svn commit: r254520 - in head/sys: kern sys

[Andre Oppermann]

On 20.08.2013 00:03, Navdeep Parhar wrote:
> On 08/19/13 14:08, Andre Oppermann wrote:
>> On 19.08.2013 19:40, Peter Grehan wrote:
>>>> I recently tried some experiments to reduce the number of mbuf and
>>>> cluster allocations in a 40G NIC driver.  M_NOFREE and EXT_EXTREF proved
>>>> very useful and the code changes to the kernel were minimal.  See
>>>> user/np/cxl_tuning.  The experiment was quite successful and I was
>>>> planning to bring in most of those changes to HEAD.  I was hoping to get
>>>> some runtime mileage on the approach in general before tweaking the
>>>> ctors/dtors for jumpbo, jumbo9, jumbo16 to allow for an mbuf+refcnt
>>>> within the cluster.  But now M_NOFREE has vanished without a warning...
>>>
>>>    I also had a virtualization work-in-progress where static mbufs were
>>> allocated in the kernel and
>>> M_NOFREE set.
>>>
>>>    Might be worth sending a prior heads-up for these type of changes.
>>
>> I'm sorry for ambushing but this stuff has to be done.  I have provided
>> an alternative way of handling it and I'm happy to help you with your
>> use case to make it good for you and to prevent the mbuf system from
>> getting bloated and hackish again.
>
> I don't know what Peter's use case is but I'm curious about the
> already-available alternative to M_NOFREE, if that's what you meant.
> Can you please elaborate?

When you supply your own (*ext_free) function you can simply omit freeing
the mbuf itself.  Should make sure not to leak it though.

-- 
Andre