I'm subscribed to debian-boot as well as debian-bugs-dist, so I already get at least two copies of every d-i bug report. I do not need another one in my inbox, so please do not CC me. Most of your recent mails seem to disregard information I sent to this bug report in <[🔎] 20041208183439.GB5834@kitenet.net> and <[🔎] 20041208184714.GD5834@kitenet.net>, so here it is again: langagechooser runs before kbd-chooser so kbd-chooser bugfixes cannot apply to it. I had thought your keyboard was working before you got to kbd-chooser; if not it's probably a bug in the kernel/lack of loading the right module for a usb keyboard. Well here's what I know: Our sparc kernels are built with the usbkbd driver built in, and that's the only USB keyboard support provided. The usb-uhci or usb-ohci drivers are not on there and not loaded, and neither does it try to load the alternative hid-based keyboard drivers (hid, keybdev). So if the usbkbd driver works, using the keyboard in what the kernel calls it "Boot Protocol mode" I think the sparc stuff should work (unsure if it still needs to load usb-uhci firt though), but if it needs to be in full HID mode, a lot of work needs to be done to get sparc to support that. Vincent.McIntyre@csiro.au wrote: > Hmm, maybe this discovery stuff is all a red herring. Obviously so, since as I said before and above, the usb-uhci and usb-ohci modules are not available on the d-i sparc initrd. > Now, usb-linux runs first, and it loads full HID support (hid.o). Not on architectures such as sparc which have no hid.o on the initrd. Rather than stumbling around the d-i source tree lost and confused as you seem to have done, I'd suggest finding a system that correctly supports the USB keyboard and find out how it does it, then apply that to d-i. -- see shy jo
Attachment:
signature.asc
Description: Digital signature