Re: drivers/leds/trigger/ledtrig-netdev.c:120:34: warning: array subscript 17 is above array bounds of 'char[16]'

[Christian Marangi <ansuelsmth@gmail.com>]

On Wed, Sep 27, 2023 at 10:44:22PM +0800, Philip Li wrote:
> On Tue, Sep 26, 2023 at 01:22:35AM +0200, Christian Marangi wrote:
> > On Tue, Sep 19, 2023 at 08:21:03PM +0800, kernel test robot wrote:
> > > tree:   https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git master
> > > head:   2cf0f715623872823a72e451243bbf555d10d032
> > > commit: 28a6a2ef18ad840a390d519840c303b03040961c leds: trigger: netdev: refactor code setting device name
> > > date:   4 months ago
> > > config: loongarch-allmodconfig (https://download.01.org/0day-ci/archive/20230919/202309192035.GTJEEbem-lkp@intel.com/config)
> > > compiler: loongarch64-linux-gcc (GCC) 13.2.0
> > > reproduce (this is a W=1 build): (https://download.01.org/0day-ci/archive/20230919/202309192035.GTJEEbem-lkp@intel.com/reproduce)
> > > 
> > > If you fix the issue in a separate patch/commit (i.e. not just a new version of
> > > the same patch/commit), kindly add following tags
> > > | Reported-by: kernel test robot <lkp@intel.com>
> > > | Closes: https://lore.kernel.org/oe-kbuild-all/202309192035.GTJEEbem-lkp@intel.com/
> > > 
> > > All warnings (new ones prefixed by >>):
> > > 
> > >    drivers/leds/trigger/ledtrig-netdev.c: In function 'set_device_name':
> > > >> drivers/leds/trigger/ledtrig-netdev.c:120:34: warning: array subscript 17 is above array bounds of 'char[16]' [-Warray-bounds=]
> > >      120 |         trigger_data->device_name[size] = 0;
> > >          |         ~~~~~~~~~~~~~~~~~~~~~~~~~^~~~~~
> > >    drivers/leds/trigger/ledtrig-netdev.c:48:14: note: while referencing 'device_name'
> > >       48 |         char device_name[IFNAMSIZ];
> > >          |              ^~~~~~~~~~~
> > > >> drivers/leds/trigger/ledtrig-netdev.c:120:34: warning: array subscript 17 is above array bounds of 'char[16]' [-Warray-bounds=]
> > >      120 |         trigger_data->device_name[size] = 0;
> > >          |         ~~~~~~~~~~~~~~~~~~~~~~~~~^~~~~~
> > >    drivers/leds/trigger/ledtrig-netdev.c:48:14: note: while referencing 'device_name'
> > >       48 |         char device_name[IFNAMSIZ];
> > >          |              ^~~~~~~~~~~
> > > 
> > > 
> > > vim +120 drivers/leds/trigger/ledtrig-netdev.c
> > > 
> > > 06f502f57d0d77 Ben Whitten       2017-12-10  106  
> > > 28a6a2ef18ad84 Andrew Lunn       2023-05-29  107  static int set_device_name(struct led_netdev_data *trigger_data,
> > > 28a6a2ef18ad84 Andrew Lunn       2023-05-29  108  			   const char *name, size_t size)
> > > 06f502f57d0d77 Ben Whitten       2017-12-10  109  {
> > > 06f502f57d0d77 Ben Whitten       2017-12-10  110  	cancel_delayed_work_sync(&trigger_data->work);
> > > 06f502f57d0d77 Ben Whitten       2017-12-10  111  
> > > d1b9e1391ab2dc Christian Marangi 2023-04-19  112  	mutex_lock(&trigger_data->lock);
> > > 06f502f57d0d77 Ben Whitten       2017-12-10  113  
> > > 06f502f57d0d77 Ben Whitten       2017-12-10  114  	if (trigger_data->net_dev) {
> > > 06f502f57d0d77 Ben Whitten       2017-12-10  115  		dev_put(trigger_data->net_dev);
> > > 06f502f57d0d77 Ben Whitten       2017-12-10  116  		trigger_data->net_dev = NULL;
> > > 06f502f57d0d77 Ben Whitten       2017-12-10  117  	}
> > > 06f502f57d0d77 Ben Whitten       2017-12-10  118  
> > > 28a6a2ef18ad84 Andrew Lunn       2023-05-29  119  	memcpy(trigger_data->device_name, name, size);
> > > 909346433064b8 Rasmus Villemoes  2019-03-14 @120  	trigger_data->device_name[size] = 0;
> > > 06f502f57d0d77 Ben Whitten       2017-12-10  121  	if (size > 0 && trigger_data->device_name[size - 1] == '\n')
> > > 06f502f57d0d77 Ben Whitten       2017-12-10  122  		trigger_data->device_name[size - 1] = 0;
> > > 06f502f57d0d77 Ben Whitten       2017-12-10  123  
> > > 06f502f57d0d77 Ben Whitten       2017-12-10  124  	if (trigger_data->device_name[0] != 0)
> > > 06f502f57d0d77 Ben Whitten       2017-12-10  125  		trigger_data->net_dev =
> > > 06f502f57d0d77 Ben Whitten       2017-12-10  126  		    dev_get_by_name(&init_net, trigger_data->device_name);
> > > 06f502f57d0d77 Ben Whitten       2017-12-10  127  
> > > e2f24cb1b5daf9 Christian Marangi 2023-04-19  128  	trigger_data->carrier_link_up = false;
> > > 06f502f57d0d77 Ben Whitten       2017-12-10  129  	if (trigger_data->net_dev != NULL)
> > > e2f24cb1b5daf9 Christian Marangi 2023-04-19  130  		trigger_data->carrier_link_up = netif_carrier_ok(trigger_data->net_dev);
> > > 06f502f57d0d77 Ben Whitten       2017-12-10  131  
> > > 06f502f57d0d77 Ben Whitten       2017-12-10  132  	trigger_data->last_activity = 0;
> > > 06f502f57d0d77 Ben Whitten       2017-12-10  133  
> > > 06f502f57d0d77 Ben Whitten       2017-12-10  134  	set_baseline_state(trigger_data);
> > > d1b9e1391ab2dc Christian Marangi 2023-04-19  135  	mutex_unlock(&trigger_data->lock);
> > > 06f502f57d0d77 Ben Whitten       2017-12-10  136  
> > > 28a6a2ef18ad84 Andrew Lunn       2023-05-29  137  	return 0;
> > > 28a6a2ef18ad84 Andrew Lunn       2023-05-29  138  }
> > > 28a6a2ef18ad84 Andrew Lunn       2023-05-29  139  
> > > 
> > > :::::: The code at line 120 was first introduced by commit
> > > :::::: 909346433064b8d840dc82af26161926b8d37558 leds: trigger: netdev: use memcpy in device_name_store
> > > 
> > > :::::: TO: Rasmus Villemoes <linux@rasmusvillemoes.dk>
> > > :::::: CC: Jacek Anaszewski <jacek.anaszewski@gmail.com>
> > > 
> > > -- 
> > > 0-DAY CI Kernel Test Service
> > > https://github.com/intel/lkp-tests/wiki
> > 
> > Can't repro so I guess this was a false-positive that got fixed? (I used
> > the repro commands and doesn't print any error)
> 
> Hi Ansuel, the issue requires to build with -Warray-bounds flag. We will fix
> the reproduce steps in earliest time. Sorry for the inconvenience.
> 

Hi again, happy to help you further.

By adding the flag I still can't repro.
I'm using this command.
COMPILER_INSTALL_PATH=$HOME/0day COMPILER=gcc-13.2.0 ~/bin/make.cross W=1 O=build_dir ARCH=loongarch SHELL=/bin/bash drivers/leds/trigger/  -Warray-bounds

Also on top of that I can see that only ledtrig-audio is compiled.
From what I can see using olddefconfig won't actually use the config
used by the bot.

By manually adding the config of the bot to the build_dir ledtrig-netdev
is correctly compiled and still no warning about this.

I'm here to any test needed to improve the repro instructions.

-- 
	Ansuel