 Kano
|
ed480de9c1
LTC text typo
|
13 years ago |
 Con Kolivas
|
132ee4c981
Do not scan other gpu platforms if one is specified.
|
13 years ago |
|
Con Kolivas
|
584fc013ab
Use a new algorithm for choosing a thread concurrency when none or no shader value is specified for scrypt.
|
13 years ago |
|
Con Kolivas
|
d0f18e83ad
Do not round up the bufsize to the maximum allocable with scrypt.
|
13 years ago |
|
Con Kolivas
|
3c3fbdce1c
Remove the rounding-up of the scrypt padbuffer which was not effectual and counter-productive on devices with lots of ram, limiting thread concurrencies and intensities.
|
13 years ago |
|
Con Kolivas
|
1c6d8a36d8
bufsize is an unsigned integer, make it so for debug.
|
13 years ago |
|
Con Kolivas
|
767d6df1a5
Whitelist AMD APP SDK 2.8 for diablo kernel.
|
13 years ago |
|
Con Kolivas
|
87b62bde43
Cope with the highest opencl platform not having usable devices.
|
13 years ago |
|
Con Kolivas
|
266d31271a
Make the numbuf larger to accept larger scrypt parameters.
|
13 years ago |
|
Con Kolivas
|
69494c12ed
BeaverCreek doesn't like BFI INT patching.
|
13 years ago |
|
Con Kolivas
|
25c39c96bb
Ease the checking on allocation of padbuffer8 in the hope it works partially anyway on an apparently failed call.
|
13 years ago |
|
Con Kolivas
|
cc3b693c6d
Minor warning fixes.
|
13 years ago |
|
Con Kolivas
|
40b747bae6
Put scrypt warning on separate line to avoid 0 being shown on windows as bufsize.
|
13 years ago |
|
Con Kolivas
|
d91af893c8
Use correct sdk version detection for SDK 2.7
|
13 years ago |
|
Con Kolivas
|
69983b778b
Revert "Pick worksize 256 with Cypress if none is specified."
|
13 years ago |
|
Con Kolivas
|
4fbe5bed15
OpenCL 1.0 does not have native atomic_add and extremely slow support with atom_add so detect opencl1.0 and use a non-atomic workaround.
|
13 years ago |
|
Con Kolivas
|
482322a4b7
Pick worksize 256 with Cypress if none is specified.
|
13 years ago |
|
Con Kolivas
|
be06cf7083
Give warning with sdk2.7 and phatk as well.
|
13 years ago |
|
Con Kolivas
|
cce19d9005
Whitelist sdk2.7 for diablo kernel as well.
|
13 years ago |
|
Con Kolivas
|
fc44b6d7a1
Use different variables for command line specified lookup gap and thread concurrency to differentiate user defined versus auto chosen values.
|
13 years ago |
|
Con Kolivas
|
97aa6ea492
Fix build error without scrypt enabled.
|
13 years ago |
|
Con Kolivas
|
43752ee58c
Limit thread concurrency for scrypt to 5xshaders if shaders is specified.
|
13 years ago |
|
Con Kolivas
|
da1b996a39
Simplify repeated use of gpus[gpu]. in ocl.c
|
13 years ago |
|
Con Kolivas
|
ea10b08dce
Find the nearest power of 2 maximum alloc size for the scrypt buffer that can successfully be allocated and is large enough to accomodate the thread concurrency chosen, thus mapping it to an intensity.
|
13 years ago |
|
Con Kolivas
|
9a6c082ad1
Make the thread concurrency and lookup gap options hidden on the command line and autotune parameters with a newly parsed --shaders option.
|
13 years ago |
|
Con Kolivas
|
3a0d60cfe1
Always create the largest possible padbuffer for scrypt kernels even if not needed for thread_concurrency, giving us some headroom for intensity levels.
|
13 years ago |
|
Con Kolivas
|
d8f81c18ee
Use the detected maximum allocable memory on a GPU to determine the optimal scrypt settings when lookup_gap and thread_concurrency parameters are not given.
|
13 years ago |
|
Con Kolivas
|
89eb1fa393
Check the maximum allocable memory size per opencl device.
|
13 years ago |
|
Con Kolivas
|
5087ff9069
Add debugging output if buffer allocation fails for scrypt and round up bufsize to a multiple of 256.
|
13 years ago |
|
Con Kolivas
|
1711b4eb77
Display size of scrypt buffer used in debug.
|
13 years ago |
