From mboxrd@z Thu Jan 1 00:00:00 1970 From: Petr Vorel Date: Mon, 12 Feb 2018 16:52:20 +0100 Subject: [LTP] [COMMITTED 1/1] Replace __BEGIN_DECLS and __END_DECLS to with expanded macros In-Reply-To: <20180212104745.GA9342@rei.lan> References: <20180209140553.17453-1-pvorel@suse.cz> <20180209155112.GA17594@rei> <20180209182030.GB1077@x230> <20180212104745.GA9342@rei.lan> Message-ID: <20180212155220.GA1105@x230> List-Id: MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: ltp@lists.linux.it Hi Cyril, > > > > testcases/kernel/syscalls/epoll2/include/epoll.h | 8 +- > > > > utils/ffsb-6.0-rc2/config.h.in | 49 +- > > > > utils/ffsb-6.0-rc2/configure | 5937 ++++++++++------------ > > > ^ > > > This does not look like intended change, can you please revert the > > > patch and apply clean version? > > > (Have you used git commit -a by mistake?) > > I'm sorry for mistake. I reverted it and commit again. > No problem at all. > > I wish there was a commit of ffsb configure files with recent/common autotools version > > as I keep reverting them often. But it's a question, what is common autotools version. > Looked shortly into the problem, it looks like the problem is Makefile > generated by automake decides to regenerate configure script because > timestamps are not preserved on git checkout and the build system thinks > that configure script is out of date since it's older than the files it > has been generated from. Looks like this is a common problem see: > https://www.gnu.org/software/automake/manual/html_node/CVS.html > One of the solutions would be to regenerate the configure with > AM_MAINTAINER_MODE() which should disable the automagic regeneration. Thanks for investigation! I'm for using it (better than have some scripts like gcc folks as mentioned there as an alternative). > Also side note, I've learned not to use git commit -a after one of the > screw ups or at least check the resulting patch for unintended changes. Well, I wasn't using '-a' switch. Custom function is sometimes even worse than '-a' switch :-). Kind regards, Petr