According to Randall (https://bugs.chromium.org/p/chromium/issues/detail?id=391827), vboot2 platforms should set fw_vboot2 to 1.

"It seems to return 0 on reef" => do you mean the command returns exit code zero (success), or "crossystem fw_vboot2" prints 1?

See below:

localhost tmp # crossystem fw_vboot2  
1localhost tmp # crossystem 'fw_vboot2?1'
localhost tmp # echo $?
0
localhost tmp # 


But really I'm wondering why we appear to pass 'fw_vboot2?1' to crossystem?

So it looks same as expected.

> why we appear to pass 'fw_vboot2?1' to crossystem?
Sorry I don't understand what do you mean?

Originally we plan to support vboot2 by updater5, but (1) turns out there was not so much difference (2) many projects using vboot1 transit to vboot2 silently at that time, so in the end we added vboot2 support to updater4, and that's why it needs to check fw_vboot2 value to decide what to do.

Or do you mean the syntax of crossystem?

  "value_name?expected_value"

is a way to tell crossystem "check if value_name is expected_value", an easier way to do shell if without substitution.

in fact you'll find such usage in many places, for example platform2/init/chromeos_startup

OK I see, thank you. The problem is just that I misunderstood what it is doing. Thank you for explaining it.