From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from cantor2.suse.de ([195.135.220.15]:60337 "EHLO mx2.suse.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1757947Ab0LTPLu (ORCPT ); Mon, 20 Dec 2010 10:11:50 -0500 Date: Mon, 20 Dec 2010 16:11:48 +0100 From: Michal Marek Subject: Re: [PATCHv2] kconfig: simplify select-with-unmet-direct-dependency warning Message-ID: <20101220151148.GC20492@sepie.suse.cz> References: <1291663669-13077-1-git-send-email-lacombar@gmail.com> <20101220151047.GB20492@sepie.suse.cz> MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Disposition: inline Content-Transfer-Encoding: 8bit In-Reply-To: <20101220151047.GB20492@sepie.suse.cz> Sender: linux-kbuild-owner@vger.kernel.org List-ID: To: Arnaud Lacombe Cc: linux-kbuild@vger.kernel.org, Catalin Marinas On Mon, Dec 20, 2010 at 04:10:47PM +0100, Michal Marek wrote: > On Mon, Dec 06, 2010 at 02:27:49PM -0500, Arnaud Lacombe wrote: > > Hi, > > > > This is an updated version of the patch I sent in mid-September to simplify the > > unmet dependency warnings. > > > > - Arnaud > > > > --- > > > > This is an attempt to simplify the expressing printed by kconfig when a > > symbol is selected but still has direct unmet dependency. > > > > First, the symbol reverse dependency is split in sub-expression. Then, > > each sub-expression is checked to ensure that it does not contains the > > unmet dependency. This removes all the false-positive symbols which > > already have the correct dependency. Finally, only the symbol doing the > > "select" is printed, instead of the full dependency tree. > > > > CC: Catalin Marinas > > Signed-off-by: Arnaud Lacombe > > --- > > scripts/kconfig/expr.c | 35 ++++++++++++++++++++++++++++++++++- > > scripts/kconfig/lkc_proto.h | 2 +- > > scripts/kconfig/symbol.c | 6 +++++- > > 3 files changed, 40 insertions(+), 3 deletions(-) > > I applied this to kbuild-2.6.git#kconfig now. If someone encounters a > warning that hides the actual culprit due to this patch, we will have to > solve it somehow, but otherwise these simplified expressions are wanted > IMO. BTW, this called for a minor warning fix: From: Michal Marek Subject: [PATCH] kconfig: Make expr_copy() take a const argument Fixes scripts/kconfig/expr.c: In function ‘expr_get_leftmost_symbol’: scripts/kconfig/expr.c:1026:2: warning: passing argument 1 of ‘expr_copy’ discards qualifiers from pointer target type scripts/kconfig/expr.c:67:14: note: expected ‘struct expr *’ but argument is of type ‘const struct expr *’ Signed-off-by: Michal Marek diff --git a/scripts/kconfig/expr.c b/scripts/kconfig/expr.c index 65531a7..0010034 100644 --- a/scripts/kconfig/expr.c +++ b/scripts/kconfig/expr.c @@ -64,7 +64,7 @@ struct expr *expr_alloc_or(struct expr *e1, struct expr *e2) return e2 ? expr_alloc_two(E_OR, e1, e2) : e1; } -struct expr *expr_copy(struct expr *org) +struct expr *expr_copy(const struct expr *org) { struct expr *e; diff --git a/scripts/kconfig/expr.h b/scripts/kconfig/expr.h index b267933..76ee319 100644 --- a/scripts/kconfig/expr.h +++ b/scripts/kconfig/expr.h @@ -193,7 +193,7 @@ struct expr *expr_alloc_two(enum expr_type type, struct expr *e1, struct expr *e struct expr *expr_alloc_comp(enum expr_type type, struct symbol *s1, struct symbol *s2); struct expr *expr_alloc_and(struct expr *e1, struct expr *e2); struct expr *expr_alloc_or(struct expr *e1, struct expr *e2); -struct expr *expr_copy(struct expr *org); +struct expr *expr_copy(const struct expr *org); void expr_free(struct expr *e); int expr_eq(struct expr *e1, struct expr *e2); void expr_eliminate_eq(struct expr **ep1, struct expr **ep2);