 Con Kolivas
|
0e56dffd92
Only queue from backup pools once we have nothing staged.
|
13 years ago |
|
Con Kolivas
|
e16b7566bd
Don't keep queueing work indefinitely if we're in opt failover mode.
|
13 years ago |
|
Con Kolivas
|
6d1949e694
Make sure we don't opt out of queueing more work if all the queued work is from one pool.
|
13 years ago |
|
Con Kolivas
|
a5f7b953fe
Set lagging flag if we're on the last of our staged items.
|
13 years ago |
|
Con Kolivas
|
ee9e621e1c
Reinstate clone on grabbing work.
|
13 years ago |
|
Con Kolivas
|
55f7c14982
Grab clones from hashlist wherever possible first.
|
13 years ago |
|
Con Kolivas
|
c3e32274ee
Cull all the early queue requests since we request every time work is popped now.
|
13 years ago |
|
Con Kolivas
|
0fbd60ae37
Keep track of staged rollable work item counts to speed up clone_available.
|
13 years ago |
|
Con Kolivas
|
83b9ddfea4
Make expiry on should_roll to 2/3 time instead of share duration since some hardware will have very fast share times.
|
13 years ago |
|
Con Kolivas
|
8697d6a2ce
Do the cheaper comparison first.
|
13 years ago |
|
Con Kolivas
|
381c56f811
Check that we'll get 1 shares' worth of work time by rolling before saying we should roll the work.
|
13 years ago |
|
Con Kolivas
|
afcfea15a7
Simplify all those total_secs usages by initialising it to 1 second.
|
13 years ago |
|
Con Kolivas
|
5fadfdb219
Overlap queued decrementing with staged incrementing.
|
13 years ago |
|
Con Kolivas
|
9f1d9ce3b7
Artificially set the pool lagging flag on pool switch in failover only mode as well.
|
13 years ago |
|
Con Kolivas
|
a6b97327e1
Artificially set the pool lagging flag on work restart to avoid messages about slow pools after every longpoll.
|
13 years ago |
|
Con Kolivas
|
44e81218fd
Factor in opt_queue value into enough work queued or staged.
|
13 years ago |
|
Con Kolivas
|
611f1cec7c
Roll work whenever we can on getwork.
|
13 years ago |
|
Con Kolivas
|
fd0be1bb51
Queue requests for getwork regardless and test whether we should send for a getwork from the getwork thread itself.
|
13 years ago |
|
Con Kolivas
|
7d77c01619
Get rid of age_work().
|
13 years ago |
|
Con Kolivas
|
d1508bd40e
Merge pull request #296 from kanoi/api
|
13 years ago |
 Kano
|
95dff7363e
API allow display/change failover-only setting
|
13 years ago |
|
Con Kolivas
|
8e20456bc0
Check we are not lagging as well as there is enough work in getwork.
|
13 years ago |
|
Con Kolivas
|
00691ababf
Merge pull request #292 from kanoi/main
|
13 years ago |
|
Con Kolivas
|
d66742a8c1
Minimise locking and unlocking when getting counts by reusing shared mutex lock functions.
|
13 years ago |
|
Con Kolivas
|
c91a95459b
Avoid getting more work if by the time the getwork thread is spawned we find ourselves with enough work.
|
13 years ago |
|
Con Kolivas
|
f27bcb8ee5
Going back to e68ecf5eb275e1cc2dc22c7db35b0bd8d9c799de
|
13 years ago |
|
Con Kolivas
|
c892ded6e0
Make sure there are true pending staged work items as well in failover only mode.
|
13 years ago |
|
Con Kolivas
|
61003df49f
In failover-only mode we need to queue enough work for the local pool and ignore the total queued count.
|
13 years ago |
|
Con Kolivas
|
8aa61f6626
Make sure we have work from the current pool somewhere in the queue in case the queue is full of requests from a pool that has just died.
|
13 years ago |
|
Con Kolivas
|
c0aaf56a8d
Since all the counts use the same mutex, grab it only once.
|
13 years ago |
