[PATCH] Search for DWZ files in debug-file-directories as well

[Sergio Durigan Junior]

On Sunday, November 15 2020, Simon Marchi wrote:

> On 2020-11-14 6:48 p.m., Sergio Durigan Junior via Gdb-patches wrote:
>> When Debian (and Ubuntu) builds its binaries, it (still) doesn't use
>> dwz's "--relative" option.  This causes their debuginfo files to
>> carry a .gnu_debugaltlink section containing a full pathname to the
>> DWZ alt debug file, like this:
>>
>>   $ readelf -wk /usr/bin/cat
>>   Contents of the .gnu_debugaltlink section:
>>
>>     Separate debug info file: /usr/lib/debug/.dwz/x86_64-linux-gnu/coreutils.debug
>>     Build-ID (0x14 bytes):
>>    ee 76 5d 71 97 37 ce 46 99 44 32 bb e8 a9 1a ef 99 96 88 db
>>
>>   Contents of the .gnu_debuglink section:
>>
>>     Separate debug info file: 06d3bee37b8c7e67b31cb2689cb351102ae73b.debug
>>     CRC value: 0x53267655
>>
>> This usually works OK, because most of the debuginfo files installed
>> via apt will be present in /usr/lib/debug anyway.  However, imagine
>> the following scenario:
>>
>> - You are using /usr/bin/cat, it crashes on you and generates a
>>   corefile.
>>
>> - You don't want/need to "apt install" the debuginfo file for
>>   coreutils from the repositories.  Instead, you already have the
>>   debuginfo files in a separate directory (e.g., $HOME/dbgsym).
>>
>> - You start GDB and "set debug-file-directory $HOME/dbgsym".
>>   You then get the following message:
>>
>>   $ gdb -ex 'set debug-file-directory ./dbgsym/usr/lib/debug' -ex 'file /bin/cat' -ex 'core-file ./cat.core'
>>   GNU gdb (Ubuntu 10.1-0ubuntu1) 10.1
>>   ...
>>   Reading symbols from /bin/cat...
>>   Reading symbols from /home/sergio/gdb/dbgsym/usr/lib/debug/.build-id/bc/06d3bee37b8c7e67b31cb2689cb351102ae73b.debug...
>>   could not find '.gnu_debugaltlink' file for /home/sergio/gdb/dbgsym/usr/lib/debug/.build-id/bc/06d3bee37b8c7e67b31cb2689cb351102ae73b.debug
>>
>> This error happens because GDB is trying to locate the build-id
>> link (inside /home/sergio/gdb/dbgsym/usr/lib/debug/.build-id) for the
>> DWZ alt debug file, which doesn't exist.  Arguably, this is a problem
>> with how dh_dwz works in Debian, and it's something I'm also planning
>> to tackle.  But, back at the problem at hand.
>>
>> Besides not being able to find the build-id link in the directory
>> mentioned above, GDB also tried to open the DWZ alt file using its
>> filename.  The problem here is that, since we don't have the distro's
>> debuginfo installed, it can't find anything under /usr/lib/debug that
>> satisfies it.
>>
>> It occurred to me that a good way to workaround this problem is to
>> actually try to locate the DWZ alt debug file inside the
>> debug-file-directories (that were likely provided by the user).  So
>> this is what the proposed patch does.
>>
>> The idea here is simple: get the filename extracted from the
>> .gnu_debugaltlink section, and manipulate it in order to replace the
>> initial part of the path (everything before "/.dwz/") by whatever
>> debug-file-directories the user might have provided.
>>
>> I talked with Mark Wielaard and he agrees this is a sensible approach.
>> In fact, apparently this is something that eu-readelf also does.
>>
>> I regtested this code, and no regressions were found.
>
> Hi Sergio,

Bonjour, Simon,

Thanks for the review.

> I don't really have an opinion on this at the moment because I don't
> know much about how dwz files are used in practice.  Is having a ".dwz"
> somewhat standard?

Yeah, both Fedora and Debian/Ubuntu generate and use DWZ files
extensively.  Fedora has the 'find-debuginfo.sh' script which is called
during rpmbuild and takes care of the debuginfo/DWZ generation:

  https://github.com/rpm-software-management/rpm/blob/HEAD/scripts/find-debuginfo.sh#L510

Debian/Ubuntu have the dh_dwz step which is invoked via debhelper during
the package build:

  https://salsa.debian.org/debian/debhelper/-/blob/master/dh_dwz

>> @@ -2238,6 +2236,61 @@ dwarf2_get_dwz_file (dwarf2_per_bfd *per_bfd)
>>    if (dwz_bfd == NULL)
>>      dwz_bfd = build_id_to_debug_bfd (buildid_len, buildid);
>>
>> +  if (dwz_bfd == nullptr)
>> +    {
>> +      /* If the user has provided us with different
>> +	 debug-file-directories, we can try them in order.  */
>> +      size_t dwz_pos = filename.find ("/.dwz/");
>> +
>> +      if (dwz_pos != std::string::npos)
>> +	{
>> +	  std::string tmpfilename = filename.erase (0, dwz_pos);
>
> "filename.erase", I think that modifies filename in place.  If so, is
> this what you intended?

Oh, you're right, both the .erase and the .insert methods modify their
arguments in-place.  This is not what I intended, but it's an easy fix.
Thanks for pointing it out.

>> +	  std::vector<gdb::unique_xmalloc_ptr<char>> debugdir_vec
>> +	    = dirnames_to_char_ptr_vec (debug_file_directory);
>> +
>> +	  for (const gdb::unique_xmalloc_ptr<char> &debugdir : debugdir_vec)
>> +	    {
>> +	      /* The idea is to iterate over the
>> +		 debug-file-directories provided by the user and
>> +		 replace the hard-coded path in the "filename" by each
>> +		 debug-file-directory.
>> +
>> +		 For example, suppose that filename is:
>> +
>> +		   /usr/lib/debug/.dwz/foo.debug
>> +
>> +		 And suppose that we have "$HOME/bar" as the
>> +		 debug-file-directory.  We would then adjust filename
>> +		 to look like:
>> +
>> +		   $HOME/bar/.dwz/foo.debug
>> +
>> +		 which would hopefully allow us to find the alt debug
>> +		 file.  */
>> +	      std::string ddir = debugdir.get ();
>> +
>> +	      if (filename.size () > ddir.size ()
>> +		  && filename.compare (0, ddir.size (), ddir) == 0)
>> +		continue;
>
> What's the intent of this condition?  Can you add a comment to make it
> explicit?

It's just a check to verify whether "filename" already refers to a file
under the "ddir" debug directory.  If it does, then there's nothing to
do here (because the function already tried to open the file before).
I'll add a comment on top of this check.

>> +
>> +	      std::string new_filename = tmpfilename.insert (0, ddir);
>
> Same question as above for "tmpfilename.insert".

Same answer as above :-).

Thanks,

-- 
Sergio
GPG key ID: 237A 54B1 0287 28BF 00EF  31F4 D0EB 7628 65FC 5E36
Please send encrypted e-mail if possible
https://sergiodj.net/