also reported as http://gcc.gnu.org/PR42672: seen with 20100107, ix86, using gold from the 2.20 branch: - omp_destroy_lock@OMP_3.0 4.4 - omp_destroy_lock_@OMP_3.0 4.4 - omp_destroy_nest_lock@OMP_3.0 4.4 - omp_destroy_nest_lock_@OMP_3.0 4.4 - omp_init_lock@OMP_3.0 4.4 - omp_init_lock_@OMP_3.0 4.4 - omp_init_nest_lock@OMP_3.0 4.4 - omp_init_nest_lock_@OMP_3.0 4.4 - omp_set_lock@OMP_3.0 4.4 - omp_set_lock_@OMP_3.0 4.4 - omp_set_nest_lock@OMP_3.0 4.4 - omp_set_nest_lock_@OMP_3.0 4.4 - omp_test_lock@OMP_3.0 4.4 - omp_test_lock_@OMP_3.0 4.4 - omp_test_nest_lock@OMP_3.0 4.4 - omp_test_nest_lock_@OMP_3.0 4.4 - omp_unset_lock@OMP_3.0 4.4 - omp_unset_lock_@OMP_3.0 4.4 - omp_unset_nest_lock@OMP_3.0 4.4 - omp_unset_nest_lock_@OMP_3.0 4.4
With the current development version of gold, I don't see any differences like this. I see symbols like omp_destroy_lock@@OMP_3.0, which is slightly different from the reported omp_destroy_lock@OMP_3.0. I'm not sure what program generated the diff in this bug report. If you still see a problem with the development version, please reopen with more details.
don't see this anymore with today's binutils trunk building the gcc trunk.