Open Pongkemon01 opened 7 years ago
I have the same problem with 12 GPUs, sometimes it works without an error.
Any idea what happens if you run multiple instances? for example, running two instances of 6 gpus/instance.
ok, it seems that according to https://bitcointalk.org/index.php?topic=826901.7915 you should be able to run multiple instances. I don't have 8+ card rigs to test if seg faults happen, but it is worth the try.
There is certainly an issue of some sort, I am using it with simplemining (linux ubuntu) and it will do this multiple times in a row (reloading ccminer) and then start mining, it will crash a little bit later. I have 11 cards trying to use it at once.
I was getting the same. Segfaulting with 13 GPU's backtrace https://pastebin.com/LZtEx3u3
Cuda error in func 'keccak256_cpu_init' at line 299 : an illegal memory access was encountered.
even tho it must be lyra2v2 and not the keccak!
I have exactly the same issue. ccminer can dig Electroneum (ethereum) with 12 GPUs (nvidia 1070ti) just fine, Lyra2Rev2 is a guaranteed instant-crash.
Same here - crashing from time to time.
Same here - Really unstable miner with 9 or more card's . Ok with 8 or less.
I compiled the source code and run it in Linux using Skunk algorithm. The program works fine from single GPU to 8 GPUs. Beyond that, the program causes segmentation-fault error