp@intel.com>
> > 
> 
> Hi Dan,
> 
> thanks for the report.
> 
> > cppcheck possible warnings: (new ones prefixed by >>, may not real problems)
> > 
> > >> drivers/firmware/arm_scmi/clock.c:242:40: warning: Variable 'msg' is not assigned a value. [unassignedVariable]
> >     struct scmi_msg_clock_describe_rates *msg;
> >                                           ^
> > 
> > vim +/msg +242 drivers/firmware/arm_scmi/clock.c
> > 
> > 7bc7caafe6b1e5 Cristian Marussi 2022-03-30  235  static int
> > 7bc7caafe6b1e5 Cristian Marussi 2022-03-30  236  scmi_clock_describe_rates_get(const struct scmi_protocol_handle *ph, u32 clk_id,
> > 7bc7caafe6b1e5 Cristian Marussi 2022-03-30  237  			      struct scmi_clock_info *clk)
> > 7bc7caafe6b1e5 Cristian Marussi 2022-03-30  238  {
> > 7bc7caafe6b1e5 Cristian Marussi 2022-03-30  239  	int ret;
> > 5f6c6430e904d2 Sudeep Holla     2017-06-06  240  
> > 
> > Please delete the blank line.
> 
> I'll do.
> 
> > 
> > 7bc7caafe6b1e5 Cristian Marussi 2022-03-30  241  	void *iter;
> > 7bc7caafe6b1e5 Cristian Marussi 2022-03-30 @242  	struct scmi_msg_clock_describe_rates *msg;
> > 
> > I was so surprised that GCC doesn't warn about this but "msg" is only
> > used to calculate the sizeof(*msg).
> > 
> 
> Probably a leftover from a previous version of the series where msg was
> usaed not only for sizeof()....I'll send a fix anyway to silence the
> bot.

... and btw there are a bunch of similar other sizeof(*msg) in a few
other SCMI protocols when calling the same response_init() helper...
..even though innocuos...I'll fix all of these to avoid needless stuff
on the stack.

Thanks,
Cristian