The internal __elf_getphdrnum_rdlock might return an inconsistent phnum.
Return a sanitized value, or return an error to users that rely on phnum
to be consistent. That way iterating over all phdrs using elf_getphdr
will return consistent results.
Signed-off-by: Mark Wielaard <mjw@redhat.com>
Since elf_strptr can fail and return NULL we should always check the result
before usage. Debug sections are only handled by section name, so make sure
the name actually exists.
Signed-off-by: Mark Wielaard <mjw@redhat.com>
If there is no e_phoff e_phnum cannot be trusted. Extended phnum can only
be gotten if we have an actual section table and a shdr for section zero,
Extended phnum can be too large to fit in the file (or a size_t).
Signed-off-by: Mark Wielaard <mjw@redhat.com>
Since ELF notes need to be properly aligned they can include padding.
Make sure the padding itself and the padding calculation doesn't overflow.
Signed-off-by: Mark Wielaard <mjw@redhat.com>
On 32bit arches size_t is too small to do size and overflow checks
of 64bit ELF files. Use the actual Elf64 types to make sure checks
on 64bit ELF files are done correctly on 32bit arches.
Signed-off-by: Mark Wielaard <mjw@redhat.com>
- The problem this is trying to solve can be seen for example thus:
$ eu-readelf -a /
eu-readelf: failed reading '/': (null)
With the fix we see a proper error number and get a reasonable error
message:
$ eu-readelf -a /
eu-readelf: failed reading '/': invalid file descriptor
Signed-off-by: Petr Machata <pmachata@redhat.com>
The --enable-mudflap configure build has been broken for 2 years without
anybody apparently noticing. GCC 4.9 removed mudflap support. Before
release we now run make distcheck with valgrind support. Removal of the
mudflap configure option simplifies the build a little.
Signed-off-by: Mark Wielaard <mjw@redhat.com>
When the elf file is mmapped ehdr can still point to ehdr_mem if the
address is unaligned. In such cases we do need to memcpy the ehdr back
to the file. We shouldn't when ehdr already comes directly from the mmapped
file, because that would cause an memcpy call with overlapping addresses
(ehdr being copied directly over itself).
Signed-off-by: Mark Wielaard <mjw@redhat.com>
When ehdr e_shoff changes, elf flags is set dirty. This indicates that
the section header moved because sections were added/removed or changed
in size.
Reported-by: Jiri Slaby <jslaby@suse.cz>
Signed-off-by: Mark Wielaard <mjw@redhat.com>
Sometimes with cross-compile toolchains, the tools are prefixed with the
target arch. Using AC_CHECK_TOOL looks for tools named like this.
Signed-off-by: Michael Forney <mforney@mforney.org>
Signed-off-by: Mark Wielaard <mjw@redhat.com>
readelf -d doesn't work if a SHT_NOBITS section is right before the actual
.dynamic section at the same offset. elflint also fails on such binaries.
So skip SHT_NOBITS sections at the same offset in [g]elf[32|64]_offscn().
The gelf_update_[eps]hdr functions didn't implicitly mark the
data structures as modified. This has always been the case and
programs should set the bit but it is somewhat inconsistent since
other update functions implicitly set the bit. Do it in these
three cases, too.
If fill_mmap() was run for a section that's past the shdr_end, but does
not immediately follow the section headers the fill start would be
determined incorrectly as shdr_end, which would wipe off contents of
sections between shdr_end and current one.
Issue was reported and triaged by Hugo Mildenberger, Graham Murray
and Peter Alfredsen. (http://bugs.gentoo.org/show_bug.cgi?id=288977)
Mark Wielaard