https://sourceware.org/bugzilla/show_bug.cgi?id=23460

zenith432 at users dot sourceforge.net changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |amodra at gmail dot com

--- Comment #1 from zenith432 at users dot sourceforge.net ---
This can be reproduced with even a single object file

The first invocation of open is done via _bfd_real_fopen which takes care that
the file descriptor is eventually closed.
== BEGIN QUOTE
openat(AT_FDCWD, "a.o", O_RDONLY) = 4
#0  0x00007ffff78fdc50 in open64 () from /lib64/libc.so.6
#1  0x00007ffff788f0f2 in __GI__IO_file_open () from /lib64/libc.so.6
#2  0x00007ffff788f29d in __GI__IO_file_fopen () from /lib64/libc.so.6
#3  0x00007ffff7883549 in __fopen_internal () from /lib64/libc.so.6
#4  0x00007ffff7ed27bf in _bfd_real_fopen () from
/lib64/libbfd-2.31.1-4.fc29.so
#5  0x00007ffff7edbd18 in bfd_fopen () from /lib64/libbfd-2.31.1-4.fc29.so
#6  0x000055555555b775 in ar_emul_default_append ()
#7  0x0000555555559ddf in replace_members ()
#8  0x0000555555556f31 in main ()
== END QUOTE

The second invocation of open is done via bfd_plugin_object_p that leaves the
file descriptor leaking
== BEGIN QUOTE
openat(AT_FDCWD, "a.o", O_RDONLY) = 7
#0  0x00007ffff78fdc50 in open64 () from /lib64/libc.so.6
#1  0x00007ffff7f6411c in bfd_plugin_open_input () from
/lib64/libbfd-2.31.1-4.fc29.so
#2  0x00007ffff7f642a9 in try_load_plugin () from
/lib64/libbfd-2.31.1-4.fc29.so
#3  0x00007ffff7f6453f in bfd_plugin_object_p () from
/lib64/libbfd-2.31.1-4.fc29.so
#4  0x00007ffff7ed4b2a in bfd_check_format_matches () from
/lib64/libbfd-2.31.1-4.fc29.so
#5  0x00007ffff7ecec88 in _bfd_write_archive_contents () from
/lib64/libbfd-2.31.1-4.fc29.so
#6  0x00007ffff7edc06e in bfd_close () from /lib64/libbfd-2.31.1-4.fc29.so
#7  0x00005555555593cd in write_archive ()
#8  0x0000555555559f7d in replace_members ()
#9  0x0000555555556f31 in main ()
== END QUOTE

-- 
You are receiving this mail because:
You are on the CC list for the bug.
_______________________________________________
bug-binutils mailing list
bug-binutils@gnu.org
https://lists.gnu.org/mailman/listinfo/bug-binutils

Reply via email to