On Mon, 15 Aug 2005, Oleg Nesterov wrote:
>
> The new patch tries to solve this, but I am very new to sparse,
> so I can't understand all implications of adding MOD_TYPEDEF to
> SYM_STRUCT->ctype.modifiers, and it looks ugly to me.
Yes, that looks more dangerous than your previous change.
The perfect solution would be to just carry the whole symbol chain along
with us, so that we could go up in the symbol stack and just report the
deepest entry that has a name. However, when we peel off one level of
symbol information, we totally just discard it (and we can't have
back-pointers, because a type can obviously be reachable through many
different other types).
Anyway, I already applied you previous patch, because it seemed to be an
improvement even if technically wrong.
Linus
-
To unsubscribe from this list: send the line "unsubscribe linux-sparse" in
the body of a message to [EMAIL PROTECTED]
More majordomo info at http://vger.kernel.org/majordomo-info.html
