Go to advanced search

by samskiter
Thu Mar 03, 2016 1:16 pm
Forum: Device Tree
Topic: device tree overlays and U-Boot
Replies: 32
Views: 12992

Re: device tree overlays and U-Boot

Thanks again, inspecting 0x000, I see that the device tree should be loaded at 0x17fed200 but that appears to be empty. Seems that U-boot is being relocated pretty close to that (0x17F4B000) so potentially stomping over the device tree. So that I can adjust U-boot, could you give me any more informa...
by samskiter
Thu Mar 03, 2016 11:07 am
Forum: Device Tree
Topic: device tree overlays and U-Boot
Replies: 32
Views: 12992

Re: device tree overlays and U-Boot

Thanks for your reply PhilE - I've since been using device_tree_address to correct the location... I'm still a bit confused though because a) my resolved device tree is reported as "size 2d35" which is under the 16KB limit and b) inspecting the memory at 0x17fed200 shows as completely empty - so my ...
by samskiter
Thu Mar 03, 2016 10:46 am
Forum: Device Tree
Topic: Why is start.elf loading my device tree to 0x17fed200?
Replies: 1
Views: 676

Why is start.elf loading my device tree to 0x17fed200?

So my understanding is that start.elf should load my device tree to 0x100. I've been playing with u-boot but never been able to see a device tree loaded by inspecting memory. After linux boots, vcdbg logs show that start.elf has loaded to 0x17fed200. Inspecting this memory from u-boot shows nothing ...
by samskiter
Thu Mar 03, 2016 10:36 am
Forum: Device Tree
Topic: device tree overlays and U-Boot
Replies: 32
Views: 12992

Re: device tree overlays and U-Boot

While u-boot doesn't directly support overlays - start.elf still does and you can have u-boot effectively 'pass through' the flattened device tree that start.elf creates. I've just managed to get this working (last night) I have a little detail here that you might find useful: https://www.raspberryp...
by samskiter
Wed Mar 02, 2016 4:35 pm
Forum: Advanced users
Topic: vcdbg doesn't seem to work on Pi Compute Module
Replies: 2
Views: 1944

Re: vcdbg doesn't seem to work on Pi Compute Module

OK, update: Firstly using device_tree=0x100 was a total mistake and red herring that I got from this site: http://dius.com.au/2015/08/19/raspberry-pi-u-boot/ The correct parameter to change the location the device tree is placed is device_tree_address. seems that device_tree=[blah] sets a filename. ...
by samskiter
Fri Feb 26, 2016 5:14 pm
Forum: Advanced users
Topic: vcdbg doesn't seem to work on Pi Compute Module
Replies: 2
Views: 1944

Re: vcdbg doesn't seem to work on Pi Compute Module

Update: I've used the mkknlimg from the source I mentioned and again tried to tag my u-boot.bin but with no success Dumping the u-boot binary shows the tags have been added: 0004e754: 00000000 00000004 78333832 00000001 ........283x.... 0004e764: 00000004 4b4f5444 0000003f 00000001 ....DTOK?....... ...
by samskiter
Fri Feb 26, 2016 4:02 pm
Forum: Advanced users
Topic: vcdbg doesn't seem to work on Pi Compute Module
Replies: 2
Views: 1944

vcdbg doesn't seem to work on Pi Compute Module

Hi, I'm trying to get a U-Boot setup working on the Pi compute module but I don't seem to be able to get device trees to be loaded by start.elf (despite using mkknlimg on my u-boot.bin). I saw that I should be able to to use vcdbg to get some information out from start.elf but I'm getting an error o...
by samskiter
Mon Feb 22, 2016 6:09 pm
Forum: Compute Module
Topic: CMIO/CM unable to rpiboot
Replies: 2
Views: 828

CMIO/CM unable to rpiboot

I have 3 CMIO boards and 3 CM's We recently purchased 2 of the CMIOs and one of the CMs I'm unable to get rpiboot to work on *either* of the new CMIO boards with *either* of the old CM's. The older CM's work fine in the older CMIO and the new CM works fine in all the CMIOs. All CM's and CMIO's seem ...
by samskiter
Wed Feb 17, 2016 5:58 pm
Forum: General discussion
Topic: NOOBS boot process
Replies: 3
Views: 543

Re: NOOBS boot process

Noobs sets the boot partition in /sys/module/bcm2708/parameters/reboot_part (x = 8 or 9 https://github.com/raspberrypi/noobs/blob/master/recovery/util.cpp#L143 then reboots https://github.com/raspberrypi/noobs/blob/master/recovery/main.cpp#L39 Amazing , thanks very much. Is recovery.elf any differe...
by samskiter
Wed Feb 17, 2016 3:31 pm
Forum: General discussion
Topic: NOOBS boot process
Replies: 3
Views: 543

NOOBS boot process

Hi, I'm looking for a little more information on the NOOB boot process. Specifically, when is the dt-blob.bin file read? In initial boot of revovery.elf or in the soft reboot to the start.elf on one of your boot partitions? How does NOOBS do this softreboot and point bootcode.bin somewhere else? Is ...
by samskiter
Tue Jan 19, 2016 4:18 pm
Forum: General discussion
Topic: WiFi Dongle incompatibility with USB socket
Replies: 0
Views: 330

WiFi Dongle incompatibility with USB socket

Hi, I've had some trouble with physical incompatibility between the official raspberry pi wifi dongle (https://www.raspberrypi.org/products/usb-wifi-dongle/) and this socket: http://uk.farnell.com/te-connectivity-amp/292303-1/usb-2-0-type-a-plug-cable/dp/1076663 Specifically, I noticed that when wig...
by samskiter
Mon Jan 11, 2016 1:17 pm
Forum: Advanced users
Topic: Linux kernel is now 4.0
Replies: 182
Views: 89543

Re: Linux kernel is now 4.0

OK, can do - can you explain a little more? Is sdhost newly introduced then? and dtoverlay=mmc gives equivalent behaviour to before?
by samskiter
Mon Jan 11, 2016 12:20 pm
Forum: Advanced users
Topic: Linux kernel is now 4.0
Replies: 182
Views: 89543

Re: Linux kernel is now 4.0

Update: Scrap that - just got another one.

https://goo.gl/photos/zvfrtQ1D5JRwACh28
by samskiter
Mon Jan 11, 2016 12:17 pm
Forum: Advanced users
Topic: Linux kernel is now 4.0
Replies: 182
Views: 89543

Re: Linux kernel is now 4.0

On second thoughts, there is something else you could try: dtoverlay=sdhost,force_pio This disables DMA support. You will see a small decrease in performance and an increase in CPU load, but it will be useful to me to know if this solves your SD card problems. Had a chance to try this today and it ...
by samskiter
Thu Jan 07, 2016 10:01 pm
Forum: Device Tree
Topic: Some GPIOs ignored in device tree on Compute Module
Replies: 11
Views: 2310

Re: Some GPIOs ignored in device tree on Compute Module

Thanks for clarifying all, sounds like I've been tripped up by the "power-down mode" terminology.

I think we've got away with it as we can probably live with those GPIOs high as the board comes on and we've got lucky on others that are more critical :)

Now on to fix the new kernel bugs :)
by samskiter
Thu Jan 07, 2016 9:52 pm
Forum: Advanced users
Topic: Linux kernel is now 4.0
Replies: 182
Views: 89543

Re: Linux kernel is now 4.0

Hi, just updated to the latest firmware (following discussion here: https://www.raspberrypi.org/forums/viewtopic.php?f=107&t=1315780) and having some issues with my wifi dongle which runs as a host. Intermittently it won't come up. Plugged an HDMI cable in and voila, turns out it's throwing up durin...
by samskiter
Wed Jan 06, 2016 5:57 pm
Forum: Device Tree
Topic: Some GPIOs ignored in device tree on Compute Module
Replies: 11
Views: 2310

Re: Some GPIOs ignored in device tree on Compute Module

Hmm, it's slowly dawning on me that I've probably misunderstood the peripherals document and then subsequently fluked a test... My understanding was that "GPIO pull-ups are maintained even in power-down mode when the core is off, when all register contents is lost." Meant that my pulling states woul...
by samskiter
Wed Jan 06, 2016 4:32 pm
Forum: Device Tree
Topic: Some GPIOs ignored in device tree on Compute Module
Replies: 11
Views: 2310

Re: Some GPIOs ignored in device tree on Compute Module

Taking an old image and doing an rpi-update will only have changed the kernel and GPU firmware. All your installed packages will have stayed the same (need an "apt-get udpate", "apt-get upgrade" to do that bit). Even so you will have jumped significantly in kernel from about 3.18.11 to 4.1.15. Inde...
by samskiter
Wed Jan 06, 2016 4:08 pm
Forum: Device Tree
Topic: Some GPIOs ignored in device tree on Compute Module
Replies: 11
Views: 2310

Re: Some GPIOs ignored in device tree on Compute Module

Yes 9 months off is a long way in RPi land I suppose. Just ran rpi-update and seeing much better control of the GPIO pins. But now need to carefully test the system after such a fundamental change. Thanks for your help. The only other change I would like to do is adjust a GPIO *before* the DTS kicks...
by samskiter
Wed Jan 06, 2016 3:26 pm
Forum: Device Tree
Topic: Some GPIOs ignored in device tree on Compute Module
Replies: 11
Views: 2310

Re: Some GPIOs ignored in device tree on Compute Module

Thanks for the quick reply. I added some information to my first post that may be relevant. Current firmware:

Linux raspberrypi 3.18.11+ #781 PREEMPT Tue Apr 21 18:02:18 BST 2015 armv6l GNU/Linux

Not sure how far behind I am, but I assume I am behind. What's the latest firmware version?
by samskiter
Wed Jan 06, 2016 3:03 pm
Forum: Device Tree
Topic: Some GPIOs ignored in device tree on Compute Module
Replies: 11
Views: 2310

Some GPIOs ignored in device tree on Compute Module

So I've run into this problem previously and just avoided the pins I couldn't adjust but now I have no choice. When changing some of the pin setups in my device tree file, the change seems to be ignored, or overwritten during device boot. It seems to be the pins associated with the Camera that cause...
by samskiter
Thu Aug 13, 2015 2:21 pm
Forum: Device Tree
Topic: GPIO Outputs on Bank 1 on CM
Replies: 8
Views: 2768

Re: GPIO Outputs on Bank 1 on CM

[comicbookguy]fastest bugfix ever[/comicbookguy]

Thanks all! :)
by samskiter
Thu Aug 13, 2015 10:48 am
Forum: Device Tree
Topic: GPIO Outputs on Bank 1 on CM
Replies: 8
Views: 2768

Re: GPIO Outputs on Bank 1 on CM

Hi, thanks for the reply. It is a bug in raspi-gpio. I just tried wiringPi gpio utility and can toggle inputs successfully! UGH that was about half a day wasted :(
by samskiter
Wed Aug 12, 2015 6:26 pm
Forum: Device Tree
Topic: GPIO Outputs on Bank 1 on CM
Replies: 8
Views: 2768

GPIO Outputs on Bank 1 on CM

Hi, I'm having some trouble getting outputs to work on Bank 1 of the compute module and I'm wondering if it is device tree related... I've managed to flick GPIO pins 12 and 13 (Bank 0) just fine but I'm getting nothing from the pins I've tried on Bank 1 - 34,35,36,37 & 44. I've tried the same wires ...

Go to advanced search