Re: ntp_init() looks like a nop

On Tue, Feb 8, 2022 at 12:33 AM Warner Losh <imp@bsdimp.com> wrote:

>
>
> On Mon, Feb 7, 2022 at 1:34 PM Poul-Henning Kamp <phk@phk.freebsd.dk>
> wrote:
>
>> --------
>> Sebastian Huber writes:
>>
>> > The ntp_init() function sets a couple of global variables to zero.
>> These=20
>> > variables should be in the .bss section. Are they not already cleared=20
>> > during the kernel loading?
>>
>> They are, but I kept the code to minimize the diff relative to Dave Mills
>> original version of the code.
>>
>> It can go no.
>>
>
> I think I lost the email race. I just pushed Sebastian's changes. Is this
> a request to not do this? If so, I'll revert.
>
> Warner
>



My reply is not an answer to your question , but I want to make a remark .


I am developing a software about "A multi-media information management
system"
( a continuation of my PhD thesis demonstration program ) having around  12
000
Pascal procedures .


For  "Record"-s , I am doing the following :

By using a script system , I am generating many Procedures   about
operations on that
"Record" , such as :
(1) Allocate
(2) Initialize
(3) ...

(n) Dispose


In your case , it is said that "in another part this record is initialized
..."

Assume that ,   "another part" is modified to ignore this initialization ,
and there is not
a call to  an initializer .

In Turkish , this is called "To load a liver to a cat"  ( I am sorry to say
that ) .

To prevent such disastrous possibilities ( after losing significant times
to understand
what is the reason ) I have developed such an approach .


My ideas are like that .



Obviously you know much better than me how to develop FreeBSD .


With my best wishes ,


Mehmet Erol Sanliturk