Looks like there's a 1 wire driver that you can
use: http://webshed.org/wiki/RaspberryPI_DS1820#Software
On Thursday, April 2, 2015 at 7:13:51 PM UTC-7, Geoffrey Carson wrote:
I am trying to connect a flow sensor directly to my BBB revC running
Ubuntu trusty. I Have Strangebrew Elsinore up
have left :-)
On Wednesday, November 5, 2014 4:24:01 PM UTC-6, Brandon I wrote:
https://github.com/adafruit/adafruit-beaglebone-io-python/
blob/master/source/c_adc.c
The adc source for bbio should be useful.
Keep in mind that these sysfs interfaces are incredibly slow compared to
memory
https://github.com/adafruit/adafruit-beaglebone-io-python/blob/master/source/c_adc.c
The adc source for bbio should be useful.
Keep in mind that these sysfs interfaces are incredibly slow compared to memory
poking since each operation requires opening, reading/writing and closing the
file.
Seems like a lot of work! Use an indirect internet connection to your
beaglebone using the USB to your computer, then set up a bridge between
your internet and the USB connection.
On Wed, Oct 29, 2014 at 11:47 PM, Lingesh Waran radhaling...@gmail.com
wrote:
Hi all...I want to perform Image
For the external source, if your
BB loses power, then you'll be powering it up through the esd does in the pin
connected to that external power source. It probably won't survive.
A general rule is to never apply voltage to a devices pin unless that device is
powered up, unless it is in the
Use one of the many remote python libraries. Rpyc could do it. Have the code
accessible from a shared drive, use module reload, and run the functions and
play with the objects as if they were local.
--
For more options, visit http://beagleboard.org/discuss
---
You received this message
All pru GPIO are accessible via a single register, so it's one tick,just like
any other register (two for loading a 32 bit immediate value).
Arm GPIO are never accessible in a single tick since there way outside the Pru
and have to go over the OCP bus. See
If you just want to poke around, you can compile
devmem2: http://free-electrons.com/pub/mirror/devmem2.c
If you want proper high speed access, see the devmem2 source, or do the
equivalent
in python
http://www.alexanderhiam.com/tutorials/beaglebone-io-using-python-mmap/,
or any other language
First google result seems useful:
http://stackoverflow.com/questions/16374670/why-i-can-not-disable-multicast-request
On Tuesday, September 16, 2014 7:14:10 AM UTC-7, Mickae1 wrote:
I've tried to deactivate the Multicast with :
ifconfig eth0 -multicast
But when i check with this command, I
know that a new data
has arrived. Then hoe to synchronize the kernel space and user-space if
there are to share a data ?
Thanks...
On Thursday, September 11, 2014 2:26:15 AM UTC+5:30, Brandon I wrote:
pseudo-interrupt from user space
There's nothing pseudo about it. Again, any usual
*
Regards,
John
What am I doing wrong?
Jan
On Wednesday, 10 September 2014 12:52:11 UTC+10, Brandon I wrote:
And, here are some compile flags you'll want to include/force:
http://www.eliteraspberries.com/blog/2013/09/cflags-for-numerical-
computing-on-the-beaglebone-black.html
On Tue
pseudo-interrupt from user space
There's nothing pseudo about it. Again, any usual way to have a userspace
application respond to an interrupt will be the exact same. The kernel will
block the userspace process until the interrupt is seen. The only real
alternative is burning up the cpu with
See UIO: https://www.kernel.org/doc/htmldocs/uio-howto/
The uio_pruss.c driver that comes with the pru package is a good example.
I have written a kernel module that registers interrupts on the rising
edge on a GPIO pin and want to relay this message to user space.
The sysfs gpio interface
You're desktop PC is 10 times faster than the Beaglebone processor.
This may help: How to Achieve 30 fps with BeagleBone Black
http://blog.lemoneerlabs.com/3rdParty/Darling_BBB_30fps_DRAFT.html
On Monday, September 8, 2014 6:03:58 PM UTC-7, janszyma...@gmail.com wrote:
Hi,
I need to
Ethernet will work without any configuration if you have dhcp on your
network. If not, you'll have to configure a static ip. If you do have dhcp
(you probably do) you can just change the hostname (
https://wiki.debian.org/HowTo/ChangeHostname) to something nice, then use
that instead of an ip
I saw the link to the reference, but I didn't see an answer to his question.
On Tue, Sep 9, 2014 at 2:47 PM, William Hermans yyrk...@gmail.com wrote:
Brandon, I guess you did not look at the link I gave. It covers every
aspect of configuring an eth device on Debian. That is when using
/etc
I'm not sure that newer open cv can be compiled on the raspberry or
beaglebone due to the limited ram. You'll probably have to cross compile.
On Tue, Sep 9, 2014 at 5:56 PM, janszymanski12...@gmail.com wrote:
Thanks for that link, it is very usefull.
In a meantime my attempt to install a newer
prag.in...@gmail.com wrote:
Hi Brandon
I read through the link, very informative thanks.I can create a thread to
do the polling and signal me when its ready.
But how to really write an ISR in arm. I see a lot of guides but they say
that it will work in Intel processors but they are not sure about
And, here are some compile flags you'll want to include/force:
http://www.eliteraspberries.com/blog/2013/09/cflags-for-numerical-computing-on-the-beaglebone-black.html
On Tue, Sep 9, 2014 at 7:16 PM, Brandon I brandon.ir...@gmail.com wrote:
I'm not sure that newer open cv can be compiled
than they are :-) But i'm very thankful for your help
so far ;-)
Am Mittwoch, 3. September 2014 22:00:16 UTC+2 schrieb Brandon I:
halfbrain, I forgot to mention, you should tie the eMMC cmd and clock
pins low on P8.20 and P8.21, as suggested by the wiki: http://elinux.org
halfbrain,
If you're using angstrom or debian, you can disable the emmc by adding this
to the optargs in uEnv.txt on the usb mass storage
partition: capemgr.disable_partno=BB-BONE-EMMC-2G
If you're not using hdmi, you can free up those
too:
halfbrain, I forgot to mention, you should tie the eMMC cmd and clock pins
low on P8.20 and P8.21, as suggested by the
wiki: http://elinux.org/Beagleboard:BeagleBoneBlack#Onboard_eMMC
On Wednesday, September 3, 2014 12:58:09 PM UTC-7, Brandon I wrote:
halfbrain,
If you're using angstrom
As a side note, you can also emulate open drain. Set the pin output state
to 0, then enable/disable the output driver. And, for performance, writing
to the gpio registers rather than doing file operations with sysfs is 30
times faster.
And, pin muxing (pullups/down and input receiver enable
I think this can only mean that a device isn't responding to the address
you're providing. Did you set the i2c device address to the same value used
with i2ctest?
On Thu, Aug 14, 2014 at 10:09 AM, Akhil Panyamparambil
akhilpana...@gmail.com wrote:
Hi all. Can you guys help me.I am using
). Do you know of a way to disable the sysfs interface /
claim those pins as being in use so that you can safely manipulate the mmap
either through software or a device tree overlay?
Thanks for your help.
On Tuesday, July 29, 2014 3:09:08 PM UTC-4, Brandon I wrote:
That's still one bit control
You have to enable the ocp master port (section 10.1.2) to access main
memory. Here's an explanation
http://nomel.tumblr.com/post/30006622413/beaglebone-tutorial-accessing-main-memory-from-the-pru
.
And, the resulting code is (if you want to do it in the pru):
// clear STANDBY_INIT bit in
unexecutable as bash script and become
interrupted as binary? know what causing that?
i've roughly put my latest configuration in
https://github.com/mpinner/Active
--matt
On Tue, Jul 29, 2014 at 1:19 PM, Brandon I brandon.ir...@gmail.com
wrote:
Don't forget, read only mount! Flash has limited
configuration, will be limited to what the pin can handle.
On Wed, Jul 30, 2014 at 10:37 PM, serge.ns...@gmail.com wrote:
On Thursday, July 31, 2014 12:39:42 AM UTC+6, Brandon I wrote:
The gpio are push pull/pseudo open drain, so there's a transistor/switch
going from 3.3V to the pin,...
At boot
The gpio are push pull/pseudo open drain, so there's a transistor/switch
going from 3.3V to the pin, and a transistor going from ground to the pin.
When you set the pin high, you're turning on only the transistor to 3.3V,
so it's sourcing the current from 3.3V, through the transistor, out of the
Source 6mA, sink 8mA, with the following pins limited to sourcing 4mA:
P9_19 gpio0[13]
· P9_20 gpio0[12]
· P9_24 gpio0[15]
· P9_26 gpio0[14]
· P9_41 gpio0[20]
· P9_42 gpio0[7]
On Monday, July 28, 2014 7:24:52 AM UTC-7, Gerald wrote:
6mA.
No.
That's still one bit control. There's going to be some unknown time between
the bits that will depend on cpu usage. For true 8 bit, you need to use
mmap to get a pointer to the gpio control block and modify the registers
directly.
Each gpio block has 32 pins, and each gpio block has a set and
Don't forget, read only mount! Flash has limited writes and is can easily
be corrupted/damaged from power failure.
On Tuesday, July 29, 2014 9:02:52 AM UTC-7, Ben Gamari wrote:
Matt Pinner mpi...@gmail.com javascript: writes:
tldr: can i run a BBB for three years?
Sure!
I'm about
There's nothing Beaglebone about this question, really. Just google send
udp packets c posix.
On Thursday, July 17, 2014 3:46:18 AM UTC-7, msc.a.f...@googlemail.com
wrote:
Hi everyone,
I am new with the Beaglebone black and I would like to make a little
program in C to send UDP packets
I don't believe the reads/writes to the GPIO registers would have worked
without the OCP master being enabled.
Thanks, I'll check out the library to see what magic is missing from that
code.
On Jun 21, 2014 8:20 AM, TJF jeli.freih...@gmail.com wrote:
A miss click and a miss post ...
The PRUSS
// readback
LBBO b, a, 0, 4
// gpio off
mov b, 0x4000
mov a, 0x44E07190
SBBO b, a, 0, 4
// readback
LBBO b, a, 0, 4
On Monday, May 19, 2014 2:36:51 PM UTC-7, Charles Steinkuehler wrote:
On 5/19/2014 4:06 PM, Brandon I wrote:
The pin mux registers require privileged memory access, which is why
match! r1 is now 0x5!*
You had me excited, even though I'd tested this before, and have read in
multiple places that it's not possible. :(
--Brandon
On Saturday, June 21, 2014 1:24:32 AM UTC-7, Brandon I wrote:
You have 100% full control over anything the PRU can access.
This pru code seems
If you want to use the triggers provided by the led interface, use the led
interface. The pin will go high or low voltage with the led interface or
gpio interface.
GPIO is for general purpose io, so it's not going to have fancy patterns
or anything, just on, off, in, out.
On Tuesday, June
The first result https://github.com/kelly/node-i2c for i2c with
bonescript refers to node-i2c. That link has some examples.
There are many good tutorials on i2c. Your transactions will contain a
device address and some number of bytes to read or write. What those bytes
are is specific to each
at 8:37 AM, Fred Basset fredbasset1...@gmail.com
wrote:
Brandon,
I'm curious as to how you calculated a lifetime of 2 yrs for the 16Gb card
with 2Gb free?
On Thu, Jun 12, 2014 at 11:55 AM, Brandon I brandon.ir...@gmail.com
wrote:
With a 16Gb card, you'll most likely get about 2 years use
I'm not sure you're ending up with a sane network configuration using the
same subnet mask for each, but did you change the udhcpd config file to
give the host a different ip for each usb connection? The host gets its ip
from the beaglebone's dhcp server. I say not sane, because the
With a 16Gb card, you'll most likely get about 2 years use before the card
fails, assuming you had 2gb free on your failing cards card, the 16Gb card
has the same number of writes until failure for the memory blocks, and the
same disk activity.
This assumes that you're have a perfect power
but you can't run the clock much faster than 1 KHz using a user-space
program under Linux.
Not true at all! You can get over 3MHz just fine with mmap to the gpio
registers. If you try to open and close a file each gpio toggle, like the
insanely inefficient sysfs interface, then yeah...you'll
Two other features of going low level with mmap:
Open drain output: By controlling the output enable registers, you can
do open drain by setting the output to drive 0, then enable output to pull
low, disable output to for high (with external pullup or internal through
pin muxing).
And a quick google search pru ethercat am335x provides a nice overview:
http://www.ti.com/lit/wp/spry187c/spry187c.pdf
On Tue, Jun 3, 2014 at 6:05 AM, Gerald Coley ger...@beagleboard.org wrote:
That is all covered in the datasheet for the processor.
http://www.ti.com/product/am3358
Gerald
For the damage question, yes, with all flash media, if you're not using a
read only mount: http://cseweb.ucsd.edu/~swanson/papers/DAC2011PowerCut.pdf
Btw, you're crazy if you're not using a read only mount (or guaranteeing no
writes) for the rootfs of an appliance. ;)
On Tuesday, May 27, 2014
You can access all regular gpio, but those will be slower than the one tick
pru gpio access.
Check out the PRU documentation at
https://github.com/beagleboard/am335x_pru_package , it explains how to use
R30 and R31, section 5.2.2.
Here's an explanation on accessing main memory from the pru:
MLC NAND enables around 10k write cycles
SLC NAND enables more than 100k write cycles.
And, of course, these are the number of cycles for whole erase
blockshttp://en.wikipedia.org/wiki/Flash_memory#Block_erasure.
If you write one byte and flush the operation to your flash disk, you're
I use this: https://github.com/nomel/beaglebone/tree/master/led-header
Makes setting up leds super easy.
On Thursday, May 8, 2014 9:35:31 AM UTC-7, Charles Steinkuehler wrote:
The way these systems are configured, I don't know if you can do what
you want without generating a custom device
In all seriousness, is there a donation link you could post so we could
send you some thanks?
On Sunday, May 4, 2014 10:56:40 AM UTC-7, RobertCNelson wrote:
On May 4, 2014 4:43 AM, Timbo tim...@gmail.com javascript: wrote:
Thanks for the info, Robert, and thanks for making these images,
The beaglebone had a power up issue. Rev A6A fixed it:
From the wikihttp://www.elinux.org/Beagleboard:BeagleBoneBlack#Revision_A6A
: Changed C24 to a 2.2uF capacitor. This extends the reset signal to
solve an issue where some boards would not boot on power up.
I have around 40 beaglebones
This is incorrect. There's nothing wrong with applying power through USB
and VDD_5V (DC plug) at the same time, from the reference manual:
The selection of either the 5VDC or the USB as the power source is handled
internally to the TPS65217C and automatically switches to 5VDC power if
both
Could always use python as a web back end and render the gui in the
browser. ;)
On Wednesday, April 9, 2014 3:28:56 AM UTC-7, Eric Palmer wrote:
In my day job I erite backend code. Not gui stuff. I'm building a large
robot and will use a BBB and display for data display and more.
I
Do you have a pullup on the i2c line?
On Monday, March 31, 2014 1:59:46 AM UTC-7, Rafael Fiebig-Bindner wrote:
Hi community,
I am currently trying to access a temperature sensor via I2C. The problem
is whenever I want to send data I get the error message: Bad address,
but the BBB
That's because your phone uses a sane filesystems that takes into account
this use case and isn't writing constantly (write one byte, the disk writes
a whole erase block). This doesn't protect you from eventual disk
corruption. The wear leveling bad-block type tables will eventually
corrupt/run
The only thing I wish for is the ability to change I2C bus speeds on the
fly.
You can do anything with a kernel module and some memory pokes. ;)
On Thursday, March 27, 2014 2:47:10 AM UTC-7, Richard-tx wrote:
I bought a BBB about 3 weeks ago. Was impressed enough that I bought two
more.
Rh, my earlier reply was to you, and that link shows that it is now a
problem with androids use of ext4.
On Thu, Mar 27, 2014 at 4:55 PM, rh_ richard_hubb...@lavabit.com wrote:
On Thu, 27 Mar 2014 13:41:24 -0500
Charles Steinkuehler char...@steinkuehler.net
wrote:
On 3/27/2014 12:26 PM,
Here's a good read: http://www.embeddedarm.com/about/resource.php?item=459
I had a lng discussion about this with a colleague of mine after we
started seeing boards die.
Basically you're eventually doomed unless you mount the whole disk as read
only since the wear leveling algorithms
For Angstrom, you can install the kernel-dev and kernel-headers packages.
This will be enough to compile kernel modules.
There have been times in the past where the packages where out of sync with
the actual kernel being used. Not sure what the current state is. Check the
stable branch...lol,
My only real issue that I have no idea what the Angstrom is really
capable of and what really comes with it.
I think the biggest hurdle for people is they have some embedded mindset
with talk of processors and whatnot. Think of this as it is, a resource
limited general purpose Linux computer
() # Close the file
No way to read the P9_24 PIN From BBB using PRUSS
May be the *WBS *does no work as I think . a week working on that and
still nothing.
El martes, 11 de marzo de 2014 00:40:04 UTC-3, Brandon I escribió:
When I do an interruput up or down the ASM keeps waiting
Along with what the others have described, since you're the arm processor
gpio rather than a pru gpio, meaning you're going all the way out to system
memory, you have to connect the pru to system memory. Here's an example of
accessing system memory with the pru:
[ 706.738323] pruss_uio 4a30.pruss: pins are not configured from
the
driver
[ 706.765286] bone-capemgr bone_capemgr.9: slot #7: Applied #3
overlays.
El lunes, 10 de marzo de 2014 20:10:55 UTC-3, Brandon I escribió:
Along with what the others have described, since you're the arm
to date.*
[1] what should i do to get my modules to work with??
On Saturday, 8 March 2014 02:42:31 UTC+5:30, Brandon I wrote:
dmesg will give you more details.
This usually means you compiled the kernel modules against a different
build of the kernel. So, the kernel source you used didn't match
Missed first line on that copy paste. Should be
KERNELDIR := /usr/src/kernel
On Sat, Mar 8, 2014 at 1:21 AM, Brandon I brandon.ir...@gmail.com wrote:
Just a regular kernel type makefile. For examlpe, here's mine for a file
phyaccess.c
BEAGLEBONE_PROJ=/systems/Projects/beaglebone/workspace
?title=Compiling_the_kernel
will this link help me to insert also to create a modules without fail ..
can you suggest your opinion..pls..
regards
siva
On Saturday, 8 March 2014 14:52:49 UTC+5:30, Brandon I wrote:
Missed first line on that copy paste. Should be
KERNELDIR := /usr
dmesg will give you more details.
This usually means you compiled the kernel modules against a different
build of the kernel. So, the kernel source you used didn't match what was
on the beaglebone.
You can install the kernel-headers and kernel-dev packages and build
directly on the
of the concept is somewhat new, and there are many
misconceptions.
On Fri, Mar 7, 2014 at 2:02 PM, John Syn john3...@gmail.com wrote:
From: Brandon I brandon.ir...@gmail.com
Reply-To: beagleboard@googlegroups.com
Date: Friday, March 7, 2014 at 1:27 PM
To: beagleboard@googlegroups.com
Cc: c
Provided, you don't interfere with their settings during power up as they
are also the boot pins.
He's saying you shouldn't interfere with their state at boot. No driving,
pullup, or pull down on any of these pins (besides the those that configure
the sysboot setting on the board).
Relays
Maybe we can become free from SD cards in the near future for BBB
development.
Doesn't uboot support network boot already?
On Thursday, February 27, 2014 11:29:23 PM UTC-8, jhg...@gmail.com wrote:
Robert,
Have you resolved any of these mysteries in the mean time?
I got into this by
Pad control registers can only be changed in kernel space. You could write a
kernel driver or do it the right way and use device tree.
See
https://github.com/nomel/beaglebone/tree/master/gpio-header
If you don't want to use it directly, you can use the generated files as an
example.
--
For
At arrow, I just asked, and they have an old A53 datasheet up, so the guy
told me it was a rev A5B. Are the arrow boards A6A?
On Monday, January 27, 2014 4:38:26 PM UTC-8, smith.wi...@gmail.com wrote:
On Monday, January 27, 2014 2:18:04 PM UTC-5, Gerald wrote:
Arrow is showing 500 boards in
Here's mine:
/dts-v1/;
/ {
compatible = ti,beaglebone, ti,beaglebone-black;
part-number = AQ-PRU-MDIO;
exclusive-use = P8.43, P8.44, P8.45, P8.46, P8.39, P8.40,
P8.41, P8.42, P8.27, P8.28, P8.29, P8.30, pru1;
fragment@0 {
target = 0xdeadbeef;
__overlay__ {
pinmux_pru_mdio_pins {
In the pru, R31 is for input, so in that table you're looking for pin
pr1_pru1_pru_r31_8. That's on P8.27, which is used by the HDMI framer. If
you *need* to use this pin, instructions on disabling the hdmi framer can
be found by searching this group.
Of course you'll have to enable the receiver,
What he means by that is, it'll cost you more in electricity to mine (with
a crud cpu setup) than to just buy some with cash from an exchange.
On Saturday, January 18, 2014 8:57:19 AM UTC-8, Michael Mullin wrote:
I'm sure you can tweak cgminer to re-enable CPU based mining... But that
And, you shouldn't rely on the pull up/down pins that are there now since
they could change with the next beaglebone (as they did between the black
and white). You can make them high impedance at boot by using something
like a bilateral switch enabled with 3V3_EXP.
On Tuesday, January 14, 2014
I think everyone (that isn'in the group would
On Wednesday, December 11, 2013 4:32:42 AM UTC-8, Giuseppe Iellamo wrote:
Thank you this is a very good news for me!
When I'll finish with RT part I'll try QT5 on BBB...
Giuseppe
Il giorno mercoledì 11 dicembre 2013 00:05:04 UTC+1, Daniel
I think everyone in the group (who isn't headless) would love to see some
benchmark results!
On Wednesday, December 11, 2013 4:32:42 AM UTC-8, Giuseppe Iellamo wrote:
Thank you this is a very good news for me!
When I'll finish with RT part I'll try QT5 on BBB...
Giuseppe
Il giorno
Google beaglebone black adc for a bunch of examples.
On Tuesday, November 26, 2013 9:24:14 PM UTC-8, Zain Dar wrote:
Hi, just like in the title, I'm trying to make a program code for making
sinewaves that work with the analog pins when I have analog accelerometers
connected to it. Is there
All of the gpio registers have dedicated set and clear registers so you
don't have to do read-modify-write operations. There wont be an issue with
collisions between separate gpio bits if you're just setting or reading pin
states.
On Sunday, November 24, 2013 11:13:21 PM UTC-8, rod calabio
You can find the memory offsets for the uart control, and then the offsets
for the specific registers in that reference manual. From there, you can
use an mmap to the registers in your program/script or use the devmem2
command to read/write the registers directly.
*mmap example with gpio
You don't need the kernel source. You can convert the compiled device tree
blob to the text version, edit it, and the convert it back to the binary
using dtc:
# backup the original .dtb
cp /boot/am335x-boneblack.dtb /boot/am335x-boneblack.dtb.orig
# generate the dts from the dtb
dtc -I dtb -O
I was wondering if it is possible to write a c/c++ code in vi through SSH
on Putty and then generate an output file and run directly on the
beaglebone?
I tried the simple helloworld.c program and it worked.
I have a network share set up, accessible by the beaglebone. I also have
the
I've connected beagleboneblack to my car power supply with 24V to 5V step
down.
24V power supply in a car? If this means you're this from the accessories
circuit, you're doomed. Accessories are turned off while the car is being
started, and a cap won't be enough to power the beaglebone, since
, 2013 9:30 PM, Brandon I brandon.ir...@gmail.com wrote:
I've connected beagleboneblack to my car power supply with 24V to 5V
step down.
24V power supply in a car? If this means you're this from the
accessories circuit, you're doomed. Accessories are turned off while the
car is being started
You can build angstrom without the gui. For some pre built images, check
out the angstrom rootfs image at
http://downloads.angstrom-distribution.org/demo/beaglebone/
On Monday, October 7, 2013 12:34:13 AM UTC-7, Rick M wrote:
Hi. I'm getting started with my BBB, having done previous
2. Why is a DT used to define the HDMI
Because this is a system-on-chip. The HDMI, gpio, gpu, pruss, etc are
all peripherals of the cpu that are located on the chip, but still separate
from the cpu.
3. If you just want to get gpio or leds working quickly, check out this
post:
and a lot of the pins are shared with the LCD/HDMI interface.
Which can be made available by disabling the hdmi framer by adding the
following to uEnv.txt on the fat32 partition:
capemgr.disable_partno=BB-BONELT-HDMI,BB-BONELT-HDMIN
On Wed, Oct 2, 2013 at 1:03 PM, Charles Steinkuehler
http://imgur.com/6W5zcb0.jpg
On Wednesday, September 25, 2013 12:20:49 PM UTC-7, Steve French wrote:
Hello! New to BBB, but lovin it so far...got two of them working on the
same network...more to come...
1) So, I used this image BBB-eMMC-flasher-2013.09.04.img
2) In order for me to do a
88 matches
Mail list logo
