# KMSDRM: using atomic mode setting breaks GPU compatibility

**URL:** <https://discourse.libsdl.org/t/kmsdrm-using-atomic-mode-setting-breaks-gpu-compatibility/28372>\
**Category:** SDL Development\
**Created:** [November 22, 2020, 5:42pm UTC](https://discourse.libsdl.org/t/kmsdrm-using-atomic-mode-setting-breaks-gpu-compatibility/28372 "2020-11-22T17:42:37Z")\
**Posts on this page:** 1\
**Showing post:** 6

<div class="post-metadata">

**Author:** ![rtrussell](https://discourse.libsdl.org/user_avatar/discourse.libsdl.org/rtrussell/32/4768_2.png) [@rtrussell](https://discourse.libsdl.org/u/rtrussell)\
**Post date:** [December 14, 2020, 10:00pm UTC](https://discourse.libsdl.org/t/kmsdrm-using-atomic-mode-setting-breaks-gpu-compatibility/28372/6 "2020-12-14T22:00:34Z")

</div>

> [@vanfanel](#):
>
> Please don’t push on trying to revert the change to ATOMIC

Delaying the release is my preference too. Wasn’t the [idea](https://discourse.libsdl.org/t/sdl-versioning-update-even-numbered-releases/26632) that even-numbered SDL2 releases should not have radical changes, to minimise the chance of regressions?

I regret that the version numbering doesn’t adhere to the [major.minor.patch](https://semver.org/) system, in which changes to the patch number should imply only “backwards compatible bug fixes”.

---

_[View the full topic](https://discourse.libsdl.org/t/kmsdrm-using-atomic-mode-setting-breaks-gpu-compatibility/28372)._