I had no luck with the dmi tools floating on the net, either they did not work at all or my hw was not supported. Most of the utils error msg was ’ Your notebook is not supported by this utility’ so clearly they are only for laptops and not workstations.
So I tried some soldering instead.
Bios chip is quite close to the sata ports.
Wetted properly.
…and removed chip and cleaned pads
I reprogrammed it with an unmodified bios file from HP.
I kind of new I would have a few problems so I decided not to resolder the chip back on but instead use the pliers trick - it works beautifully as the pliers press point fits and holds the chip in place.
Machine starts up and AMT option is back in bios, sadly mac lan address was FFFFFFFFFFFF but using Intel’s Eeupdate /nic number /mac=123456789123 in dos fixed it back to what I earlier had. Intel’s hw test passes in windows so seems to be ok.
Ctrl+A opens additonal fields in Security>System IDs menu so I successfully added the proper asset tag, UUID, chassis sn, build ID and featurebytes. Remember that the build ID has a few ** on your chassis label so take picture of bios system IDs menu before you reflash and start experimenting.
The new problem now is ‘Management platform (ME) in manufacturing mode’ and I can’t seem to fix it.
Ahh plutomaniac posted som nice information about ME manufacturing mode - link
I ran ‘fptw -closemnf’ in windows and computer restarted and ‘Management platform (ME) in manufacturing mode’ was gone. I don’t know if it’s needed but I used -greset after reebot.
Before I solder back the chip does anyone know if this looks correct ? Not provisioned and Pre provisioning I not sure about.
Very nice, you found a different solution and detailed it as well. You can use MEInfo tool to see more robust AMT state info instead of Windows System Info. You should also be able to control AMT provisioning via MEBx while the system boots. Generally, if you don’t use AMT, it should stay unprovisioned.
You’ll see even more info at MEInfo if you add -verbose command. These Slot info are probably taken from the DMI so they are completely useless. Either way though, the firmware is now healthy so you can commit to the current SPI state. If you want you can configure AMT via the MEBx, it might be possible to completely disable it there as well.
MEInfo -verbose lists ‘No error’ under ErrorCode, AMT option is back in bios and I have no yellow exclamation mark in device manager so I think I’m happy, I’ll just resolder the chip and call it a success.
Wait, I saw something. Can you run “fptw -greset” command and check if after the restart you can now see “Valid OEM data, ICC programmed” next to ICC? That shouldn’t say “No valid OEM data, ICC not programmed”.
ICC is Integrated Clock Controller, an important regulator for system clocks. There is no upgrade path from ME8 to ME9-10, only to the latest ME8 firmware, as provided at the (CS)ME thread. Since HP provides a full SPI image for your workstation, what you need to do is to make sure that you have the latest HP BIOS v3.92 applied.
Given that, I think I know what’s going on now and if I’m right, it should affect all Z420 users. HP messed up while updating the ME firmware at the newer SPI/BIOS updates. Normally, to update it they need to do everything from scratch which takes a lot of time so what the community does (and probably OEMs as well) is to follow the Engine CleanUp Guide to clean and update the firmware at the same time. However, from experience we (the community) know that Flash Image Tool v8.1.40.1456 or earlier versions are bugged when it comes to Patsburg (C600) SPI images and the ME settings are not detected when following the CleanUp Guide (Step 5 > 2nd bullet). So I think that HP did the same procedure as depicted in that guide but was not aware that their (probably outdated) Flash Image Tool was bugged on the Z420’s Patsburg platform and did not do the workaround.
Flash the attached (hopefully) fixed SPI image with the programmer, hold it with the clip (nice idea btw ), do “fptw -closemnf” and then “fptw -greset”. Afterwards, check if ICC has valid OEM data.
Still has ‘No valid OEM data, ICC not programmed’ after the new flash.
What I did notice is that when running “fptw -greset” the command prints in yellow ‘Could not set the GlobalReset bit’ but finishes of with ‘FPT Operation Passed’ in green writing.
I’ve been looking at setting up ICC profiles in the bring up guides and it looks too complicated so I think I’m going to try an old bios like 1.14 and see if that changes anything, if not I need a Z420 or Z620 bios dump.
Ah if you have kept such older BIOS then great. At the HP site I couldn’t find anything old enough (same wrongfully upgraded ME firmware). Finding another dump won’t help because we might not be able to clean it due to the FIT 8.1.40 bug with C600 SKUs. And yes, there is no point in trying to mess with ICC as it’s very complicated.
We’re not doing something wrong. I’m certain that the guide was followed by HP because of a harmless side-effect it has in which the Flash Image Tool version is set to be the same as the actual ME firmware (as seen in MEA) even though that may not actually be the case. For example, the firmware/tools/documentation packages from 8.1.40.1416 up until at least 8.1.65.1586 (last we’ve found) included Flash Image Tool 8.1.40.1456. However, BIOS 3.92 includes ME 8.1.60.1561 (in between 8.1.40 and 8.1.65 packages) but shows Flash Image Tool version of 8.1.60.1561 which makes no sense as it doesn’t exist. If they had done everything manually (I don’t blame them, only ME 8 C600 is bugged) the firmware would show as 8.1.60.1561 but FIT as 8.1.40.1456. Hopefully you get what I’m trying to say.
Anyway, it doesn’t matter. We don’t want very old BIOS because these had ME 7 which got later upgraded to ME 8. I think BIOS 2.55 with ME 8.1.10.1286 is a good candidate to flash as it is and test if the error goes away. FIT 8.1.10.1286 does exist. Don’t forget to do a manual greset by removing all power (AC, RTC not needed) for 1 minute after re-flashing the SPI chip.
Well I don’t think we can try something else, especially with the bugged FIT 8.1.40 in combination with the Patsburg platform. I suggest you flash the latest BIOS from HP and then proceed to close ME manufacturing mode to stop the error from showing. Then do a “-greset” and call it a day. I don’t think you’ll see any issue during everyday use.
Agree, it’s not like we didn’t try - AMT option is back in bios and I have no yellow exclamation mark in device manager and no manufacturer mode so I’ll solder it back on.
Hi plutomatiac and Twist, I know that this is an old thread, i was following every step that you did, but at the end, how did you manage to get the AMT options back in BIOS, since I lost them and I am not able to get them back.
I have two identical laptops, HP Elitebook 2170p, same CPU, exactly the same, but one of them has AMT options in BIOS and has Intel ME in windows, but other one lost AMT BIOS options (MeBX) and had locked ME FW and vpro. I managed to unlock ME and vpro and AMT, but it is still missing AMT options in BIOS.
Is there a way to pull all BIOS data/ME from working one and apply it to non working one?
Not much familiar wit entire SPI/BIOS/ME FW story, pretty much learned all from reading your posts. Since BIOS chip on this board is removable and I have bought new BIOS chips and CH341A programmer, i took BIOS from working one and placed it in non working one, but its still not showing AMT options in BIOS.
), do “fptw -closemnf” and then “fptw -greset”. Afterwards, check if ICC has valid OEM data.
