 ckolivas
|
b19ee2f5f5
Make sure a GPU doesn't work on a block longer than opt_scantime.
|
14 years ago |
|
ckolivas
|
26546ad5b9
Make the optimisations per-gpu card and update code to work properly with multiple cards.
|
14 years ago |
|
Con Kolivas
|
8253f1414b
Use some line breaks in the kernel.
|
14 years ago |
|
Con Kolivas
|
4257deafdb
Convert abcd... to an array.
|
14 years ago |
|
Con Kolivas
|
75cf5ccda6
Replace Ws with an array.
|
14 years ago |
|
Con Kolivas
|
852e78e7b4
Fix mutex unlocking with only one thread and opt_log_interval.
|
14 years ago |
|
Con Kolivas
|
70f7357692
Make the output display the 5 second and total average Mhash/s.
|
14 years ago |
|
Con Kolivas
|
debe77767a
Use cpu_from_thr_id when binding threads.
|
14 years ago |
|
Con Kolivas
|
afc83cb25b
Need gnu_source for memmem to work properly.
|
14 years ago |
|
Con Kolivas
|
4cd5f47efa
Revert "Multiple compiler warning fixes."
|
14 years ago |
|
Con Kolivas
|
88761e6cf2
Multiple compiler warning fixes.
|
14 years ago |
|
ckolivas
|
d28f526b1e
Need memmem in advance() to work properly for patching.
|
14 years ago |
|
ckolivas
|
19eea9067f
Implement code detecting max work size and optimal vector width.
|
14 years ago |
|
ckolivas
|
33352213eb
Report preferred vector width.
|
14 years ago |
|
ckolivas
|
237a5067c3
Skip trying to start thread of GPUs that don't successfully initcl().
|
14 years ago |
|
ckolivas
|
14ca888373
Update help.
|
14 years ago |
|
ckolivas
|
c08be809f2
Fix the setting of number of processors.
|
14 years ago |
|
Con Kolivas
|
407df0f4b1
Use correct boolean testing and long enough string for cl_amd_media ops testing and kernel changing.
|
14 years ago |
|
ckolivas
|
c2e4165590
Detect platforms that support cl_amd_media_ops and decide whether to BFI_INT patch or not.
|
14 years ago |
|
ckolivas
|
2ab6180d55
Reset count once all threads are started to avoid slow rate being shown initially.
|
14 years ago |
|
Con Kolivas
|
932ff72f70
The gpuminer thread uses very little cpu and needs to keep the gpu busy with as few delays as possible. Don't nice it.
|
14 years ago |
|
Con Kolivas
|
f54d2cc0ed
Make poclbm use 4 vectors and decrease worksize to keep pipelines fullish.
|
14 years ago |
|
ckolivas
|
b4d2733cfc
Convert to poclbm kernel.
|
14 years ago |
|
ckolivas
|
662404812a
Fix deref.
|
14 years ago |
|
ckolivas
|
fa2f6b19c1
Unwind.
|
14 years ago |
|
ckolivas
|
79fec01a46
Remove the input buffer and just pass args to the kernel as per plugin design.
|
14 years ago |
|
Con Kolivas
|
f117675ac2
Optimise work loop to make cl calls asynchronous where possible.
|
14 years ago |
|
Con Kolivas
|
f05270b8c8
Optimise loop and make debug debug only.
|
14 years ago |
|
Con Kolivas
|
fa4c10b1d9
Implement ma macro for amd bytealign that gets patched into bfi_int as well.
|
14 years ago |
|
Con Kolivas
|
91e5cef3a5
Actually get first BFI_INT patch working.
|
14 years ago |
