<!-- Some styling for better description lists --><style type='text/css'>dt { font-weight: bold;float: left;display:inline;margin-right: 1em} dd { display:block; margin-left: 2em}</style>

   mchehab: right now, Jenkis build will be very noisy:
   <br> https://lore.kernel.org/linux-media/20200903064113.28864-1-jenkins@linuxtv.org/T/#u
   <br> (that's because I added some hacks there to suppress atomisp sparse/smatch noise)
   <br> I'll exclude the vivid patches, as Hans won't be able to review
   <br> I'll also drop a patch that is a duplicated version of another one at fixes
   <br> hi all
   sailus: Hyvää huomenta!
   syoung: hi all
   mchehab: anything for today?
   <br> <u>syoung</u>: there's a pending PR from you...
   <br> my plan is to merge it, but I'd like to fix first some atomisp warnings when W=1
   <br> in order to be able to test your PR with Jenkins
   <br> I'm working on such patchset right now
   <br> there aren't many - most are trivial to solve.... scratching my head to see a way to get rid of this one:
   <br> drivers/staging/media/atomisp//pci/atomisp_compat_css20.c: In function ‘atomisp_css2_dbg_print’:
   <br> drivers/staging/media/atomisp//pci/atomisp_compat_css20.c:164:2: warning: function ‘atomisp_css2_dbg_print’ might be a candidate for ‘gnu_printf’ format attribute [-Wsuggest-attribute=form&gt;
   <br> 164 |  vprintk(fmt, args);
   <br> |  ^~~~~~~
   <br> &amp; related friends
   <br> one of the bad things of gcc new warning system is that it produces multiple lines
   syoung: <u>mchehab</u>: sure
   mchehab: so, suppressing such warnings via a regex-matching file is not trivial
   sailus: Yeah, it seems harder indeed.
   <br> I've been just picking the usual sensor and some other patches, nothing unusual here.
   mchehab: I was able to cleanup a few of those gnu_printf warnings, but there are still some that I can't solve...
   <br> also, the best would be to return to old gcc behavior of not printing 3 lines for each warning
   <br> but I don't know how to disable this new "hints" part of warnings
   <br> well, I could teach the script to guess (for instance, discarding lines that start with spaces), but seems riskier to lose something
   <br> it sounds that this would do the trick: -fno-diagnostics-show-caret
   <br> need to check how to pass it though Kernel's makefile
   <br> make CCOPTIONS="-fno-diagnostics-show-caret"
   <br> doesn't work
   <br> KBUILD_CFLAGS=
   <br> doesn't work
   <br> (neither CFLAGS/CCFLAGS)
   <br> anyway, anything for today?
   <br> make CC="gcc -fno-diagnostics-show-caret"
   <br> this worked
   <br> drivers/staging/media/atomisp//pci/atomisp_compat_css20.c: In function ‘atomisp_css2_dbg_ftrace_print’:
   <br> drivers/staging/media/atomisp//pci/atomisp_compat_css20.c:164:2: warning: function ‘atomisp_css2_dbg_ftrace_print’ might be a candidate for ‘gnu_printf’ format attribute [-Wsuggest-attribute=format]
   <br> drivers/staging/media/atomisp//pci/atomisp_compat_css20.c:164:2: warning: function ‘atomisp_css2_dbg_ftrace_print’ might be a candidate for ‘gnu_printf’ format attribute [-Wsuggest-attribute=format]
   <br> drivers/staging/media/atomisp//pci/atomisp_compat_css20.c: In function ‘__set_css_print_env’:
   <br> drivers/staging/media/atomisp//pci/atomisp_compat_css20.c:860:50: warning: assignment left-hand side might be a candidate for a format attribute [-Wsuggest-attribute=format]
   <br> drivers/staging/media/atomisp//pci/atomisp_compat_css20.c: In function ‘atomisp_css_load_firmware’:
   <br> drivers/staging/media/atomisp//pci/atomisp_compat_css20.c:894:49: warning: assignment left-hand side might be a candidate for a format attribute [-Wsuggest-attribute=format]
   <br> anything else for today?
   <br> (sorry if I repeated the message... my Internet connection has dropped)
   <br> ok, all warnings got fixed
   <br> we're now free again of W=1 warnings
   <br> there are still 3 smatch warnings... all related to big functions that causes it to take a long time to process, hitting a timeout at the toolset
   <br> (two at vivid, one at em28xx)
   <br> I also changed the Jenkins script to be more sparse at the error output, with CC="gcc -fno-diagnostics-show-caret"