Skip site navigation (1)Skip section navigation (2)
Date:      Tue, 22 May 2012 12:22:54 +0200
From:      =?utf-8?Q?Dag-Erling_Sm=C3=B8rgrav?= <des@des.no>
To:        Baptiste Daroussin <bapt@FreeBSD.org>
Cc:        svn-src-head@freebsd.org, svn-src-all@freebsd.org, src-committers@freebsd.org
Subject:   Re: svn commit: r235723 - in head: contrib/byacc tools/regression/usr.bin/yacc usr.bin/yacc usr.bin/yacc/test
Message-ID:  <86likkpla9.fsf@ds4.des.no>
In-Reply-To: <201205211331.q4LDVRvB060077@svn.freebsd.org> (Baptiste Daroussin's message of "Mon, 21 May 2012 13:31:27 %2B0000 (UTC)")
References:  <201205211331.q4LDVRvB060077@svn.freebsd.org>

next in thread | previous in thread | raw e-mail | index | archive | help
Baptiste Daroussin <bapt@FreeBSD.org> writes:
> Log:
>   Import byacc from invisible island, it brings us lots of compatibilitie=
s with
>   bison, keeping full compatibility with our previous yacc
>   implementation.

This commit broke the build, in large part because Baptiste tested with
Clang instead of GCC, and GCC generates a warning when compiling the
generated code.  Baptiste has a patch ready to commit, currently
undergoing review by myself and other interested parties.

I chose not to ask Baptiste to revert the commit, because it is a
complex commit that copies some files from a vendor branch into contrib
and removes others from the usr.bin/yacc, and I wasn't sure that
reverting it wouldn't cause trouble with merge tracking.

Lessons learned:

1) always test a *stock* build before committing:

   % cd /usr/src && env __MAKE_CONF=3D/dev/null make buildworld

2) the contrib part and the usr.bin part should probably have been
   committed separately.

DES
--=20
Dag-Erling Sm=C3=B8rgrav - des@des.no



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?86likkpla9.fsf>