07:33 <thasti> is it expected that I get a usb1.USBErrorAccess: LIBUSB_ERROR_ACCESS [-3] on the first time I run `glasgow run ...` after plugging in the glasgow? second and any consecutive calls always work fine...

07:41 <d1b2> <dragonmux> it is not - what OS are you running?

08:00 <thasti> arch linux

08:01 <thasti> behavior is the same on two different machines (same OS)

08:04 <d1b2> <zyp> let me guess; first run reloads the fx2 causing it to reenumerate, and then after reenumeration it's accessed before udev have had time to fix permissions

08:05 <d1b2> <Darius> sounds plausible

08:09 <d1b2> <dragonmux> we'll check if our Glasgow exhibits that behaviour tomorrow as we also run Arch

08:10 <d1b2> <dragonmux> it didn't use to but we can't rule out that a recent kernel of systemd update hasn't screwed things up in the suggested way

08:12 <d1b2> <zyp> if it's a race condition against udev, there could be a bunch of factors contributing to whether it'll fail or not

08:15 <d1b2> <zyp> if my hunch is right, adding a sleep somewhere around here might help: https://github.com/GlasgowEmbedded/glasgow/blob/main/software/glasgow/device/hardware.py#L143

08:20 <thasti> cool, thanks for checking