> Le 7 oct. 2012, 00:15, Paul Eggert wrote : > > On 10/06/2012 10:46 AM, Akim Demaille wrote: > > Please, install in maint > > Unfortunately I can't build 'maint', even without the patch. > 'make V=V' reports: > > rm -f calc++-parser.tmp > touch calc++-parser.tmp > ../../tests/bison -d -ra -S lalr1.cc -o ./calc++-parser.cc \ > ./calc++-parser.yy > make[3]: *** [calc++-parser.stamp] Error 141 > make[3]: Leaving directory `/home/eggert/src/gnu/bison-maint/examples/calc++' > > I suppose I could try to get to the bottom of it, but how > about if you install the patch instead?
Sure, done. With a fixlet (you have put the call to YYLVAL_INITIALIZE under the control of b4_locations_if). > The build failure is not a good sign, but that's a different > story.... Yes, indeed, it is. Unless you use a single checkout for both maint and master, and go over the full ./bootstrap cycle? Is V=V some kind of VERBOSE=x? Is it used somewhere? I have installed the following patches. Akim Demaille (3): skeletons: style changes tests: no longer disable -O compiler options warnings: avoid warnings from clang Paul Eggert (1): yacc.c: initialize yylval in pure-parser mode NEWS | 13 +++++++++++++ data/c.m4 | 2 +- data/glr.c | 25 +++++++++++++------------ data/yacc.c | 47 +++++++++++++++++++++++++++++++++++++---------- src/uniqstr.h | 2 +- tests/atlocal.in | 11 ++--------- 6 files changed, 67 insertions(+), 33 deletions(-) -- 1.7.12.2
