Hi Samuel, On 26/6/23 02:41, Samuel Thibault wrote: >> This simple change allows hurd to be bootable off usb! > > Well, yes and no :) > > We also need /dev entries. It happens that this re-uses /dev/sd* > names, so we need to care about compatibility. We probably want > rumpdisk_device_open to forward to the kernel when 'disabled' is true. OK
>> It is not ideal to have entire usb stack with the mass storage driver >> and combined with SATA, but there is no easy way to separate >> the usb stack into host/device yet. > > Ok, but can't we build a separate rumpusbdisk that only has the USB > drivers? Each process (SATA+IDE and USB) can easily drive its own PCI > card. Yes that is certainly possible, I can work on that next. >> Caveats: USB seems to share IRQs with network device, >> which does not seem to work well currently. >> I am not sure of the status of irq sharing but it seems broken. > > That was supposed to be working, but evil lies in the details as usual. > I have a feeling that ACPI provides a list of possible IRQs per device and it's up to the host to figure out a set of unconflicting configurations and to set them using _SRS methods. Where can I find this info? Damien
