On Wed, 11 Jun 2014 06:59:52 +0900, Joerg Sonnenberger <joerg%britannica.bec.de@localhost> wrote:
On Tue, Jun 10, 2014 at 07:39:53PM +0900, OBATA Akio wrote:On Tue, 10 Jun 2014 05:44:34 +0900, Joerg Sonnenberger <joerg%britannica.bec.de@localhost> wrote: >On Sat, Jun 07, 2014 at 07:22:51PM +0000, Aleksey Cheusov wrote: >>Module Name: pkgsrc >>Committed By: cheusov >>Date: Sat Jun 7 19:22:51 UTC 2014 >> >>Modified Files: >> pkgsrc/net/libfetch: Makefile >> pkgsrc/net/libfetch/files: ftp.c http.c >> >>Log Message: >>Define _GNU_SOURCE by default for glibc-based systems (vasprintf(3)). >>This fixes bootstrap failure on AltLinux. > >This is wrong, please revert. "This is wrong" has no information why it is wrong, and resolve nothing.About as much information as the original commit about what goes wrong then...
I don't feel so, "_GNU_SOURCE is required for glibc-based systems to use vasprintf(3)", and "This is wrong". As far as man vasprintf(3), former is correct then...
* using vasprintf(3) from glibc is wrong? * libnbcompat miss to detect vasprintf(3) from glibc, it's wrong and should be fixed instead? * ...? And will you apply "right fix" for the issue?Well, look at the changes and it should be pretty obvious that this is not the correct apprroach.
You should have said as "This approach is wrong" Will your "correct approach" be finished before pkgsrc-2014Q2 branch? It is "bootstrap stopper". -- OBATA Akio / obata%lins.jp@localhost