| |
| (One intermediate revision by one other user not shown) |
| Line 1: | Line 1: |
| − | Unclear computation?
| |
| | | | |
| − | The text says:
| |
| − | ''Each turf and thread combination has its own stack memory block called a stacklet.
| |
| − | [...] the upper part of the address space is reserved for stacklets in a 2 dimensional array indexed by thread ID and turf ID. A system that allows 1M turfs and 1M threads and 8k stacklets reserves the upper 8GB of the address space.''
| |
| − |
| |
| − | As far as I can see, 1e6 × 1e6 × 8KB is not 8GB.
| |
| − | 1000 turfs × 1000 threads would be more plausible.
| |
| − | [[User:Afranke|Afranke]] ([[User talk:Afranke|talk]]) 22:22, 23 March 2015 (UTC)
| |
| − |
| |
| − |
| |
| − | == Grant / Revoke and Collusion ==
| |
| − |
| |
| − | The text states that:
| |
| − |
| |
| − | ''Granting and revoking rights are hardware operations. They can be explicitly done to a specific turf or thread with grant and revoke, which puts a new region entry into the PLB or removes it, respectively. ''
| |
| − |
| |
| − | However, it does not specify how (if it all) it is determined that I have the right to communicate with that specific turf or thread, in fact, if both of the fields can be wildcarded then I can grant to all of them. If that is the case, is it possible to truly isolate two processes and prevent collusion?
| |
| − |
| |
| − | It would be useful to state something to this effect or, if I've misunderstood, some clarification.
| |
| − |
| |
| − | --[[User:Williammlleslie|Williammlleslie]] ([[User talk:Williammlleslie|talk]]) 05:56, 17 June 2015 (UTC)
| |
