Ex Parte Agarwala et al

Board of Patent Appeals and Interferences

Aug 30, 2012

10301900 (B.P.A.I. Aug. 30, 2012)

UNITED STATES PATENT AND TRADEMARK OFFICE
________________

BEFORE THE BOARD OF PATENT APPEALS
AND INTERFERENCES
________________

Ex parte MANISHA AGARWALA and JOHN M. JOHNSEN
________________

Appeal 2010-002593
Application 10/301,900
Technology Center 2100
________________

Before MAHSHID D. SAADAT, ERIC S. FRAHM, and
JASON V. MORGAN, Administrative Patent Judges.

MORGAN, Administrative Patent Judge.

DECISION ON APPEAL

Appeal 2010-002593
Application 10/301,900

2
STATEMENT OF THE CASE
Introduction
This is an appeal under 35 U.S.C. § 134(a) from the Examiner’s final
rejection of claims 6 – 18. Claims 1 – 5 are canceled. App. Br. 2. We have
jurisdiction under 35 U.S.C. § 6(b).
We affirm-in-part.
Invention
The invention is in the technical field of emulation hardware for
highly integrated digital signal processing systems. See Spec. 1, ll. 2 – 4.
The invention includes a trace collection method in which a data processor
begins trace data collection even if a trace trigger is received during an
interval when a central processing unit is stalled. Trace data collection is
deferred if a trace trigger is received during an interval of an invalid
instruction boundary until a valid instruction boundary. See Abstract.
Exemplary Claims
6. A method of trace collection, comprising:
executing stall cycles on a data processor;
receiving a trace trigger during an interval in which the
data processor is executing stall cycles;
in response to said trace trigger, commencing collection
of trace data associated with operation of the data processor;
and
providing the collected trace data to support analysis of
operation of the data processor.
(Emphasis added).
7. The method of Claim 6, wherein the collected trace data is
indicative of a length of said interval.
(Emphasis added).
Appeal 2010-002593
Application 10/301,900

3
14. The method of Claim 6, wherein said receiving includes
receiving said trace trigger during a temporal overlap of said
interval and a further interval of an invalid instruction
boundary.
(Emphasis added).
Rejection
The Examiner rejects claims 6 – 18 under 35 U.S.C. § 102(e) as being
anticipated by Wygodny (US 2003/0088854 A1; May 8, 2003; filed Jan. 24,
2002). Ans. 3 – 4.
ISSUES
1. Did the Examiner err in finding that Wygodny discloses “receiving
a trace trigger during an interval in which the data processor is executing
stall cycles,” as recited in claim 6?
2. Did the Examiner err in finding that Wygodny discloses “the
collected trace data is indicative of a length of said interval,” as recited in
claim 7?
3. Did the Examiner err in finding that Wygodny discloses “said
receiving includes receiving said trace trigger during a temporal overlap of
said interval and a further interval of an invalid instruction boundary,” as
recited in claim 14?
ANALYSIS
Claim 6
The Examiner finds that Wygodny, which is directed to a system and
method for conditional tracing of computer programs, discloses all of the
recitations of claim 6, including “receiving a trace trigger during an interval
in which the data processor is executing stall cycles.” See Ans. 3.
Appeal 2010-002593
Application 10/301,900

4
Specifically, the Examiner finds that “an interval in which the data processor
is executing stall cycles” includes the processor executing instructions when
the main program is stalled, such as when an interrupt is taken in Wygodny.
See Ans. 6 – 7. Appellants argue that executing instructions of the
BugTrapper exception handler (i.e., the handler for the INT3 interrupt,
herein referred to as the “breakpoint interrupt”) is not the same as executing
stall cycles. See App. Br. 11. However, we agree with the Examiner that a
broad, but reasonable interpretation of “executing stall cycles” includes
executing instructions when the main program is stalled. See Ans. 6 – 7.
Appellants further contend that the Examiner erred in finding that a
trace trigger is received during this interval because the trace point, the
breakpoint instruction, is executed before the breakpoint interrupt is taken,
as opposed to being executing after the breakpoint interrupt is taken. See
App. Br. 9 – 11. However, we agree with the Examiner that entering the
trace mode, which happens after the breakpoint instruction is executed and
after the breakpoint interrupt is taken, includes “receiving a trace trigger.”
See Ans. 6. In Wygodny, the BugTrapper execution handler determines
whether a breakpoint (i.e., a breakpoint instruction) occurred “as a result of
the tracing or for another reason.” Wygodny ¶ [0198] (emphasis added). It
is “[w]hen an exception is the result of the tracing [that] the handler notifies
the appropriate routines in the tracing library.” Id. (emphasis added). This
notification provides for “receiving a trace trigger.” Since this notification
occurs as part of the BugTrapper execution handler (i.e., after the breakpoint
interrupt is taken and while the processor is executing stall cycles), we agree
with the Examiner that Wygodny discloses “receiving a trace trigger during
Appeal 2010-002593
Application 10/301,900

5
an interval in which the data processor is executing stall cycles,” as recited
in claim 1.
Accordingly, we sustain the Examiner’s 35 U.S.C. § 102(e) rejection
of claim 6. We also sustain the Examiner’s rejection of claim 11 – 13, which
are not argued separately with sufficient specificity. See App. Br. 13 – 15.
Claim 7
Claim 7 recites “wherein the collected trace data is indicative of a
length of said interval.” Appellants argue that Wygodny’s trace data
collection is data associated with executing the client program, not data
associated with executing the interrupt (e.g., the length of the interval in
which the data processor is executing stall cycles). See App. Br. 12.
The Examiner finds that Wygodny describes this limitation because
Wygodny includes a buffer that fills up, thus indicating that the length of the
stall interval was sufficient to fill up the buffer. See Ans. 3 and 7 – 8 (citing
Wygodny ¶ [0194]). However, Wygodny merely describes a circular buffer
storing an execution trace during a time window that is dependent upon the
size of the trace buffer. See Wygodny ¶ [0194]. Such a buffer would fill up
after the time window has been exceeded, leading to older data being
overwritten by newer data. See id. The Examiner has not shown how filling
up this buffer is indicative of the length of the stall cycle. Therefore, we find
that the Examiner erred in finding that Wygodny discloses “wherein the
collected trace data is indicative of a length of said interval,” as recited in
claim 7.
Accordingly, we do not sustain the Examiner’s 35 U.S.C. § 102(e)
rejection of claim 7. We also do not sustain the Examiner’s 35
Appeal 2010-002593
Application 10/301,900

6
U.S.C. § 102(e) rejection of claims 8 – 10, which depend on claim 7 and
thus contain the same disputed limitation.
Claim 14
Claim 14 recites “wherein said receiving includes receiving said trace
trigger during a temporal overlap of said interval and a further interval of an
invalid instruction boundary.” The Examiner finds that Wygodny, which
discloses performing a screen capture when a system crash occurs, describes
this recitation. See Ans. 4; see also Wygodny ¶ [0226].
Appellants contend that the Examiner erred because Wygodny does
not teach the occurrence of the tracing library notification (i.e., the trigger
that occurs during a stall cycle) and an interval of a system crash (i.e., an
interval of an invalid instruction boundary). See App. Br. 16. The Examiner
finds that because Wygodny’s screen capture is disclosed as an example of
conditional tracing, Wygodny describes receiving the trace trigger during the
temporal overlap. See Ans. 8 – 9 (citing Wygodny ¶¶ [0017] and [0240] –
[0253]). However, the Examiner does not show that conditional tracing must
necessarily be triggered during a stall cycle interval. That is, Wygodny
discloses (1) receiving a trace trigger during a stall cycle interval (i.e., when
a breakpoint interrupt is taken because of a breakpoint) and (2) receiving a
trace trigger during an invalid instruction boundary interval (i.e., a system
crash). However, the Examiner does not show that Wygodny describes
receiving a trace trigger during a temporal overlap of a stall cycle interval
and an invalid instruction boundary interval. Therefore, we find that the
Examiner erred in finding that Wygodny describes “wherein said receiving
includes receiving said trace trigger during a temporal overlap of said
Appeal 2010-002593
Application 10/301,900

7
interval and a further interval of an invalid instruction boundary,” as recited
in claim 14.
Accordingly, we do not sustain the Examiner’s 35 U.S.C. § 102(e)
rejection of claim 14. We also do not sustain the Examiner’s 35
U.S.C. § 102(e) rejection of dependent claims 15 – 18, which contain similar
recitations.
DECISION
The Examiner’s decision to reject claims 6 and 11 – 13 is affirmed.
The Examiner’s decision to reject claims 7 – 10 and 14 – 18 is
reversed.
No time period for taking any subsequent action in connection with
this appeal may be extended under 37 C.F.R. § 1.136(a)(1)(iv).

AFFIRMED-IN-PART

ELD