Hello World Project

Setting up Rust on Windows 10 with MSYS2 without installing Visual Studio

2021-01-07T12:40:00.004-08:00

Setting up Rust on Windows 10 with MSYS2 without installing Visual Studio

1. Install MSYS2

2. Update MSYS2 and run : pacman -Sy && pacman -Syu and install git and gcc: pacman -S git && pacman -S mingw-w64-x86_64-toolchain

3. Open the MSYS2 terminal and install the rust compiler with : pacman -S mingw-w64-x86_64-rust

You may stop here but you will miss the rustup package (which is not part of MSYS2) to manage the Rust compiler environment. To install rustup, follow the next steps.

4. Clone the rustup git repository :

git clone git@github.com:rust-lang/rustup.git --depth=1

5. Compile and run rustup:

cd rustup

cargo run --release

6. Once rustup detects the previously installed rust compiler and asks to uninstall it before you can continue. So, open a new MSYS2 terminal and uninstall rust : pacman -R mingw-w64-x86_64-rust

7. Now back to the rustup terminal, choose yes to continue the installation.

On the next question choose : custom installation and chose as host triple : x86_64-pc-windows-gnu

Now, press ENTER to finish the setup

Installing Tensorflow with GPU support on Ubuntu 16.04

2016-12-05T12:27:00.001-08:00

Installing Docker for running learning resources (eg Udacity MOOC)

1. Head to the Docker installation webpage and install Docker https://docs.docker.com/engine/installation/linux/ubuntulinux/

2. If you have a large second HDD, you might want to make sure all the docker images you use are stored on the large disk to preserve your precious SSD space.
For this, I found that creating a symbolic link to a new location on the second hard-drive disk worked best for me.

a. Make sure that the large hard drive disk is mounted automatically when booting the disk
https://help.ubuntu.com/community/Fstab

b. Setup a symbolic link to a directory inside your large hard drive disk. Setting DOCKER_OPTS did not work for me, but the symbolic link method worked:
https://forums.docker.com/t/how-do-i-change-the-docker-image-installation-directory/1169

Installing Python Anaconda

1. Follow the instructions here https://www.continuum.io/downloads

2. Download this python script (It will update the PIP libraries useful to download and easily install Tensorflow later-on) https://bootstrap.pypa.io/ez_setup.py

3. Run the script in a Bash Terminal "python ez_setup.py "

Installing the GPU computing libraries

1. Install CUDA 8.0 following these instructions
http://askubuntu.com/questions/799184/how-can-i-install-cuda-on-ubuntu-16-04

2. At the end of the process run the command "ls /usr/local/cuda/" to check that the installation has been successfully completed. You should see the following directories: "include" and "lib64"

3. Restart your machine

4. To install the NVIDIA CuDNN v5 library follow these instructions (it is mostly about creating an NVIDIA dev account, downloading an archive and copying the files inside ls /usr/local/cuda/"):
http://askubuntu.com/questions/767269/how-can-i-install-cudnn-on-ubuntu-16-04

5. Add the libraries to your environment variable PATH by adding at the end of the ".bashrc" file:

# Added support for CUDA and CUDnn (NVIDIA GPU-accelerated library of primitives for deep neural networks)
LD_LIBRARY_PATH="/usr/local/cuda/lib64:/usr/local/cuda/extras/CUPTI/lib64"
export CUDA_HOME="/usr/local/cuda"
export LD_LIBRARY_PATH
export PATH="$LD_LIBRARY_PATH:$PATH"

Installing Tensorflow with GPU support

To install Tensorflow, run the following:
# Ubuntu/Linux 64-bit, GPU enabled, Python 3.5
# Requires CUDA toolkit 8.0 and CuDNN v5. For other versions, see "Installing from sources" below.
$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/gpu/tensorflow_gpu-0.12.0rc0-cp35-cp35m-linux_x86_64.whl
$ sudo pip install --upgrade $TF_BINARY_URL

To test your installation of Tensorflow, run the following in a Bash Terminal

$ python
...
>>> import tensorflow as tf
>>> hello = tf.constant('Hello, TensorFlow!')
>>> sess = tf.Session()
>>> print(sess.run(hello))
Hello, TensorFlow!
>>> a = tf.constant(10)
>>> b = tf.constant(32)
>>> print(sess.run(a + b))
42
>>>

Installing Ubuntu 16.04 on an Alienware Aurora R5

2016-11-23T16:13:00.000-08:00

Installing Ubuntu 16.04 LTS on an Alienware Aurora R5 with NVIDIA GPU support (Dual boot Windows)

I have compiled these instructions from my succesful installation on Nov 25.

Would you be happy to live without Windows? If not use Dual Boot.

This procedure is inspired from this DELL support page : http://www.dell.com/support/article/uk/en/ukbsdt1/SLN301754/en

You can refer to the link above for additional steps and to see useful screenshots. However, please, make sure to take care about the GPU-related steps below.

Re-installing Windows 10 with BIOS Secure Boot to OFF

First, check that have only one disk installed in your machine. I have found that having one disk only makes it easier to re-install Windows.

Secondly, an important step is to create two USB Recovery disks. Use two USB memory sticks bigger than 16 GB supporting the USB 3.0 standard (file copy will be much faster).
Use the first one to create a recovery disk for Windows 10, as a backup if the installation does not go as expected or if you want to get rid of Ubuntu.
And use the second USB stick to create a USB installation stick for Ubuntu. I used Rufus to create it.
Do not hesitate to backup your important files from windows.

Shrink the windows partition to split the HDD space in half.

If Windows was not installed with BIOS secure boot enabled, then, you can safely skip these steps to the "Installing Linux Ubuntu" section below.

In the BIOS (accessed by pressing F2 on computer startup) under SATA operation set AHCI. Also, set the BIOS to use the UEFI Boot mode, Secure Boot to Off and set the Boot setting option Legacy to Off.

Plug the USB stick into the machine and reboot the computer.
On startup, press F12 to choose to boot over the USB stick and re-install Window 10.

Installing Linux Ubuntu 16.04 LTS next to Windows 10

Shutdown the computer and remove the NVIDIA GPU from the Desktop machine and plug your computer monitor into the only available Display Port at the back of the Central Unit.

This will allow us to use the default CPU display drivers when installing Ubuntu.

Now use the USB memory stick to install Ubuntu.
You can refer to the link to the DELL on how to partition your Hard-drive disk.
When installing Ubuntu, I created 3 other partitions inside the partition I freed under windows: "boot", "/" and "swap".

Configuring NVIDIA GPU Support in Ubuntu

After Ubuntu is installed, restart the machine and choose to use Ubuntu. Now install the NVIDIA drivers using this command:

sudo add-apt-repository ppa:graphics-drivers/ppa

sudo apt-get update

sudo apt-get install nvidia-375

(Note that the NVIDIA-CUDA-TOOLKIT package installs the Cuda compilation tools, release 8.0. This is useful for developing high performance computing routines. I am installing this because I am interested in improving my Machine Learning and Deep Learning skills)

Shutdown the computer and insert your NVIDIA GPU back into the machine.

Reboot the machine again to Ubuntu and run the following command to validate the GPU installation:
nvidia-settings

This command should open a configuration window listing your current GPU.

Et voila!

Shutdown Hangs Issue

If your PC hangs when shutting down, press ESC on your keyboard and look at the displayed lines. If you recognise the following lines, in particular the bug from i2c_dw_isr in i2c-designware-core, then this solution can apply to you:

[OK] Reached target shutdown
 [ 143.1 ] : BUG: Unable to handle kernel NULL pointer dereference at     (null)
 [ 143.1 ] : IP:  [] i2c_dw_isr+0x349/0x730      [i2c-designware-core]
 [ 143.1 ] : PGD 0
 [ 143.1 ] : Ooops : 0000 [#1] SMP

Create a new file in order to deactivate the module:

 $ sudo gedit /etc/modprobe.d/blacklist-i2c-designware-core.conf

and copy the following line in the conf file: install i2c-designware-core /bin/false

Reboot the machine once, and then try shutting down again, now the computer should shutdown ok.

You can delete /etc/modprobe.d/blacklist-i2c-designware-core.conf if this fix does not work for you.

Feedback appreciated :+1: on Github !

How to create an executable from a Scala project

2011-06-19T12:29:00.000-07:00

I started to have fun with scala and I wanted to be able to package and launch a Scala executable.

It is an eclipse project and it is already using Maven for dependency resolution.

I found the following solutions (to confirm)

1. Use SBT as a build tool. SBT seems to support the Maven pom.xml files for dependency resolution.

2. Use Gary Boone's solution by wrapping the Scala project entry point into a Java static main function

3. Use mvn assembly:single plugin as described on stackoverflow

Use Python to login to a secured website

2011-04-14T09:00:00.000-07:00

This Python 2 class logs in and connect to a secured website. The class inherit from thread so that it is possible to run multiple instance of the class.

class CRReader(threading.Thread):

data = None
headers = None

def __init__(self, crNo, semaphore):
threading.Thread.__init__(self, name=crNo)
self.loggedin = False self.crNo = crNo self.semaphore = semaphore
self.opener = None
self.status = None

def login(self):
if self.loggedin is False:
#print 'Login ' + self.crNo
url = "https://example.com/login.php"
opts = {
'username': 'John',
'password': 'pass1',
}

CRReader.headers = {
'User-Agent': 'Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv:1.8.1.12) Gecko/20080201 Firefox/2.0.0.12',
'Accept': 'text/xml,application/xml,application/xhtml+xml,text/html;q=0.9,text/plain;q=0.8,image/png,*/*;q=0.5',
'Accept-Language': 'en-gb,en;q=0.5',
'Accept-Charset': 'ISO-8859-1,utf-8;q=0.7,*;q=0.7',
'Connection': 'keep-alive'
}

CRReader.data = urllib.urlencode(opts)
request = urllib2.Request(url, CRReader.data, CRReader.headers)

cookies = cookielib.CookieJar()
response = urllib2.urlopen(url)
cookies.extract_cookies(response, request)

cookie_handler = urllib2.HTTPCookieProcessor(cookies)
redirect_handler = urllib2.HTTPRedirectHandler()
self.opener = urllib2.build_opener(redirect_handler, cookie_handler)

response = self.opener.open(request)
self.loggedin = True
return self.opener

/etc/X11/xorg.conf

2011-04-03T15:32:00.000-07:00

#To setup bigger resolutions with Ubuntu in a Virtual Machine

#Section "Device"

#Identifier "VESA Framebuffer"

#Driver "vesa"

#VideoRam 4096

# Insert Clocks lines here if appropriate

#EndSection

Section "Device"

Identifier "Configured Video Device"

#Driver "vboxvideo"

EndSection

Section "Monitor"

Identifier "My Monitor"

HorizSync 31.5 - 150.0

VertRefresh 75-85

EndSection

Section "Screen"

Identifier "Screen 1"

Device "VESA Framebuffer"

Monitor "My Monitor"

# If your card can handle it, a higher default color depth (like 24 or 32)

# is highly recommended.

# DefaultDepth 8

# DefaultDepth 16

DefaultDepth 24

# DefaultDepth 32

# "1024x768" is also a conservative usable default resolution. If you

# have a better monitor, feel free to try resolutions such as

# "1152x864", "1280x1024", "1600x1200", and "1800x1400" (or whatever your

# card/monitor can produce)

Subsection "Display"

Depth 8

Modes "1680x1050" "1280x1024" "1024x768" "800x600"

EndSubsection

Subsection "Display"

Depth 16

Modes "1680x1050" "1280x1024" "1024x768" "800x600"

EndSubsection

Subsection "Display"

Depth 24

Modes "1680x1050" "1600x1200" "1600x1000" "1280x1024" "1024x768" "800x600"

EndSubsection

Subsection "Display"

Depth 32

Modes "1680x1050" "1600x1200" "1600x1000" "1280x1024" "1024x768" "800x600"

EndSubsection

EndSection

Section "ServerLayout"

# The Identifier line must be present

Identifier "Simple Layout"

# Each Screen line specifies a Screen section name, and optionally

# the relative position of other screens. The four names after

# primary screen name are the screens to the top, bottom, left and right

# of the primary screen. In this example, screen 2 is located to the

# right of screen 1.

Screen "Screen 1"

EndSection

# commented out by update-manager, HAL is now used and auto-detects devices

# Keyboard settings are now read from /etc/default/console-setup

#Section "InputDevice"

#Identifier "Configured Mouse"

#Driver "vboxmouse"

#Option "CorePointer"

#EndSection

iPhone Email Hack for data retrieval

2011-01-04T14:49:00.000-08:00

I wanted to send an email I typed in my iphone from a cybercafe without wifi access.

I just setup a adhoc wifi network and sshfs the whole file system. For ssh setup, see my previous blog post.

I discovered that email are clearly stored in : /private/var/mobile/Library/Mail

And emails are stored as .emlx files.

To read those files use a mail client just like thunderbird or use a simple text editor.

You can use a base64 decoder to decode encrypted message part when necessary.

With an iPhone jailbroken ios 4, how to setup tethering with ssh for free

2010-09-22T16:50:00.000-07:00

Free tethering on iPhone with jailbroken IOS 4.0.1 and Ubuntu Linux. That is also possible on windows with putty.

I. Preparing the operation, setup of the necessary programs

1. Install openSSH with cydia then reboot the iPhone

2. Install sshfs to browse all the iphone files easily (sudo apt-get install sshfs)

3. Download the fixed terminal (step not necessary)

http://code.google.com/p/mobileterminal/downloads/list which is useful for changing the default password right on the iphone.

4. Get the iphone address in the wifi settings

(let's say it is 192.168.0.101)

5. Using sshfs, mount the iPhone filesystem :

mkdir myiphone && sshfs root@192.168.0.101:/ myiphone

6. Copy the unziped Terminal application in

/var/stash/Applications folder of the iPhone filesystem then reboot the iPhone

7. Change the default password (alpine). See : http://iphoneoverdose.com/2008/how-to-change-the-default-alpine-ssh-password-on-your-iphone/

II. Tether now!

1. Create an AdHoc connection on Ubuntu with the following parameters:

security 40/128 bit key (That is the maximal security setting I was able to set)

ipv4 : link-local only

2. Connect Ubuntu to that connection

3. Connect the iphone to the Ad Hoc connection

4. On the iPhone open a new safari page then return to the settings to note the ip address

5. On Ubuntu, open the terminal then enter : ssh -ND 9999 root@192.168.0.101

6. On Ubuntu open firefox then fill in the iPhone's IP address (which you memorized earlier) into the SOCKS Host field, and put 1080 as the port number. Make sure all other proxy fields are blank.

7. The proxy (the iphone) doesn't forward DNS, you will also have to change an advanced option in Firefox. In the Firefox URL bar, enter "about:config"In Filter, type "socks" and then double-click "network.proxy.socks_remote_dns" to make its value "true".

That's all! Viva Internet everywhere!

Toshiba NB 200

2009-06-16T19:35:00.000-07:00

It's 2009, I've acquired a brand new TOSHIBA NB 200 NetBook.
The first thing i did : install Ubuntu 04.09.
I had hard time to setup the wifi, but finally, i succeded. I have wireless internet now with my netbook! My wireless card is an Atheros 9285.

Just followed the procedure described in the page : http://wireless.kernel.org/en/users/Download
And it worked like a charm.
1. Dowload
2. Compilation
3. Installation
4. Reboot
5. Connexion via ubuntu GUI

Update, it's 2016, now, I have restarted my netbook which now uses Ubuntu 14.04. Problem, the sound does not work.
This is how I fixed it using the advice from this forum page :

sudo apt-get install alsa-base
sudo apt-get install pulseaudio

sudo gedit /etc/modprobe.d/alsa-base.conf
then add at the end of the file:
options snd-hda-intel model=auto

Home network, IP adress behind a router, how it works?

2008-05-28T16:10:00.000-07:00

If you're using a router to access the Internet, then, maybe, you have noticed that your computer IP address is not the same seen from a computer or server outside your local network. Your router is hiding your local adress and permit the sharing of one single access to the Internet.

How is it working?

The router uses a technique named NAPT (Network Address Port Translator).
NAP ... What?
Typically, several computers are linked to one router in a Small or Home Office (SOHO). This router has only one IP address and is providing this single access to the SOHO computers.
Let's say that, in fact, SOHO is Pr. John Doe's local network. Pr. John Doe has one nice black and silver brand new laptop featuring a huge hard drive disk and the latest processor. He owns as well, one slow and an old dusty computer he uses from time to time to browse the net. Let call the first computer, computer B like Brilliant and the second one, computer C like Crappy.
A is the name of the router. Computers B and C are linked to the router A. A,B,C constitute a local network. In this network, Pr. Doe's choose the following addresses:
A: 255.168.1.1
B: 255.168.1.2
C: 255.168.1.3

The address seen by an external device: 23.45.67.86. This address has been given by John Doe's ISP (Internet Service Provider).
John Doe is now connecting to the Internet using his powerful laptop. His home page is google.com (IP : 64.233.169.99).
The router A has already assigned "identification numbers" to B and C, named ports.
So, every communication initiated by B will be stamped "port 1".
And as well, every communication initiated by C will be stamped "port 2".
B requests a connection to google.com (IP : 64.233.169.99).
IP datagram contains (unit of information on the Internet)

IP address source
Port source (after the colon in the example)
IP address destination
Port destination (after the colon in the example)

the message
{src: 255.168.1.2 : 80 | dest: 64.233.169.99 : 80 | HTTP/1.0 GET} is sent to the router A.
Then, router A replaces the IP address source and the Port source by his external adress. And, in order to trace the communication, A replaces the source port by 1. The message emitted is then:
{src: 23.45.67.86 : 1 | dest: 64.233.169.99 : 80 | HTTP/1.0 GET}
When the reply comes from google, the router knows that the response must be sent to A because the message contains destination port 1.

This is how Pr. Doe's brand new computer communication is not mixed up with old and crappy computer's messages. This is how we are all able to share only one IP adress with a router.

source : http://www.faqs.org/rfcs/rfc3022.html

Sockets

2008-05-28T14:32:00.000-07:00

For a software engineer, a socket is a network interface. Programs talking to each other are sharing the same thread.

There are two communication modes :
connected : Stream Sockets
unconnected : Datagram Sockets

In connected mode, the excellent quality of transmission is achieved using TCP.
Web browsing with HTTP, for example, is actually using TCP. TCP is informationlossless! The information reach the final destination in the original order and no data is lost.

The protocol used by Datagram Sockets is UDP. A packet retrieving information must be implemented.

A discussion about "delete this" in C++

2008-05-19T11:28:00.000-07:00

To test your C++ programming experience and curiosity, a pop quiz question could be : what can you say about the C++ expression "delete this"?
In C++, "this" keyword is used in object member functions and is a pointer to the curent object.
As a beginner, the answer seemed to be obvious to me: how the language can allow an object to commit suicide? Impossible, I thought.
However, it's possible to "delete this". Deleting the current object does not cause compilation errors. Although "delete this" is accepted by compilation, its usage can lead to logic errors. Indeed, if this suicide is absolutely necessary, the following rules must be observed:

1. The object must be created via dynamic allocation using new operator.
2. No member function of the object is invoked on "this" object after suicide.
3. No member object of "this" object is called after suicide.
4. Nothing will use "this" object after suicide.

PS: I once tried to violate rule #3, and the result was not surprising: Wonderful Death Blue Screen crash followed by an automatic restart! Nice!

Begginners : How to learn VHDL!

2008-03-25T15:38:00.000-07:00

VHDL is one of the most Hardware Description Language used. This Description Language has many advantages over the others:
VHDL has powerful synthesis functions allowing a strong and friendly design reuse methodology.
This single advantage is so strong designers should not look aside. Good digital designs are generic a parameterizable. They are easy to maintain, reuse and help to produce complex system quicker.

So, How to learn VHDL?

1. Learn Boolean algebra/ Boolean logic
This will give you a better comprehension about the logic of a digital circuit and an overview on how computation is done on basic circuits.
http://en.wikibooks.org/wiki/Digital_Circuits

2. Learn Finite state machine and Sequential Logic.
http://en.wikipedia.org/wiki/Finite_state_machine

Those two notions are absolute prerequisite to start coding in VHDL.

This done, you can start a VHDL tutorial and then you're ready for the next chapters.