Appeal 2007-2127
Reexamination Control No. 90/006,621
of the operating system. Since the claims are all directed to "preemptive
multithreading," "cooperative multithreading" is not relevant.
It is argued that one skilled in the art would understand "[t]hat all
major multithreading operating systems implement 'threads' that may
voluntarily relinquish control of the CPU instead of being interrupted or
preempted including IBM's OS/2 operating system, Microsoft's Windows
NT, Windows 95, Windows 98, Windows 2000 and Windows XP, and Sun
Microsystems' Solaris and Java systems (Reiffin, ¶ 32)" (Br. 33).
Again, these arguments are misleading. The fact that a thread can
voluntarily relinquish control if it finishes before the end of its timeslice, i.e.,
before it is interrupted, is irrelevant to the issue of whether a thread has to be
capable of being preempted if it is still executing at the end of its timeslice.
"Preemptive multithreading" requires that all threads are interruptible when
the timeslice expires. To the extent Mr. Reiffin relies on "cooperative
multithreading," such as Solaris, this is not what is claimed.
For the reasons stated above, the argued portions of the Ligler and
Reiffin declarations do not persuade us that threads in a "preemptive
multithreading" system do not have to be interruptible.
c. More that one thread must be interrupted
Patent Owner argues that "multithreading" only requires one thread to
be interrupted (Br. 34-36). It is argued that "multithreading" is defined in
the '603 patent as follows ('603 patent, col. 1, lines 24-38):
The term "multithreading" is used in this application in its
ordinary generally understood sense to mean the concurrent time-
sliced preemptive execution of a plurality of threads of instructions
located within the same single operator-selected application program,
58
Page: Previous 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 Next
Last modified: September 9, 2013