Diff for /site/data/goals/Attic/threads.cgi between versions 1.6 and 1.7

version 1.6, 2004/02/22 15:34:35 version 1.7, 2004/03/01 03:39:08
Line 3 Line 3
 # $DragonFly$  # $DragonFly$
   
 $TITLE(DragonFly - Light Weight Kernel Threading Model)  $TITLE(DragonFly - Light Weight Kernel Threading Model)
<CENTER>The Light Weight Kernel Threading Model</CENTER><H1>The Light Weight Kernel Threading Model</H1>
 <P>  <P>
 DragonFly employs a light weight kernel threading (LWKT) model at its core.  DragonFly employs a light weight kernel threading (LWKT) model at its core.
 Each process in the system has an associated thread, and most kernel-only  Each process in the system has an associated thread, and most kernel-only
Line 62  create an abstraction for FAST software Line 62  create an abstraction for FAST software
 well, which will allow traditional hardclock() and statclock() distribution  well, which will allow traditional hardclock() and statclock() distribution
 to operate across all cpus.  to operate across all cpus.
 <P>  <P>
<CENTER>The IPI Messaging Subsystem</CENTER><H1>The IPI Messaging Subsystem</H1>
 <P>  <P>
 The LWKT model implements an asynchronous messaging system for communication  The LWKT model implements an asynchronous messaging system for communication
 between cpus.  Basically you simply make a call providing the target cpu with  between cpus.  Basically you simply make a call providing the target cpu with
Line 77  and then whatever was running before res Line 77  and then whatever was running before res
 are not allowed to block in any manner whatsoever.  IPI messages are used  are not allowed to block in any manner whatsoever.  IPI messages are used
 to do things like schedule threads and free memory belonging to other cpus.  to do things like schedule threads and free memory belonging to other cpus.
 <P>  <P>
<CENTER>The IPI-based CPU Synchronization Subsystem</CENTER><H1>The IPI-based CPU Synchronization Subsystem</H1>
 <P>  <P>
 The LWKT model implements a generalized, machine independant cpu  The LWKT model implements a generalized, machine independant cpu
 synchronization API.  The API may be used to place target cpu(s) into a   synchronization API.  The API may be used to place target cpu(s) into a 
Line 101  synchronization interface operates in a Line 101  synchronization interface operates in a
 back functions tend to work just like the callback functions used in the  back functions tend to work just like the callback functions used in the
 IPI messaging subsystem.  IPI messaging subsystem.
 <P>  <P>
<CENTER>Serializing Tokens</CENTER><H1>Serializing Tokens</H1>
 
 <P>  <P>
 A serializing token may be held by any number of threads simultaneously.  A serializing token may be held by any number of threads simultaneously.
 A thread holding a token is guaranteed that no other thread also  A thread holding a token is guaranteed that no other thread also

Removed from v.1.6  
changed lines
  Added in v.1.7