when compiling with the following use flags: Calculating dependencies ...done! [ebuild U ] sys-apps/shadow-4.0.4.1 [4.0.4.1] +nls +pam -(selinux) I get the following error: .....beginning deleted 14:58 <tamran> gcc -DHAVE_CONFIG_H -I. -I. -I.. -I../lib -O3 -pipe -msse -msse2 -ftracer -fweb -march=k8 -MT xmalloc.lo -MD -MP -MF .deps/xmalloc.Tpo -c xmalloc.c -o xmalloc.o 14:58 <tamran> xmalloc.c:16: error: conflicting types for 'malloc' 14:58 <tamran> xmalloc.c:16: error: conflicting types for 'malloc' 14:58 <tamran> make[2]: *** [xmalloc.lo] Error 1 14:58 <tamran> make[2]: Leaving directory `/var/tmp/portage/shadow-4.0.4.1/work/shadow-4.0.4.1/libmisc' 14:58 <tamran> make[1]: *** [all-recursive] Error 1 14:58 <tamran> make[1]: Leaving directory `/var/tmp/portage/shadow-4.0.4.1/work/shadow-4.0.4.1' 14:58 <tamran> make: *** [all] Error 2 14:58 <tamran> !!! ERROR: sys-apps/shadow-4.0.4.1 failed. 14:58 <tamran> !!! Function src_compile, Line 79, Exitcode 2 14:58 <tamran> !!! compile problem
Created attachment 29043 [details, diff] Removes extern char *malloc ();
My attachment should fix this problem.
Just out of curiosity, is there some gcc-3.4 package in portage I'm not seeing?
with gcc 3.4 in pre-release, there's certainly nothing wrong with getting a head start. ;) I've actually played around with 3.4 myself and have been messing around with fixes for various apps. gcc 3.4 will be pretty important for amd64... there are some extreme performance gains to be had. I sent tamran an e-build for it and he went nuts with it, probably for this exact reason. ^_^ my glib 1.2.10 fix is already in portage, which suprised me... seeing how we dont actually have gcc 3.4 in the tree yet. but nothing breaks 3.3 so far, so it cant really hurt to be prepared.
oh yeah, btw, gcc 3.4 is in portage now. ^_^
Created attachment 29801 [details, diff] correct the declaration of malloc
Created attachment 29802 [details, diff] correct the declaration of malloc - ebuild patch
in the tree. give it some time to reach rsync mirrors