| 322 |
|
* situations in which we cannot guarantee to make node s |
| 323 |
|
* unreachable in this way: (1) If s is the trailing node of list |
| 324 |
|
* (i.e., with null next), then it is pinned as the target node |
| 325 |
< |
* for appends, so can only be removed later when other nodes are |
| 325 |
> |
* for appends, so can only be removed later after other nodes are |
| 326 |
|
* appended. (2) We cannot necessarily unlink s given a |
| 327 |
|
* predecessor node that is matched (including the case of being |
| 328 |
|
* cancelled): the predecessor may already be unspliced, in which |
| 344 |
|
* When these cases arise, rather than always retraversing the |
| 345 |
|
* entire list to find an actual predecessor to unlink (which |
| 346 |
|
* won't help for case (1) anyway), we record a conservative |
| 347 |
< |
* estimate of possible unsplice failures (in "sweepVotes"). We |
| 348 |
< |
* trigger a full sweep when the estimate exceeds a threshold |
| 349 |
< |
* indicating the maximum number of estimated removal failures to |
| 350 |
< |
* tolerate before sweeping through, unlinking cancelled nodes |
| 351 |
< |
* that were not unlinked upon initial removal. We perform sweeps |
| 352 |
< |
* by the thread hitting threshold (rather than background threads |
| 353 |
< |
* or by spreading work to other threads) because in the main |
| 354 |
< |
* contexts in which removal occurs, the caller is already |
| 355 |
< |
* timed-out, cancelled, or performing a potentially O(n) |
| 356 |
< |
* operation (i.e., remove(x)), none of which are time-critical |
| 357 |
< |
* enough to warrant the overhead that alternatives would impose |
| 358 |
< |
* on other threads. |
| 347 |
> |
* estimate of possible unsplice failures (in "sweepVotes"). |
| 348 |
> |
* We trigger a full sweep when the estimate exceeds a threshold |
| 349 |
> |
* ("SWEEP_THRESHOLD") indicating the maximum number of estimated |
| 350 |
> |
* removal failures to tolerate before sweeping through, unlinking |
| 351 |
> |
* cancelled nodes that were not unlinked upon initial removal. |
| 352 |
> |
* We perform sweeps by the thread hitting threshold (rather than |
| 353 |
> |
* background threads or by spreading work to other threads) |
| 354 |
> |
* because in the main contexts in which removal occurs, the |
| 355 |
> |
* caller is already timed-out, cancelled, or performing a |
| 356 |
> |
* potentially O(n) operation (e.g. remove(x)), none of which are |
| 357 |
> |
* time-critical enough to warrant the overhead that alternatives |
| 358 |
> |
* would impose on other threads. |
| 359 |
|
* |
| 360 |
|
* Because the sweepVotes estimate is conservative, and because |
| 361 |
|
* nodes become unlinked "naturally" as they fall off the head of |
| 428 |
|
} |
| 429 |
|
|
| 430 |
|
/** |
| 431 |
< |
* Creates a new node. Uses relaxed write because item can only |
| 432 |
< |
* be seen if followed by CAS. |
| 431 |
> |
* Constructs a new node. Uses relaxed write because item can |
| 432 |
> |
* only be seen after publication via casNext. |
| 433 |
|
*/ |
| 434 |
|
Node(Object item, boolean isData) { |
| 435 |
|
UNSAFE.putObject(this, itemOffset, item); // relaxed write |
| 877 |
|
} |
| 878 |
|
|
| 879 |
|
/** |
| 880 |
< |
* Unlinks matched nodes encountered in a traversal from head. |
| 880 |
> |
* Unlinks matched (typically cancelled) nodes encountered in a |
| 881 |
> |
* traversal from head. |
| 882 |
|
*/ |
| 883 |
|
private void sweep() { |
| 884 |
|
for (Node p = head, s, n; p != null && (s = p.next) != null; ) { |
| 885 |
< |
if (p == s) // stale |
| 886 |
< |
p = head; |
| 886 |
< |
else if (!s.isMatched()) |
| 885 |
> |
if (!s.isMatched()) |
| 886 |
> |
// Unmatched nodes are never self-linked |
| 887 |
|
p = s; |
| 888 |
|
else if ((n = s.next) == null) // trailing node is pinned |
| 889 |
|
break; |
| 890 |
+ |
else if (s == n) // stale |
| 891 |
+ |
// No need to also check for p == s, since that implies s == n |
| 892 |
+ |
p = head; |
| 893 |
|
else |
| 894 |
|
p.casNext(s, n); |
| 895 |
|
} |
