Oracle America, Inc. v. Google Inc.
Filing
509
DECLARATION of RUCHIKA AGRAWAL in Opposition to #496 MOTION in Limine No. 5, #494 MOTION in Limine No. 3, #492 MOTION in Limine No. 1, #493 MOTION in Limine NO. 2, #495 MOTION in Limine No. 4 filed byGoogle Inc.. (Attachments: #1 Exhibit 1-1, #2 Exhibit 1-2, #3 Exhibit 1-3, #4 Exhibit 1-4, #5 Exhibit 1-5, #6 Exhibit 1-6, #7 Exhibit 1-7, #8 Exhibit 1-8, #9 Exhibit 1-9, #10 Exhibit 1-10, #11 Exhibit 1-11, #12 Exhibit 1-12, #13 Exhibit 2-1, #14 Exhibit 2-2, #15 Exhibit 2-3, #16 Exhibit 2-4, #17 Exhibit 2-5, #18 Exhibit 2-6, #19 Exhibit 2-7, #20 Exhibit 2-8, #21 Exhibit 2-9, #22 Exhibit 2-10, #23 Exhibit 2-11, #24 Exhibit 2-12, #25 Exhibit 2-13, #26 Exhibit 2-14, #27 Exhibit 2-15, #28 Exhibit 2-16, #29 Exhibit 2-17, #30 Exhibit 3-1, #31 Exhibit 3-2, #32 Exhibit 3-3, #33 Exhibit 3-4, #34 Exhibit 3-5, #35 Exhibit 3-6, #36 Exhibit 3-7, #37 Exhibit 3-8, #38 Exhibit 3-9, #39 Exhibit 3-10, #40 Exhibit 3-11, #41 Exhibit 4-1, #42 Exhibit 4-2, #43 Exhibit 4-3, #44 Exhibit 5-1, #45 Exhibit 5-2, #46 Exhibit 5-3, #47 Supplement 5-4)(Related document(s) #496 , #494 , #492 , #493 , #495 ) (Kamber, Matthias) (Filed on 10/7/2011)
EXHIBIT 2-10
Attorneys' Eyes Only
1
UNITED STATES DISTRICT COURT
2
NORTHERN DISTRICT OF CALIFORNIA
3
SAN FRANCISCO DIVISION
4
5
------------------------
6
ORACLE AMERICA, INC.,
)
7
Plaintiff,
)
8
vs.
) No. CV 10-03561 WHA
9
GOOGLE, INC.,
)
10
11
Defendant.
)
------------------------
12
13
ATTORNEYS' EYES ONLY
14
15
Videotaped Deposition of EREZ LANDAU,
16
taken at 42 Chauncy Street, Boston,
17
Massachusetts, commencing at 9:50 a.m.,
18
Wednesday, September 14, 2011, before
19
Jill Shepherd, RPR, MA-CSR No. 148608,
20
NH-CSR No. 128, CA-CSR No. 13275, CLR,
21
Notary Public.
22
23
24
25
PAGES 1 - 113
Page 1
Veritext National Deposition & Litigation Services
866 299-5127
EXHIBIT 2-10
Attorneys' Eyes Only
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
tape two. We are back on the record.
Q. To continue where we left off before the
break, let's take a look at paragraph 27 on
page six. You indicate that you modified
two files for your testing. One is
RuntimeInit.java and the second one is
AndroidRuntime.cpp; is that right?
A. Yes.
Q. Is RuntimeInit that Java part of the accused
functionality?
MR. PETERS: Objection. Form.
A. I want to open now, which is the accused
functionality basically is the fork,
disabling the fork -- disabling the usage of
fork as a standalone operation and move it
into fork/exec combination. When you're
using a new process in fork/exec folk, you
get disabling of the copy-on-write
functionality. So just looking right now
which one of the files was that.
One file contains -- both of them I
use. One file is calling the other
function, which is the defining the first
one.
Let me describe it for you. On
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
one place, and it was sufficient.
Q. Do you know which files are part of the
accused functionality?
MR. PETERS: Objection. Form.
A. I did not study it.
Q. Do you know if RuntimeInit.java is part of
the accused functionality?
MR. PETERS: Objection. Form.
A. As I was saying, I did not study it. My
task was only to disable it.
Q. You modified RuntimeInit.java, correct?
A. Yes.
Q. Do you know if that file was part of the
accused functionality before you modified
it?
MR. PETERS: Objection. Form.
A. You asked me this question -- this is the
third time you ask me. I say I don't know.
Q. Okay.
Did your modifications disable the
whole preloader?
MR. PETERS: Objection. Form.
A. What do you mean by "preloader"? In which
context?
Q. When the zygote starts up, it preloads a
Page 46
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
page one of Appendix A, we added a new JNI
function, which is in the RuntimeInit_exec,
which do the exec operation. The exec
operation disable the fork as a standalone,
and, by this, disable copy-on-write. So
together they are doing this operation. The
actual disabling is inside B and A is
calling B.
Q. A is RuntimeInit.java?
A. Yes.
Q. And B is AndroidRuntime.cpp?
A. Yes. No, B is AndroidRuntime.cpp. This
is A and this is B [indicating].
Q. I'm sorry. Right.
Are any other files part of the
accused functionality?
MR. PETERS: Objection. Form.
A. Can you -- are there any other files what?
Q. That are part of the accused functionality
in Android.
MR. PETERS: Objection. Form.
A. I was not looking for the accused
functionality. I was looking how to disable
it. It's the actual functionality is
composed many places, but I disable only in
Page 48
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
number of frameworks into memory; is that
correct?
MR. PETERS: Objection. Form.
A. You also do that. This is not the only
thing it does.
Q. It is one of the functions that it performs;
is that correct?
MR. PETERS: Objection. Form.
A. Correct.
Q. And those frameworks that are loaded can be
cloned using copy-on-write in Android; is
that your understanding?
MR. PETERS: Objection. Form.
A. Your description of the situation is not
precise.
Q. Let me start over then.
Does Android have a preloader?
MR. PETERS: Objection. Form.
A. Preloader is used in several places in
different contexts. This is why I ask you
what context you are referring.
Q. Do Java virtual machines have a class
preloader?
MR. PETERS: Objection. Form.
A. Again, same answer. Preloader is not a
Page 47
Page 49
13 (Pages 46 - 49)
Veritext National Deposition & Litigation Services
866 299-5127
EXHIBIT 2-10
Attorneys' Eyes Only
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
come to Israel.
MR. PETERS: I have a few follow-up
questions, Mr. Landau.
*****
EXAMINATION BY MR. PETERS
Q. If you look at Exhibit 496, which is a
document entitled "MtaskClassLoading."
A. The list of class loading? Yes. Found it.
Q. This is your program that will load in the
classes that are named here in the static
stream classes list on the first page?
A. Yes.
Q. If I could ask you to turn to page 21,
please.
A. Yes.
Q. And here at the bottom of page 21, there's a
series of classes that are in the package
space
org.apatchi.harmony.xnet.provider.jsse.
Do you see those?
A. Yes.
Q. So when your test program loads, it will
load those classes; is that correct?
A. It can only load them. It cannot execute
them.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
Q. Is there any relationship between your test
program loading that class and the
bluetooth.ScoSocket entries -- in the log
entry of Exhibit 499?
A. In the Bluetooth case it's likely that you
are right, but we have to verify this in the
code.
Q. So it could be not accidental that these log
entries relating to bluetooth.ScoSocket
appear in the log because of the loading of
the class ScoSocket?
A. It could be. We have to check it.
Q. Speaking more generally about the tests that
you and your team ran, would you say that
the tests that are described in your report
and you've testified about today, are they
representative of the kind of performance
testing that you and your team members do in
your everyday work?
A. Yes. This is a typical scenario of work
that we are doing.
Q. Okay.
And when you do your work, do you
normally rely on your colleagues like Seeon
to help you do your work?
Page 106
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
Q. But what will it do with any sort of static
initialization methods that may be present
when it initializes the classes, when the
classes are loaded?
A. In the 4.8, I don't think it's been done.
I'm not sure about this. Anyway, this is
not a message you get over there -- from the
log. The message in the log is not due to
that.
Q. So there's -- so in your view, there's no
relationship between the loading of the
classes and the log message in Exhibit 499?
A. It could be. I would say it's not. It
could be.
Q. Okay.
If I could ask you to turn to page two
of Exhibit 496, MtaskClassLoading.java.
A. Which one? The same one?
Q. Yes. Page two.
So here do you see a series of classes
all beginning the name "android.bluetooth"?
A. Yes.
Q. And the last one on that list is
"android.bluetooth.ScoSocket"?
A. Yes.
Page 108
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
A. I assist a lot of and -- I am being assisted
a lot with hands-on work with other of
the -- I'm supervising quite multiple
projects.
Q. Is there anything about the projects that
you did in your analysis of Android that you
would say is not typical of the kinds of
testing you do in the Java environment?
A. No. We use normal methods that -- normal
procedure.
Q. Okay.
MR. PETERS: Thank you, Mr. Landau.
THE VIDEOGRAPHER: The time is
1:34 p.m. This is the end of tape three,
and the deposition is concluded, and we are
off the record.
(Whereby the proceedings were
concluded at 1:34 p.m.)
Page 107
Page 109
28 (Pages 106 - 109)
Veritext National Deposition & Litigation Services
866 299-5127
EXHIBIT 2-10