sgminer

Commit Graph

Author	SHA1	Message	Date
unknown	f7f582732a	Fixed bufsize type.	11 years ago
Noel Maersk	c6a27709f8	core: use global algorithm structure instead of opt_nfactor. Also squashed: config: add log messages to set_algo() and set_nfactor(). algorithm: use set_algorithm_nfactor() when setting default nfactor in set_algorithm(). Otherwise algorithm->n defaults to 0. P.S. Did I already mention how this could have been C++?..	11 years ago
Noel Maersk	c2c8231b9e	kernel: zuikkis supports all worksizes. Closes https://github.com/veox/sgminer/issues/130	11 years ago
Noel Maersk	e18bcc2582	kernel: use N-factor table instead of passing parameter N directly.	11 years ago
Noel Maersk	3afaaf1bfa	ocl: rename nfactor to N where appropriate. N = 2^nfactor, so `nfactor` was a bad name choice here.	11 years ago
Noel Maersk	4b3b2ef5a6	ocl: use 'nf' instead of 'n' in compiled kernel binary name. It is used as a factor number here (e.g. 10, not 1024).	11 years ago
Noel Maersk	5b42d38f33	ocl/misc: spacing.	11 years ago
Noel Maersk	7217da3078	ocl: vectors are hard-set to 1, add appropriate "optimisation".	11 years ago
Noel Maersk	ab9fcb1a14	ocl: use same type convention for nfactor (`cl_uint` instead of `int`).	11 years ago
Teemu Suikki	73bb1504e9	Scrypt-nfactor support! Added new configuration parameter "nfactor", which defaults to 10 (normal scrypt). Use 11 for vertcoin. Kernels modified accordingly.	11 years ago
Brandon Tuomela	bda1070259	kernel: add psw kernel.	11 years ago
Noel Maersk	e123cc20c3	kernel: search in `pwd`/kernel/ for kernels. Closes https://github.com/veox/sgminer/issues/49.	11 years ago
Sanjin Trošelj	5d5769156f	VS2010 build: Prepare ocl.c Cherry-picked by veox.	11 years ago
Noel Maersk	e8978db55f	kernel: slight refactor, add warnings about new kernels, and known limitations of zuikkis. Limitations of alexkar* kernels are not known to me. Closes https://github.com/veox/sgminer/issues/28	11 years ago
Noel Maersk	5f326d64f5	kernel: allow choosing Alexey Karimov's kernels. Am I the only one who gets a segfault?..	11 years ago
Noel Maersk	7950c10374	kernel: integrate Zuikkis' kernel, selectable with `--kernel=zuikkis`. Kernel ckolivas is still the default. Needs documentation, too. https://github.com/veox/sgminer/issues/4	11 years ago
Noel Maersk	9d1db95a28	misc: move scrypt130511.cl kernel file to kernel/ckolivas.cl Prepare for more kernels. :) Get rid of the kernel date, since this fudges up commit history. Named after Con Kolivas, the most recent committer.	11 years ago
Martin Danielsen	7aeae40af2	EXPERIMENTAL: A new way of setting intensity; introducing xintensity! All of this is credited to ArGee of RGMiner, he did the initial ground work for this setting. This new setting allows for a much finer grained intensity setting and also opens up for dual gpu threads on devices not previously able to. Note: make sure to use lower thread-concurrency values when you increase cpu threads. Intensity is currently used to spawn GPU threads as a simple 2^value setting. I:13 = 8192 threads I:15 = 32768 threads I:17 = 131072 threads I:18 = 262144 threads I:19 = 524288 threads I:20 = 1048576 threads Notice how the higher settings increase thread count tremendously. Now enter the xintensity setting (Yes, I am a genius with my naming convention!). It is simply a shader multiplier, obviously based on the amount of shaders you got on a card, this should allow the same value to scale with different card models. 6970 with 1536 shaders: xI:64 = 98304 threads R9 280X with 2048 shaders: xI:64 = 131072 threads R9 290 with 2560 shaders: xI:64 = 180224 threads R9 290X with 2816 shaders: xI:64 = 163840 threads 6970 with 1536 shaders: xI:300 = 460800 threads R9 280X with 2048 shaders: xI:300 = 614400 threads R9 290 with 2560 shaders: xI:300 = 768000 threads R9 290X with 2816 shaders: xI:300 = 844800 threads It's now much easier to control thread intensity and it potentially allows for a uniform way of setting the intensity on your system. I'm very interested in constructive feedback, as I do not have access to a lot of different card models. This change has been tested on 6970, R9 290, R9 290X - all with equal or a little better speeds than regular intensity setting after a little tuning, but your mileage may vary. Don't fret it, if this doesn't work for you, the regular intensity setting is still available. Conflicts: driver-opencl.c sgminer.c	11 years ago
Noel Maersk	50066cf3c7	misc: Replace all remaining instances of 'cgminer' with 'sgminer'. sed again.	11 years ago
Noel Maersk	199cb10e18	HAVE_OPENCL is mandatory, remove checks form code an build system.	11 years ago
Noel Maersk	95586f72f0	Squash-merge branch dead-end with unnecessary check removals.	11 years ago
Noel Maersk	d0070c0424	HARD RESET to: Remove Bitcoin detection code. I'm pretty sure no one was working on this but me. The code was unusable (gave 100% hardware errors) due to me not testing it sufficiently. I hope this will not happen again. ADL doesn't work for some reason, too.	11 years ago
Con Kolivas	578fabe07c	Do not perform bfi int patching for opencl1.2 or later.	11 years ago
Con Kolivas	466d3beca0	Provide a --disable-libcurl config option to build support for stratum mining only.	11 years ago
Kano	383eeabc71	ocl.c fix applog warnings on windows	12 years ago
Con Kolivas	b196d4fa0c	Support more shares to be returned for scrypt mining.	12 years ago
Kano	54d5d26095	ocl correct applog typing	12 years ago
Con Kolivas	2643ad1b22	Use only the one jump in ocl.c to bypass binary saves for osx opencl.	12 years ago
Con Kolivas	0a8f584909	Initialise variables not set on OSX in ocl.c.	12 years ago
Con Kolivas	9aae2256d3	Bypass attempting to read and save binary files on OSX to avoid crashes on >1 GPU.	12 years ago
Con Kolivas	57e5bfbb25	Set default ocl work size for scrypt to 256.	12 years ago
ckolivas	6ffba7e9d8	Convert error getting device IDs in ocl code to info log level only since multiple platforms may be installed and the error is harmless there.	12 years ago
ckolivas	a797898fc3	Unnecessary extra array in ocl code.	12 years ago
Kano	ed480de9c1	LTC text typo	12 years ago
Con Kolivas	132ee4c981	Do not scan other gpu platforms if one is specified.	12 years ago
Con Kolivas	584fc013ab	Use a new algorithm for choosing a thread concurrency when none or no shader value is specified for scrypt.	12 years ago
Con Kolivas	d0f18e83ad	Do not round up the bufsize to the maximum allocable with scrypt.	12 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.	12 years ago
Con Kolivas	1c6d8a36d8	bufsize is an unsigned integer, make it so for debug.	12 years ago
Con Kolivas	767d6df1a5	Whitelist AMD APP SDK 2.8 for diablo kernel.	12 years ago
Con Kolivas	87b62bde43	Cope with the highest opencl platform not having usable devices.	12 years ago
Con Kolivas	266d31271a	Make the numbuf larger to accept larger scrypt parameters.	12 years ago
Con Kolivas	69494c12ed	BeaverCreek doesn't like BFI INT patching.	12 years ago
Con Kolivas	25c39c96bb	Ease the checking on allocation of padbuffer8 in the hope it works partially anyway on an apparently failed call.	12 years ago
Con Kolivas	cc3b693c6d	Minor warning fixes.	12 years ago
Con Kolivas	40b747bae6	Put scrypt warning on separate line to avoid 0 being shown on windows as bufsize.	12 years ago
Con Kolivas	d91af893c8	Use correct sdk version detection for SDK 2.7	12 years ago
Con Kolivas	69983b778b	Revert "Pick worksize 256 with Cypress if none is specified." This reverts commit `482322a4b7`. Worksize 256 was only helpful on cypress with ultra-low memory speeds with old SDKs and the new kernels require higher memory clocks, having the opposite net effect.	12 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.	12 years ago
Con Kolivas	482322a4b7	Pick worksize 256 with Cypress if none is specified.	12 years ago

1 2 3 4 5

236 Commits (43f96783cd2dccffc6ea28b56a4d783b75c955c0)