Luke Dashjr
|
2b7661d39a
Bugfix: Use a new POOL_STALE flag to handle stale-on-arrival disabling, instead of piggybacking on POOL_REJECTING
|
13 years ago |
Luke Dashjr
|
24fafd0f58
Bugfix: Don't warn for stale-on-arrival work if the pool is already disabled for it
|
13 years ago |
Luke Dashjr
|
40400be160
Merge branch 'ocl_dyn_timeBeginPeriod' into bfgminer-2.6.x
|
13 years ago |
Luke Dashjr
|
de73e0fcf4
opencl: Use timeBeginPeriod on Windows to ensure gettimeofday has sufficient precision for dynamic intensity
|
13 years ago |
Luke Dashjr
|
00cd89b7ae
Revert "Windows' timer resolution is limited to 15ms accuracy. This was breaking dynamic intensity since it tries to measure below this."
|
13 years ago |
Luke Dashjr
|
5373383d2c
Merge branch 'bugfix_adl_fanpercent_deadlock' into bfgminer-2.6.x
|
13 years ago |
Luke Dashjr
|
ccf77ca1ca
Bugfix: opencl: Move ADL fanspeed warning messages to a new thread to get around summary-update deadlocking
|
13 years ago |
Luke Dashjr
|
06bcf60781
Merge branch 'readme_fpgagrp' into bfgminer-2.6.x
|
13 years ago |
Luke Dashjr
|
8b056c0b5d
README: Note that user groups don't get updated until re-login
|
13 years ago |
Con Kolivas
|
90824e1461
Initialise cnt in libztex.c
|
13 years ago |
Con Kolivas
|
d2e93ca636
Don't try to start devices that don't support scrypt when scrypt mining.
|
13 years ago |
Con Kolivas
|
7f11d28c13
Repeating on timeout in ztex could make the code never return.
|
13 years ago |
Con Kolivas
|
2b4def7c8e
Offset libusb reads/writes by length written as well in ztex.
|
13 years ago |
Con Kolivas
|
04f1c88734
Cope with timeouts and partial reads in ztex code.
|
13 years ago |
Con Kolivas
|
e5ee9f0f8e
If there are more devices than nDevs, don't iterate over them as they may overwrite devices mapped below that with the mapping option.
|
13 years ago |
Con Kolivas
|
826175c3a1
Fix README faq on bfl auto-detect.
|
13 years ago |
Con Kolivas
|
50bdf850a9
Set memory clock based on memdiff if present from with engine changes, allowing it to parallel manual changes from the menu as well.
|
13 years ago |
Con Kolivas
|
8fc2983614
Windows' timer resolution is limited to 15ms accuracy. This was breaking dynamic intensity since it tries to measure below this.
|
13 years ago |
Luke Dashjr
|
4c0ae57af1
Merge commit '70c57e7' into bfgminer
|
13 years ago |
Luke Dashjr
|
00b5e07b28
Merge commit '618ef0c' into bfgminer
|
13 years ago |
Luke Dashjr
|
1f68cad7e2
Merge branch 'bugfix_ocl_noadl_statline' into bfgminer
|
13 years ago |
Luke Dashjr
|
920166a41f
Bugfix: opencl: Show blank device-info statline area if GPU doesn't have ADL, to fix column alignment
|
13 years ago |
Con Kolivas
|
cadc7d762a
Bump version number to 2.6.5
|
13 years ago |
Con Kolivas
|
1e36f19ad5
Update NEWS.
|
13 years ago |
Con Kolivas
|
2480cf23cc
Don't try to get bitforce temperature if we're polling for a result to minimise the chance of interleaved responses.
|
13 years ago |
Con Kolivas
|
ec4a864649
Fix README faq on bfl auto-detect.
|
13 years ago |
Con Kolivas
|
83675099b9
Revert "Only get bitforce temperature at a time when we have not requested any other responses to minimise risk of interleaved responses."
|
13 years ago |
Con Kolivas
|
e34105a47a
Set memory clock based on memdiff if present from with engine changes, allowing it to parallel manual changes from the menu as well.
|
13 years ago |
Con Kolivas
|
c433f234c5
Increase the timeout on bitforce as per Paul Sheppard's suggestion to account for throttling + work time + excess.
|
13 years ago |
Con Kolivas
|
2a474fe914
Fix ADL gpu-map not working when there are more ADL devices than openCL. Patch supplied and tested by Nite69.
|
13 years ago |