|
e2b7c934
|
2012-03-16T23:52:20
|
|
removed 8-component vector support from kernel, as this is not supported in CGMINER anyway
|
|
|
782fc63c
|
2012-03-16T23:38:39
|
|
forgot to update kernel modification date, fixed ;)
|
|
|
c3261587
|
2012-03-16T23:34:15
|
|
reordered an addition in the kernel, which results in less instructions used in the GPU ISA code for GCN
|
|
|
aa52e7df
|
2012-03-01T17:24:38
|
|
optimized nonce-check and output code for -v 2 and -v 4
|
|
|
77f177b6
|
2012-02-24T15:38:41
|
|
only small code formating changes
|
|
|
d9d4831b
|
2012-02-24T15:32:19
|
|
removed vec_step() as this could lead to errors on older SDKs
|
|
|
d7d07979
|
2012-02-24T08:24:36
|
|
unified code for generating nonce in kernel and moved addition of base to the end -> faster
|
|
d32cd583
|
2012-02-24T17:37:09
|
|
Reinstate the old output mechanism setting output[FOUND] per vector.
|
|
|
fb077c6d
|
2012-02-24T13:27:15
|
|
Pass vectors * worksize to kernel to avoid one op.
|
|
|
70e8ade5
|
2012-02-24T13:22:42
|
|
Revert behaviour to old nonce init code.
|