When I have a cluster already running, I sometimes want to re-run the setup commands even if there's no changes, without having to shut down the cluster. For example if I am installing a package via a Github repo, I want to trigger the install again after pushing new changes.

However, calling ray up doesn

How you are using LightGBM?

LightGBM component: R package

Environment info

Operating System: macOS 10.14

C++ compiler version: gcc 8.1.0

CMake version: 3.17.3

R version: 4.0.2

LightGBM version or commit hash: https://github.com/microsoft/LightGBM/tree/c07644d1d71540204a9b56f26667e8180bd009e2

Reproducible example(s)

Thanks to @Laurae2 for sharing this with m

PR #6447 adds a public API to get the maximum number of registers per thread (numba.cuda.Dispatcher.get_regs_per_thread()). There are other attributes that might be nice to provide - shared memory per block, local memory per thread, const memory usage, maximum block size.

These are all available in the FuncAttr named tuple: https://github.com/numba/numba/blob/master/numba/cuda/cudadrv/drive

See optuna/optuna#1913.

Some type hints are not complying to the latest mypy == 0.790 and should be addressed.

Version of Singularity:

3.0.3

Expected behavior

Some progress or update information during the creation of the SIF file.

Actual behavior

During "Creating SIF file..." step, the console does not produce any visible output for the user - but does create the resulting file eventually.

Steps to reproduce behavior

I was executing a few “larger” builds through the bui

Is your feature request related to a problem? Please describe.

If I want to read data from a webDav/http endpoint using an oidc token, I'd need to be able to pass the token somehow in ROOT to the Davix client wrapper. Possibly through env variable and rootrc.

Describe the solution you'd like

Make the Davix client wrapper able to take the token from env variable and rootrc.

Addi

Feature request

The asynchronous functionality of parallel-ssh is very much appreciated for management tasks. Thank you for this!

For the feature request:

It seems that the open_session() can keep worker allocated for much longer than expected in failure cases. Would it be reasonable to stretch the timeout option of wait_socket() to open_session() ?