Advice on a multithreaded netisr patch?

[Robert Watson]

On Mon, 6 Apr 2009, Julian Elischer wrote:

> while this is true, m_pullup ALWAYS does things so in fact you want to 
> always put it in a test to see if it is really needed..

Then m_pullup() should be fixed?  Keeping the expression of the pullup short 
makes the network code a lot more compact, which is a significant benefit.

Robert N M Watson
Computer Laboratory
University of Cambridge

>
> from memory it is something like:
>
> if (m->m_len < headerlen && (m = m_pullup(m, headerlen)) == NULL) {
>       log(LOG_WARNING,
>          "nglmi: m_pullup failed for %d bytes\n", headerlen);
>             return (0);
> }
> header = mtod(m, struct header *);
>
>
>>> 
>>> m_pullup() here ensures that the first sizeof(*w) bytes of mbuf data are
>>> contiguously stored so that the cast of w to m's data will point at a
>>> complete structure we can use to interpret packet data.  In the common 
>>> case
>>> in the receipt path, m_pullup() should be a no-op, since almost all 
>>> drivers
>>> receive data in a single cluster.
>>> 
>>> However, there are cases where it might not happen, such as loopback 
>>> traffic
>>> where unusual encapsulation is used, leading to a call to M_PREPEND() that
>>> inserts a new mbuf on the front of the chain, which is later m_defrag()'d
>>> leading to a higher level header crossing a boundary or the like.
>>> 
>>> This issue is almost entirely independent from things like the cache line
>>> miss issue, unless you hit the uncommon case of having to do work in
>>> m_pullup(), in which case life sucks.
>>> 
>>> It would be useful to use DTrace to profile a number of the workfull 
>>> m_foo()
>>> functions to make sure we're not hitting them in normal workloads, btw.
>> 
>> I highly suspect m_pullup will take any real effect on RX path, given
>> how most of drivers allocate the mbuf for RX ring (all RX mbufs should
>> be mclusters).
>> 
>> Best Regards,
>> sephe
>> 
>
>