Ex Parte BITAR et al - Page 8




          Appeal No. 2002-0792                                                        
          Application 08/801,646                                                      


          with the Examiner that the suggestion to use shared memory                  
          between the user-level scheduler and the kernel as a                        
          communication venue is of interest.  However, we fail to find               
          that Polychronopoulos has described any further details of such a           
          communication mechanism other than this one suggestion.                     
          Therefore, we fail to find that Polychronopoulos teaches the                
          above limitations as recited in Appellants’ claims 1, 19 and 23.            
               Appellants also argue that Polychronopoulos and Anderson               
          fail to teach the use of shared arena as a communication                    
          mechanism for conveying register context in the scheduling of               
          threads and the use of a number requested variable and a number             
          allocated variable, both of which are stored in the shared arena,           
          to allocate threads to the processor as recited in Appellants’              
          claim 13.  See page 10 of the brief.                                        
               We note that Appellants’ claim 13 recite:                              
                    defining a shared arena within the memory, wherein the            
               shared arena includes a register save area for each of the             
               plurality of threads;                                                  
                    starting the program, wherein the step of starting the            
               program includes the step of setting, via the user level               
               scheduler, a number requested variable within the shared               
               arena requesting that one or more processors from the                  
               plurality of processors be assigned to the program and                 
               setting, via the kernel level scheduler, a number allocated            
               variable within the shared arena indicating how many                   
               processors from the plurality of processors are assigned to            
               the program; and                                                       
                                          8                                           





Page:  Previous  1  2  3  4  5  6  7  8  9  10  11  12  13  Next 

Last modified: November 3, 2007