summaryrefslogtreecommitdiffstats
path: root/xmrstak/backend/amd
Commit message (Collapse)AuthorAgeFilesLines
...
* Merge pull request #115 from fireice-uk/topic-net-extpsychocrypt2017-11-161-1/+1
|\ | | | | Network extensions
| * Pass threadno up the food chainUnknown2017-11-141-1/+1
| |
* | Fix misc bugs (#117)fireice-uk2017-11-152-1/+6
|/ | | | | | * CONF_NO_TLS fix * gpu error fix + always return to sched after pin
* add backend typepsychocrypt2017-11-101-0/+1
| | | | | - add type of the backend to each backend-plugin - add `gteName` to `iBackend` to get the name of the backend
* tune VEGA auto suggestionpsychocrypt2017-10-271-3/+15
| | | | | | - allow to use up to 2024 threads on VEGA gpus fix #5
* increase safety memory for autosuggestionpsychocrypt2017-10-271-2/+2
| | | | | - increase safty memory from 64 to 128 MiB - NVIDIA: increase lmem reserve per thread to 1kiB
* rename `xmr` to `monero`psychocrypt2017-10-273-11/+11
| | | | | | - rename all `xmr` to `monero` - be insensitive while check for set currency - add function to compate two strings insensitive
* add eon support to amd backendpsychocrypt2017-10-274-16/+48
| | | | | - add compile parameter to support aeon and xmr - update auto suggestion to handle aeon
* Merge pull request #69 from psychocrypt/topic-threadAffinityfireice-uk2017-10-272-10/+27
|\ | | | | thread affinity for non cpu backends
| * check if affinity must be setpsychocrypt2017-10-271-2/+3
| | | | | | | | - check if affinity vale before calling `thd_setaffinity`
| * thread affinity for non cpu backendspsychocrypt2017-10-262-10/+26
| | | | | | | | | | | | | | | | This is a follow up of #43 and use the some mechanism to set the thread affinity for non cpu backends correct. - use cpu affinity workflow for nvidia and amd - cpu: move messages of thread spawning before thread creation
* | amd: use 64bit targetpsychocrypt2017-10-244-12/+21
|/ | | | | - remove 32bit target value - use always 64bit target for amd backend
* Merge pull request #45 from fireice-uk/topic-nonce-allocfireice-uk2017-10-233-10/+16
|\ | | | | Implement pool-controlled nonce allocation
| * Fix GPU nicehash supportfireice-uk2017-10-221-1/+3
| |
| * move nicehash check into nonce calcfireice-uk2017-10-223-7/+6
| |
| * Implement changes suggestedfireice-uk2017-10-221-1/+1
| |
| * Implement pool-controlled nonce allocationfireice-uk2017-10-221-7/+12
| |
* | Proper removal of gpu_thread_numfireice-uk2017-10-221-10/+1
|/
* fix assert in AMD backendpsychocrypt2017-10-201-1/+1
| | | | | | fix bug announced in #52 If the miner is compiled in debug mode the assert condition results in a compile time error.
* Merge pull request #44 from fireice-uk/fix-uninit-accessfireice-uk2017-10-171-1/+1
|\ | | | | Make sure all singletons are set to null and make env a global ptr
| * Make sure all singletons are set to null and make env a global ptrfireice-uk2017-10-121-1/+1
| |
* | remove the gpu_thread_num safetyfireice-uk2017-10-122-7/+0
|/
* Merge pull request #24 from psychocrypt/fix-amdEarlyReturnfireice-uk2017-10-111-335/+376
|\ | | | | remove early returns
| * remove early returnspsychocrypt2017-10-101-335/+376
| | | | | | | | | | | | | | | | | | | | | | | | Early returns within a kernel can have unexpected behavior, this strongly depends on the opencl runtime compiler. To avoid errors all early returns are removed. There is no negative effect(e.g.performance) if the threads stay alive up to the end of the kernel. One source for the early return is: http://al-key-opencl.blogspot.de/2014/09/be-careful-not-to-mix-early-return-and.html This pull request also fix a race condition where all fill the shared memory. Also a bug introduced with #16 is fixed, because of the early return not was possible that the last block works with an wrong initilized shared memory (result should be a wrong hash if the result target size is valid).
* | improve AMD auto suggestionpsychocrypt2017-10-101-2/+7
|/ | | | | - use always a multiple of the number of compute units - add more device information to the config file
* avoid possible illegal memory accesspsychocrypt2017-10-062-8/+31
| | | | | Is is possible that the number of results are greater than the result output array, in this case invalid memory can be access within the device and on the host side.
* Merge pull request #16 from psychocrypt/fix-invalidKernelCallfireice-uk2017-10-062-24/+64
|\ | | | | fix invalid kernel call parameter
| * add assert to validate number of global threadspsychocrypt2017-10-051-0/+5
| | | | | | | | check that the number of global threads is a multiple of the workgroup size
| * fix invalid kernel call parameterpsychocrypt2017-10-052-24/+59
| | | | | | | | | | The number of threads within a kernel must be a multiple of the worksize. If not it can crash on some systems.
* | fix amd auto suggestionpsychocrypt2017-10-041-3/+5
|/ | | | | | Platform index zero was always used to search for the device detection. Fix: use correct detected AMD platform index
* readd nicehash pool supportpsychocrypt2017-10-011-1/+6
| | | | use nicehash start nonce in all backends if nicehash is enabled
* nvidia: fix return value if no device is foundpsychocrypt2017-09-301-4/+2
|
* fix windows compilepsychocrypt2017-09-301-1/+2
| | | | - string literal can only be 16kbyte
* fix clang CUDA compilepsychocrypt2017-09-301-1/+1
| | | | | | - add clang compile section - cleanup CMake file - fix missing return value
* rename `Params` to `params`psychocrypt2017-09-303-4/+4
|
* fix compilepsychocrypt2017-09-303-27/+19
|
* cleanup includespsychocrypt2017-09-307-42/+38
|
* group filespsychocrypt2017-09-3014-0/+3410
- move source code to `src` - categorize files and move to group folder - change upper case class files to lower case - change C++ header to `*.hpp`
OpenPOWER on IntegriCloud