On Thu, Jan 29, 2015 at 12:27 PM, Vinícius Tinti <[email protected]> wrote:
> On Thu, Jan 29, 2015 at 3:07 AM, Greg KH <[email protected]> wrote: > > On Thu, Jan 29, 2015 at 02:16:51AM -0200, Vinícius Tinti wrote: > >> On Thu, Jan 29, 2015 at 2:08 AM, Greg KH <[email protected]> wrote: > >> > On Thu, Jan 29, 2015 at 01:48:43AM -0200, Vinícius Tinti wrote: > >> >> This is a simple patch that initializes a function with NULL to > avoid some > >> >> compiler warnings. In such cases should I proceed as a normal patch > or it is > >> >> better to send to another ML like to one for trivial patches? > >> >> > >> >> Thanks, > >> >> > >> >> Tinti > >> > > >> >> >From a391789bf44afbdbe2a7b3c76301b5ece9f72475 Mon Sep 17 00:00:00 > 2001 > >> >> From: =?UTF-8?q?Vin=C3=ADcius=20Tinti?= <[email protected]> > >> >> Date: Thu, 29 Jan 2015 01:35:34 -0200 > >> >> Subject: [PATCH] x86: LLVMLinux: Fix uninitialized function do_reloc > >> >> MIME-Version: 1.0 > >> >> Content-Type: text/plain; charset=UTF-8 > >> >> Content-Transfer-Encoding: 8bit > >> >> > >> >> Explicit initializes do_reloc function with NULL. Later the function > is > >> >> either proper initialized of an error issued. > >> >> > >> >> Signed-off-by: Vinícius Tinti <[email protected]> > >> >> --- > >> >> arch/x86/tools/relocs.c | 2 +- > >> >> 1 file changed, 1 insertion(+), 1 deletion(-) > >> >> > >> >> diff --git a/arch/x86/tools/relocs.c b/arch/x86/tools/relocs.c > >> >> index 0c2fae8..1d533f1 100644 > >> >> --- a/arch/x86/tools/relocs.c > >> >> +++ b/arch/x86/tools/relocs.c > >> >> @@ -971,7 +971,7 @@ static void emit_relocs(int as_text, int > use_real_mode) > >> >> int i; > >> >> int (*write_reloc)(uint32_t, FILE *) = write32; > >> >> int (*do_reloc)(struct section *sec, Elf_Rel *rel, Elf_Sym > *sym, > >> >> - const char *symname); > >> >> + const char *symname) = NULL; > >> > > >> > I think you need to get an updated version of the compiler as this > patch > >> > should not be needed at all. It doesn't cause a warning here for me > >> > without it. > >> > >> In fact it causes a warning on Clang which complains that: > >> > >> arch/x86/tools/relocs.c:977:6: warning: variable 'do_reloc' is used > >> uninitialized whenever 'if' condition is false > >> [-Wsometimes-uninitialized] > > > > I suggest you file a bug with clang, gcc doesn't have this problem at > > all as obviously, if you look at the code, that variable can never be > > used uninitialized. > > I can simply turn down this kind of warning. > > >> I think there is not a problem on the current code but to avoid > >> further problems I believe it is worth to initialize this function > >> with NULL. > >> What do you think? > > > > Don't paper over bugs in the compiler with kernel code changes for no > > good reason :) > > Agreed. But whenever I find a warning in GCC during the build what > should I do with it? > Can I simply send it to the main ml? > If you believe your patch is correct, and if checkpatch.pl doesn't complain about your patch, then get_maintainers.pl will tell you the people and mailing lists you should send your patch to. > > > thanks, > > > > greg k-h > > > > -- > Simplicity is the ultimate sophistication > > _______________________________________________ > Kernelnewbies mailing list > [email protected] > http://lists.kernelnewbies.org/mailman/listinfo/kernelnewbies > -- Peter
_______________________________________________ Kernelnewbies mailing list [email protected] http://lists.kernelnewbies.org/mailman/listinfo/kernelnewbies
