From mboxrd@z Thu Jan 1 00:00:00 1970 From: David Gibson Subject: Re: [PATCH] Enhance source position implementation. Date: Sat, 4 Oct 2008 13:53:20 +1000 Message-ID: <20081004035320.GG30184@yookeroo.seuss> References: <1223066853-5978-1-git-send-email-jdl@jdl.com> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Return-path: Content-Disposition: inline In-Reply-To: <1223066853-5978-1-git-send-email-jdl-CYoMK+44s/E@public.gmane.org> List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: devicetree-discuss-bounces+gldd-devicetree-discuss=m.gmane.org-mnsaURCQ41sdnm+yROfE0A@public.gmane.org Errors-To: devicetree-discuss-bounces+gldd-devicetree-discuss=m.gmane.org-mnsaURCQ41sdnm+yROfE0A@public.gmane.org To: Jon Loeliger Cc: devicetree-discuss-mnsaURCQ41sdnm+yROfE0A@public.gmane.org List-Id: devicetree@vger.kernel.org On Fri, Oct 03, 2008 at 03:47:33PM -0500, Jon Loeliger wrote: > Implemented some print and copy routines. > Made empty srcpos objects that will be used later. > Protected .h file from multiple #include's. > Added srcpos_error() and srcpos_warn(). > > Signed-off-by: Jon Loeliger > --- > > This patch is a refactoring of the yyerror() and yyerrorf(() > changes out of my proposed IR patch set back to the current > master branch. It should be pretty close to one Gibson's [3/5] > patch where he was headed in the same direction. This one > anticipates srcpos_warn() too. > > BTW, I realize my earlier reply to his [3/5] patch was sort > of meant to be in response to his following patch as well. > Oh well. > > As he was also in favor of this change with his patch set, > I'm just going to apply this one straight up. Unfortunately, it differs from my 3/5 in exactly the aspects that prompted me to refactor your srcpos stuff into my series in the first place. Your srcpos_empty variable and srcpos_copy() and srcpos_dump() routines are not used here, which means they don't belong in this patch - they belong in the patch that actually does something with them. In fact, srcpos_copy() isn't needed at all. In your IR stuff, the ir_srcpos has exactly the same lifetime as the surrounding ir structure, if you make it an embedded structure rather than a pointer srcpos_copy() disappears. [snip] > extern int yylex(void); > +extern void yyerror(char const *s); This should be static, not extern, now. [snip] > diff --git a/srcpos.c b/srcpos.c > index d5a4b2a..8bb0c02 100644 > --- a/srcpos.c > +++ b/srcpos.c > @@ -17,18 +17,40 @@ > * USA > */ > > +#define _GNU_SOURCE Sigh. asprintf() is really handy, but dtc is supposed to be portable. Anyway, it's applied now, so I'll rebase my series on top of this. -- David Gibson | I'll have my music baroque, and my code david AT gibson.dropbear.id.au | minimalist, thank you. NOT _the_ _other_ | _way_ _around_! http://www.ozlabs.org/~dgibson