Andreas Klöckner comments

Results 957 comments of


                                            Andreas Klöckner

generate_loop_schedule_v2

I was just rebasing this. You beat me to it! :)

generate_loop_schedule_v2

Unsubscribing... @-mention or request review once it's ready for a look or needs attention. But I'm eager to get this in soon!

Add pyopencl-like GenericScanKernel

Transplanted here for CI: https://gitlab.tiker.net/inducer/pycuda/merge_requests/11

Add pyopencl-like GenericScanKernel

* You (I assume inadvertently) changed the `bpl-subset` submodule back to an old version. (see the CI failure) * I may not have been fully clear on what I was...

Add pyopencl-like GenericScanKernel

> Ah! I think I misunderstood earlier. But just to be clear, you want not only the SCAN_INTERVALS_SOURCE and the UPDATE_SOURCE to be identical but the entire GenericScanKernel class, right?...

invoker lock-file conflict on nfs cluster

Thanks for the report! I'm currently chasing a deadline (Sunday)--I'll worry about this next week, likely by deriving all cache dirs (binary and invoker) from the one passed to the...

feat: allow capturing the compiler output

Happy to consider a PR. Functionality is in `pyopencl._cl.Program._get_build_logs()`.

pycuda uses unavailable compute capabilities on older versions of CUDA with new hardware

You can force an `arch` by passing an argument to `SourceModule`: https://documen.tician.de/pycuda/driver.html#pycuda.compiler.SourceModule I'd be happy to take a patch/pull request that reads an environment variable like `PYCUDA_DEFAULT_JIT_ARCH`. (e.g.)

pycuda uses unavailable compute capabilities on older versions of CUDA with new hardware

Short of parsing `nvcc` output, I don't think so.