On Mon, Mar 17, 2014, at 04:38 AM, Michael Haberler wrote:
> please review:
> 
>    
> http://git.linuxcnc.org/gitweb?p=linuxcnc.git;a=shortlog;h=refs/heads/hal-revised-link-semantics
> 
> - Michael
> 

Looks good to me.

The part I'm not able to verify is the pointer stuff.  That SHMPTR mess 
is required because user space components (and halcmd, running in
user space) have different addresses for the same shared memory
locations compared to real-time components in kernel space.  The test
case (if I understand it correctly) has a user space component only.

Have you also tested the behavior with a kernel space component,
like PID?

loadrt pid
show pin pid.0.pgain  (shows compiled default)
setp pin pid.0.pgain 23
show pin pid.0.pgain   (shows 23)
net testnet pid.0.pgain
show sig testnet  (shows 23)
show pin pid.0.pgain  (shows 23)
sets testnet 42
show sig testnet  (shows 42)
show pin pid.0.pgain  (shows 42)
unlinkpin pid.0.pgain
show sig testnet  (shows 42)
show pin pid.0.pgain (shows 42)
sets testnet 55
show sig testnet  (shows 55)
show pin pid.0.pgain (shows 42)

If the above test works, then all is good.  I don't really see a need
to do that as a permanent runtest, just a one-time thing to make 
sure the pointer stuff is right.


-- 
  John Kasunich
  [email protected]

------------------------------------------------------------------------------
Learn Graph Databases - Download FREE O'Reilly Book
"Graph Databases" is the definitive new guide to graph databases and their
applications. Written by three acclaimed leaders in the field,
this first edition is now available. Download your free book today!
http://p.sf.net/sfu/13534_NeoTech
_______________________________________________
Emc-developers mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/emc-developers

Reply via email to