ARM LIMITED

Patent Trials and Appeals Board

Aug 10, 2021

2020002110 (P.T.A.B. Aug. 10, 2021)

UNITED STATES PATENT AND TRADEMARK OFFICE
UNITED STATES DEPARTMENT OF COMMERCE
United States Patent and Trademark Office
Address: COMMISSIONER FOR PATENTS
P.O. Box 1450
Alexandria, Virginia 22313-1450
www.uspto.gov
APPLICATION NO. FILING DATE FIRST NAMED INVENTOR ATTORNEY DOCKET NO. CONFIRMATION NO.
14/953,201 11/27/2015 Alexander Alfred HORNUNG JRL-550-1937 7950
73459 7590 08/10/2021
NIXON & VANDERHYE, P.C.
901 NORTH GLEBE ROAD, 11TH FLOOR
ARLINGTON, VA 22203
EXAMINER
DOMAN, SHAWN
ART UNIT PAPER NUMBER
2183
NOTIFICATION DATE DELIVERY MODE
08/10/2021 ELECTRONIC
Please find below and/or attached an Office communication concerning this application or proceeding.
The time period for reply, if any, is set in the attached communication.
Notice of the Office communication was sent electronically on above-indicated "Notification Date" to the
following e-mail address(es):
PTOMAIL@nixonvan.com
pair_nixon@firsttofile.com
PTOL-90A (Rev. 04/07)
UNITED STATES PATENT AND TRADEMARK OFFICE
BEFORE THE PATENT TRIAL AND APPEAL BOARD
Ex parte ALEXANDER ALFRED HORNUNG and
ADRIAN VIOREL POPESCU
Appeal 2020-002110
Application 14/953,201
Technology Center 2100
Before JOSEPH L. DIXON, JOHN A. EVANS, and JAMES W. DEJMEK,
Administrative Patent Judges.
DIXON, Administrative Patent Judge.
DECISION ON APPEAL
STATEMENT OF THE CASE
Pursuant to 35 U.S.C. § 134(a), Appellant1 appeals from the
Examiner’s decision to reject claims 1–7 and 9–13. Claim 8 has been
canceled. We have jurisdiction under 35 U.S.C. § 6(b).
We AFFIRM.

1 We use the word Appellant to refer to “applicant” as defined in 37 C.F.R.
§ 1.42(a) (2019). Appellant identifies the real party in interest as ARM
Limited. Appeal Br. 3.
Appeal 2020-002110
Application 14/953,201
2
CLAIMED SUBJECT MATTER
The claims are directed to an apparatus and method for branch
prediction. Claim 1, reproduced below, is illustrative of the claimed subject
matter:
1. Apparatus comprising:

branch target storage to store entries comprising branch
instruction source addresses and indications of branch
instruction target addresses, wherein the entries each further
comprise a bias weight and metadata;

history storage to store history-based weights for branch
instruction source addresses, wherein a history-based weight is
dependent on whether a branch to a branch instruction target
address from a branch instruction source address has previously
been taken for at least one previous encounter with the branch
instruction source address; and

prediction generation circuitry to receive the bias weight
and the history-based weight of the branch instruction source
address and to generate a taken-or-not taken prediction that is
either a taken prediction or a not-taken prediction for the
branch,

selection circuitry to receive the taken-or-not taken
prediction from the prediction generation circuitry, the
metadata, and a takenness value generated based on a plurality
of history tables, and to select between the taken-or-not taken
prediction and the takenness value based on the metadata.
Appeal 2020-002110
Application 14/953,201
3
REFERENCES
The prior art relied upon by the Examiner is:
Name Reference Date
Tago et al. US 2002/0199091 A1 Dec. 26, 2002
Cantin et al. US 2013/0007425 A1 Jan. 3, 2013
Williams et al. US 2015/0121050 A1 Apr. 30, 2015
Kozyrakis EE382A Lecture 5: Branch Prediction,
Department of Electrical Engineering,
Stanford University,
http://eeclass.stanford.edu/ee382a
Autumn 2009

REJECTIONS
Claims 1, 5–7, and 11–13 stand rejected under 35 U.S.C. § 103 as
being unpatentable over Tago in view of Cantin in view of Kozyrakis.
Claims 2–4, 9, and 10 stand rejected under 35 U.S.C. § 103 as being
unpatentable over Tago in view of Cantin in view of Kozyrakis in view of
Williams.
OPINION
35 U.S.C. § 103
With respect to claims 1, 5–7, and 11–13, Appellant does not set forth
separate arguments for patentability. Appeal Br. 8. As a result, we select
independent claim 1 as the representative claim for the group. See 37 C.F.R.
§ 41.37(c)(1)(iv). Therefore, claims 1–7 and 9–13 will stand or fall with
representative independent claim 1.2 Arguments that Appellant could have

2 Appellant does not set forth any heading for dependent claims 2–4, 9, and
10, and no separate arguments for patentability of dependent claims 2–4, 9,
and 10.
Appeal 2020-002110
Application 14/953,201
4
made but did not make in the Brief are deemed to be waived. See 37 C.F.R.
§ 41.37(c)(1)(iv).
Applicable Law
“[R]ejections on obviousness grounds cannot be sustained by mere
conclusory statements; instead, there must be some articulated reasoning
with some rational underpinning to support the legal conclusion of
obviousness.” KSR Int’l Co. v. Teleflex Inc., 550 U.S. 398, 418 (2007)
(citing In re Kahn, 441 F.3d 977, 988 (Fed. Cir. 2006)). The Examiner’s
burden of proving non-patentability is by a preponderance of the evidence.
See In re Caveney, 761 F.2d 671, 674 (Fed. Cir. 1985) (“[P]reponderance of
the evidence is the standard that must be met by the PTO in making
rejections.”).
“If all elements of a claim are found in the prior art, as is the case here,
the factfinder must further consider the factual questions of whether a person
of ordinary skill in the art would be motivated to combine those
references . . . .” Dome Patent L.P. v. Lee, 799 F.3d 1372, 1380 (Fed. Cir.
2015); see also KSR, 550 U.S. at 418 (“[I]t can be important to identify a
reason that would have prompted a person of ordinary skill in the relevant
field to combine the elements in the way the claimed new invention does.”).
The PTAB must make a finding of a motivation to combine when it is
disputed. See In re NuVasive, 842 F.3d 1376 (Fed. Cir. 2016). “The factual
inquiry whether to combine references must be thorough and searching and
the need for specificity pervades our authority on the findings on motivation
to combine.” Id. at 1381–82 (bracketed alterations, internal quotation
marks, and citation omitted).
The USPTO “must examine the relevant data and articulate a
satisfactory explanation for its action including a rational connection
Appeal 2020-002110
Application 14/953,201
5
between the facts found and the choice made.” Motor Vehicle Mfrs. Ass’n v.
State Farm Mut. Auto. Ins. Co., 463 U.S. 29, 43 (1983) (internal quotation
marks and citations omitted); see also Synopsys, Inc. v. Mentor Graphics
Corp., 814 F.3d 1309, 1322 (Fed. Cir. 2016) (stating that, as an
administrative agency, the PTAB “must articulate logical and rational
reasons for [its] decisions” (internal quotation marks and citations omitted)).
Where “the problem is known, the possible approaches to solving the
problem are known and finite, and the solution is predictable through use of
a known option,” a solution that is obvious to try may indeed be obvious.
Abbott Labs. v. Sandoz, Inc., 544 F.3d 1341, 1351 (Fed. Cir. 2008) (citing
KSR, 550 U.S. at 421). Moreover, a finding of obviousness under the
“obvious to try” standard “does not require absolute predictability of success
. . . all that is required is a reasonable expectation of success.” In re Kubin,
561 F.3d 1351, 1360 (Fed. Cir. 2009) (emphasis, internal quotation, and
bracketed alteration omitted) (quoting In re O’Farrell, 853 F.2d 894, 903–04
(Fed. Cir. 1988)).
With respect to representative independent claim 1, Appellant argues
that the Tago and Cantin references are fundamentally opposed and there is
no reasonable basis to combine the teachings of the Tago and Cantin
references. Appeal Br. 9; Reply Br. 2.
Tago describes using either a pattern history table (PHT) or a branch
destination buffer (BTB) for branch prediction. Tago ¶ 21. Cantin also
describes use of a history table (i.e., a branch instruction history table, or
BHT) and a branch instruction target address cache (BTAC) for branch
prediction. Cantin ¶¶ 8–9. Cantin further describes branch prediction is
based on a first signal from the history table (BHT), a second signal from the
Appeal 2020-002110
Application 14/953,201
6
branch instruction target address cache (BTAC)—including an override bit
(i.e., a third signal).
Appellant asserts that Tago teaches, for branch prediction purposes,
the PHT has a higher priority to the BTB. Appeal Br. 9 (citing Tago ¶ 41).
By contrast, Appellant argues Cantin teaches the opposite priority—namely,
the signal(s) from the BTAC have a higher priority as compared to the BHT.
Appeal Br. 10. As such, Appellant argues the Examiner’s proposed
combination would render either Tago or Cantin inoperable for their
intended purpose. Appeal Br. 10.
Appellant also argues that the Tago and Cantin references fail to
address the problem solved by the claimed technology. Appeal Br. 11;
Reply Br. 4. Appellant also argues that there is no motivation to combine
the Tago and Cantin references. Appeal Br. 11; Reply Br. 4.
Appellant argues that the two prediction operations cannot be
reasonably be treated separately where the generation and selection
operations cannot reasonably be treated separately. Reply Br. 3.
Finally, Appellant contends that Cantin does not teach the claimed
selection circuitry, and that the three references fail to generate both a
“taken-or-not-taken” value and a “takenness value generated based on a
plurality of history tables” and selecting between them. Appeal Br. 12–13.
We find that the Examiner has addressed all the issues involved in this
appeal. Accordingly, we adopt the Examiner’s factual findings, reasoning,
and conclusions (Final Act. 3–11; Ans. 3–8) as our own. See, e.g., In re
Paulsen, 30 F.3d 1475, 1478 n.6 (Fed. Cir. 1994).
The Examiner addresses each of Appellant’s arguments in the
Examiner’s Answer. Ans. 3–8. The Examiner explains that each of the
Tago and Cantin references teach the use of two known types of information
Appeal 2020-002110
Application 14/953,201
7
and the fact that the priorities used by the two references to select between
multiple types of branch predictions are different is irrelevant as the
priorities are for two different operations and the differing priorities do not
actually represent a conflict. Ans. 3–5. The Examiner further finds that the
Tago and Cantin references address the problem to which the claimed
technology is directed. Ans. 5. With regards to the combination, the
Examiner finds that whether to use indices or full addresses represents one
of the many well-known tradeoffs in the art between size and speed versus
accuracy and that skilled artisans would have been well aware of this
fundamental trade-off. Ans. 5–6. Moreover, the Examiner further finds that
the use of an override bit disclosed by Cantin implicitly discloses the
claimed selection circuitry relying upon MPEP § 2144.01, “in considering
the disclosure of a reference, it is proper to take into account not only
specific teachings of the reference but also the inferences which one skilled
in the art would reasonably be expected to draw therefrom.” Ans. 6.
We agree with the Examiner that there would be implicitly some
circuitry required to generate the override bit, and we further find that the
broadest reasonable interpretation of the claimed selection circuitry does not
further distinguish from that disclosed by the Cantin reference. We further
note claim terms are to be given their broadest reasonable interpretation, as
understood by those of ordinary skill in the art and taking into account
whatever enlightenment may be had from the Specification. In re Morris,
127 F.3d 1048, 1054 (Fed. Cir. 1997). “In the patentability context, claims
are to be given their broadest reasonable interpretations . . . limitations are
not to be read into the claims from the specification.” In re Van Geuns, 988
F.2d 1181, 1184 (Fed. Cir. 1993) (citations omitted).
Appeal 2020-002110
Application 14/953,201
8
The Examiner finds that the Cantin reference discloses evaluating a
value of the override bit, where the override bit is stored in a BTAC entry in
a processor, and selecting which prediction controls in accordance with the
value of the bit. Ans. 6. The Examiner finds a person having ordinary skill
in the art would be reasonably expected to infer that the evaluating of the
override bit and resulting selection are performed in circuitry, as they are
performed by a processor. The Examiner also finds that given that the
override bit is used by circuitry to select which prediction controls, that
circuitry is selection circuitry. Ans. 6. The Examiner further finds that
given that Cantin’s selection circuitry selects between a first branch
prediction and a second branch prediction, the Examiner maintains that
Cantin discloses the claimed selection circuitry to select between the
taken/not taken prediction (a first branch prediction) and a takenness value
(a second branch prediction). Ans. 6–7.
We find that the Examiner’s reliance upon the Tago reference for one
policy/priority and reliance upon the Cantin reference for a second
policy/priority where each reference discloses use of two policies/priorities
to be reasonable and the Examiner provides a line of reasoning thereto. See
Ans. 7.
Additionally, the Examiner responds to Appellant’s arguments and
identifies that the Appellant is arguing “aliasing” but that is not specifically
recited in the language of the claims and that skilled artisan would have
appreciated that there are trade-offs between size and speed versus accuracy
of determinations. Ans. 5–6 (see Appeal Br. 11).
Appellant further argues that the three references failed to generate
both of the values used and selecting between them. See Reply Br. 6–7;
Appeal 2020-002110
Application 14/953,201
9
Appeal Br. 13–14. Appellant also argues that the Examiner is piecing
together the rejection based upon hindsight reconstruction. Reply Br. 6–7.
We disagree with Appellant and find that Appellant is arguing the
individual references rather than the combination as applied by the Examiner
and where the Tago and Cantin references each teach use of two known
values and selection, but the Examiner relies upon a different pairing of the
values and selection of using this different combination of known values. In
reviewing the record here, we are not persuaded that the Examiner’s
rejection is based on improper hindsight because Appellant does not point to
any evidence of record that shows combining the teachings of the cited
references in the manner suggested by the Examiner would have been
“uniquely challenging or difficult for one of ordinary skill in the art” or
would have “represented an unobvious step over the prior art.” Leapfrog
Enters., Inc. v. Fisher-Price, Inc., 485 F.3d 1157, 1162 (Fed. Cir. 2007)
(citing KSR, 550 U.S. at 418).
In addition, Appellant has not identified any knowledge gleaned only
from the present application that was not within the level of ordinary skill at
the time the claimed invention was made. See In re McLaughlin, 443 F.2d
1392, 1395 (CCPA 1971). Moreover, Appellant has not provided any
objective evidence of secondary considerations (e.g., long-felt but unmet
need, commercial success, or unexpected results), which our reviewing court
guides “operates as a beneficial check on hindsight.” Cheese Sys., Inc. v.
Tetra Pak Cheese & Powder Sys., Inc., 725 F.3d 1341, 1352 (Fed. Cir.
2013).
As a result, Appellant’s has not shown error in the Examiner’s factual
findings or the ultimate conclusion of obviousness of representative
Appeal 2020-002110
Application 14/953,201
10
independent claim 1 and independent claim 12 not separately argued and
dependent claims 2–7, 9–11, and 13.
We sustain the Examiner’s obviousness rejections of claims 1–7 and
9–13.
DECISION SUMMARY
In summary:
Claims
Rejected
35 U.S.C.
§
Reference(s)/Basis Affirmed Reversed
1, 5–7, 11–
13
103 Tago, Cantin,
Kozyrakis
1, 5–7, 11–
13

2–4, 9, 10 103 Tago, Cantin,
Kozyrakis,
Williams
2–4, 9, 10
Overall
Outcome
1–7, 9–13

TIME PERIOD FOR RESPONSE
No time period for taking any subsequent action in connection with
this appeal may be extended under 37 C.F.R. § 1.136(a). See 37 C.F.R.
§ 1.136(a)(1)(iv).
AFFIRMED