https://bz.apache.org/bugzilla/show_bug.cgi?id=63503

--- Comment #20 from Yann Ylavic <[email protected]> ---
(In reply to Ruediger Pluem from comment #19)
> Any feedback on the proposed approach? If the concept looks fine I would
> create a merged patch for testing.
I share the same analysis, DNS resolution races with constructors hence pool
corruption and further crash.

Regarding the volatile(s) added to structs, I think the compatibility issue can
be addressed by using volatile accesses instead of volatile declaration, i.e.
define something like:

#define AP_VOLATILIZE_T(T, x) (*(T volatile *)&(x))

and use it wherever we access worker->local_status or worker->cp->addr (which
then need to be declared volatile).

-- 
You are receiving this mail because:
You are the assignee for the bug.
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to