-march=native -c -o liblash_la-comm_event.lo `test -f 'comm_event.c' || echo './'`comm_event.c libtool: compile: x86_64-pc-linux-gnu-gcc -DHAVE_CONFIG_H -I. -I.. -I.. -DLASH_BUILD -DHAVE_LASH -O2 -pipe -march=native -c client.c -fPIC -DPIC -o .libs/liblash_la-client.o lash.c: In function ‘lash_init’: lash.c:203:19: error: storage size of ‘max_fds’ isn’t known lash.c:204:5: error: unknown type name ‘rlim_t’ lash.c:206:15: error: ‘RLIMIT_NOFILE’ undeclared (first use in this function) lash.c:206:15: note: each undeclared identifier is reported only once for each function it appears in make[2]: *** [liblash_la-lash.lo] Error 1 make[2]: *** Waiting for unfinished jobs.... libtool: compile: x86_64-pc-linux-gnu-gcc -DHAVE_CONFIG_H -I. -I.. -I.. -DLASH_BUILD -DHAVE_LASH -O2 -pipe -march=native -c comm_event.c -fPIC -DPIC -o .libs/liblash_la-comm_event.o make[2]: Leaving directory `/var/tmp/portage/media-sound/lash-0.5.4-r1/work/lash-0.5.4/liblash' Reproducible: Always Google found patch at http://permalink.gmane.org/gmane.linux.redhat.fedora.extras.cvs/822346, to include #include <sys/resource.h> between existing #include <pthread.h> and #include <sys/types.h>. It works for me. Could you kindly please apply it? Thank you in advance.
I've seen the same bug and have applied the same fix to my /usr/local/portage, would appreciate the addition up-stream. Looks like it should be safe/harmless on older instances?
fix in portage
