Andrew Beverley wrote:
My latest revision of the patch for netfilter marking will follow soon.
Before I post it, I will reply to comments on the previous version.
Omitting bits I'm happy with.
Question number 2: what is stubQosConfig.cc? Does that also need
updating for this patch?
I've added stubs for all the non-inline functions. They just all call
fatal(). If I've not got this quite right then please let me know.
Fine.
I've moved what I consider to be the most appropriate functions inline.
If you disagree with my choices then please tell me.
Okay. will see when the patch arrives. Common sense is usually the key.
<snip>
The reconfigure straightening will change your dependency logic model a
bit. Specifically such that the absence of --with-netfilter-conntrack
(implicit / auto-detect case) results in the same auto-detect currently
only done by explicit "yes" results, but which does not terminate in
AC_MSG_ERROR.
Done.
The above configure concept would tie in with removing the --enable-qos
option altogether. There's no reason for the QOS code not to be included
that I can see (it has no dependencies, apart from the optional upstream
kernel patch), and with this patch and the isTosActive(), it's enabled
at runtime only when needed anyway. Is there any reason to keep the
option?
This argument has bee put forward for other small features. Sometimes it
succeeds, sometimes not. I know that a mere 1KB saving on binary size
can be extremely useful for embeded devices so the option is likely to
be wanted by somebody.
That said the device I see most wanting cut-down is the WRT where your
feature is probably on the keeper list of those to build.
Amos
--
Please be using
Current Stable Squid 2.7.STABLE9 or 3.1.8
Beta testers wanted for 3.2.0.2
