Details
- Reviewers
rnk
Diff Detail
Event Timeline
lgtm
lib/AST/VTableBuilder.cpp | ||
---|---|---|
2795 | So this fix (I think I wrote it?) was incorrect. We need to look back at the chain of method overrides in the current vftable, rather than all possible overrides. | |
2885–2886 | I would say something like: // Once a chain of method overrides adds a return adjusting vftable slot, all subsequent overrides will also use an extra method slot. Would you agree with that statement? |
lib/AST/VTableBuilder.cpp | ||
---|---|---|
2795 | Yep, you wrote it in r198080. | |
2885–2886 | Yes, this statement seems to be correct. In my comment I wanted to emphasize that we have to do it even if this is suboptimal vftable-size-wise, but apparently made the main point a bit too complex. Do you think I should just use your wording or merge the two? |
So this fix (I think I wrote it?) was incorrect. We need to look back at the chain of method overrides in the current vftable, rather than all possible overrides.