Appeal No. 2002-0792 Application 08/801,646 allocating, at the user level scheduler, one or more threads of the plurality of threads to each of the processors assigned to the program, wherein the step of allocating includes the step of reading register context from one of the plurality of register save areas. For the reasons as we have pointed out above, we fail to find that Polychronopoulos or Anderson teaches the above limitations. For claims 27, 40 and 52, Appellants argue that the Examiner has failed to show how the references cite and teach or suggest all the limitations recited in these claims. In particular, Appellants argue that the Examiner has not shown how the references teach the use of a first and second kernel of scheduler to schedule threads across a first and second kernel. We note that Appellants’ claim 23 recites: a kernel level scheduler, executing within the protected kernel mode of one of the plurality of processors, for allocating processors to a program; first program code executing in one of the plurality of processors for creating a shared arena within the memory, wherein the shared arena includes a register save area for each of the plurality of threads; and second program code executing in the processor, wherein the second program code includes scheduling code for scheduling threads from the plurality of threads, wherein the scheduling code includes program code for selecting a thread from the plurality of threads and for switching to the selected thread by reading register context associated with the selected thread from one of the plurality of register save areas. 9Page: Previous 1 2 3 4 5 6 7 8 9 10 11 12 13 NextLast modified: November 3, 2007