云微

Results 169 comments of 云微

I think that's a good approach! Please do it

Yes, I think so! We are working to fix it in our next version, this is a high priority.

We already have some GPU specific maps and helpers, like in https://github.com/eunomia-bpf/bpftime/tree/master/runtime/src/bpf_map/gpu This is per GPU thread, no lock and store in the GPU memory. You can also find some...

Does this complete?

Hi YRXING, could you share your system environment? I'm also curious about the results...why so slow (

https://eunomia-bpf.github.io/bpftime/benchmark/uprobe/results.html We have a standard ci that runs in github action every time we have a commit. The result is similar to the example_results.md in our repo now, and different...

Thanks! By the way, are you running the benchmarks the same as it? https://github.com/eunomia-bpf/bpftime/blob/master/.github%2Fworkflows%2Fbenchmarks.yml

We can add some like SPIR-V backend. I already has some experiment in https://github.com/eunomia-bpf/XDP-on-GPU