Wayne Gao comments

Results 45 comments of


                                            Wayne Gao

performance tuning help need when I issue read on fast NVMe device with example code

> Seems you were corrupting memory. You must ensure the `service.read_fixed`s are finished before `std::vector buf` gets destroyed. > > See https://github.com/CarterLi/liburing4cpp#taskhpp hello I read your comment and code again....

performance tuning help need when I issue read on fast NVMe device with example code

> > service.read_fixed(0, buf.data(), buf.size(), offset, 0, IOSQE_FIXED_FILE ) | panic_on_err("read_fixed(1)", false); > > This is an async operation, which returns immediately without waiting the I/O to be finished. That...

performance tuning help need when I issue read on fast NVMe device with example code

> Don't talk about performance before you get things correct. OK, I will pre-allocate memory buffer for this experiment. but I feel one global buf does not impact the performance...

nvmf-fio can connect SPDK nvmf target, but failed to start up with libnvmf malloc error

> TCP has been tested a lot, RDMA was tested by RXE and kernel target only. Recently I have no enough time to fix this(I hope that I can fix...

run nvmf-fio never stop after the specified runtime

I found nvme connect kernel way works fine. only libnvmf has some problem dmesg log: ``` [2æœˆ25 11:33] nvmet: creating nvm controller 1 for subsystem nvmet-always-rdma for NQN nqn.2014-08.org.nvmexpress:libnvmf:uuid:44f3e7b5-5076-3652-8f9c-33d4e5afac0. [...

error run the rnn speech workload, failed to process data after enter docker

I install host OS 22.04, now try run this, I got out of space error when building container image. ``` => ERROR [ 1/10] FROM docker.io/pytorch/pytorch:1.7.0-cuda11.0-cudnn8-devel@sha256:837e6964e5db6e5b35f4d5e98e9cac073ab757766039b9503f39c14beafb0e98 182.4s => => resolve...

failed to build object_detection container with below error on FedoraOS37

I run this in ubuntu22.04, the same error below ``` 1.392 self.run_command(cmd) 1.392 File "/opt/conda/lib/python3.7/distutils/dist.py", line 985, in run_command 1.392 cmd_obj.run() 1.392 File "/opt/conda/lib/python3.7/distutils/command/build.py", line 135, in run 1.392 self.run_command(cmd_name)...

docker run error for image_segmentation/pytorch test following the guide

guys, I installed nvidia docker in fedora, now I can start container, but when I run next step it shows me error like below. how to fix this? ``` root@6ec7b9c99e06:/#...

docker run error for image_segmentation/pytorch test following the guide

guys, I install host OS with Ubuntun22.04, I still see this error, could you please shed some light? ``` dcg@oq1:/mnt/nvme1n1/mlperf/ubuntu/training/image_segmentation/pytorch$ sudo docker run --ipc=host -it --rm --runtime=nvidia -v /mnt/nvme1n1/mlperf/ubuntu/training/image_segmentation/pytorch/raw-data-dir:/raw_data -v...

No build support for Fedora37 OS

thank you so much. I am trying manually build it. here is error I found. :) I am using ubuntu 22.04 now ``` runtime library [libglog.so.0] in /usr/lib/x86_64-linux-gnu may be...