Re: [PATCH v3] filter local function prefixed by $L

[Wu Zhangjin <wuzhangjin@gmail.com>]

On Mon, 2009-06-15 at 12:30 -0400, Steven Rostedt wrote:
> On Mon, 15 Jun 2009, David Daney wrote:
> 
> > Wu Zhangjin wrote:
> > > From: Wu Zhangjin <wuzj@lemote.com>
> > > 
> > > this patch fixed the warning as following:
> > > 
> > > mipsel-linux-gnu-objcopy: 'fs/proc/.tmp_gl_devices.o': No such file
> > > mipsel-linux-gnu-ld: fs/proc/.tmp_gl_devices.o: No such file: No such
> > > file or directory
> > > rm: cannot remove `fs/proc/.tmp_gl_devices.o': No such file or directory
> > > rm: cannot remove `fs/proc/.tmp_mx_devices.o': No such file or directory
> > > 
> > > the real reason of above warning is that the $Lxx local functions will
> > > be treated as global symbols, so, should be filtered.
> > > 
> > > Signed-off-by: Wu Zhangjin <wuzj@lemote.com>
> > > ---
> > >  scripts/recordmcount.pl |    4 ++++
> > >  1 files changed, 4 insertions(+), 0 deletions(-)
> > > 
> > > diff --git a/scripts/recordmcount.pl b/scripts/recordmcount.pl
> > > index 533d3bf..542cb04 100755
> > > --- a/scripts/recordmcount.pl
> > > +++ b/scripts/recordmcount.pl
> > > @@ -343,6 +343,10 @@ sub update_funcs
> > >   if (!$use_locals) {
> > >       return;
> > >   }
> > > + # filter $LXXX tags
> > > + if ("$ref_func" =~ m/\$L/) {
> > > +         return;
> > > + }
> > 
> > Certainly this is true for mips.  I doubt it is for all architectures targed
> > by Linux.
> 

have tried to use function_regex instead, but for I'm poor in playing
with regular expression of perl. at last, i use something easier like
above. and I _guess_ this problem maybe exist in some other platforms,
so just put it there.

> Yes, that should probably go into a mips only change. Unless you can 
> reproduce it on all other archs, or at least x86.
> 
> You could also do this in the function_regex variable.
> 
>  "^([0-9a-fA-F]+)\\s+<(.|[^\$]L.*?|\$[^L].*?|[^\$][^L].*?)>:"
> 

works well, and this seems mips64-specific, so, I moved this specific
function_regex to "if ($bits == 64) { ... }".

> There may even be a better way, but I'm not in the mood to look it up ;-)

seems not that easy to understand :-)

-- Wu Zhangjin