Now, the get_data_block function has clearer INFO messages
about the block number, channel number of number of valid
samples.

On finding a new DEVIO, spawn as most associated RFFE
as possible. For this, we loop through all possible
RFE hash entries looking for valid RFFE endpoints.

On calling recv () a return value of 0 means that the peer
(endpoint) has disconnected. In our case, we should return an
error as to avoid infinitely looping over the receiving function

Some trace messages, such as "sending x bytes" and "receiving x bytes"
were using the INFO flag and the more appropriate one would be TRACE

Currently, there is a bug in the FPGA firmware in that we can't
access DDR3 addresses past 0x00FFFFFF.

This is a known issue, github #39.

Now we have a new acquisition channel, called
ADCSWAP which is the ADC RAW data after the switching
module.

We had a problem when compiling the actual code in the Virtual Machine,
where the compiler wouldn't initialize all the struct fields with 0.
It's probably because we have an older version of gcc there

Devel luc

Conflicts:
	hal/hal_utils/hal_utils.h

Our MDP worker is not a socket type and can't be
poll'ed agains. Now, we fixed this by modifying
the MDP library allowing us to exit mdp_workder_recv
internal poll () and check for incoming messages from
other sockets, such as the PIPE one.

This fixes #24 github issue

Now, each SMIO will deal with the traditional
Majordomo socket and the additional PIPE socket
for monitoring messages. For now, every message
is interpreted as a self-destruct one.

This fixed #79 github issue

Mounting the LOG directory is not needed anymore

Previously we were not calling the appropriate functions to cleanly
destroy all alloc'ed objects.

These are just convenient wrappers for standard C strtoul ()
function