Thursday, May 14, 2015

Running VMware Workstation 10.0.0.5 on Fedora 21

A number of changes in the Linux kernel near the 3.18->3.19 transition prevent VMware Workstation from building the modules that it needs to run.

I've found bits and pieces of the required patches from all over the net, including, but not limited to:
The basic idea is to unpack the sources in /usr/lib/vmware/modules/source, patch them, then repack them into updated .tar files.

Start by downloading the files.

Then you'll need to do the following things as root:
  • Execute: cd /usr/lib/vmware/modules/source
  • Move the downloaded files to this directory
  • Take a look at fix.sh and the associated patches to convince yourself that they aren't going to hurt your computer.
  • Execute: bash ./fix.sh
  • You can optionally remove the .tar.orig files that are left over

Monday, January 12, 2015

LG G3 Android adb USB debugging access

I have spent much time trying to develop software on my Fedora 21 machine with this phone.  I would always get "No devices found" from "adb devices" or no text at all from fastboot.  Here is what it took to get it to work:

USB debugging for this phone is done in the usual way: via Settings, About Phone, Software Information, then tapping Build number repeatedly.

Ethernet mode must also be enabled via Settings, More..., Mobile Broadband Connect to ON.

Finally, to install software via adb: Settings, Security, Unknown Sources must be enabled.

Friday, January 9, 2015

Accessing the Rigol DS1054Z Oscilloscope via TCP

After several attempts at searching for how to access my DSO via TCP, I finally broke down and did a port scan.

The first thing I noticed was that the scan locked up my scope... but not before finding several open ports.  Most notably, port 5555.  I reset the scope and carried on.

Armed with the DS1000Z Programming Guide from their website, I was very quickly controlling and reading from the scope!
$ telnet 192.168.1.40 5555
Trying 192.168.1.40...
Connected to 192.168.1.40.
Escape character is '^]'.
:STOP
:WAV:FORMAT ASCII
:WAV:DATA?
#90000159933.999684e-02,-4.000316e-02,-3.159046e-06...
The Programming Guide has all the details.  Most notably, it appears that data capture is always returned as a subsample of 1200 points; you don't get the whole buffer.

On the plus side, it seems reasonably fast, so it's likely to be good enough for most purposes... you'll just need to get the USB stick out when you want the complete buffer contents.