'Should parse compilation unit: UnexpectedEof' running dwarfdump on a binary with a SHF_COMPRESSED .debug_info

[luser]

I tried running gimli's dwarfdump on a binary from an Ubuntu package that was crashing Breakpad's dump_syms, but dwarfdump panics as well. A copy of the binary in question is here. It's from this deb file.

luser@eye7:/build/gimli$ RUST_BACKTRACE=1 cargo run --example dwarfdump -- /tmp/deb-contents/usr/lib/debug/.build-id/02/163988533d4192aa610a0d6ec03c349a2d10b9.debug
    Finished dev [unoptimized + debuginfo] target(s) in 0.2 secs
     Running `target/debug/examples/dwarfdump /tmp/deb-contents/usr/lib/debug/.build-id/02/163988533d4192aa610a0d6ec03c349a2d10b9.debug`

.debug_info
thread 'main' panicked at 'Should parse compilation unit: UnexpectedEof', /buildslave/rust-buildbot/slave/stable-dist-rustc-linux/build/src/libcore/result.rs:868
stack backtrace:
   1:     0x55880364e66c - std::sys::imp::backtrace::tracing::imp::write::hf33ae72d0baa11ed
                        at /buildslave/rust-buildbot/slave/stable-dist-rustc-linux/build/src/libstd/sys/unix/backtrace/tracing/gcc_s.rs:42
   2:     0x558803650b8e - std::panicking::default_hook::{{closure}}::h59672b733cc6a455
                        at /buildslave/rust-buildbot/slave/stable-dist-rustc-linux/build/src/libstd/panicking.rs:351
   3:     0x558803650794 - std::panicking::default_hook::h1670459d2f3f8843
                        at /buildslave/rust-buildbot/slave/stable-dist-rustc-linux/build/src/libstd/panicking.rs:367
   4:     0x558803650f6b - std::panicking::rust_panic_with_hook::hcf0ddb069e7beee7
                        at /buildslave/rust-buildbot/slave/stable-dist-rustc-linux/build/src/libstd/panicking.rs:555
   5:     0x558803650e04 - std::panicking::begin_panic::hd6eb68e27bdf6140
                        at /buildslave/rust-buildbot/slave/stable-dist-rustc-linux/build/src/libstd/panicking.rs:517
   6:     0x558803650d29 - std::panicking::begin_panic_fmt::hfea5965948b877f8
                        at /buildslave/rust-buildbot/slave/stable-dist-rustc-linux/build/src/libstd/panicking.rs:501
   7:     0x558803650cb7 - rust_begin_unwind
                        at /buildslave/rust-buildbot/slave/stable-dist-rustc-linux/build/src/libstd/panicking.rs:477
   8:     0x55880367c5cd - core::panicking::panic_fmt::hc0f6d7b2c300cdd9
                        at /buildslave/rust-buildbot/slave/stable-dist-rustc-linux/build/src/libcore/panicking.rs:69
   9:     0x5588035984b6 - core::result::unwrap_failed::h9c0a209799863b53
                        at /buildslave/rust-buildbot/slave/stable-dist-rustc-linux/build/src/libcore/macros.rs:29
  10:     0x55880358d9a5 - <core::result::Result<T, E>>::expect::h33b0d09147a2e563
                        at /buildslave/rust-buildbot/slave/stable-dist-rustc-linux/build/src/libcore/result.rs:769
  11:     0x5588035f5f09 - dwarfdump::dump_info::h446d1756de9d6452
                        at /build/gimli/examples/dwarfdump.rs:144
  12:     0x5588035f56d0 - dwarfdump::dump_file::hda1b626be36ebb40
                        at /build/gimli/examples/dwarfdump.rs:105
  13:     0x5588035f51b0 - dwarfdump::main::hde97895e853bae8c
                        at /build/gimli/examples/dwarfdump.rs:83
  14:     0x558803657dfa - __rust_maybe_catch_panic
                        at /buildslave/rust-buildbot/slave/stable-dist-rustc-linux/build/src/libpanic_unwind/lib.rs:98
  15:     0x5588036516d6 - std::rt::lang_start::hd7c880a37a646e81
                        at /buildslave/rust-buildbot/slave/stable-dist-rustc-linux/build/src/libstd/panicking.rs:436
                        at /buildslave/rust-buildbot/slave/stable-dist-rustc-linux/build/src/libstd/panic.rs:361
                        at /buildslave/rust-buildbot/slave/stable-dist-rustc-linux/build/src/libstd/rt.rs:57
  16:     0x55880360c912 - main
  17:     0x7fca291523f0 - __libc_start_main
  18:     0x55880356dd29 - _start
  19:                0x0 - <unknown>

I'm using the latest master (d591dc4).

[fitzgen]

Thanks for the bug report!

[luser]

FWIW, dwarfdump on my system (Ubuntu 16.10) reads this file without issue:

.debug_info

COMPILE_UNIT<header overall offset = 0x00000000>:
< 0><0x0000000b>  DW_TAG_compile_unit
                    DW_AT_producer              GNU C11 6.2.0 20160901 -mtune=generic -march=x86-64 -g -O2 -fstack-protector-strong -fno-strict-aliasing -fstack-protector-strong
                    DW_AT_language              DW_LANG_C99
                    DW_AT_name                  dbus-binding-tool-glib.c
                    DW_AT_comp_dir              ./dbus
<...>

[tromey]

I think the issue is that the .debug_info section is SHF_COMPRESSED, but neither xmas-elf nor gimli decompresses the contents.

[philipc]

Yes, gimli's dwarfdump works after using objcopy --decompress-debug-sections.

This is probably best to fix after implementing #182, assuming we still plan to do that.

[luser]

Thanks for the info @tromey! Looking at the readelf output for the binary I now notice that it has a C in the flags column:

  [29] .debug_info       PROGBITS         0000000000000000  00000370
       000000000000e133  0000000000000000   C       0     0     1

With a new-enough binutils it's trivially easy to generate a binary for testing. I'm on Ubuntu 16.10:

luser@eye7:/build$ gcc --version
gcc (Ubuntu 6.2.0-5ubuntu12) 6.2.0 20161005
Copyright (C) 2016 Free Software Foundation, Inc.
This is free software; see the source for copying conditions.  There is NO
warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.

luser@eye7:/build$ objcopy --version
GNU objcopy (GNU Binutils for Ubuntu) 2.27
Copyright (C) 2016 Free Software Foundation, Inc.
This program is free software; you may redistribute it under the terms of
the GNU General Public License version 3 or (at your option) any later version.
This program has absolutely no warranty.
luser@eye7:/build$ cat hello.c
#include <stdio.h>

int main(int argc, char** argv)
{
  int x;
  printf("Hello world\n");
  return 0;
}
luser@eye7:/build$ gcc -o hello hello.c -g
luser@eye7:/build$ objcopy --compress-debug-sections hello hello-compressed
luser@eye7:/build/gimli$ cargo run --example dwarfdump -- /build/hello-compressed
    Finished dev [unoptimized + debuginfo] target(s) in 0.0 secs
     Running `target/debug/examples/dwarfdump /build/hello-compressed`

.debug_info
thread 'main' panicked at 'Should parse compilation unit: UnexpectedEof', /buildslave/rust-buildbot/slave/stable-dist-rustc-linux/build/src/libcore/result.rs:868
note: Run with `RUST_BACKTRACE=1` for a backtrace.

[luser]

Oh, also, there's a good blog post describing SHF_COMPRESSED.

[fitzgen]

Thanks for the info @luser !

[philipc]

This was fixed with the object update in #301.