dkentw / 0xbench Goto Github PK
View Code? Open in Web Editor NEWAutomatically exported from code.google.com/p/0xbench
Automatically exported from code.google.com/p/0xbench
This task to to integrate existing file system benchmark suite like IOzone[1]
or Bonnie++[2].
[1] IOzone: http://www.iozone.org/
[2] Bonnie++: http://www.coker.com.au/bonnie++/
Original issue reported on code.google.com by [email protected]
on 18 Aug 2010 at 3:15
Hi,
I tried to run the UnixBench tests of 0xbench. However, I didn't get any result
for that because I don't have the binaries for the tests in the /system/bin
directory. Hence, I tried to build the apk and binaries using SDK, eclipse, and
NDK as described in the Wiki. However, when I create the Benchmark project
using eclipse, I always get the following errors:
1. Benchmark cannot be resolved to a
variable Case.java /Benchmark/src/org/zeroxlab/zeroxbenchmark line 42 Java
Problem
2. Case cannot be
resolved Tester.java /Benchmark/src/org/zeroxlab/benchmark line 60 Java Problem
and so on.
I guess the above problem is for the following line in Case.java
abstract public ArrayList<Scenario> getScenarios ();
here the compiler cannot resolve what is Scenario.
I would be very much grateful if you would please help me in this regard.
Thanks.
Syed Rafiul Hussain
Original issue reported on code.google.com by [email protected]
on 19 Jun 2012 at 4:08
Every submission should attach extra system information for better reference.
We can import the existing implementation from system-information project:
http://code.google.com/p/system-information/
Original issue reported on code.google.com by [email protected]
on 12 Jun 2011 at 9:33
Issue #15 initially merged SunSpider JavaScript benchmark suite into 0xbench.
However, we should conserve one copy of that and split it from directory src/
to somewhere like directory external/ .
Original issue reported on code.google.com by [email protected]
on 28 Jul 2011 at 2:06
0xbench is expected to be useful to devices without GPU for the sake of cost
and popularity. It is not sufficient to perform benchmarking over Canvas in
Android framework. This task is to improve 0xbench's 2D coverage.
Hopefully, it could cover the existing/corresponding 2D items in Rowboat
General Runner.
Original issue reported on code.google.com by [email protected]
on 29 Aug 2010 at 6:23
What steps will reproduce the problem?
1. run oxbench on froyo
2. in Native execution emulator is crashing.
3.
What is the expected output? What do you see instead?
What version of the product are you using? On what operating system?
Please provide any additional information below.
Original issue reported on code.google.com by [email protected]
on 24 Sep 2010 at 4:51
What steps will reproduce the problem?
1. Open 0xbench on a device in landscape mode
2. It appears in portrait mode
What is the expected output? What do you see instead?
Run 0xbench in landscape mode
What version of the product are you using? On what operating system?
1.1.1 on android 2.2
Please provide any additional information below.
We need landscape mode on Android-x86.
Original issue reported on code.google.com by cwhuang%[email protected]
on 2 Dec 2010 at 6:16
for those, that cannot or do not wish to usu google. market
Original issue reported on code.google.com by [email protected]
on 8 Sep 2010 at 12:04
What steps will reproduce the problem?
1. Run a 2D benchmark on the phone.
2. Login to get API key on the phone.
3. Upload the results to a new page with specific name such as
new-result-2d-run1
4. Visit http://0xbenchmark.appspot.com and login. The page exists with name
new-result-2d-run1 but when you open the page, no data is shown and the
following message is shown - "No data available."
What is the expected output? What do you see instead?
I expect to see the benchmark results but see a page with "No data available"
message.
What version of the product are you using? On what operating system?
0xBenchmark 1.1.5 on Android 2.3.6
Please provide any additional information below.
Original issue reported on code.google.com by [email protected]
on 6 Mar 2013 at 6:30
For native benchmark items such as libmicro, we execute the programs and
collect the results through UNIX pipe. However, we should always perform
sanity checks in advance.
At least, we need to avoid executing native benchmarks when programs are not
present.
Original issue reported on code.google.com by [email protected]
on 15 Jul 2011 at 5:11
This task is to improve the descriptions in wiki page:
http://code.google.com/p/0xbench/wiki/Source
At least, it should mention the following:
(1) Java part
(2) JNI
(3) standalone C/C++ programs
Original issue reported on code.google.com by [email protected]
on 18 Aug 2010 at 3:22
Reported by George Yuan:
the graphics tests seem to be hitting vsync on the more modern devices, such as
the new Droids and Galaxy phones. Basically, the frame rate of all your tests
are artifically capped at 55-60fps, making it difficult to discern the
performance differences between higher-end devices.
Two potential workarounds are to either increase the number of
iterations until you're running very low frame rates or modify the
benchmark to increase the amount of work drawn every frame until it
starts missing vsync, then use the “amount of work” as the benchmark
result.
Original issue reported on code.google.com by [email protected]
on 12 Nov 2010 at 2:48
We want to integrate it in the FOSS-Market f-droid. Unfortunately it doesn't
build.
http://f-droid.org/forums/topic/0xbench-open-source-benchmark-suite-for-android/
Please give instructions how to get it run.
Original issue reported on code.google.com by [email protected]
on 26 Sep 2012 at 4:25
SunSpider is a JavaScript benchmark. This benchmark tests the core JavaScript
language only, not the DOM or other browser APIs. This task is using WebView
to launch SunSpider tests and then obtain results.
Original issue reported on code.google.com by [email protected]
on 1 Jul 2011 at 4:29
What steps will reproduce the problem?
1. adb shell am start -n org.zeroxlab.zeroxbenchmark/.Benchmark --ez misc true
--ez autorun true
What is the expected output? What do you see instead?
Expected behaviour is that 0xbench ui start and begin to run misc category testes.
Current behaviour is that 0xbench ui start but doesn't run misc category testes.
What version of the product are you using? On what operating system?
clone of git repository. On android.
Please provide any additional information below.
I attach a patch that fix the problem.
Original issue reported on code.google.com by [email protected]
on 20 Feb 2012 at 8:18
Attachments:
0xbench is indeed a framework for Android benchmarking purpose, and it would be
great if we could integrate external system benchmark suite like
byte-unixbench[1] or freebench[2]
[1] byte-unixbench: http://code.google.com/p/byte-unixbench/
[2] freebench: http://code.google.com/p/freebench/
Original issue reported on code.google.com by [email protected]
on 18 Aug 2010 at 3:12
What steps will reproduce the problem?
1. Linpack benchmark
2. Scimark benchmark
3.
What is the expected output? What do you see instead?
Expected 90 for single thread or 332.676 for multithread via linpack app in
market.
0xbenchmark gives linpack result of 15
I did not test scimark, but it is way off
What version of the product are you using? On what operating system?
1.1.5
Please provide any additional information below.
you can see 0xbenchmark result:
http://0xbenchmark.appspot.com/run/[email protected]/PublicPage
Am I just using the application wrong or did I miss a setting somewhere?
Thanks
Original issue reported on code.google.com by [email protected]
on 14 Nov 2014 at 9:16
Attachments:
What steps will reproduce the problem?
1. Initiate 0xBenchmark on android platform;
2. Check the Native from batch select;
3. Press Run button;
4. Check the output after test done;
What is the expected output? What do you see instead?
A report should be auto generated, but nothing happens after test done.
What version of the product are you using? On what operating system?
0xbench 1.1.1, Android 2.1
Please provide any additional information below.
Other options works well such as Math, 2D, 3D, VM.
Original issue reported on code.google.com by [email protected]
on 14 Feb 2011 at 7:24
What steps will reproduce the problem?
1. Start Canvas Redraw and Draw Circle
What is the expected output? What do you see instead?
Expected test are visually ok.
But a code inspection show that they have both the same logic flaw. They both did not test what they
should be which should be their onDraw() method speed.
For example for DrawCircle:
After a new view invalidate in DrawCircle.java:oneRound() we return to
Tester.java:lazyLoop() polling
for end of test and calling again oneRound() until the thread is preempted and
ui main thread is
scheduled. At this point we execute DrawCircle.java:onDraw() method and
decrease loop counter. Then
we return to Tester.java:lazyLoop() to invalidate again view and looping until
new preemption.
So the test spent the main part of it's time into Tester.java:lazyLoop() and not into DrawCircle.java:onDraw()
as it should do.
Note that in case we have a dual processor core then test correctly measure DrawCircle.java:onDraw() execution
time.
What version of the product are you using? On what operating system?
git version on Android
Please provide any additional information below.
Original issue reported on code.google.com by [email protected]
on 20 Feb 2012 at 9:21
2D and 3D bench items are not run in full screen mode right now. there may be
status bar affecting the bench result.
Run rendering bench inclusive of 2D and 3D items in full screen.
Original issue reported on code.google.com by [email protected]
on 26 Jan 2011 at 2:30
This task is to measure Dalvik Virtual Machine (JVM) performance on mobile
devices by Caffeine Mark.
Original issue reported on code.google.com by [email protected]
on 1 Jul 2011 at 4:32
What steps will reproduce the problem?
1. Start program
2. Select all benchmarks
3. Click run
What is the expected output? What do you see instead?
Infinite loop
What version of the product are you using? On what operating system?
TAZ Pad with 832M memory available
Android 4.0.4
Please provide any additional information below.
Problem is due to activity state not being saved.
So on memory constrained systems the initial activity which contains the "Case"
objects is destroyed then recreated when the first test activity ends. The Case
object state used to determine which test to run is thus lost and the first
test is rerun (even if not selected).
Attached patch fixes this
Original issue reported on code.google.com by [email protected]
on 23 Jan 2014 at 3:10
Attachments:
At present, the native benchmark is implemented as standalone applications,
which are not limited by Dalvik VM and optimized by platform-specific
configurations (such as armv7-a and NEON). However, it would be nice if there
are some corresponding JNI items as baseline. Thus, contributors could put the
results to web service without "root".
Original issue reported on code.google.com by [email protected]
on 29 Aug 2010 at 1:48
This task aims to improve description of web service:
(1) Add descriptive text to welcome page during loading uploaded results
(0xbench page URL or brief introduction)
(2) marker -> manufacturer
(3) model -> hardware model
Original issue reported on code.google.com by [email protected]
on 28 Sep 2010 at 4:32
Inspired by Issue #9, 0xbench should implement accurate graphics calculation
methods to fit more devices with different resolutions. We would like to call
it after "strict mode" to reflect the fine-grained calculation methods for
Android graphics.
Original issue reported on code.google.com by [email protected]
on 27 Jun 2011 at 8:53
as Linaro mentioned ( https://bugs.launchpad.net/linaro-android/+bug/803566 )
To provide a configurable path to 0xbench is necessary.
Original issue reported on code.google.com by [email protected]
on 6 Jul 2011 at 11:02
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google ❤️ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.