findnonce.c


Log

Author Commit Date CI Message
Con Kolivas b196d4fa 2013-05-11T16:19:02 Support more shares to be returned for scrypt mining.
Con Kolivas c1ff90a7 2013-04-18T15:07:32 Do testing for HW errors on submit nonce for both scrypt and sha.
Con Kolivas 6b714dd4 2013-04-18T14:47:46 Increment hardware error count from the one site.
Con Kolivas 5412323e 2013-03-09T15:12:41 Fix memory leak with share submission on GPU work structures as discovered by twobitcoins.
Kano a344deb6 2013-01-05T23:47:26 rename device_api -> device_drv and all related api -> drv and add a device_drv->drv enum for identifying which driver each is
Con Kolivas c99636f2 2012-12-07T13:21:46 Begin tearing down the old workio command queues by removing submit commands from there and submit them asynchronously via their own threads.
Con Kolivas 8bd86968 2012-10-26T23:28:46 The work struct pointer in struct pc_data in findnonce is never freed yet there is no need to allocate it separately so make struct work a static part of the struct pc_data. s
Con Kolivas 2b6e3676 2012-10-08T22:10:15 Count an invalid nonce count as a hardware error on opencl.
Con Kolivas 2ff16103 2012-10-06T14:58:26 Prevent corrupt values returned from the opencl code from trying to read beyond the end of the buffer by masking the value to a max of 15.
Kano 307d8da0 2012-09-04T15:57:11 HW: error counter auto for all devices - ztex code not fixed
Con Kolivas 86d5377e 2012-08-23T10:48:15 Use atomic ops to never miss a nonce on opencl kernels, including nonce==0, also allowing us to make the output buffer smaller.
Con Kolivas e49bd981 2012-08-07T19:56:43 Use the scrypt CPU code to confirm results from OCL code, and mark failures as HW errors, making it easier to tune scrypt parameters.
Con Kolivas b9e5f8e5 2012-07-21T10:28:41 Revert "Use cpu scrypt code to check if an scrypt share is below target before submitting it." This reverts commit 808f403a14267263066b0c257c262a1be6ef47ca. Conflicts: findnonce.c
Con Kolivas d9eba43a 2012-07-20T23:49:43 Fix build.
Con Kolivas 808f403a 2012-07-20T23:44:43 Use cpu scrypt code to check if an scrypt share is below target before submitting it.
Con Kolivas d72add9a 2012-07-20T16:16:18 Send correct values to scrypt kernel to get it finally working.
Con Kolivas 6ac14f42 2012-07-16T13:14:21 Don't enqueuewrite buffer at all for pad8 and pass work details around for scrypt in dev_blk.
Con Kolivas ea444d02 2012-07-14T00:59:38 Fix nonce submission code for scrypt.
Con Kolivas 8fd3bf74 2012-07-13T20:40:37 Build fix for opt scrypt.
Con Kolivas 1aacfe52 2012-07-13T20:38:26 Don't check postcalc nonce with sha256 in scrypt.
Con Kolivas b085c338 2012-07-13T20:28:36 Make scrypt buffers and midstate compatible with cgminer.
Con Kolivas 8f08a775 2012-02-22T14:10:04 Use any() in kernel output code and revert breakage of diakgcn kernel.
Con Kolivas d1cddf8b 2012-02-21T22:23:07 Update licensing to GPL V3.
Con Kolivas 810ad045 2012-02-18T23:16:08 More copyright updates.
Con Kolivas 83dde50f 2012-02-18T15:17:50 Updates to diakgcn kernel courtesy of Philip Kaufmann <phil.kaufmann@t-online.de>
zefir 713e8be6 2012-02-04T14:22:47 move CPU chipset specific optimization into device-cpu
Con Kolivas 52bb99c3 2012-02-11T21:42:02 Fix double definition of A0 B0 to zeroA zeroB.
Con Kolivas 405a2120 2012-02-11T20:11:18 Remove unnecessary check for opt_debug on every invocation of applog at LOG_DEBUG and place the check in applog().
ckolivas 85e9a841 2012-02-08T16:24:46 Use K array explicitly to make it clear what is being added.
ckolivas 033913ca 2012-02-08T15:40:10 First working port of the diakgcn kernel.
Philip Kaufmann c91155ac 2012-02-09T19:59:35 made code a bit more consistent and removed 2 unneeded brackets
ckolivas 1bfa82c3 2012-02-10T17:02:13 Revert "Search for nonce if it is not found to be correct as a debugging check." This reverts commit d9e07385daa59f945016850f075ef55d1784799f. Purely a debugging exercise.
ckolivas d9e07385 2012-02-10T15:19:45 Search for nonce if it is not found to be correct as a debugging check.
Philip Kaufmann 68e36c65 2012-02-09T19:59:35 made code a bit more consistent and removed 2 unneeded brackets
ckolivas 7a069b93 2012-02-08T16:24:46 Use K array explicitly to make it clear what is being added.
ckolivas 53d33c75 2012-02-08T15:40:10 First working port of the diakgcn kernel.
Con Kolivas 8171ceab 2012-01-15T21:39:49 Simplify submit_nonce loop and avoid potentially missing FOUND - 1 entry. Reported by Luke-Jr.
Con Kolivas a51514d9 2011-12-27T10:30:04 White space cleanup.
Con Kolivas 4d13f8ad 2011-09-07T09:02:28 ByteReverse is not used and the bswap opcode breaks big endian builds. Remove it.
Con Kolivas cf54f9b8 2011-08-17T16:07:15 Move to 256 sized buffers and don't risk overwrite by using only 127 mask.
Con Kolivas 0f782ba6 2011-08-17T15:47:18 Update poclbm kernel to FF sized mask and only check that range.
Phateus d15d225a 2011-08-16T23:19:46 Changed phatk version to 2.2
Con Kolivas 116a9dc0 2011-07-23T14:17:25 Update phatk kernel to one with new parameters for slightly less overhead again. Make the queue kernel parameters call a function pointer to select phatk or poclbm.
Con Kolivas 1c070475 2011-07-17T23:38:29 Cycling over 1026 nonces is a remnant of old code not relevant now and can just lead to repeat submission of the same block increasing rejects. Remove it.
Con Kolivas b6ac9f18 2011-07-17T20:06:43 Revert "Make sure not to try and submit work for nonces higher than already submitted in current work item." This reverts commit a2dd410e362116a12ac5609da77abb3a5ee1b736. Incorrect fix. Missed real shares that are accepted.
Con Kolivas a2dd410e 2011-07-17T19:33:53 Make sure not to try and submit work for nonces higher than already submitted in current work item. This reduces rejects substantially.
Con Kolivas cb13e2cf 2011-07-05T19:47:03 Make it possible to build without opencl for cpu mining only.
Con Kolivas 13b43cfa 2011-07-03T00:28:51 Update copyright and authors.
ckolivas 7ae9afc4 2011-07-01T14:16:41 Profile points and warning clean ups.
Con Kolivas d5d4d1da 2011-06-30T14:41:01 Don't want to free the work data out of the transient structs.
Con Kolivas 2b6e8416 2011-06-29T23:38:16 Use a buffer of up to 512 * 4 integers when retrieving work from the GPU. This allows each local thread id to have one slot to put any positive results into, thus making overlapping results far less likely. Thus races will be much rarer, allowing more threads. It should also pick up blocks close to each other more reliably and hopefully decrease the number of rejects and opencl errors. Do the search over the buffer entirely in a separate thread to allow the GPU to stay as busy as possible. Detach threads from themselves to prevent unlucky even where dereferencing occurs by freeing the data that stores the thread info.
Con Kolivas 6374e0fa 2011-06-28T21:11:04 Import the phatk kernel. Enable it only for hardware with amd media ops for now since it crashes nvidia et. al. Fallback to the poclbm kernel for the rest. Try harder to avoid stale blocks around longpoll detecting new blocks.
Con Kolivas a45c54aa 2011-06-27T11:31:05 Make postcalc_hash asynchronous as well.
ckolivas f490143a 2011-06-29T09:22:21 Add local thread count to info, store hw error count, and make share submission debug only.
ckolivas 08a78210 2011-06-29T10:12:00 Make the log show what the thread is: cpu or gpu and what number.
Con Kolivas 88761e6c 2011-06-23T21:04:29 Multiple compiler warning fixes.
Con Kolivas f117675a 2011-06-22T10:15:23 Optimise work loop to make cl calls asynchronous where possible.
ckolivas 6b77d850 2011-06-17T14:00:41 Fixes.
Con Kolivas 4cd5f47e 2011-06-23T21:09:22 Revert "Multiple compiler warning fixes." This reverts commit a5cbfbde2610e9f60e14b41a4e0595bcb34c772a. Broke.
Con Kolivas dde70397 2011-06-14T10:32:54 Merge gpumining from oclmine. Unstable.