Hi Mathias, On Wed, Jun 29, 2022 at 05:39:49PM +0300, Mathias Nyman wrote: > Anyway, maybe flushing the CRCR register by reading it back after writing it would help: Thanks for your help and sorry for the delay. I tried the patch but it didn't help, the io fault still appears and the device does not work. I added some more debugging, actually printing the cmd_ring address read by the calls you patch adds. Unfortunately they all returned 0, looks like the register is write-only RAZ on this hardware. Is it possible that the device actually can't handle a 64 bit DMA mask and needs a smaller one, say 40 bits? Regards, -- Jörg Rödel jroedel@xxxxxxx SUSE Software Solutions Germany GmbH Frankenstraße 146 90461 Nürnberg Germany (HRB 36809, AG Nürnberg) Geschäftsführer: Ivo Totev, Andrew Myers, Andrew McDonald, Boudien Moerman
