there's only one valgrind report that looks bad to me, the rest are essentially nuisance message from libunwind (msync_validate):
==15636== Invalid read of size 8
==15636== at 0x58DF33D: access_mem (in /auto/nlg-05/dy_052/julia/v0.4.0-rc3/lib/julia/libjulia-debug.so)
==15636== by 0x58DD3FF: is_plt_entry (in /auto/nlg-05/dy_052/julia/v0.4.0-rc3/lib/julia/libjulia-debug.so)
==15636== by 0x58DD59B: _ULx86_64_step (in /auto/nlg-05/dy_052/julia/v0.4.0-rc3/lib/julia/libjulia-debug.so)
==15636== by 0x4FAF6CE: rec_backtrace_ctx (task.c:661)
==15636== by 0x4FAF5F9: rec_backtrace (task.c:645)
==15636== by 0x4FC7030: profile_bt (signals-linux.c:17)
==15636== by 0x63F770F: ??? (in /lib64/
libpthread-2.12.so)
==15636== by 0xD70B14F: ??? (in /auto/nlg-05/dy_052/julia/v0.4.0-rc3/lib/julia/sys-debug.so)
==15636== by 0x10FEFFE2BF: ???
==15636== by 0xD77AC1F: jlcall_finalizer_2641 (in /auto/nlg-05/dy_052/julia/v0.4.0-rc3/lib/julia/sys-debug.so)
==15636== by 0x4F005C1: jl_apply (julia.h:1324)
==15636== by 0x4F067EB: jl_apply_generic (gf.c:1684)
==15636== Address 0x10feffe2c0 is not stack'd, malloc'd or (recently) free'd
after all.