Closed Amjad50 closed 8 months ago
Oops, indeed!
I partially reverted this in 3e4aa185d16ef3b6535febc51b5656411a57ca06 because we don't actually need the data to be cloneable. We ask for the ranges multiple times, but we only ask for the section data once, so moving it is fine.
The macOS code inside samply was running into the Clone bound.
This was triggering a bug in
DwarfCfiIndex
as:try_new_eh_frame
will callbase_addresses_for_sections
which will take the values, then the second timebase_addresses_for_sections
is called, it will get zeros bases, which is invalid and will not produce any frames