|
f6d2853c
|
2012-08-18T19:53:43
|
|
Summary layout difference.
|
|
|
ed3eb577
|
2012-08-18T15:44:48
|
|
Introduce a new statistic, Work Utility, which is the number of difficulty 1 shares solved per minute. This is useful for measuring a relative rate of work that is independent of reject rate and target difficulty.
|
|
|
26ab7346
|
2012-08-18T15:27:53
|
|
Implement a new pool strategy, BALANCE, which monitors work performed per pool as a rolling average every 10 minutes to try and distribute work evenly over all the pools.
Do this by monitoring diff1 solutions to allow different difficulty target pools to be treated equally, along with solo mining.
Update the documentation to describe this strategy and more accurately describe the load-balance one.
|
|
|
d8c76bbd
|
2012-08-18T11:26:11
|
|
Getwork fail was not being detected. Remove a vast amount of unused variables and functions used in the old queue request mechanism and redefine the getfail testing.
|
|
|
8fa70809
|
2012-08-18T11:07:32
|
|
Don't try to start devices that don't support scrypt when scrypt mining.
|
|
|
b814b42c
|
2012-08-16T18:55:09
|
|
Consider us lagging only once our queue is almost full and no staged work.
|
|
|
afcfea15
|
2012-08-16T09:30:49
|
|
Simplify all those total_secs usages by initialising it to 1 second.
|
|
|
5fadfdb2
|
2012-08-16T09:19:45
|
|
Overlap queued decrementing with staged incrementing.
|
|
|
9f1d9ce3
|
2012-08-16T00:16:13
|
|
Artificially set the pool lagging flag on pool switch in failover only mode as well.
|
|
|
a6b97327
|
2012-08-16T00:09:50
|
|
Artificially set the pool lagging flag on work restart to avoid messages about slow pools after every longpoll.
|
|
|
44e81218
|
2012-08-15T22:56:18
|
|
Factor in opt_queue value into enough work queued or staged.
|
|
|
611f1cec
|
2012-08-15T22:48:08
|
|
Roll work whenever we can on getwork.
|
|
|
b60f9da4
|
2012-08-16T17:24:04
|
|
Simplify the enough work algorithm dramatically.
|
|
|
0e56dffd
|
2012-08-16T17:01:25
|
|
Only queue from backup pools once we have nothing staged.
|
|
|
e16b7566
|
2012-08-16T15:49:11
|
|
Don't keep queueing work indefinitely if we're in opt failover mode.
|
|
|
6d1949e6
|
2012-08-16T15:46:41
|
|
Make sure we don't opt out of queueing more work if all the queued work is from one pool.
|
|
|
a5f7b953
|
2012-08-16T11:47:00
|
|
Set lagging flag if we're on the last of our staged items.
|
|
|
ee9e621e
|
2012-08-16T10:59:49
|
|
Reinstate clone on grabbing work.
|
|
|
55f7c149
|
2012-08-16T10:55:57
|
|
Grab clones from hashlist wherever possible first.
|
|
|
c3e32274
|
2012-08-16T10:41:01
|
|
Cull all the early queue requests since we request every time work is popped now.
|
|
|
0fbd60ae
|
2012-08-16T10:24:52
|
|
Keep track of staged rollable work item counts to speed up clone_available.
|
|
|
83b9ddfe
|
2012-08-16T10:18:30
|
|
Make expiry on should_roll to 2/3 time instead of share duration since some hardware will have very fast share times.
|
|
|
8697d6a2
|
2012-08-16T10:16:58
|
|
Do the cheaper comparison first.
|
|
|
381c56f8
|
2012-08-16T09:53:36
|
|
Check that we'll get 1 shares' worth of work time by rolling before saying we should roll the work.
|
|
|
fd0be1bb
|
2012-08-15T22:28:09
|
|
Queue requests for getwork regardless and test whether we should send for a getwork from the getwork thread itself.
|
|
|
7d77c016
|
2012-08-15T21:26:50
|
|
Get rid of age_work().
|
|
|
d1508bd4
|
2012-08-14T14:45:34
|
|
Merge pull request #296 from kanoi/api
API allow display/change failover-only setting (API v1.16)
|
|
95dff736
|
2012-08-14T12:54:27
|
|
API allow display/change failover-only setting
|
|
|
8e20456b
|
2012-08-14T09:13:59
|
|
Check we are not lagging as well as there is enough work in getwork.
|
|
|
00691aba
|
2012-08-11T20:42:05
|
|
Merge pull request #292 from kanoi/main
Escape " and \ when writing json config file
|
|
|
d66742a8
|
2012-08-12T11:04:51
|
|
Minimise locking and unlocking when getting counts by reusing shared mutex lock functions.
|
|
|
c91a9545
|
2012-08-12T10:55:18
|
|
Avoid getting more work if by the time the getwork thread is spawned we find ourselves with enough work.
|
|
|
f27bcb8e
|
2012-08-12T10:45:59
|
|
Going back to e68ecf5eb275e1cc2dc22c7db35b0bd8d9c799de
|
|
|
c892ded6
|
2012-08-12T10:40:58
|
|
Make sure there are true pending staged work items as well in failover only mode.
|
|
|
61003df4
|
2012-08-12T10:33:26
|
|
In failover-only mode we need to queue enough work for the local pool and ignore the total queued count.
|
|
|
8aa61f66
|
2012-08-11T16:47:49
|
|
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.
|
|
|
c0aaf56a
|
2012-08-11T16:11:53
|
|
Since all the counts use the same mutex, grab it only once.
|
|
|
4f9394be
|
2012-08-11T14:39:35
|
|
When popping work, grab cloned work first if possible since original work can be reused to make further clones.
|
|
|
8085ae68
|
2012-08-11T14:04:55
|
|
Further simplify the queue request mechanism.
|
|
|
f83863a9
|
2012-08-11T13:30:41
|
|
Keep total queued count as a fake pending staged count to account for the period a queue is in flight before it is staged.
|
|
|
e47dc873
|
2012-08-10T21:44:15
|
|
Clone work at the time of requesting it if an existing work item can be rolled.
|
|
|
e68ecf5e
|
2012-08-10T20:18:04
|
|
Queue one request for each staged request removed, keeping the staged request count optimal at all times.
|
|
|
52e5524d
|
2012-08-09T18:01:35
|
|
Escape " and \ when writing json config file
|
|
|
3dd1658e
|
2012-08-07T12:10:01
|
|
We may as well leave one curl still available per pool instead of reaping the last one.
|
|
|
c7bcad65
|
2012-08-07T11:59:54
|
|
Need to recheck the pool->curls count on regaining the pool lock after the pthread conditional wait returns.
|
|
|
ad8c4b77
|
2012-08-07T11:52:37
|
|
Revert "Only add to the pool curlring and increment the counter under mutex lock."
This reverts commit 8897e0657507072448f4e951ce1caadca15b4b30.
Wrong fix. Recursive locking now.
|
|
|
145f04cc
|
2012-08-07T11:10:59
|
|
Display reaped debug message outside mutex lock to avoid recursive locking.
|
|
|
8897e065
|
2012-08-07T11:05:23
|
|
Only add to the pool curlring and increment the counter under mutex lock.
|
|
|
25fd6cd0
|
2012-08-05T15:35:37
|
|
Correct writing of scrypt parameters to config file based on command line parameters only.
|
|
|
ac131d31
|
2012-08-04T22:34:47
|
|
Merge pull request #279 from tyeken8/tydus
Add scrypt support while writing conf
|
|
|
7611499b
|
2012-08-05T14:57:46
|
|
Queue a request on pool switch in case we have no work from the new pool yet.
|
|
|
4efa31d2
|
2012-08-05T14:44:11
|
|
Display failover only mode in pool menu and allow it to be toggled live.
|
|
|
1a041668
|
2012-08-05T14:37:49
|
|
Reinstate check for system queueing lag when the current pool's queue is maxed out, there is no staged work, and the work is needed now.
|
|
|
499c594c
|
2012-08-05T14:29:21
|
|
There is no need for pool active testing to be mandatory any more with queue request changes.
|
|
|
e4326e3c
|
2012-08-05T14:24:31
|
|
Fix harmless warnings.
|
|
|
9a45a6d9
|
2012-08-05T14:19:49
|
|
Check the current staged and global queued as well before queueing requests.
Discard stales before ageing work in the watchdog thread.
Queue requests after discarding and ageing work in watchdog thread.
Display accurate global queued in curses output.
Reuse variable in age_work().
|
|
|
bf9c3247
|
2012-08-05T13:40:46
|
|
The queueing mechanism has become a complex state machine that is no longer predictable. Rewrite it from scratch watching only current queues in flight and staged work available on a pool by pool basis.
|
|
|
b66e0241
|
2012-08-04T09:52:18
|
|
Queue an extra request whenever staged work drops below mining thread count in hash_pop.
|
|
|
b70ec3cb
|
2012-08-02T17:17:53
|
|
Merge pull request #283 from kanoi/icarus
ICA support 57600 baud rate, up to 8 FPGA and partial working FPGA boards
|
|
|
15dc4bb3
|
2012-08-03T10:10:53
|
|
Scrypt mining does not support block testing yet so don't try to print it.
|
|
|
2953aa25
|
2012-08-02T22:46:49
|
|
We dropped the temporary stopping of curl recruiting on submit_fail by mistake, reinstate it.
|
|
|
9410875e
|
2012-08-02T22:46:13
|
|
Make threads report in either side of the scanhash function in case we miss reporting in when restarting work.
|
|
|
d5dffa6a
|
2012-08-02T16:45:12
|
|
Don't make mandatory work and its clones last forever.
|
|
|
aa52db45
|
2012-08-01T23:42:00
|
|
Make test work for pool_active mandatory work items to smooth out staged work counts when in failover-only mode.
|
|
|
e067be42
|
2012-08-01T22:50:30
|
|
ICA support 57600 baud rate, up to 8 FPGA and partial working FPGA boards
|
|
|
8414a9a7
|
2012-08-01T22:48:38
|
|
Add debugging output when work is found stale as to why.
|
|
|
d37d044f
|
2012-08-01T20:05:30
|
|
Add some headroom to the number of curls available per pool to allow for longpoll and sendwork curls.
|
|
fd51e5ba
|
2012-07-30T18:27:33
|
|
Add scrypt support while writing conf
|
|
|
99204bd5
|
2012-07-30T14:05:16
|
|
Display scrypt as being built in as well.
|
|
|
7a07c7d0
|
2012-07-30T14:03:15
|
|
Fix build warning about KL_SCRYPT when built without scrypt support.
|
|
|
c91148f1
|
2012-07-30T09:02:23
|
|
Remove the low hash count determinant of hardware being sick. A low hash rate can be for poor network connectivity or scrypt mining, neither of which are due to sick hardware.
|
|
|
d8ec6d3e
|
2012-07-29T21:35:24
|
|
Display kilohash when suitable, but store the global mhash value still truly in megahashes to not break the API output.
|
|
|
b5517af0
|
2012-07-29T21:07:24
|
|
Don't try and print curses output for devices that won't fit on the screen.
|
|
|
500ed85b
|
2012-07-29T03:46:49
|
|
Merge pull request #271 from luke-jr/cg_logfixes
Logging bugfixes
|
|
|
28ac422a
|
2012-07-29T03:45:10
|
|
Merge pull request #270 from luke-jr/bugfix_procname
Bugfix: Copy argv[0] given to dirname()
|
|
|
b40c8b84
|
2012-07-29T19:10:40
|
|
Limit total number of curls recruited per pool to the number of mining threads to prevent blasting the network when we only have one pool to talk to.
|
|
|
e87ff7c1
|
2012-07-29T13:15:39
|
|
Keep a counter of enabled pools and use that instead of iterating over the pool list. Use that value to ensure we don't set the last remaining active pool to the rejecting state.
|
|
|
5148502c
|
2012-07-25T22:30:37
|
|
Reinstate help information for lookup gap and thread concurrency since tc will often need to be manually set for optimal scrypt performance.
|
|
|
9a6c082a
|
2012-07-24T20:27:37
|
|
Make the thread concurrency and lookup gap options hidden on the command line and autotune parameters with a newly parsed --shaders option.
|
|
da4ff2bd
|
2012-07-23T20:22:01
|
|
Bugfix: Use a mutex to control non-curses output
Without this, there is no guarantee writes won't overlap (and it happens quite a bit on Windows with the bitforce driver)
|
|
|
d98e561a
|
2012-07-23T20:15:45
|
|
Simplify code to a single vprintf path for curses-less printing
|
|
|
80593c47
|
2012-07-23T19:56:03
|
|
Move opt_quiet check to my_log_curses, so it works for curses-less builds
|
|
|
a1cb0d06
|
2012-07-23T15:48:00
|
|
Bugfix: Copy argv[0] given to dirname()
Per manpage, dirname can (and does on GNU/Linux!) modify its argument
|
|
|
94c94d65
|
2012-07-23T07:38:58
|
|
Nonce testing for btc got screwed up, leading to no accepted shares. Fix it.
|
|
|
39f7d2fa
|
2012-07-21T17:31:06
|
|
Allow lookup gap and thread concurrency to be passed per device and store details in kernel binary filename.
|
|
|
b9e5f8e5
|
2012-07-21T10:28:41
|
|
Revert "Use cpu scrypt code to check if an scrypt share is below target before submitting it."
This reverts commit 808f403a14267263066b0c257c262a1be6ef47ca.
Conflicts:
findnonce.c
|
|
|
808f403a
|
2012-07-20T23:44:43
|
|
Use cpu scrypt code to check if an scrypt share is below target before submitting it.
|
|
|
d72add9a
|
2012-07-20T16:16:18
|
|
Send correct values to scrypt kernel to get it finally working.
|
|
|
dd254545
|
2012-07-16T22:03:43
|
|
Detach pthread from within the api thread in case it is terminated due to not being instantiated before pthread_cancel is called from main, leading to a segfault.
|
|
|
4abecc26
|
2012-07-16T22:03:43
|
|
Detach pthread from within the api thread in case it is terminated due to not being instantiated before pthread_cancel is called from main, leading to a segfault.
|
|
|
77b7ed4b
|
2012-07-16T20:25:16
|
|
Debug output per thread hashrate is out by a factor of 1000.
|
|
|
76c37891
|
2012-07-16T13:19:42
|
|
Handle KL_SCRYPT in config write.
|
|
|
5cd4bbd2
|
2012-07-16T13:18:43
|
|
Get rid of stuff.
|
|
|
6ac14f42
|
2012-07-16T13:14:21
|
|
Don't enqueuewrite buffer at all for pad8 and pass work details around for scrypt in dev_blk.
|
|
|
a1edc7db
|
2012-07-15T05:37:49
|
|
Bugfix: Fix build without curses but with OpenCL
|
|
|
6c6c2852
|
2012-07-14T22:25:41
|
|
Show Khash hashrates when scrypt is in use.
|
|
|
54f1b808
|
2012-07-14T22:19:55
|
|
Free the scratchbuf memory allocated in scrypt and don't check if CPUs are sick since they can't be. Prepare for khash hash rates in display.
|
|
|
a5ebb712
|
2012-07-14T22:01:20
|
|
Add cpumining capability for scrypt.
|
|
|
41daf995
|
2012-07-14T09:45:55
|
|
Calculate midstate in separate function and remove likely/unlikely macros since they're dependent on pools, not code design.
|
|
|
8230ab05
|
2012-07-14T01:10:50
|
|
Display in debug mode when we're making the midstate locally.
|