Tarun Thammisetty

Is there any update regarding this feature? What is currently the best way to look at events like cache-misses and branch mispredictions with magic-trace?

The results in the post were surprising, so I ran a benchmark of my own using ```fio-3.28```. ### System Config - intel i9-14900K 5GHz - NVMe SSD over PCIe4 -...

Thank you for submitting this pull request. I've conducted testing on Ubuntu 24.04 LTS to evaluate the performance. The io_uring-based implementation shows approximately 6% lower throughput compared to the standard...

Here are my results with the latest code: ### Setup - Ubuntu 24.04 - liburing 2.5 - Mainthread bound to 6 - Backend bound to 5 With ```io_uring```: ``` ~/quill/build/benchmarks/backend_throughput$...

I've tried using ``` io_uring_register_iowq_aff ``` as used here [wq-aff.c](https://github.com/axboe/liburing/blob/66e7618a900b172f464cde132548fecbfc09f8cd/test/wq-aff.c#L106C8-L106C34) to set the cpu affinity for worker threads, but couldn't get it to work. The change was to add these...