> 5 developers just for clang , how many do we have? Not for clang, just for project debian clang. (about 17 000 packages) 99% of all clang fixes is very helpful for us every friday i'm looking for closed bugs in clang.debian bugzilla and integrates patches in cooker.
> You know how many srpms are in contrib? 8218 > Ok so contrib is useless , and pointless to maintain? Not useless. Just Augean stables ;) And in future we need to look deeper in contrib and drop all stuff older than 4 years (exlcude games) >take you figures as good, so another 10-13%, means 20-26% to me.... Old stuff in contrib is a main reason for 20-26% May we need to discuss to drop very old crap. >But since contrib is useless I guess my part ends here . Contrib is good. Don't take it so close to the heart. I said only about *OLD STUFF*. 2014-09-09 23:47 GMT+04:00 [email protected] < [email protected]>: > I don't care about lives, so debian have 5 developers just for clang , > how many do we have? > Ok so contrib is useless , and pointless to maintain? > You know how many srpms are in contrib? > 6517 in release, and and let's say that you are right and "few thousands > packages in contrib totally outdated and not maintained by upstream for > years" > how about the remainig srpms, because this are srpms, not packages.each > one generates at least 2 rpms, > i'll let you to do the math. > From where I'm stying you said 10-13% in main, let's just say that that > reaming part of contrib is in a fucking good shape as main and let's > take you figures as good, so another 10-13%, means 20-26% to me.... > But since contrib is useless I guess my part ends here . > Take care . > > > Il 09/09/2014 21:24, Alexander Khryukin ha scritto: > > >Ok you know what is the manpower in debian? > In http://clang.debian.net/ 5 persons > > >please take a look again at the contrib... > If contrib will be wiped accidentally I don't even remark > few thousands packages in contrib totally outdated and not maintained by > upstream for years > if you want to compare gcc vs clang you need to compare comparable > things (500-1000 core packages in main repo) > Now you compares a dog vs red color > > > As TPG said yesterday best and simple solution to fix clang issue is > export CC=gcc > > But you must read what i said, issue with LIVES related only to binutils. > > 2014-09-09 22:02 GMT+04:00 [email protected] < > [email protected]>: > >> >> Il 09/09/2014 01:19, Alexander Khryukin ha scritto: >> >>> Clang shows not a perfect results, but it's not disaster. >>> As i saw on debian clang project and percent of failed packages is about >>> 7-10% >>> for us i think 10-13%, >>> Very GOOD results. >>> And repairing binutils issue will fix all errors with bad invocation (as >>> for lives) >>> Real clang fail results is about 5-7% like for gcc between branches 4.6 >>> vs 4.7, 4.7 vs 4.8 etc >>> >>> >>> Ok you know what is the manpower in debian? >> >> 10-13 % huh ? Optimistic... please take a look again at the contrib... >> -- >> Greetings >> >> ___________________ >> >> MRB ain't no shit >> >> Rosalinux.Ro >> >> Mandrivausers.Ro >> >> ________________ >> >> Talk is cheap, show me the code. >> >> _______________________________________________ >> OM-Cooker mailing list >> [email protected] >> http://ml.openmandriva.org/listinfo.cgi/om-cooker-openmandriva.org >> > > > > _______________________________________________ > OM-Cooker mailing > [email protected]http://ml.openmandriva.org/listinfo.cgi/om-cooker-openmandriva.org > > > -- > Greetings > > ___________________ > > MRB ain't no shit > > Rosalinux.Ro > > Mandrivausers.Ro > ________________ > > Talk is cheap, show me the code. > > _______________________________________________ > OM-Cooker mailing list > [email protected] > http://ml.openmandriva.org/listinfo.cgi/om-cooker-openmandriva.org >
_______________________________________________ OM-Cooker mailing list [email protected] http://ml.openmandriva.org/listinfo.cgi/om-cooker-openmandriva.org
