XPost: comp.os.linux.networking   
   From: rick.jones2@hp.com   
      
   Les Cargill  wrote:   
   > Mark wrote:   
   > > given a chip with ARM7 core and L2 8-port switch. There is Linux   
   > > 2.4 kernel ported on this chip. Suppouse there is a kernel routine   
   > > accepting. for example, IGMP packets; my question is - if packets   
   > > come in the switch via different ingress ports simultaneously,   
   > > does it mean that the routine might be invoked also simultaneously   
   > > and cause race conditions conflicts, so semaphores or similar   
   > > mechanisms should be applied ?   
   > >   
      
   > Do you have a senior person on the ground that can help you with   
   > this?  You really need thirty minutes with somebody who understands   
   > this architecture in detail and a whiteboard.   
      
   > If you don't have that, you have a much larger problem and you need   
   > to mind your savings carefully.   
      
   > Short answer is - there is one switch, and it serializes all flows   
   > that terminate at the ARM. That means that all IGMP packets will   
   > either be delivered intact to the ARM or dropped. In terms of   
   > defending the context of an IGMP packet, that depends on the Linux   
   > kernel in question. % find . -name *.[ch] | xargs grep   
   > "CONFIG_IP_MULTICAST" for starters.   
      
   And if he has further questions we probably should point him towards   
   comp.os.linux.networking   
      
   rick jones   
   --   
   firebug n, the idiot who tosses a lit cigarette out his car window   
   these opinions are mine, all mine; HP might not want them anyway... :)   
   feel free to post, OR email to rick.jones2 in hp.com but NOT BOTH...   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)