david-macmahon

These planning methods in CUDA.CUFFT are implementations of the planning functions defined in `AbstractFFTs`. All the planning functions in `AbstractFFTs` take optional keyword arguments. I think it would be better...

The cuFFT docs say that a complex-to-real out-of-place transform always overwrites its input buffer, but it doesn't mention any other case when that happens. This corresponds to `plan_irfft` and `plan_brfft`....

Great, it sounds like we've converged! I opened issue https://github.com/JuliaMath/AbstractFFTs.jl/issues/71 regarding `flags`. I'll be happy to work on a PR for adding `flags` to the CUDA.CUFFT planning functions, but it...

I think this is a low priority in terms of features, but I think a brief discussion in the memo would be nice for completeness (maybe in a short appendix?).

In reply to your question in #614 about how important this feature is, I think the answer is that it very important in certain circumstances. For example, we currently use...

> I was having another look at the documentation, and: > > > NOTE:The callback API is available in the statically linked cuFFT library only, and only on 64 bit...

I don't know that much about the Driver API's module handling, but does `CuGlobal()` end up calling `cuModuleGetGlobal()`? Not sure how that differs from `cuModuleGetFunction()`. FWIW, the C program I...

I noticed the sample you linked to also uses `cudaMemcpyFromSymbol()`. I also noticed its `Makefile` has: ``` LIBRARIES += -lcufft_static -lculibos ``` so it too is using the statically linked...

> I checked my C program that uses CuFFT callbacks and it looks like it does statically link the CuFFT library. I'll try to build it with dynamically linking to...

Sorry-not-sorry for reviving this 18+ month old issue, but I recently encountered another use case where cuFFT callbacks could (I think) really boost performance. My application needs to multiply each...