RTTI draft proposal

[Jim Dehnert]

| From: Daveed Vandevoorde <daveed at edg.com>
| 
| Jason Merrill wrote:
| > 
| > >>>>> Daveed Vandevoorde <daveed at edg.com> writes:
| [...]
| >  > Assuming that after linking and loading only one type_info structure is
| >  > active for any particular type symbol
| > 
| > We've found that we can't always rely on that, when people build shared
| > libraries with -B symbolic.  So we fall back on strcmp of the mangled name.
| 
| Is there a -B symbolic option on IA-64? (Is it a base ABI issue, or can
| every vendor do what they please?) If so, that is quite unfortunate I 
| suppose. (Unless we cab "escape" from the -B somehow.)

As a command line option, -B symbolic is up to the linker vendor.
It is (or used to be) specified in the SVID, for whatever that's worth.
The IA-64 base ABI (indeed the new gABI) now has a representation
("protected" symbols) which provide the same semantics on a per-symbol
basis, and should ultimately cut down on the usage of -B symbolic.  This
is useful functionality, and we can't avoid it.

However, I see no reason why we couldn't specify that either (a)
type_info structure names may not be "protected" and they may not be
made so by -B symbolic, or maybe (b) if they are, they are treated as
distinct types in the different DSOs, with whatever consequences that
has.

-	    Jim Dehnert		dehnert at sgi.com
				(650)933-4272