Package: valgrind Version: 1:3.19.0-1 Severity: normal Hi! Valgrind reports a false positive during glibc initialization on mips64el:
Conditional jump or move depends on uninitialised value(s) at 0x400746C: cached_fpabi_reject_phdr_p (dl-machine-reject-phdr.h:57) by 0x400746C: elf_machine_reject_phdr_p (dl-machine-reject-phdr.h:217) by 0x40079E0: open_verify.constprop.0 (dl-load.c:1793) by 0x400ABA8: _dl_map_object (dl-load.c:2211) by 0x401DA2C: map_doit (rtld.c:647) by 0x401B3C8: _dl_catch_exception (dl-error-skeleton.c:208) by 0x401B478: _dl_catch_error (dl-error-skeleton.c:227) by 0x401D96C: do_preload (rtld.c:822) by 0x401F058: handle_preload_list (rtld.c:898) by 0x4021BC4: dl_main (rtld.c:1859) by 0x401CD98: _dl_sysdep_start (dl-sysdep.c:140) by 0x401E3D0: _dl_start_final (rtld.c:497) by 0x401E684: _dl_start (rtld.c:586) This makes all packages that run valgrind during their testsuite fail if they consider this error to be fatal. Also, all tools report the following warning: WARNING: unhandled mips64-linux syscall: 5327 You may be able to write your own handler. Read the file README_MISSING_SYSCALL_OR_IOCTL. Nevertheless we consider this a bug. Please report it at http://valgrind.org/support/bug_reports.html. The syscall in question is rseq, added in 2018. For a package to fail it'd need to do so on unexpected stderr -- but that's the default for some test engines.