Kaczorek wrote: What you're saying is the problem does not occur on x86/x64 architecture, right?
I would be surprised if there is any difference between vanilla and raspberry kernel for drivers/usb/serial branch. Could you do diff between the two?
Kaczorek wrote: I have started to review linux kernel tree for usb serial - see: github.com/torvalds/linux/tree/master/drivers/usb/serial
As for now I cannot find a commit that introduces access rights issue on usb bus. Maybe you can find it.
SteveBz wrote: - Sync almost never works, and this is my main problem. It almost never works with Ekos, it never works with PyIndi-Client.
JCJ wrote: My idea was to use PHD2 : Brain / Algorithms / Predictive PEC : is Period Length (PHD2)= Major Period (EKOS) ?
Using PHD2 with KStars 3.4.3 works fine for me. I see in the error message it's trying to connect on port 0. That is incorrect. It should be port 4400. In your profile editor, make sure the port number is set correctly.
Avocette wrote: You are quite correct - nothing is yet off the hook. Attached is my lsusb list from the original working version, pre-kernel-upgrade. The PPB is Bus 001 Device 004. The EQdirect cable has a Proflific brand serial adapter which I recall caused a number of problems in the past - we were all advised to avoid Prolific and choose FTDI instead. I guess I have to get used to running lsusb before starting up KStars and counting the entries before resorting to replugging everything!
$ lsusb Bus 002 Device 004: ID 03c3:1604 Bus 002 Device 003: ID 174c:3074 ASMedia Technology Inc. ASM1074 SuperSpeed hub Bus 002 Device 002: ID 174c:3074 ASMedia Technology Inc. ASM1074 SuperSpeed hub Bus 002 Device 001: ID 1d6b:0003 Linux Foundation 3.0 root hub Bus 001 Device 004: ID 0403:6001 Future Technology Devices International, Ltd FT232 Serial (UART) IC Bus 001 Device 006: ID 0bda:b812 Realtek Semiconductor Corp. Bus 001 Device 009: ID 03c3:1f01 Bus 001 Device 008: ID 04b4:6572 Cypress Semiconductor Corp. Bus 001 Device 010: ID 1618:0921 Bus 001 Device 005: ID 174c:2074 ASMedia Technology Inc. ASM1074 High-Speed hub Bus 001 Device 003: ID 174c:2074 ASMedia Technology Inc. ASM1074 High-Speed hub Bus 001 Device 002: ID 2109:3431 VIA Labs, Inc. Hub Bus 001 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
Did a little research, and it seems the Pegasus PPB is controlled via USB-to-Serial internally (you plug a USB cable into it, but internally it has a USB-to-Serial adapter, and it shows up on the host computer as a serial device). So it seems USB-to-serial adapters aren't off the hook yet!
Hmm, well I guess it isn't as simple as I had hoped. It seems each of us has "that one device" that breaks things, but it isn't the same device for all of us.
rdricks wrote: I’m using an EQDirect cable, but mine has always been plugged directly into the RPi. My camera s plugged into the hub, and the focuser into the camera. The focuser is my only other serial device.
So an interesting development. I tried a lot of different things, but I narrowed it down to the USB-to-Serial adapter (EQDirect cable). If I leave it unplugged, or plugged directly into the Pi, everything works perfectly fine. But, as soon as I plug it into the USB3 hub, that's when things get wonky.
Now my question for the group is, is it just me, or is it the same for you guys as well? If you're having the same problems, are you using a USB-to-Serial cable (for example, an EQDirect cable)? Even if you have a mount with a USB port on it, those are often USB-to-Serial internally. And if you do have one of those, does unplugging it (and rebooting) make things work again?
Link to the raspberrypi.org forum post:
Just for fun (because the alternative is to actually do some real work) I'm updating my Linux laptop to a 5.4 kernel, to see if the problem also exists there.
knro wrote: So that doesn't work in my case. So I tried changing what they're pointed to:
chmod -R +rx /sys/devices/platform/scb/fd500000.pcie
So while that makes lsusb works, when you try to use drivers they cannot access libusb still (e.g. QHY). So it's still broken.