czw., 2 kwi 2020 o 14:55 Ferruh Yigit <ferruh.yi...@intel.com> napisał(a): > > On 4/1/2020 3:21 PM, Michal Krawczyk wrote: > > ENA device is using 48-bit memory for IO. because of that, the upper > > limit had to be updated. > > What is the impact of this change, and what is the reason of the change? Can > you > please explain in the commit log? >
It's just a cosmetic change in our case and to align it with what the device is in fact doing. Structure 'ena_common_mem_addr' could be misleading - it was defining 64 bits for address, while the device in fact could read up to 48 bits. This value (address) is being verified before being set in 'ena_com_mem_addr_set()' function, so the functionality remains still the same. I'll update the commit log in v3.
