Con Kolivas
cad84c6f2c
Change poclbm version number.
2012-02-11 11:51:15 +11:00
Con Kolivas
3d4c78ca89
Leave diakgcn and diablo kernels as silent options for now since they don't work.
2012-02-11 11:48:58 +11:00
Con Kolivas
4f1676f67f
One array is faster than 2 separate arrays so change to that in poclbm kernel..
2012-02-11 11:48:58 +11:00
Con Kolivas
f5903e609d
Microoptimisations to poclbm kernel which increase throughput slightly.
2012-02-11 11:48:58 +11:00
ckolivas
f2d5db0c77
Use only working kernels by default.
2012-02-11 11:48:58 +11:00
ckolivas
b7fa2b50e0
Make vectors compile in diablokernel.
2012-02-11 11:48:12 +11:00
ckolivas
451b82df97
Convert #if to #ifdef in diablo .cl.
2012-02-11 11:48:12 +11:00
ckolivas
59d3d0112b
Implement diablo kernel support and try to make it work.
2012-02-11 11:48:12 +11:00
ckolivas
e2a9c667d6
Import diablominer kernel.
2012-02-11 11:48:12 +11:00
ckolivas
95a989da4d
Conflicting entries of cl_kernel may have been causing problems, and automatically chosen kernel type was not being passed on. Rename the enum to cl_kernels and store the chosen kernel in each clState.
2012-02-11 11:48:12 +11:00
ckolivas
604b2f7645
Set cl_amd_media_ops with the BITALIGN flag and allow non-bitselect devices to build.
2012-02-11 11:48:12 +11:00
ckolivas
e6cf96adfa
ALlow much longer filenames for kernels to load properly.
2012-02-11 11:48:12 +11:00
ckolivas
4822cca72b
Allow different kernels to be used by different devices and fix the logic fail of overcorrecting on last commit with !strstr.
2012-02-11 11:48:11 +11:00
Con Kolivas
196e8a0f76
Fix kernel selection process and build error.
2012-02-11 11:48:11 +11:00
Philip Kaufmann
ec2dfeacc1
instead of uint *, int * was used as type for the output buffer (self-made bug from a former commmit)
2012-02-11 11:48:11 +11:00
Philip Kaufmann
c91155ac55
made code a bit more consistent and removed 2 unneeded brackets
2012-02-11 11:48:11 +11:00
Philip Kaufmann
17feecac8c
queue_phatk_kernel now uses CL_SET_VARG() for base-nonce(s), too
2012-02-11 11:48:11 +11:00
Philip Kaufmann
47a09ceab8
added OpenCL >= 1.1 detection code, in preparation of OpenCL 1.1 global offset parameter support
2012-02-11 11:48:11 +11:00
Philip Kaufmann
c43dd734cb
renamed all rot-functions to rotr to be clear that it is a rotate right
2012-02-11 11:48:11 +11:00
Philip Kaufmann
ab848057da
changed base-nonce generation for non GOFFSET paths to use get_local_size(0) and left shift
2012-02-11 11:48:11 +11:00
Philip Kaufmann
1398a805a7
leave out kernel parameter "const u base", if GOFFSET (not yet implemented) is defined
2012-02-11 11:48:11 +11:00
Philip Kaufmann
dcd286182e
changed format of output buffer from ulong * to uint * to be cgminer compatible
2012-02-11 11:48:11 +11:00
Philip Kaufmann
15cc3582d8
First commit - only changed date in diakgcn120208.cl
2012-02-11 11:48:11 +11:00
Con Kolivas
80cf0fd407
Diakgcn should be using the modified PreVal4 which includes T1.
2012-02-11 11:48:11 +11:00
ckolivas
cce08b6e6d
Reinstate original code as much as possible for comparison.
2012-02-11 11:48:11 +11:00
ckolivas
9e9db012fb
Convert to Unix EOL.
2012-02-11 11:48:11 +11:00
ckolivas
85e9a841d5
Use K array explicitly to make it clear what is being added.
2012-02-11 11:48:11 +11:00
ckolivas
49cff57098
Use correct variable on output in diakgcn kernel.
2012-02-11 11:48:11 +11:00
ckolivas
033913ca8e
First working port of the diakgcn kernel.
2012-02-11 11:48:11 +11:00
ckolivas
cb7145b179
Add basic build ability with diakgcn and put all kernel names in configure.ac to avoid changing them in mutliple places.
2012-02-11 11:48:11 +11:00
ckolivas
3af3cbe96a
Import diakgcn kernel.
2012-02-11 11:48:11 +11:00
Con Kolivas
6e3ad7ad25
Logic fail on queue request.
2012-02-09 22:40:33 +11:00
Con Kolivas
d40979d173
No point discarding a share even if opt_fail is enabled.
2012-02-09 22:28:59 +11:00
Con Kolivas
435e5c85f6
Work items have a tendency to expire at exactly the same time and we don't queue extra items when there are plenty in the queue, regardless of age.
...
Allow extra work items to be queued if adequate time has passed since we last requested work even if over the limit.
2012-02-09 22:27:20 +11:00
Con Kolivas
ffbf15ad82
Discard work when failover-only is enabled and the work has come from a different pool.
2012-02-09 22:00:04 +11:00
Con Kolivas
b54d462c85
Missing include to build on newer mingw32.
2012-02-09 21:56:07 +11:00
Con Kolivas
5b4761003c
Move from the thread safe localtime_r to regular localtime which is the only one supported on newer pthread libraries on mingw32 to make it compile with
...
the newer ming. Thread safety is of no importance where localtime is used in this code.
2012-02-09 21:54:23 +11:00
Con Kolivas
63777c9d30
Bump version to 2.2.3 with NEWS.
2012-02-06 21:37:22 +11:00
Con Kolivas
e0540bd44a
Revert "Rewrite the convoluted get_work() function to be much simpler and roll work as much as possible with each new work item."
...
This reverts commit dec99ab739d16f2dd4f48482e713a25ebaef8e66.
This seems to cause a race on work in free_work(). Presumably other threads are still accessing the structure.
2012-02-06 21:35:24 +11:00
Con Kolivas
2db367e732
Bump version to 2.2.2
2012-02-06 18:41:50 +11:00
Con Kolivas
04b6ec948f
Curses needs to be included before stdbool to compile properly with pdcurses on mingw32.
2012-02-06 18:41:15 +11:00
Con Kolivas
aa8552a146
Update NEWS.
2012-02-06 18:24:38 +11:00
Con Kolivas
4301351402
Merge pull request #104 from kanoi/master
...
Return last accepted share pool/time for devices
2012-02-05 23:22:52 -08:00
Con Kolivas
de4c6c29d7
Provide support for the submitold extension on a per-pool basis based on the value being detected in a longpoll.
2012-02-06 18:12:22 +11:00
Con Kolivas
1dff4297fa
Don't send a ping to a dynamic device if it's not enabled as that will just enable it for one pass and then disable it again.
2012-02-06 17:43:24 +11:00
Con Kolivas
dec99ab739
Rewrite the convoluted get_work() function to be much simpler and roll work as much as possible with each new work item.
2012-02-06 17:36:51 +11:00
Con Kolivas
a3e77937c8
Roll as much work as possible from the work returned from a longpoll.
2012-02-06 16:11:53 +11:00
Con Kolivas
eff9fe43b2
Rolling work on each loop through the mining thread serves no purpose.
2012-02-06 10:42:52 +11:00
Con Kolivas
d42d0c8ebd
Allow to stage more than necessary work items if we're just rolling work.
2012-02-06 10:41:05 +11:00
Con Kolivas
b9612715aa
Replace divide_work with reuse_work function used twice.
2012-02-06 10:40:04 +11:00
