Adding a few folks working on Clang's DWARF5 functionality to see if this
is a known bug (David Anderson mentioned it may've come from Clang r327823)
and/or has been fixed.

Perhaps a minimal example that produces this behavior/bug would be useful
for David to have (not sure if he wants/needs it if this is already fixed
and was never released in an official clang release? David?)

On Sun, Apr 15, 2018 at 7:53 PM David Blaikie <> wrote:

> Does your example have a DW_AT_producer string by chance?
> On Sat, Apr 14, 2018 at 4:03 PM David Anderson via Dwarf-Discuss <
>> wrote:
>> I've just pushed a new version of libdwarf/dwarfdump to sourceforge.
>> I mention this here  because at least one compiler version is emitting
>> an incorrect .debug_str_offsets section (I believe).
>> You can get the source with
>>    git clone git://
>> The new  dwarfdump option --print-str-offsets
>> will cause dwarfdump to access new libdwarf functions to
>> print   the DWARF5 .debug_str_offsets data (headers and everything)
>> independently of any other
>> section.
>> The example I have in hand (the only one so far) seems to have
>> the content of the section a bit wrong: the header of the only
>> table in it has an incorrect length value.  Dwarfdump and
>> libdwarf report the apparent botch.  Hmm.
>> The compiler creating  this example may have
>> already been fixed (I don't know).
>> Any smallish object files with DWARF5 .debug_str_offsets that
>> can be put into the public regression tests base would be most
>> welcome.  Especially one with multiple headers, such as
>> would most likely be created by a link step.
>> For information:
>> and click on Work In Progress for what has
>> happened recently.
>> Regrettably, at present using cmake to build is not
>> working.    We hope to fix that before long.
>>     ./configure ; make
>> works, of course.
>> DavidA.
>> _______________________________________________
>> Dwarf-Discuss mailing list
Dwarf-Discuss mailing list

Reply via email to