I know that an i7-820QM (Clarksfield) is perfectly capable of working with 8 GB DIMMs and probably up to 32 GB or even more (the Thinkpad W510 has 4 DIMM slots and I have tested it with 20 GB) and that is from around the same time as the Arrendale chips - which does not mean anything but* I still refuse to accept that Nehalem is that limited*.
*The* *> official specs are not trustworthy IMHO and cpuid(1) and /proc/cpuinfo* *> show the same physical address width of 36 bits (which would indicate a* *> **maximum of 64 GB)*.
And let me add... Stefan, you are perfectly correct, *to the highlighted RED* marked by me in your original text/email. Nehalem is perfectly capable of addressing MORE than 8GB of DDR. And it can support for sure 16GB, maybe even 32GB of DDR.
And I even might know (actually, certainly, I do) why LENOVO 410s is NOT able/capable to support 8GB DIMMs.
Let me throw this challenge to INTEL IOTG. If they (after all, it is INTEL technology, isn't it!?) know what is the root cause of this problem, and then how to fix this problem. ;-)
INTEL IOTG, floor all yours!
Zoran
On Sat, Jan 21, 2017 at 11:55 AM, Stefan Tauner < stefan.tauner@alumni.tuwien.ac.at> wrote:
Hi Vladimir,
since you have REed the raminit for Nehalem I'd like you to ask if you have any knowledge, information or pointers about using 8 GB DIMMs with it or even using more than 8 GB in total. In my case it is about an Arrandale i5-520M (in a Thinkpad 410s).
I know that an i7-820QM (Clarksfield) is perfectly capable of working with 8 GB DIMMs and probably up to 32 GB or even more (the Thinkpad W510 has 4 DIMM slots and I have tested it with 20 GB) and that is from around the same time as the Arrendale chips - which does not mean anything but I still refuse to accept that Nehalem is that limited. The official specs are not trustworthy IMHO and cpuid(1) and /proc/cpuinfo show the same physical address width of 36 bits (which would indicate a maximum of 64 GB).
The current raminit for Nehalem in coreboot is not able to train the two 8 GB DIMMs I have tested so far. I have added a debug output to choose_reg178 in the first loop before the margins are compared to STANDARD_MIN_MARGIN that shows that all margins are 0. If there is anything I could try or information I can provide, please let me know.
The (ancient) vendor firmware I've been using on the T410s does sometimes manage to boot Linux with an 8 GB DIMM (dmesg is attached including the e820 map), but it is quite broken and memtest86 locks up or reboots within seconds so that's probably not a good target for RE efforts. :)
-- Kind regards/Mit freundlichen Grüßen, Stefan Tauner
-- coreboot mailing list: coreboot@coreboot.org https://www.coreboot.org/mailman/listinfo/coreboot