Raspberry Pi Forums

The easiest way is to get udev to change the permissions for you. In an ideal world you would create a new group - "leds", say - arranging that the user belongs to it and that the /sys/class/leds entries are owned by it. We can make life easier for ourselves by reusing an existing group wh...

So there's no misunderstanding, the firmware in our rpi-firmware repo (which is installed by "sudo rpi-update") does include support for the HAT map but does not include the HAT map itself - you'll need to add it from the Google drive link above.

Note that if you have access to rpi-update in Recalbox then you can now run "sudo rpi-update pulls/5610" to install the latest kernel including the mapping for your HAT.

i2c10 is the numeric ID for I2C0 on the camera GPIOs, but in DT it's called "i2c_csi_dsi". You can probably get over many hurdles like this by reading one of the existing camera overlays, e.g. https://github.com/raspberrypi/linux/blob/rpi-6.1.y/arch/arm/boot/dts/overlays/imx708-overlay.dts...

Documentation is available on all the overlays: pi@raspberrypi:~$ dtoverlay -h spi1-3cs Name: spi1-3cs Info: Enables spi1 with three chip select (CS) lines and associated spidev dev nodes. The gpio pin numbers for the CS lines and spidev device node creation are configurable. N.B.: spi1 is only acce...

The latest rpi-update release includes firmware that understands the HAT map. The map itself will appear in future kernel releases, but for now you can download a version from here: https://drive.google.com/file/d/1VGNyPKsCuOWwxevNqEEGCGnlQUej-Cfi/view?usp=sharing The file is called "hat_map.dt...

I think I've got this mechanism working in test builds. If you give me the UUID of your HAT and the name of an overlay to load instead I can get you a trial firmware build and new "hat_map.dtb" to make it all work.

[ Moved to Interfacing, since this is not about Device Tree configuration ] Users of this overlay are required to provide an external mechanism to switch off the power supply when signalled - failure to do so results in a kernel BUG, increased power consumption and undefined behaviour. This was writ...

Yes, unfortunately that's just the way it is. There's a reason the dtparam to enable I2C0 is called "i2c_vc" - it's normally owned by the VideoCore VPU. Freeing up that bus for use by the ARM cores is one of many advantages of the switch to libcamera.

An I2C-attached GPIO then?

There isn't a published list, and there is no test pad on the ACT_LED signal.

You can suppress the PCIe link down message by adding "loglevel=3" to your cmdline.txt. Note that you should remove "quiet", or put it before "loglevel=3", because "quiet" seems to be effectively "loglevel=4" and the options are processed in reading ...

1. What is the failure mode - no /dev/i2c-1? Or just no response? 2. What does "raspi-gpio get 0-3" return? 3. What happens when you use the same config.txt in a regular RPiOS image? Posting a kernel boot log might help to identify the problem. You really shouldn't need to create a custom ...

I think we're going to do the same and add a third parameter to all affected overlays.

Why doesn't it throw an error when compiling the overlay

Because the overlay is compiled in isolation, without reference to any/all of the possible base DTB/DTS files. If the "pwm:" label existed in the overlay itself then you might get a complaint of some description.

Code: Select all

What is the reason that #pmw-cells had to be increased to '3'? Only because of polarity?

It would seem so.

Where to find the dt-bindings related to this

In the usual place: Documentation/devicetree/bindings/pwm/pwm-bcm2835.txt

Yes - this upstream change has broken it: https://github.com/raspberrypi/linux/co ... b307b66754

We can restore the old #pwm-cells value in our tree, but I've asked the patch author and reviewer what they suggest.

I guess you are right about the cost - we have previously used FTDI UART cables and FTDI JTAG adaptors, and in comparison to both of those the Debug Probe is super cheap, but there's no comparison to something that costs a dollar if all you want is a UART.

Because not everyone likes the idea of discarding what seemed like perfectly good hardware until one small problem was found? Because the case isn't large enough? There are many possible reasons.

I'm afraid that what you are asking for is not possible - BCM2835 (and BCM2836 and BCM2837) only has one position in the range GPIO 0 to GPIO 27 (the GPIOs that appear on the 40-pin header) for each UART TX and RX, i.e. TXD0, RXD0, TXD1 and RXD1, and those positions are the same: TXD0 and TXD1 are b...

It's a pity you didn't just put the name of the overlay in the EEPROM, but you are where you are. Yours isn't the only HAT with this problem, and I'm thinking we could add a table of exceptions that lists known stale HAT UUIDs (or perhaps hashes of the overlays) and the names of overlays to load ins...

Although the Amazon product page says to connect the red wire to 5V, I think it's actually a voltage reference (the UART itself gets its power from the USB lead) - try connecting it to 3V3. The STA pin is actually used to deliver a pulse to request a shutdown. The software resources include a Python...

Unfortunately all the pulls and levels set by the SoC are lost at a shutdown because part of the shutdown is a full chip reset. If you want a non-standard pull (where standard is high for GPIOs 0-8 and low for 9-27} that survives a reset then it will have to come from an external resistor that is sm...

That should be compiled to 'vc4-kms-dsi-com35h3p7.dtbo' hence the config.txt line for calling it would be 'dtoverlay=vc4-kms-dsi-com35h3p7'

Correct, but the fact that the line is commented out makes it moot.

Code: Select all

idnore_lcd=1

This is a typo - it should be "ignore_lcd=1".

Post the contents of (or at least the modifications to) config.txt.

Search found 4870 matches

Re: Permission denied - /sys/class/leds/led0/trigger

Re: Kernel 6.1 and dts: brcm,disable-headphones removed

Re: Kernel 6.1 and dts: brcm,disable-headphones removed

Re: Adding a new i2c image sensor to RPI 4B

Re: what is /dev/spidev0.1 and where is SPI1?

Re: Kernel 6.1 and dts: brcm,disable-headphones removed

Re: Kernel 6.1 and dts: brcm,disable-headphones removed

Re: Using gpio-poweroff with global_en for controlled power up and shutdown

Re: Videocore I2C

Re: RPI 4 test pad list

Re: RPI 4 test pad list

Re: Hide start up screens

Re: cmProvisioner, scriptexecute.img, Passing devicetree for I2c

Re: Kernel 6.6.0 - PWM seems not working/available

Re: Kernel 6.6.0 - PWM seems not working/available

Re: Kernel 6.6.0 - PWM seems not working/available

Re: Kernel 6.6.0 - PWM seems not working/available

Re: Attaching both UARTS to GPIO pins on a Raspberry Pi Zero v1.3

Re: Attaching both UARTS to GPIO pins on a Raspberry Pi Zero v1.3

Re: Attaching both UARTS to GPIO pins on a Raspberry Pi Zero v1.3

Re: Kernel 6.1 and dts: brcm,disable-headphones removed

Re: USB alternative for UART to GPIO 14/15/18

Re: Best way to insure certain GPIO pins start high or low at boot time?

Re: VC4 not working with DSI overlays loaded

Re: VC4 not working with DSI overlays loaded