Re: [peterz-queue:x86/core 3/3] arch/x86/kernel/static_call.c:98 __static_call_transform() error: uninitialized symbol 'op'.

[Dan Carpenter <error27@gmail.com>]

On Mon, Jan 23, 2023 at 04:23:21PM +0100, Peter Zijlstra wrote:
> On Mon, Jan 23, 2023 at 03:02:54PM +0300, Dan Carpenter wrote:
> > tree:   https://git.kernel.org/pub/scm/linux/kernel/git/peterz/queue.git x86/core
> > head:   700f4c6d925823018a2d09f92de5405065d2ffec
> > commit: 700f4c6d925823018a2d09f92de5405065d2ffec [3/3] x86/static_call: Add support for Jcc tail-calls
> > config: i386-randconfig-m021 (https://download.01.org/0day-ci/archive/20230121/202301211834.noGVyk06-lkp@intel.com/config)
> > compiler: gcc-11 (Debian 11.3.0-8) 11.3.0
> > 
> > If you fix the issue, kindly add following tag where applicable
> > | Reported-by: kernel test robot <lkp@intel.com>
> > | Reported-by: Dan Carpenter <error27@gmail.com>
> > 
> > smatch warnings:
> > arch/x86/kernel/static_call.c:98 __static_call_transform() error: uninitialized symbol 'op'.
> > 
> > vim +/op +98 arch/x86/kernel/static_call.c
> > 
> > c27c753ea6fd12 Thomas Gleixner 2022-07-12   53  static void __ref __static_call_transform(void *insn, enum insn_type type,
> > c27c753ea6fd12 Thomas Gleixner 2022-07-12   54  					  void *func, bool modinit)
> > e6d6c071f22de2 Josh Poimboeuf  2020-08-18   55  {
> > 3f2a8fc4b15de1 Peter Zijlstra  2021-01-18   56  	const void *emulate = NULL;
> > 452cddbff74b6a Peter Zijlstra  2020-08-18   57  	int size = CALL_INSN_SIZE;
> > 452cddbff74b6a Peter Zijlstra  2020-08-18   58  	const void *code;
> > 700f4c6d925823 Peter Zijlstra  2023-01-20   59  	u8 op, buf[6];
> > 700f4c6d925823 Peter Zijlstra  2023-01-20   60  
> > 700f4c6d925823 Peter Zijlstra  2023-01-20   61  	if ((type == JMP || type == RET) && (op = __is_Jcc(insn)))
> > 700f4c6d925823 Peter Zijlstra  2023-01-20   62  		type = JCC;
> > 
> > Smatch is concerned that type can start as "JCC".
> 
> It can't. What's the normal way to shut this kind of warnings up?
> 
> Something like BUG_ON(type==JCC); should squash that value I suppose,
> but it does seems a bit crude.

We could just ignore it.  These are one time only emails.

This code hasn't hit linux-next yet, so I don't have it on my system.
It may be that using the cross function DB silences the warning.

regards,
dan carpenter