This is supported by AMDVLK (as well as the NVIDIA proprietary driver). This can be useful for things like resource streaming and memory defragmentation. It'd be good if RADV could add support for this as well.
FYI I've had a branch and posted patches for starting this on gfx8 and gfx9. https://github.com/airlied/mesa/tree/radv-wip-transfer-queue-4 There are lots of corner cases and I hadn't resolved all the CTS tests on gfx9 I don't think. I'm not sure when I'll loop back around to it, it would be good to have something more to test it with than CTS.
-- GitLab Migration Automatic Message -- This bug has been migrated to freedesktop.org's GitLab instance and has been closed from further activity. You can subscribe and participate further through the new bug through this link to our GitLab instance: https://gitlab.freedesktop.org/mesa/mesa/issues/850.
Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.