| | 40 | * make the share layout more deterministic, to get rid of the RTT that's |
| | 41 | needed to fetch the offset table before we can fetch the UEB. Maybe |
| | 42 | reserve space for the UEB at the start of the share (but that makes it hard |
| | 43 | to have an extensible open-ended UEB). |
| | 44 | * since we probably do need to stick with the offset table, see if there's a |
| | 45 | way to guarantee having the same offsets across all shares. This would allow |
| | 46 | the table pulled (and validated) from one share to be used to compute read |
| | 47 | vectors for all other shares. We probably already have identical tables, |
| | 48 | but it'd be nice to make the guarantee explicit. The tables need to be |
| | 49 | covered by the UEB, else we can't share them: the first read might be from |
| | 50 | a corrupted share, and that would malign (hah!) the other shares. |
