Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751108AbdLaPqJ (ORCPT ); Sun, 31 Dec 2017 10:46:09 -0500 Received: from 178.115.242.59.static.drei.at ([178.115.242.59]:38015 "EHLO mail.osadl.at" rhost-flags-OK-FAIL-OK-OK) by vger.kernel.org with ESMTP id S1750854AbdLaPqH (ORCPT ); Sun, 31 Dec 2017 10:46:07 -0500 Date: Sun, 31 Dec 2017 15:45:48 +0000 From: Nicholas Mc Guire To: Masahiro Yamada Cc: Lukas Bulwahn , Linux Kbuild mailing list , sil2review@lists.osadl.org, Michal Marek , Linux Kernel Mailing List Subject: Re: [PATCH v2] fixdep: exit with error code in error branches of do_config_file() Message-ID: <20171231154548.GA27372@osadl.at> References: <1513801623-19069-1-git-send-email-lukas.bulwahn@gmail.com> <1513883429-9527-1-git-send-email-lukas.bulwahn@gmail.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: User-Agent: Mutt/1.5.23 (2014-03-12) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 3656 Lines: 113 On Sun, Dec 31, 2017 at 01:51:33AM +0900, Masahiro Yamada wrote: > 2017-12-22 4:10 GMT+09:00 Lukas Bulwahn : > > do_config_file() should exit with an error code, and not return if it fails > > as then the error in do_config_file() would go unnoticed in the current > > code and allow the build to continue. The exit with error code will make > > the build fail in those very exceptional cases. If this occurs, this > > actually indicates a deeper problem in the execution of the kernel build > > process. > > > > Now, that the function exists, we do not explicitly free memory and close > > the file handlers in do_config_file(), as this is covered by exit(). > > > > This issue in the fixdep script was introduced with its initial > > implementation back in 2002 by the original author Kai Germaschewski with > > this commit 04bd72170653 ("kbuild: Make dependencies at compile time"). > > "04bd72170653" is just confusing. > > If you really want to mention this hash, > please clearly say it is in the history repository > outside of this. > > > > This issue was identified during the review of a previous patch that > > intended to address a memory leak detected by a static analysis tool. > > > > Link: https://lkml.org/lkml/2017/12/14/736 > > > > Fixes: 04bd72170653 ("kbuild: Make dependencies at compile time") > > Please drop this pointless Fixes tag > because that commit is not reachable from this patch. > > > > > Suggested-by: Nicholas Mc Guire > > Suggested-by: Masahiro Yamada > > Signed-off-by: Lukas Bulwahn > > --- > > compile tested on top of next-20171220 with clang and gcc > > Change in v2: > > - no code change; only include proper Fixes tag and explain it > > > > scripts/basic/fixdep.c | 11 +++++------ > > 1 file changed, 5 insertions(+), 6 deletions(-) > > > > diff --git a/scripts/basic/fixdep.c b/scripts/basic/fixdep.c > > index bbf62cb..4274610 100644 > > --- a/scripts/basic/fixdep.c > > +++ b/scripts/basic/fixdep.c > > @@ -284,19 +284,18 @@ static void do_config_file(const char *filename) > > exit(2); > > } > > if (st.st_size == 0) { > > - close(fd); > > - return; > > + fprintf(stderr, "fixdep: error empty file config file: "); > > + perror(filename); > > + exit(2); > > } > > No. This is correct as-is. > > do_config_file() does not parse .cmd files > but parse source files (.c .h .S etc.) > What case for a legitimately empty source file would there be ? the files being checked are assumed to be mandatory for the given configuration right ? if so how can a configuration depend on an empty file ? would that not point to a config problem and rather be fixed there ? or am I misunderstanding this here ? > Having an empty source file is rare, but possible. > putting a printf in the empty case handling and running a set of randconfigs did not reveal such an event (10 runs). thx! hofrat > > > > > > > map = malloc(st.st_size + 1); > > if (!map) { > > perror("fixdep: malloc"); > > - close(fd); > > - return; > > + exit(2); > > } > > if (read(fd, map, st.st_size) != st.st_size) { > > perror("fixdep: read"); > > - close(fd); > > - return; > > + exit(2); > > } > > map[st.st_size] = '\0'; > > close(fd); > > -- > > 2.7.4 > > > > These two changes are OK. > > > > -- > Best Regards > Masahiro Yamada