Appeal 2007-2127
Reexamination Control No. 90/006,621
it finished executing. This is neither disclosed nor contemplated. Even if
the editor was interruptible, e.g., if a programmer inserted an enable
interrupt (EI) instruction at the beginning of the editor interrupt service
routine (see previous description of why the editor is not interruptible), a
clock-activated interrupt would always cause the system to go to the editor
interrupt service routine (the interrupt would interrupt the interrupt), never
back to the compiler. It is not clear that interrupting the editor in the middle
of its operation to run another instance of the editor would even work. But,
in any case, there is no switching back and forth between the compiler and
the editor as required for "preemptive multithreading." For this additional
reason, we find that there is no written description of "preemptive
multithreading" in the 1982, 1985, or 1990 applications, and, therefore, the
'604 patent is not entitled to the priority filing date of the 1982 application.
e. Editor and compiler are not executed concurrently
The 1982 application also fails to disclose "multithreading" because
the editor and compiler programs do not execute "concurrently" in the
technical sense required by the definition of "multithreading." The '604
patent's definition of "multithreading" requires "concurrent . . . execution of
a plurality of threads of instructions located within the same software
program." "Concurrent execution" means that, at a given time, two or more
threads within the same program are executing at some point between their
starting and finishing points at the same time, not just that two sets of
instructions execute closely in time.. If there are two threads and two
processors, both threads could run simultaneously (at exactly the same time);
however, because there is only one processor, the threads take turns
75
Page: Previous 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 Next
Last modified: September 9, 2013