Ex Parte Canut et al

Board of Patent Appeals and Interferences

Feb 26, 2009

09765916 (B.P.A.I. Feb. 26, 2009)

UNITED STATES PATENT AND TRADEMARK OFFICE

__________

BEFORE THE BOARD OF PATENT APPEALS
AND INTERFERENCES
__________

Ex parte FREDERIC CANUT and MUSTAPHA DERRAS
__________

Appeal 2008-3796
Application 09/765,916
Technology Center 2100
__________

Decided:1 February 27, 2009
__________

Before JAMES D. THOMAS, JOSEPH L. DIXON, and
ST. JOHN COURTENAY III, Administrative Patent Judges.

COURTENAY, Administrative Patent Judge.

DECISION ON APPEAL

1 The two-month time period for filing an appeal or commencing a civil
action, as recited in 37 CFR § 1.304, begins to run from the decided date
shown on this page of the decision. The time period does not run from the
Mail Date (paper delivery) or Notification Data (electronic delivery).

Appeal 2008-3796
Application 09/765,916

2
This is a decision on appeal under 35 U.S.C. § 134(a) from the
Examiner’s rejection of claims 1-26. We have jurisdiction under 35 U.S.C.
§ 6(b). We affirm.
STATEMENT OF THE CASE
THE INVENTION
Appellants’ invention relates to the field of software development.
More particularly, Appellants’ invention relates to a system and methods for
software code optimization. (Spec. 1).
Independent claim 1 is illustrative:
1. A method of optimizing a software program for a target
processor to meet performance objectives, where the software
program is coded in a high-level language, the method
comprising the steps of:

(a) optimizing the software program such that a resulting
first optimized form of the software program is completely
independent of the target processor and is at least partially
coded in the high-level language, determining a first
performance profile for the first optimized form of the software
program, and comparing the first performance profile with the
performance objectives;

(b) based on the results of comparing the first
performance profile with the performance objectives, if the
performance objectives are not met by the first optimized form
of the software program, then optimizing the first optimized
form of the software program such that a resulting second
optimized form of the software program includes at least one
portion that is dependent on the target processor and is coded in
the high-level language; and

Appeal 2008-3796
Application 09/765,916

3
(c) flagging the at least one portion to indicate that the at
least one portion is dependent on the target processor if the first
optimized form of the software program is optimized to create
the second optimized form of the software program.

PRIOR ART
Pieper US 2003/0005419 A1 Jan. 2, 2003

Cain et al., "Portable Software Library Optimization," MITRE
TECHNICAL REPORT, Feb., 1998. (hereinafter "Cain").

Kum et al., “A Floating-Point To Integer C Converter with Shift
Reduction for Fixed-Point Digital Signal Processors,” IEEE, 0-7803-
5041-3/99, 1999. (hereinafter "Kum").

THE REJECTIONS
1. Claims 1-3, 8-16, and 21-26 stand rejected under 35 U.S.C. § 103(a)
as being unpatentable over the combination of Pieper and Cain.

2. Claims 4-7 and 17-20 stand rejected under 35 U.S.C. § 103(a) as
being unpatentable over the combination of Pieper, Cain, and Kum.

GROUPING OF CLAIMS
In the principal Brief, Appellants present arguments for independent
claim 1 as being representative of claims 2-3, 8-16, and 21-26 (App. Br. 6-
7). Regarding the second obviousness rejection, Appellants submit that
“claims 4-7 and 17-20 are patentable over Pieper in view of Cain and Kum
under 35 U.S.C. §103 due to their respective dependence on present
independent claims 1 and 14 . . . .” (App. Br. 7). We will, therefore, treat
claims 1-26 as standing or falling with representative claim 1. See 37 C.F.R.
§ 41.37(c)(1)(vii).
Appeal 2008-3796
Application 09/765,916

4
APPELLANTS’ CONTENTIONS
Regarding representative claim 1, Appellants contend that Pieper does
not disclose or suggest the limitation of “a resulting first optimized form of
the software program is completely independent of the target processor and
is at least partially coded in the high-level language.” (App. Br. 5). In
particular, Appellants contend that Pieper’s teaching of optimized code 60
(that is substantially independent of the architecture of the target processor)
does not meet the language of claim 1 that requires a first optimized form of
the software program that is completely independent of the target processor
(App. Br 5; see also Pieper, para. [0030]).

EXAMINER’S RESPONSE
The Examiner avers that “[o]ne skilled in the pertinent art
would know that a high-level language compiler forms intermediate code
which is machine independent and optimizes the intermediate code in the
middle end of the compilation stage, while formulating the target object
code, which is also optimized by using machine-specific optimizations at a
later stage of the compilation.” (Ans. 11, ¶2, ll. 4-8).

ISSUE
Based upon our review of the administrative record, we have
determined that the following issue is dispositive in this appeal:
Have Appellants shown the Examiner erred in finding that the
combination of Pieper and Cain teaches and/or suggests the argued
limitation of "a resulting first optimized form of the software
Appeal 2008-3796
Application 09/765,916

5
program is completely independent of the target processor?" (See
independent claims 1 and 14).

PRINCIPLES OF LAW
“What matters is the objective reach of the claim. If the claim extends
to what is obvious, it is invalid under § 103.” KSR Int’l Co. v. Teleflex, Inc.,
127 S. Ct. 1727, 1742 (2007). To be nonobvious, an improvement must be
“more than the predictable use of prior art elements according to their
established functions.” Id. at 1740.

FINDINGS OF FACT
In our analysis infra, we rely on the following findings of fact (FF)
that are supported by a preponderance of the evidence:

THE PRIMARY PIEPER REFERENCE
1. Pieper teaches that “[c]omputer programmers typically develop
computer programs for conventional processors using relatively
high-level source code computer languages (e.g., C++, Pascal,
Fortran, etc.).” (Para. [0009]).
2. Pieper teaches that “During this compilation process, the
prefetch instructions may be explicitly inserted into an
intermediate level, machine-independent code that is first
generated by the process from the input source code. (Para.
[0020]).
Appeal 2008-3796
Application 09/765,916

6
3. Pieper teaches that “[a] later machine code-generation process
may then translate/convert the intermediate level code,
including the prefetch instructions, into machine-specific
program instructions that are intended to be executed by the
target processor.” (Para. [0020]).

THE SECONDARY CAIN REFERENCE
4. Cain teaches "The BLAS library has been written so that it is
portable to several platforms. It has also been optimized to the
extent possible in high-level source code." (Page 1, ¶ 2, ll. 6-8,
emphasis added).

ANALYSIS
We decide the question of whether Appellants have shown the
Examiner erred in finding that the combination of Pieper and Cain teaches
and/or suggests the argued limitation of "a resulting first optimized form of
the software program is completely independent of the target processor."
(See independent claims 1 and 14).
In the principal Brief, Appellants point to page 6, lines 7-8 of the
Specification as providing support for the argued “first optimized form of
the software program [that] is completely independent of the target
processor" limitation. (App. Br. 3).
Lines 7-8 of page 6 of Appellants’ Specification disclose: "Preferably,
the programming language is one that is completely portable between all
Appeal 2008-3796
Application 09/765,916

7
probable DSP targets." To put this statement in context, for convenience we
reproduce lines 3-11 of page 6 of Appellants’ Specification, as follows:

FIG. 1 is a flow diagram generally depicting steps in a process
of optimizing a software program for a target processor. In the
embodiment represented in FIG. 1, an optimization process 100
comprises three optimization steps. In a first optimization step
102, the software for the DSP processor target is written in a
high-level language such as C, C+ + or Ada. Preferably, the
programming language is one that is completely portable
between all probable DSP targets. In coding the software,
optimization techniques particular to the language are
preferably used. The code optimization in this step 102
preferably does not employ any optimization tools that depend
on the processor that is meant to host the application.
(Spec. 6, ll. 3-11).

We particularly note that the aforementioned portion of the
Specification discloses a first optimization step 102 where software for the
target processor is written in a high level language where “[i]n coding the
software, optimization techniques particular to the language are preferably
used.” (Spec. ll. 8-9). Thus, as supported by Appellants’ Specification, the
claimed “first optimized form of the software” is created by writing code in
a high-level language using target independent optimization techniques
particular to the language.
However, the Examiner counters that “[o]ne skilled in the pertinent art
would know that a high-level language compiler forms intermediate code
which is machine independent and optimizes the intermediate code in the
middle end of the compilation stage, while formulating the target object
Appeal 2008-3796
Application 09/765,916

8

code, which is also optimized by using machine-specific optimizations at a
later stage of the compilation. (Ans. 11, ¶2, ll. 4-8).
We find the Examiner’s statement is supported by Pieper’s teaching of
“machine-independent code that is first generated by the process from the
input source code . . . .” (FF. 2). Pieper also teaches the use of high-level
languages (FF 1). Moreover, Pieper teaches that “[a] later machine code-
generation process may then translate/convert the intermediate level code,
including the prefetch instructions, into machine-specific program
instructions that are intended to be executed by the target processor.”
(FF 3).
However, from our review of Pieper, we agree with Appellants that
the machine independent code in Pieper is not the output of optimizer
processes 58 that is clearly shown in Figure 2 as receiving an input from
target machine model 80. Thus, the output of Pieper’s optimizer 58 is code
optimized for a particular target machine. We also agree with Appellants
that intermediate program code that is substantially independent of the target
processor architecture is not completely independent of the target processor
(see Pieper, ¶ [0030]).
Nevertheless, we examine the totality of the evidence before us for
indicia of obviousness.2 In particular, we find the secondary Cain reference

2 “Invention or discovery is the requirement which constitutes the
foundation of the right to obtain a patent . . . unless more ingenuity and skill
were required in making or applying the said improvement than are
possessed by an ordinary mechanic acquainted with the business, there is an
absence of that degree of skill and ingenuity which constitute the essential
Appeal 2008-3796
Application 09/765,916

9

clearly teaches a library of portable software programs that have been
optimized in high-level source code: "The BLAS library has been written so
that it is portable to several platforms. It has also been optimized to the
extent possible in high-level source code." (FF 4).
Appellants have pointed to the disclosure in the Specification of a
completely portable programming language (Spec. 6, lines 7-8) as providing
support for the argued first optimized form of the software program that is
completely independent of the target processor. (App. Br. 3). Thus, when
we apply a broad but reasonable construction fully consistent with
Appellants’ Specification, we find Cain’s portable, optimized high-level
source code clearly teaches the argued limitations of “a resulting first
optimized form of the software program [that] is completely independent of
the target processor and is at least partially coded in the high-level
language.” (Claim 1).
While our reading of Appellants’ claim 1 on the references has
departed from some aspects of the Examiner’s reading, it is our view that
Appellants have been given full and fair notice of the references. Appellants
are responsible for all that is disclosed by the collective teachings of the
cited references. See In re Zenitz, 333 F.2d 924, 926 (CCPA 1964) (“This
court has held in a number of decisions that a United States patent speaks for

elements of every invention.” Dunbar v. Myers, 94 U.S. 187, 197 (1876)
(citing Hotchkiss v. Greenwood, 52 U.S. 248, 267 (1850)). We note that
Hotchkiss v. Greenwood was cited with approval by the Supreme Court in
KSR, 127 S. Ct. at 1734, 1739, 1746.

Appeal 2008-3796
Application 09/765,916

10
all it discloses as of its filing date, even when used in combination with other
references.”). Our reviewing court has stated: “[t]he use of patents as
references is not limited to what the patentees describe as their own
inventions or to the problems with which they are concerned. They are part
of the literature of the art, relevant for all they contain.” In re Heck, 699
F.2d 1331, 1333 (Fed. Cir. 1983) (quoting In re Lemelson, 397 F.2d 1006,
1009 (CCPA 1968)). It is our view that this reasoning is equally applicable
to non patent references, such as the secondary Cain reference we have
addressed supra.
Even though, in some instances, we sustain rejections for different
reasons than those advanced by the Examiner, our position is still based
upon the collective teachings of the references and does not constitute a new
ground of rejection. See In re Bush, 296 F.2d 491, 496 (CCPA 1961); In re
Boyer, 363 F.2d 455, 458 n.2 (CCPA 1966).
For at least the aforementioned reasons, it is our view that the
Examiner has set forth a sufficient initial showing of obviousness, and we
find that Appellants have not shown error in the Examiner’s rejection of
representative claim 1 as being unpatentable over the combination of Pieper
and Cain. Accordingly, we sustain the Examiner’s rejection of
representative claim 1, and of claims 2-26 which fall therewith.

CONCLUSION
Based on the findings of facts and analysis above, we conclude that
Appellants has not established that the Examiner erred in rejecting claims
1-26 under 35 U.S.C. § 103(a) for obviousness.
Appeal 2008-3796
Application 09/765,916

11
DECISION
We affirm the Examiner’s decision rejecting claims 1-26.
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

pgc

VISTA IP LAW GROUP LLP
1885 Lundy Avenue
Suite 108
SAN JOSE, CA 95131