On Thu, 14 Dec 2017 10:43:33 +0100, Angelo Compagnucci wrote: > On Thu, Dec 14, 2017 at 9:36 AM, Jean Delvare <jdelvare@xxxxxxx> wrote: > > I suppose that BUILD_DYNAMIC_LIB=0 should imply USE_STATIC_LIB=1. > > Something like: > > > > --- a/Makefile > > +++ b/Makefile > > @@ -43,6 +43,8 @@ endif > > ifeq ($(BUILD_DYNAMIC_LIB),0) > > ifeq ($(BUILD_STATIC_LIB),0) > > $(error BUILD_DYNAMIC_LIB and BUILD_STATIC_LIB cannot be disabled at the same time) > > +else > > +USE_STATIC_LIB := 1 > > endif > > endif > > > > What do you think? > > I think you are totally right. If BUILD_DYNAMIC_LIB is 0 the static > variant should be built and used too. Thanks. Both fixes committed. > Could you realease a new version of i2c-tools with these patches? I'm > in the phase of updating the package in buildroot and I do really > prefer to have a new stable release than an old release with a bunch > of patches on top. I don't have such a plan, sorry. Version 4.0 is still relatively new, I expect more problems to be found and fixed in the next weeks, I don't think it makes sense to rush a new release. There are only 13 commits since 4.0, it's not that many. Backporting patches for distribution packages is the usual process, and 6 patches is not a bunch. If the patch count really bothers you, you can always consolidate the fix-ups into the original patch and you'll be down to 4, that seems easily manageable. Quilt is your friend ;-) -- Jean Delvare SUSE L3 Support
