First time poster, sorry if I'm not following any rules. I've googled and I've searched the forum before posting this.

I'm using a ccminer that says its 2.2:

*** ccminer 2.2 for nVidia GPUs by tpruvot@github ***
    Built with VC++ 2013 and nVidia CUDA SDK 8.0 64-bits

  Originally based on Christian Buchner and Christian H. project
  Include some algos from alexis78, djm34, sp, tsiv and klausT.

BTC donation address: 1AJdfCpLWPNoAMDfHF1wD5y8VgKSSTHxPo (tpruvot)


I'm using it with algorithm skunk.

The machine has a single Gigabyte GTX 1080 GPU and a number of EVGA GTX 1070. Each has 8GB of memory. They all work great using EWBF equihash.

However, when using skunk on ccminer, I get these errors:

[2017-08-06 23:50:17] GPU #5: EVGA GTX 1070, 19.66 MH/s
[2017-08-06 23:50:17] GPU #0: scanhash_skunk out of memory
[2017-08-06 23:50:17] GPU #1: EVGA GTX 1070, 19.23 MH/s
[2017-08-06 23:50:18] GPU #0: out of memory

Not on every block, but almost every 3rd block. OC settings do not seem to change the result.  For the time being, I'm having GPU 0 work on another problem with another set of software. That works fine.
 
Any advice? Thanks in advance.