Thanx for the investigation! I'll update the package. Yes something has to be done about what upstream distribution we use. Unfortunately there are atmel specific patches that people rely on that was not accepted upstream, so switching is not as straight forward as it might seem...
On Mon, Aug 31, 2020 at 1:48 PM Gregor Riepl <onit...@gmail.com> wrote: > > > elf.c: In function ‘setup_group’: > > > elf.c:706:35: error: overflow in conversion from ‘unsigned int’ to > ‘int’ changes value from ‘num_group = 4294967295’ to ‘-1’ [-Werror=overflow] > > > 706 | elf_tdata (abfd)->num_group = num_group = -1; > > Upstream bug is here: > https://sourceware.org/bugzilla/show_bug.cgi?id=25717 > > This was fixed upstream by: > > https://sourceware.org/git/?p=binutils-gdb.git;a=commit;h=cda7e5603f6efd7c3716e45cc6ea11b70dd8daae > > Looks like it's caused by GCC 10 being more picky about signedness. > > > > In file included from elf.c:45: > > > elf.c: In function ‘elfcore_write_linux_prpsinfo32’: > > > elf-linux-psinfo.h:73:7: warning: ‘strncpy’ output may be truncated > copying 16 bytes from a string of length 16 [-Wstringop-truncation] > > This and the rest of the errors are false positives occurring from GCC 8 > onwards. They have been silenced upstream: > > > https://sourceware.org/git/?p=binutils-gdb.git;a=commit;h=5a6312e8c015d4a98020038f3b6e144db230f3ca > > > https://sourceware.org/git/?p=binutils-gdb.git;a=commit;h=b9f26d2e29bb56a0404216c5612d6d7fee54a769 > > > https://sourceware.org/git/?p=binutils-gdb.git;a=commit;h=d99b4b92c8ed0f7ef98f370bbf65a360ed66ad7b > > > https://sourceware.org/git/?p=binutils-gdb.git;a=commit;h=602f16570454a1597c2af28af66852133432d1f2 > > (there may be other patches as well) > > Related bug reports: > https://sourceware.org/bugzilla/show_bug.cgi?id=23146 > (can't find any that describe the original problem) > > Also, please note that elf-linux-psinfo.h was renamed to > elf-linux-core.h upstream: > > https://sourceware.org/git/?p=binutils-gdb.git;a=commit;h=de64ce13a78669f094d6909fce51d210e2f9d2c0 > > Perhaps it's time to replace the outdated Atmel versions of avr-gcc > avr-binutils and avr-libc in favour of the upstream versions? > -- Håkan Ardö
