15 Commits (7807eb6f2df6bced8742e44a746527e1dce0c4c3)

Author	SHA1	Message	Date
jsteube	68662f0062	replace swap32 and swap64 with faster versions for NV	9 years ago
jsteube	94d6fd3693	revert switch to rotate ()	9 years ago
jsteube	212395168f	Testchange for speed	9 years ago
jsteube	84568e5b3d	Increase bcrypt speed for NV	9 years ago
jsteube	21891bf8db	Fix 64 bit based rotates for use with ForceWare >= 358.x	9 years ago
jsteube	964ece53a3	add __bfe stub for AMD	9 years ago
jsteube	6c22de104e	Slightly increase blowfish based algorithms for NV	9 years ago
Jens Steube	0e5b37a537	Fix AMD speeds @ sha512crypt	9 years ago
jsteube	7be2c2fd8d	Fixed some speeds: ... 1800 3200 7100 7200 7400 8200 9300 9000 9600 11300 11600	9 years ago
jsteube	9d9097a12c	Fix -m 7800 by adding amd_bytealign emulation	9 years ago
jsteube	9c392b472e	Removed useless comments	9 years ago
jsteube	c29c7c093f	Fix use of LOP3	9 years ago
jsteube	6a65c7b1ce	Prepare for LOP3 integration	9 years ago
jsteube	2283d5c843	Fix more append_* functions in kernels	9 years ago
jsteube	0bf4e3c34a	- Dropped all vector code since new GPU's are all scalar, makes the code much easier ... - Some performance on low-end GPU may drop because of that, but only for a few hash-modes - Dropped scalar code (aka warp) since we do not have any vector datatypes anymore - Renamed C++ overloading functions memcat32_9 -> memcat_c32_w4x4_a3x4 - Still need to fix kernels to new function names, needs to be done manually - Temperature Management needs to be rewritten partially because of conflicting datatypes names - Added code to create different codepaths for NV on AMD in runtime in host (see data.vendor_id) - Added code to create different codepaths for NV on AMD in runtime in kernels (see IS_NV and IS_AMD) - First tests working for -m 0, for example - Great performance increases in general for NV so far - Tested amp_* and markov_* kernel - Migrated special NV optimizations for rule processor	9 years ago