Linux Kernel Network Subsystem Patching

[Lennart Sorensen]

On Sat, Feb 01, 2014 at 07:01:23PM -0500, David Thornton wrote:
> But I think your missing the point of the question: "What is the effect of
> hyperthreading versus multiple core, with respect to dead locks?". I think
> the answer is nothing. From a dead lock perspective, it's the same problem
> : O'Leary is dead and O'reily don't know it, O'reily is dead and O'Leary don't
> know it ( http://www.youtube.com/watch?v=YZWpffswMnM ).  Two threads /
> Processes depend on each-other to get something done,
> http://en.wikipedia.org/wiki/Deadlock . It doesn't matter of those two
> threads / process are in their own thread or core, they are still waiting
> on each other.

Yes it certainly is the same.  SMT versus SMP makes no difference to
software running on the system as far as dead locks are concerned.
You still have multiple things happening at the same time.

-- 
Len Sorensen
--
The Toronto Linux Users Group.      Meetings: http://gtalug.org/
TLUG requests: Linux topics, No HTML, wrap text below 80 columns
How to UNSUBSCRIBE: http://gtalug.org/wiki/Mailing_lists