#userspace

[nixcraft]

[eilidhmccormickcurating-blog]

Yuri Pattinson at the Chisenhale

Spending time at London's silicone roundabout, the focal point for tech development in east London, Pattison has utilised his artists residency to explore the relationship between work and leisure. In his large exhibition ‘User, Space’ at the Chisenhale Gallery, he has pulled much of his experience over the past year into an immersive experience for the viewer. In this space technology dominates, influencing the entire room, particularly with the robotic addition and subtraction of natural light, the outside world. Although the tech is alienating and somewhat subtracted from the everyday, it's setting is recognisable with the addition of work environment design that is becoming commonplace. Office plants, relaxation space, floating barriers and a single lengthy table at which to potentially work with others are all the modern office standards. Work that operates behind the scenes of this polished world of technology is also addressed with wires and metal racking that exist with and because of a human presence. He is interested in the unspoken people in the automated, interface driven world and the exploration of the spaces in which this work takes place. A staple of the Chisenhale at the moment appears to be art that reaches out to influence the staff. The previous exhibition examined the working relationship in the modern day as Maria Eichhorn shut the gallery for 5weeks. Now Pattison has pulled the front of house into the gallery with an intrusive transformation of this working environment into the same aesthetic of the exhibition and ultimately employing a transparency to the workings of the gallery by making the day to day work a prop in his practice. The entirety of the space is also littered with screens and webcams that resonate the idea of documenting and sharing simultaneously. On the multiple screens are further rendered office spaces that enforce communal activity while others illustrate the idea of time. Time in relation to work and play is an interesting and prevalent concept alone but combined with the introduction of technology, labour and community, the active space initiates an interesting conversation. This work is a look into our time, where the future is now. 

[slightlyxdrifting]

[jellyjellyrobot]

Configuring gpio on pandaboard linaro part 2

We left off at page 3719 of the technical reference manual for OMAP 4460. According to this page, we would want to set the following parameters.

register | description | value 8 | INPUTENABLE | 1 4 | PULLTYPESELECT | 1 3 | PULLUDENABLE | 1 2 - 0 | MUXMODE | 011

counting from the right, we have register "3-0" has the value of 1011 (base 2) = 11 (base 10) = b (base 16/hex) register "7-4" has the value of 0001 (base 2) = 1 (base 10) = 1 (based 16/hex) register "11-8" has the value of 0001 (base 2) = 1 (base 10) = 1 (based 16/hex)

so, we need to change the mux settings for this gpio to 0x11b.

--- start here ---

echo 0x11b > gpmc_ad13 cat gpmc_ad13

--> name: gpmc_ad13.gpio_37 (0x4a10005a/0x05a = 0x011b), b d18, t NA --> mode: OMAP_PIN_INPUT_PULLUP | OMAP_MUX_MODE3 --> signals: gpmc_ad13 | kpd_col1 | c2c_data10 | gpio_37 | NA | sdmmc1_dat5 | NA | NA

ls /sys/class

--- stop here ---

Oh darn, I can't find /sys/class/gpio. Looks like I have to recompile the kernel. Information on how to do just that...

More on this on the next issue.

[jellyjellyrobot]

Configuring gpio on pandaboard linaro part 1

I don't think the linaro build for pandaboard has GPIO switched on by default. So this article is written to chronicle my journey to enabling GPIO on the Pandaboard ES. I'm a noob so please forgive me for not using the phrase, gpio interfacing from userspace.

Alright here goes.

Firstly read this literature and this. This should help you get going.

So according to these websites, you need to do the following...

--- start here ---

sudo bash grep GPIOLIB /boot/config-`uname -r` --> CONFIG_ARCH_REQUIRE_GPIOLIB=y --> CONFIG_GPIOLIB=y grep GPIO_SYSFS /boot/config-`uname -r` --> CONFIG_GPIO_SYSFS=y cd /sys/class/gpio

--- pause here ---

YIKES! If there is no directory called /sys/class/gpio... then probably your pin mux mode is not set to the GPIO functions. Here are links 1 and 2 to help you in do these sort of stuff.

Referring to the system reference manual, we find that the GPIOs have different mux modes. (Page 55 in the ES manual) I have GPIOs connected at GPMC_AD14 and GPMC_AD13 (pins 9 & 10 of J6).

Using this info, we can proceed with the mux config.

--- start here ---

cd /sys/kernel/debug/omap_mux #this is the place in userspace that magic (multiplexing) happens (at least for pandaboard) ls #to display all available interfaces cat gpmc_ad13 #change the latter to whatever port you wish to configure

--> name: gpmc_ad13.gpio_37 (0x4a10005a/0x05a = 0x000b), b d18, t NA --> mode: OMAP_PIN_OUTPUT | OMAP_MUX_MODE3 --> signals: gpmc_ad13 | kpd_col1 | c2c_data10 | gpio_37 | NA | sdmmc1_dat5 | NA | NA

### Okay this is a lot to handle, but notice that the 3 row of the output closely resembles the one that is reflected in the pandaboard system reference manual.

### according to page 198 of the OMAP 4460 data manual, GPMC is the General Purpose Memory Controller. Its an OMAP unified memory controller that interfaces ext. memory devices. Hmm....

### by the way "cat" means "concatenate" the file(s) and print on the standard output.

### We need to write the mux configuration to this "device", in HEXADECIMAL! Using the OMAP 4460 technical reference manual (5872 pages long!), we find (on page 3719)  the pad configuration register functionality of the TI processor.

--- stop here ---