Claudio André

> @claudioandre-br it now hits me I believe we're not currently calling clFlush() from set_salt(). Should we possibly add that? I do. I see no problem. Why are you worried?...

> the salt possibly isn't really transfered until the kernel actually needs it, I guess so. No flush, no issue, send, emit, transmit, ...

> Transfer and CPU computation in parallel, yes. This is about CPU and GPU processing in parallel. You are right. But you are right because I have nothing to do...

Host code: ``` if (gpu_nvidia(device_info[sequential_id]))``` In kernel code ```if amd_gcn(DEVICE_INFO) && DEV_VER_MAJOR < 2500``` So: ``` if (gpu_amd(device_info[sequential_id]) && major > xxxx) #major is available somewhere avoid async transfers else...

Last I read about it, WSL doesn't support GPU Acc/OpenCL. It is unclear whether this is still the case at https://docs.nvidia.com/cuda/wsl-user-guide/index.html#features-not-yet-supported. However, if `clinfo` fails, the problem is not related...

mscash2 can't be run by `--stress-test`. The patch below fixes the segfault, but it continues to fail because OpenCL initialization is not done correctly. Or shared code is not able...

> Are we confident this issue only affects stress-test? I'm not aware of (other) use cases that restart the execution environment like this: ``` opencl-init() format-init() format-done() opencl-done() openl-init() and...

> @claudioandre-br I assume this problem went under your radar because you disable this format for being extremely slow? I used to have `--internal` enabled for all CI executions. It...

> @claudioandre-br I have plenty on my to-do list so if you have the time, please bite on this I'm going to add it to my christmas to-do list.

I need to re-sync with sha512crypt (wait the PR and my post merge tests). ``` Testing: raw-SHA512-opencl [SHA512 OpenCL]... FAILED (cmp_one(0)) Testing: XSHA512-opencl, Mac OS X 10.7 salted [SHA512 OpenCL]......