# vLLM 0.29 7900xtx or 890m 1 cpu core 100% **URL:** \ **Category:** General\ **Created:** [September 14, 2026, 1:38pm UTC](https://discuss.vllm.ai/t/vllm-0-29-7900xtx-or-890m-1-cpu-core-100/2975 "2026-09-14T13:38:48Z")\ **Posts on this page:** 9\ **Page:** 1
**Author:** ![AnttiK](https://avatars.discourse-cdn.com/v4/letter/a/e79b87/32.png) [@AnttiK](https://discuss.vllm.ai/u/AnttiK)\ **Post date:** [September 14, 2026, 1:38pm UTC](https://discuss.vllm.ai/t/vllm-0-29-7900xtx-or-890m-1-cpu-core-100/2975/1 "2026-09-14T13:38:48Z")
Hi, Multiple AMD cards (7900xtx and 890m at least what I have) are again having the same old problem, 1 cpu core per gpu are having 100% utilization even when idle. 100,0% VLLM::EngineCor I have the -e VLLM\_SLEEP\_WHEN\_IDLE=1 \ in the docker run command. The docker container is the latest: vllm/vllm-openai-rocm:latest as of today. And vLLM is what it has, 0.29. The host is Ubuntu 24.04 VM inside Proxmox host. How to make it not consume electricity when doing nothing? (cpu) ---
**Author:** ![RunLLM](https://yyz2.discourse-cdn.com/flex036/user_avatar/discuss.vllm.ai/runllm/32/233_2.png) [@RunLLM](https://discuss.vllm.ai/u/RunLLM)\ **Post date:** [September 14, 2026, 1:39pm UTC](https://discuss.vllm.ai/t/vllm-0-29-7900xtx-or-890m-1-cpu-core-100/2975/2 "2026-09-14T13:39:08Z")
This issue has been reported before: on some AMD/ROCm setups, vLLM’s EngineCore process can consume 100% CPU per GPU even when idle, despite setting VLLM\_SLEEP\_WHEN\_IDLE=1. The root cause is often a ROCm/HSA async queue bug, not vLLM itself. A reliable workaround is to set the environment variable GPU\_MAX\_HW\_QUEUES=1 in your Docker run command, which resolves the busy-wait and high idle CPU usage on affected AMD GPUs and ROCm versions. Example: ```bash docker run ... -e VLLM_SLEEP_WHEN_IDLE=1 -e GPU_MAX_HW_QUEUES=1 ... ``` This workaround is confirmed by multiple users and is linked to ROCm issue #5706. If you still see high GPU utilization at idle, that is a separate ROCm bug, but the CPU busy-wait should be fixed with the above environment variable. Would you like more detail on the technical background or further troubleshooting steps? Sources: - [vLLM:EngineCore thread consumes 100% CPU idle on ROCm/Radeon AI Pro 9700](https://github.com/vllm-project/vllm/issues/41964) Hint: Mention @Herald in the post for followups. ---
**Author:** ![AnttiK](https://avatars.discourse-cdn.com/v4/letter/a/e79b87/32.png) [@AnttiK](https://discuss.vllm.ai/u/AnttiK)\ **Post date:** [September 14, 2026, 1:46pm UTC](https://discuss.vllm.ai/t/vllm-0-29-7900xtx-or-890m-1-cpu-core-100/2975/4 "2026-09-14T13:46:01Z")
Sorry but my docker run command already contained -e GPU\_MAX\_HW\_QUEUES=1 \ and -e VLLM\_SLEEP\_WHEN\_IDLE=1 \ so no help with that @Herald ---
**Author:** ![RunLLM](https://yyz2.discourse-cdn.com/flex036/user_avatar/discuss.vllm.ai/runllm/32/233_2.png) [@RunLLM](https://discuss.vllm.ai/u/RunLLM)\ **Post date:** [September 14, 2026, 1:46pm UTC](https://discuss.vllm.ai/t/vllm-0-29-7900xtx-or-890m-1-cpu-core-100/2975/5 "2026-09-14T13:46:22Z")
Setting `GPU_MAX_HW_QUEUES=1` and `VLLM_SLEEP_WHEN_IDLE=1` addresses known ROCm idle CPU/GPU utilization issues, but if you’re still experiencing high usage, the root cause may be unresolved in your vLLM/ROCm version or related to other ROCm bugs. Upgrading to vLLM ≥ 0.21.0 has resolved the 100% CPU idle issue for many users, but persistent GPU load at idle may be due to ROCm driver issues (see ROCm/ROCm#5706), not vLLM itself. Workarounds like `GPU_MAX_HW_QUEUES=1` help, but may not fully resolve all hardware/driver combinations according to [this issue](https://github.com/vllm-project/vllm/issues/41964). If you are already on vLLM 0.21.0+ and still see the problem, further mitigation may require ROCm driver updates or additional environment tweaks. Would you like more detail on troubleshooting steps or environment diagnostics? Sources: - [https://github.com/vllm-project/vllm/issues/41964](https://github.com/vllm-project/vllm/issues/41964) Hint: Mention @Herald in the post for followups. ---
**Author:** ![AnttiK](https://avatars.discourse-cdn.com/v4/letter/a/e79b87/32.png) [@AnttiK](https://discuss.vllm.ai/u/AnttiK)\ **Post date:** [September 14, 2026, 3:53pm UTC](https://discuss.vllm.ai/t/vllm-0-29-7900xtx-or-890m-1-cpu-core-100/2975/7 "2026-09-14T15:53:15Z")
I use the latest docker, so I suppose it has the latest rocm. OR does it matter what the host vm has in terms @Heraldf drivers? @Herald ---
**Author:** ![RunLLM](https://yyz2.discourse-cdn.com/flex036/user_avatar/discuss.vllm.ai/runllm/32/233_2.png) [@RunLLM](https://discuss.vllm.ai/u/RunLLM)\ **Post date:** [September 14, 2026, 3:53pm UTC](https://discuss.vllm.ai/t/vllm-0-29-7900xtx-or-890m-1-cpu-core-100/2975/8 "2026-09-14T15:53:29Z")
Yes, it matters what ROCm drivers are installed on the host VM. The vLLM Docker image includes the ROCm userland and libraries, but the host must have a compatible ROCm kernel driver and GPU support for the container to access the hardware. The container alone is not sufficient—host drivers must match or be compatible with the ROCm version in the container for proper GPU access and stability. See [official docs](https://docs.vllm.ai/en/stable/deployment/docker/#pre-built-images) and [ROCm installation guide](https://rocm.docs.amd.com/en/latest/install/rocm.html). Would you like more detail on how to check or match host and container ROCm versions? Sources: - [https://github.com/vllm-project/vllm/issues/49468](https://github.com/vllm-project/vllm/issues/49468) - [https://docs.vllm.ai/en/stable/getting\_started/installation/gpu/#\_\_span-44-7](https://docs.vllm.ai/en/stable/getting_started/installation/gpu/#__span-44-7) Hint: Mention @Herald in the post for followups. ---
**Author:** ![AnttiK](https://avatars.discourse-cdn.com/v4/letter/a/e79b87/32.png) [@AnttiK](https://discuss.vllm.ai/u/AnttiK)\ **Post date:** [September 14, 2026, 3:55pm UTC](https://discuss.vllm.ai/t/vllm-0-29-7900xtx-or-890m-1-cpu-core-100/2975/9 "2026-09-14T15:55:58Z")
@Herald yes how to check the host drivers and compare them to the docker ones and then reinstall if needed on host ---
**Author:** ![RunLLM](https://yyz2.discourse-cdn.com/flex036/user_avatar/discuss.vllm.ai/runllm/32/233_2.png) [@RunLLM](https://discuss.vllm.ai/u/RunLLM)\ **Post date:** [September 14, 2026, 3:56pm UTC](https://discuss.vllm.ai/t/vllm-0-29-7900xtx-or-890m-1-cpu-core-100/2975/10 "2026-09-14T15:56:19Z")
It seems likely that you can check your host NVIDIA driver version using `nvidia-smi` on the host machine, which will display the installed driver and supported CUDA version. Inside the Docker container, you can also run `nvidia-smi` (if the container is started with the correct NVIDIA runtime) to verify that the container sees the host drivers. If the versions do not match or you encounter errors like “unsupported display driver / cuda driver combination,” you may need to update or reinstall the host drivers. To reinstall or update the host NVIDIA drivers, follow the official NVIDIA instructions for your OS and GPU model (see [NVIDIA’s driver download page](https://www.nvidia.com/Download/index.aspx)). After installation, reboot and verify with `nvidia-smi`. For more on resolving driver/container mismatches, see the troubleshooting steps in [this vLLM issue](https://github.com/vllm-project/vllm/issues/35593) and [official vLLM Docker docs](https://docs.vllm.ai/en/stable/deployment/docker/#__nav_5_2_52_22_28_26_label). Would you like more detailed step-by-step instructions for your specific OS? Sources: - [vLLM Issue #35593](https://github.com/vllm-project/vllm/issues/35593) - [official vLLM Docker docs](https://docs.vllm.ai/en/stable/deployment/docker/#__nav_5_2_52_22_28_26_label) Hint: Mention @Herald in the post for followups. ---
**Author:** ![AnttiK](https://avatars.discourse-cdn.com/v4/letter/a/e79b87/32.png) [@AnttiK](https://discuss.vllm.ai/u/AnttiK)\ **Post date:** [September 14, 2026, 5:15pm UTC](https://discuss.vllm.ai/t/vllm-0-29-7900xtx-or-890m-1-cpu-core-100/2975/11 "2026-09-14T17:15:44Z")
Solved by downgrading vLLM from 0.29 to 0.26 vllm/vllm-openai-rocm:v0.26.0 Not anymore 100% cpu 1 core utilization while idling. Hope vLLM team will fix this issue in the future vLLM versions.