Re: [rfc] physname cross-check [Re: [RFA] Typedef'd method parameters [0/4]]

[Jan Kratochvil <jan.kratochvil@redhat.com>]

On Tue, 17 May 2011 20:15:18 +0200, Keith Seitz wrote:
> Is reverting dwarf2_physname better or worse than DW_AT_MIPS_linkage_name?

In which single case can be dwarf2_physname better than
DW_AT_MIPS_linkage_name?  That's the question.  dwarf2_physname is AFAIK to
give the linkage name and DW_AT_MIPS_linkage_name always matches that.

Apparently the testsuite has regressions with demangled + canonicalized
DW_AT_MIPS_linkage_name so I am wrong but I do not understand why.


> Using DW_AT_MIPS_linkage_name will not pass cpexprs.exp without some
> hacking; the demangled name will need to be re-parsed (to remove
> typedefs

DW_AT_MIPS_linkage_name already has all the typedefs removed - it is the
linkage name.


> I really see this as an even bigger risk than keeping the current
> code. And then there's constructors -- no version of GCC that I've
> seen outputs DW_AT_MIPS_linkage_name for ctors, so they would still
> have to be computed in some way.

Yes, I agree for ctors/dtors we should use the dwarf2_physname computation.


Thanks,
Jan