Strangeness in 842 Fix

[Fred Wright]

On Mon, 14 Apr 2025, Hal Murray wrote:

>> I do know that the SO_* symbols are for the socket options and the SCM_*
>> symbols are for the CMSG types, so I don't see how this could possibly be
>>  correct.  Note the code immediately above it.
>
> The OLD stuff is a mess.
>
> I did it the way you expect, but that didn't work because
> SCM_TIMESTAMPNS_OLD didn't exist.
>
> All the SCM symbols are setup useing the SO sysmbols.
>  #define SCM_foo SO_foo
> So I use the SO_ symbols.
>
> The code works for the test case and still works on quite a few normal
> cases.

If the values are identical, then it's not functionally incorrect, but 
it's certainly conceptually incorrect to compare an SO_* value to a 
cmsg_type field.  And if the values are identical, it wouldn't change the 
behavior to use the correct name.

Fred Wright