This is the mail archive of the libc-alpha@sourceware.org mailing list for the glibc project.
Index Nav: | [Date Index] [Subject Index] [Author Index] [Thread Index] | |
---|---|---|
Message Nav: | [Date Prev] [Date Next] | [Thread Prev] [Thread Next] |
Other format: | [Raw text] |
On Monday 30 December 2013 09:12:22 Joseph S. Myers wrote: > On Mon, 30 Dec 2013, Mike Frysinger wrote: > > it looks like a limitation in the libunwind logic (part of gcc) ? the > > ia64 > > I believe ia64 might use either libgcc unwind code or separate libunwind. libunwind is packaged w/gcc itself and that's what my system is using. so it's technically a sep lib (i.e. it's a dedicated shared lib), but it's still somewhat in the realm of the gcc project considering the code is shipped in the gcc tarball. > > backtrace code uses the x86_64 file and that just calls _Unwind_* funcs > > that come via gcc. which means i should bounce this to upstream gcc and > > not worry about the test failure in glibc right ? i've never poked into > > the unwind lib too closely before ... > > I don't think it can usefully be reported to GCC until you've traced the > backtracing process yourself and identified either (a) a case where you > can describe exactly what's wrong with the unwind information generated by > GCC compared to what you say it should generate that would allow > backtracing here or (b) a case where the unwind information is sufficient > but one of the _Unwind_* functions, that you've identified as coming from > libgcc, is handling it in an identified incorrect or suboptimal way. when i attach with gdb, the backtrace command shows the full stack trace. but calling backtrace() at the same point shows the limited output i mentioned earlier. if this glibc test isn't sufficient (and it's not like it's complicated code -- can easily replace the test-skeleton.c usage with a main() that calls do_test() and get the same result), then i'll probably bail at this point and just mark it as a known failure. learning the inner details of libunwind and the generated info is something i'll put off until next x-mas break. -mike
Attachment:
signature.asc
Description: This is a digitally signed message part.
Index Nav: | [Date Index] [Subject Index] [Author Index] [Thread Index] | |
---|---|---|
Message Nav: | [Date Prev] [Date Next] | [Thread Prev] [Thread Next] |