Re: [gpsd-dev] Updated docs on NTP segment management

[Hal Murray]

address@hidden said:
> I see no reason to make it not writeable by the client.  So no need, yet,
> for a compatibility break. 

The current setup doesn't work with more than one client.  I can't run a 
debugging monitor in parallel with ntpd.

It also simplifies security.  If the client side is read-only, a layer of 
attacks go away without any sysadmin setup complications.

-------

> Sadly read order can not be fixed on many architectures, like x86.  
> So X and Y may have been flushed by the cache to main RAM, but not the data
> it is guarding. 

I'm not a wizard on this area.  I have worked with people who are.  When I 
asked one of them, he gave me that recipe.  When I asked about cache flushes 
and such, he said that x86 hardware does the right thing.

I think the actual code needs volatiles in the right places (so the compiler 
will do the right thing) and a few memory barriers on some architectures.

Do you know of a good web page covering this area?  I've looked before 
without finding anything.  I just tried again, this looks like great 
background, but I haven't found the section that discusses specific 
architectures.
  Documentation/memory-barriers.txt
I think that's part of the kernel sources.


-- 
These are my opinions.  I hate spam.