When and when not to use CTLFLAG_MPSAFE with the SYSCTL macros..?

[NGie Cooper]

On Fri, Aug 7, 2015 at 1:53 PM, Oliver Pinter
<oliver.pinter at hardenedbsd.org> wrote:
> On 8/7/15, John Baldwin <jhb at freebsd.org> wrote:
>> On Wednesday, August 05, 2015 08:38:18 AM Hans Petter Selasky wrote:
>>> On 08/05/15 00:47, Garrett Cooper wrote:
>>> > Hi,
>>> >    I’ve been trying to figure out (because sysctl(9) is lacking) when to
>>> > use CTLFLAG_MPSAFE. Is it strictly when dealing with SYSCTL_PROC
>>> > handlers that do proper locking of shared resources, or are there other
>>> > nuances that need to be handled?
>>> >    I’m also asking because SYSCTL_UQUAD, for instance, explicitly uses
>>> > CTLFLAG_MPSAFE in the handler, which is a bit confusing.
>>> > Thanks!
>>> > -NGie
>>>
>>> Hi,
>>>
>>> This flag decides if you have Giant automatically locked or not around
>>> the sysctl proc. Your functions should have their own locks basically. I
>>> believe it is a leftover from many years ago, when the FreeBSD kernel
>>> was going multi threaded.
>>
>> It's only about 2-3 years old actually.
>>
>> To answer your question Garrett: yes it is really only for SYSCTL_PROC
>> handlers.  The existing "simple" handlers like sysctl_handle_int are
>> as atomic as they can be regardless of Giant, so they don't need Giant.
>
> For the basic types the CTLFLAG_MPSAFE flag are always added:
> https://github.com/freebsd/freebsd/blob/master/sys/sys/sysctl.h#L322 .
>
>>
>> If you have a variable that you want to control access to via locking you
>> need to use a custom handler, even if it is a simple int.

Thank you, all -- I see it now.
-NGie