ckolivas
830f2902b9
Numerous style police clean ups in cgminer.c
13 years ago
ckolivas
1e9421475c
Timersub is supported on all build platforms so do away with custom timerval_subtract function.
13 years ago
Paul Sheppard
efaa7398fb
Tweak sick/dead logic
...
(remove pre-computed time calculations)
13 years ago
Paul Sheppard
86c8bbe57e
Need to run Hashmeter all the time.
...
and not just if logging/display is enabled
13 years ago
Paul Sheppard
75a651c13f
Revert "Check for submit_stale before checking for work_restart"
...
Makes no sense to continue working on the old block whether submit_stale is enabled or not.
13 years ago
Paul Sheppard
f225392990
Add low hash threshold in sick/dead processing
...
Add check for fd in comms procedures
13 years ago
Con Kolivas
3267b534a8
Implement rudimentary X-Mining-Hashrate support.
13 years ago
Con Kolivas
24316fc7fc
Revert "Work is checked if it's stale elsewhere outside of can_roll so there is no need to check it again."
...
This reverts commit 5ad58f9a5c
.
13 years ago
Con Kolivas
5ad58f9a5c
Work is checked if it's stale elsewhere outside of can_roll so there is no need to check it again.
13 years ago
Con Kolivas
eddd02fea1
Put upper bounds to under 2 hours that work can be rolled into the future for bitcoind will deem it invalid beyond that.
13 years ago
Con Kolivas
bcec5f5102
Revert "Check we don't exhaust the entire unsigned 32 bit ntime range when rolling time to cope with extremely high hashrates."
...
This reverts commit 522f620c89
.
Unrealistic. Limits are bitcoind related to 2 hours in the future.
13 years ago
Con Kolivas
522f620c89
Check we don't exhaust the entire unsigned 32 bit ntime range when rolling time to cope with extremely high hashrates.
13 years ago
Kano
c21fc06560
define API option --api-groups
13 years ago
ckolivas
21a23a45d7
Work around pools that advertise very low expire= time inappropriately as this leads to many false positives for stale shares detected.
13 years ago
Paul Sheppard
d3e2b62c54
Change sick/dead processing to use device pointer, not gpu array.
...
Change BFL timing to adjust only when hashing complete (not error/idle etc.).
13 years ago
Con Kolivas
68a3a9ad10
There is no need for work to be a union in struct workio_cmd
13 years ago
ckolivas
b198423d17
Don't keep rolling work right up to the expire= cut off. Use 2/3 of the time between the scantime and the expiry as cutoff for reusing work.
13 years ago
ckolivas
6e80b63bb8
Revert "Increase the getwork delay factored in to determine if work vs share is stale to avoid too tight timing."
...
This reverts commit d8de1bbc5b
.
Wrong fix.
13 years ago
ckolivas
d8de1bbc5b
Increase the getwork delay factored in to determine if work vs share is stale to avoid too tight timing.
13 years ago
Paul Sheppard
1ef52e0bac
Check for submit_stale before checking for work_restart
...
(to keep Kano happy)
13 years ago
Paul Sheppard
90d82aa61d
Revert to pre pool merge
13 years ago
Con Kolivas
c027492fa4
Make the pools array a dynamically allocated array to allow unlimited pools to be added.
13 years ago
Con Kolivas
5cf4b7c432
Make the devices array a dynamically allocated array of pointers to allow unlimited devices.
13 years ago
Con Kolivas
17ba2dca63
Logic fail on queueing multiple requests at once. Just queue one at a time.
13 years ago
Con Kolivas
42ea29ca4e
Use a queueing bool set under control_lock to prevent multiple calls to queue_request racing.
13 years ago
Con Kolivas
63dd598e2a
Queue multiple requests at once when levels are low.
13 years ago
Con Kolivas
757922e4ce
Use the work clone flag to determine if we should subtract it from the total queued variable and provide a subtract queued function to prevent looping over locked code.
13 years ago
Con Kolivas
49dd8fb548
Don't decrement staged extras count from longpoll work.
13 years ago
Con Kolivas
d93e5f710d
Count longpoll's contribution to the queue.
13 years ago
Con Kolivas
05bc638d97
Increase queued count before pushing message.
13 years ago
Con Kolivas
32f5272123
Revert "With better bounds on the amount of work cloned, there is no need to age work and ageing it was picking off master work items that could be further rolled."
...
This reverts commit 5d90c50fc0
.
13 years ago
Con Kolivas
5d90c50fc0
With better bounds on the amount of work cloned, there is no need to age work and ageing it was picking off master work items that could be further rolled.
13 years ago
Con Kolivas
47f66405c0
Alternatively check staged work count for rolltime capable pools when deciding to queue requests.
13 years ago
Con Kolivas
efa9569b66
Test we have enough work queued for pools with and without rolltime capability.
13 years ago
Con Kolivas
1bbc860a15
Don't count longpoll work as a staged extra work.
13 years ago
Con Kolivas
ebaa615f6d
Count extra cloned work in the total queued count.
13 years ago
Con Kolivas
74cd6548a9
Use a static base measurement difference of how many items to clone since requests_staged may not climb while rolling.
13 years ago
Con Kolivas
7b57df1171
Allow 1/3 extra buffer of staged work when ageing it.
13 years ago
Con Kolivas
53269a97f3
Revert "Simplify the total_queued count to those staged not cloned and remove the locking since it's no longer a critical value."
...
This reverts commit 9f811c528f
.
13 years ago
Con Kolivas
a05c8e3fd9
Revert "Take into account total_queued as well when deciding whether to queue a fresh request or not."
...
This reverts commit b20089fdb7
.
13 years ago
Con Kolivas
750474bcfb
Revert "Further simplify the total_queued counting mechanism and do all dec_queued from the one location."
...
This reverts commit 790acad9f9
.
13 years ago
Con Kolivas
d2c1a6bd6b
Revert "Make sure to have at least one staged work item when deciding whether to queue another request or not and dec queued in free work not discard work."
...
This reverts commit c860172275
.
13 years ago
Con Kolivas
c860172275
Make sure to have at least one staged work item when deciding whether to queue another request or not and dec queued in free work not discard work.
13 years ago
Con Kolivas
790acad9f9
Further simplify the total_queued counting mechanism and do all dec_queued from the one location.
13 years ago
Con Kolivas
b20089fdb7
Take into account total_queued as well when deciding whether to queue a fresh request or not.
13 years ago
Con Kolivas
ded16838db
Add the getwork delay time instead of subtracting it when determining if a share is stale.
13 years ago
Con Kolivas
b5757d124b
Don't count getwork delay when determining if shares are stale.
13 years ago
Con Kolivas
9f811c528f
Simplify the total_queued count to those staged not cloned and remove the locking since it's no longer a critical value.
...
Clone only anticipated difference sicne there will be a lag from the value returned by requests_staged().
Keep 1/3 buffer of extra work items when ageing them.
13 years ago
Con Kolivas
411784a99d
As work is sorted by age, we can discard the oldest work at regular intervals to keep only 1 of the newest work items per mining thread.
13 years ago
Con Kolivas
359635a8cf
Only roll enough work to have one staged work for each mining thread.
13 years ago