# 590 release feedback & discussion

**URL:** <https://forums.developer.nvidia.com/t/590-release-feedback-discussion/353310>\
**Category:** Linux\
**Created:** [December 2, 2025, 3:33pm UTC](https://forums.developer.nvidia.com/t/590-release-feedback-discussion/353310 "2025-12-02T15:33:15Z")\
**Posts on this page:** 1\
**Showing post:** 136

<div class="post-metadata">

**Author:** ![vtkc2016](https://sea2.discourse-cdn.com/nvidia/user_avatar/forums.developer.nvidia.com/vtkc2016/32/457980_2.png) [@vtkc2016](https://forums.developer.nvidia.com/u/vtkc2016)\
**Post date:** [December 19, 2025, 1:32am UTC](https://forums.developer.nvidia.com/t/590-release-feedback-discussion/353310/136 "2025-12-19T01:32:46Z")

</div>

I am experiencing severe system latency and frequent audio Xruns (stuttering) on an **Arch-based system** running **Kernel 6.13** with an **RTX 3070**.

**Technical Problem:** The GPU is currently forced onto **IRQ 16** , which is shared with multiple PCIe Root Ports on my **Z690** motherboard. This shared legacy interrupt line is causing significant processing delays, especially when using the **NVIDIA Open Kernel Modules**.

**Key Details:**

- **Hardware:** Intel i7-13700KF | RTX 3070 | Z690 Chipset.

- **Driver:** 590.xx (Open Modules).

- **Architecture:** System is optimized for **x86-64-v3**.

- **Symptoms:** Audio dropouts and micro-stuttering under GPU load.

**Request:** It appears that MSI (Message Signaled Interrupts) is not being correctly utilized or remapped for the GPU on this specific chipset with the current open drivers.

1. Is there a known workaround to force MSI on the open modules for Z690?

2. Will the upcoming **v590 driver** address IRQ sharing and improve interrupt handling for the open-source kernel modules?

---

_[View the full topic](https://forums.developer.nvidia.com/t/590-release-feedback-discussion/353310)._