Hi, I've built the llvm-dwarfdump tool from the 3.3 branch and ran it on simple binary compiled with dwarf4 debug information. The tool seem to read the file correctly including dwarf information (DIContext::getDWARFContext returns a valid context), but when trying to access line information, all I get is "<invalid>" string for the file name and 0's for line and column information. I've seen that there was a talk in the recent Europoe lldb-dev meeting but couldn't figure out the current status from the slides. Am I missing something or there still no support for parsing dwarf4 info? Thanks, Benjamin. --------------------------------------------------------------------- Intel Israel (74) Limited This e-mail and any attachments may contain confidential material for the sole use of the intended recipient(s). Any review or distribution by others is strictly prohibited. If you are not the intended recipient, please contact the sender and delete all copies. -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.llvm.org/pipermail/llvm-dev/attachments/20130708/cc023876/attachment.html>
> I've built the llvm-dwarfdump tool from the 3.3 branch and ran it on simple > binary compiled with dwarf4 debug information. > > > > The tool seem to read the file correctly including dwarf information > (DIContext::getDWARFContext returns a valid context), but when trying to > access line information, all I get is "<invalid>" string for the file name > and 0's for line and column information. > > > > I've seen that there was a talk in the recent Europoe lldb-dev meeting but > couldn't figure out the current status from the slides. > > > > Am I missing something or there still no support for parsing dwarf4 info? >You're missing something. Basically there is some support for parsing it in llvm-dwarfdump, but that tool is largely used for testing the code emission of llvm and clang. If there's something missing/needed we've been adding it as we go along. Line information is one of those that needs to be added. llvm support of dwarf4 itself is pretty decent with just a few features of dwarf4 that aren't used. -eric> Thanks, > > Benjamin. > > --------------------------------------------------------------------- > Intel Israel (74) Limited > > This e-mail and any attachments may contain confidential material for > the sole use of the intended recipient(s). Any review or distribution > by others is strictly prohibited. If you are not the intended > recipient, please contact the sender and delete all copies. > > > _______________________________________________ > LLVM Developers mailing list > LLVMdev at cs.uiuc.edu http://llvm.cs.uiuc.edu > http://lists.cs.uiuc.edu/mailman/listinfo/llvmdev >
On Mon, Jul 8, 2013 at 12:45 PM, Eric Christopher <echristo at gmail.com> wrote:>> I've built the llvm-dwarfdump tool from the 3.3 branch and ran it on simple >> binary compiled with dwarf4 debug information. >> >> >> >> The tool seem to read the file correctly including dwarf information >> (DIContext::getDWARFContext returns a valid context), but when trying to >> access line information, all I get is "<invalid>" string for the file name >> and 0's for line and column information. >> >> >> >> I've seen that there was a talk in the recent Europoe lldb-dev meeting but >> couldn't figure out the current status from the slides. >> >> >> >> Am I missing something or there still no support for parsing dwarf4 info? >> > > You're missing something. Basically there is some support for parsing > it in llvm-dwarfdump, but that tool is largely used for testing the > code emission of llvm and clang. If there's something missing/needed > we've been adding it as we go along. Line information is one of those > that needs to be added. >Slight clarification: We can dump the line information we output, but I've definitely seen binaries output by gcc that we can't dump. I'm not sure what the issue is and haven't looked yet. -eric> llvm support of dwarf4 itself is pretty decent with just a few > features of dwarf4 that aren't used. > > -eric > >> Thanks, >> >> Benjamin. >> >> --------------------------------------------------------------------- >> Intel Israel (74) Limited >> >> This e-mail and any attachments may contain confidential material for >> the sole use of the intended recipient(s). Any review or distribution >> by others is strictly prohibited. If you are not the intended >> recipient, please contact the sender and delete all copies. >> >> >> _______________________________________________ >> LLVM Developers mailing list >> LLVMdev at cs.uiuc.edu http://llvm.cs.uiuc.edu >> http://lists.cs.uiuc.edu/mailman/listinfo/llvmdev >>