On Tue, 16 Apr 2019 15:12:26 +0100, Dreamcat4 wrote: > A small further development: > > Found out there was actually a BIOS setting to enable SPD write after > all. Because this is an overclocking board. It was just buried in a > mess of menus / almost impossible to find. > > I enabled the option and rebooted. Unfortunately enabling the SPD > write, it did not change any of the bad behaviour. > > dmesg -w > [ 147.799024] i801_smbus 0000:00:1f.4: SPD Write Disable is set Somehow the SPD protection is still enabled. Either the BIOS option is doing something else, or you need a cold boot to reset this configuration bit. > [ 147.799056] i801_smbus 0000:00:1f.4: SMBus using PCI interrupt > > [ 351.051412] ee1004 2-0050: 512 byte EE1004-compliant SPD EEPROM, read-only > [ 351.051425] i2c i2c-2: new_device: Instantiated device ee1004 at 0x50 > > [ 362.268407] ee1004 2-0052: 512 byte EE1004-compliant SPD EEPROM, read-only > [ 362.268426] i2c i2c-2: new_device: Instantiated device ee1004 at 0x52 > > [ 378.888376] ee1004 2-0050: Failed to select page 0 (-6) > > > So my suspicion / hunch is that: > > * Perhaps this is something missing or broken in the ee1004 driver, > that fails with these ASUS motherboards > > Is it that this ee1004 driver was not used for quite a while? And only > came back recently for the DDR4 support. Than maybe that would make The ee1004 driver is specific to DDR4 memory. It has been developed in November 2017 and went upstream in October 2018 (v4.20). It probably has not seen too much testing so far. > some sense. And if ASUS was not specifically tested as a hardware > configuration. Could be doing some strange thing perhaps. Standing Asus was not tested by me. I know at least one more user but I don't know on what hardware. > in-between mapping the smbus to access the RAM. It's just a hunch. > Really I have no idea! Best guess. I'm at the same point. -- Jean Delvare SUSE L3 Support
