Not until a long time later, when i tried to use my dock again and discovered it didn't work. I think i just applied updates as they were offered by my system (Fedora) and didn't do or change anything special.
│ │ Update Error: could not validate firmware: FW image image not compatible to this controller (Native). LVFS page says it's for X1 Carbon 7th, but filename suggests it's for 5th? It doesn't help that a few of the released firmware images (either for Windows or on LVFS) are newer than the version in the "Use version of higher" column of Lenovo's primary Thunderbolt critical update page, so you'd have to click through to each Windows download page. There's at least one (X1 Carbon 7th) where the filename doesn't appear to match the entry name.
The correct approach is probably to match Lenovo's Windows release code with the equivalent code on LVFS. Does anyone know what _AssistMode is I think you're right - I put together the table by string searching for the machine name on the LVFS webpage, but some of the names don't quite match. *: Some devices appear to have two firmware entries, ?.firmware and ?_AssistMode.firmware. None? (X390 has version 20, but nothing for X390 Yoga) None? (X380 has version 20, but nothing for X380 Yoga) Do the fwupd developers have to deal with this, or is it something that Lenovo provides when they add firmware to LVFS?
Have we not been updating Thunderbolt firmware until now? Maybe it didn't seem like there was a compelling reason to do so.? There were one or two cases where LVFS had a version released a month or two earlier, but none that were really old.
I noticed that for each of the affected machines, LVFS either has Thunderbolt firmware version 20 / 43 (the version that fixes the bug), or no Thunderbolt firmware at all. Is the process of writing the firmware image to the correct location dramatically different on different machine types? haphazardly populated for a bug that can allegedly brick a machine, especially since Lenovo already has Windows utilities to update the firmware on each of the affected machine types.
I have more experience with embedded device firmware than PC firmware, so apologies if this comes off as naive: This table seems. The table below is a summary of firmware availability for affected machines on LVFS, as of. Failure of this ROM can apparently cause error messages, Thunderbolt chipset failure, or even boot delay / failure. It's not clear what the actual problem is, but some of the individual updates mentioned "ROM wear out," so it might be related to excessive writes to flash somewhere.