[PATCH] Add PRIV_TOPREQ for per "top request" / req->top state

[Poul-Henning Kamp]

--------
In message <54FDDE5E.5090303 at schokola.de>, Nils Goroll writes:

>+* ``PRIV_TOPREQ`` "per top-request" private pointers live for the
>+  duration of one request and all its ESI-includes. They are only
>+  defined for the client side. When used from backend VCL subs, a NULL
>+  pointer will be passed.

So how does locking of a PRIV_TOPREQ work if we implement parallel-ESI ?

We might have two different child-fetches running in parallel, asking for
a PRIV_TOPREQ at the exact same instant from the same VMOD ?

-- 
Poul-Henning Kamp       | UNIX since Zilog Zeus 3.20
phk at FreeBSD.ORG         | TCP/IP since RFC 956
FreeBSD committer       | BSD since 4.3-tahoe    
Never attribute to malice what can adequately be explained by incompetence.