Attention is currently required from: Arthur Heymans, Jérémy Compostella.
1 comment:
Commit Message:
Patch Set #1, Line 9: We most likely operate on regions that end exactly at the 4GiB boundary
would it make sense to document that 64bit + overflow is still possible?
Well, the idea of the region_create() functions is to avoid overflows.
I realized only later that region_end() still can overflow. This is most
annoying, and only because region_end() doesn't return an inclusive value.
I'd prefer to discuss the implementations further.
Maybe, if nobody objects, we could make region_end() return an inclusive
offset. It's return value doesn't seem to be used much. My only concern
right now is that it could confuse developers that are used to the current
behaviour.
To view, visit change 79906. To unsubscribe, or for help writing mail filters, visit settings.