Andreas Schwab <[email protected]> writes: > The ltrace.main/dwarf test produces this output on ppc64: > > static .opd translation of 0x3fffb7fad190: no error > couldn't translate address of @libdwarftest.so: Success > static .opd translation of 0x3fffb7fad1b8: no error > couldn't translate address of @libdwarftest.so: Success
[...] > The first few errors come from section symbols, the rest are STT_OBJECT > symbols. Shouldn't populate_this_symtab generally ignore section > symbols? Should it exempt STT_OBJECT symbols from going through > arch_translate_address? It probably should ignore section symbols. I can't decide on the account of object symbols. I don't know off hand if anything that ltrace can actually trace might be represented as STT_OBJECT and we should therefore let backends see it. But it seems that we should ignore them also. Thanks, Petr _______________________________________________ Ltrace-devel mailing list [email protected] http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/ltrace-devel
