Looks good besides what kib@ already mentioned.

I see, thanks for the explanation. It would be perfect to add a short version of it to the commit message as well.

This looks good, but I'm a little bit confused as to how it is different from rc.local? Does it provide anything which can't be achieved by using rc.local which seems to serve the same purpose?

Looks good besides the SYSCTL_INT vs UINT issue.

Looks good to me.

Gleb, if we real world testing is needed we'd be happy to test it in our environment.

Another thing we discovered is that FreeBSD actually used to handle this case correctly up until r182851 where the relevant check was dropped by accident during the mtu handling refactoring. It used to check for the received MTU size and return early:

if (tp->t_maxopd <= mss)
    return (inp);

Looks good, thanks!

Looks good, mostly nitpicks for now. Will look deeper tonight.

This was submitted as r282866.

In D2518#46456, @jhb wrote:

Ok. I think the longer term fix is to update the MAP_IN message to include the relative offset of the mapping to the backing object, then it wouldn't have to parse the header to see what offset of the file is mapped.

Example running unixbench:

With this change pmcstat(8) still does not resolve symbols for some applications, but at least they show up in pmcstat -T and in the callgraph (before that pmcstat would mark their frames as "dubious").

Looks good. Besides things ngie already commented on I don't see any issues.

Looks good, thanks!