| Summary: | qemu machines use Cirrus graphics, should use std or virtio | ||
|---|---|---|---|
| Product: | [Build System, Metadata & Runtime] OE-Core | Reporter: | Ross Burton <ross.burton> |
| Component: | core | Assignee: | Kevin Hao <kexin.hao> |
| Status: | RESOLVED FIXED | QA Contact: | |
| Severity: | normal | ||
| Priority: | Medium+ | CC: | akuster, alex.kanavin, meta.mr.watcher, meta.watcher, randy.macleod, richard.purdie |
| Version: | 0.0.0 | ||
| Target Milestone: | 2.8 M4 | ||
| Hardware: | x86 | ||
| OS: | Multiple | ||
| Whiteboard: | |||
| OS type for building Yocto: | --- | Type of Regression: | --- |
| Verified: | Documentation change: | No (bug/feature does not impact docs) | |
|
Description
Ross Burton
2019-08-01 15:52:07 UTC
Some context: https://www.kraxel.org/blog/2014/10/qemu-using-cirrus-considered-harmful/ That predates virtio drivers. Plan: 1) Unify all boards to at least -vga=std, remove cirrus driver from oe-core. Needs CONFIG_DRM_BOCHS in the kernel. 2) See if virtio works, enable drivers on all platforms, use that. 3) If (2) fails, try vmware on platforms where vmware fails Some incomplete prior art at poky-contrib:ross/virtio +1 from me: let's standardize on virtio (you do not need working virgl to have it). Kevin, If you could do this before M3, that would be great. Let us know your plans and availability. Patch for step 1 sent: https://lists.yoctoproject.org/pipermail/linux-yocto/2019-August/007877.html Note that this enables bochs for qemux86(-64); it is already enabled for qemuarm(64) and qemurisc(32/64). Once this merges and propagates to oe-core, we can drop the vmware/cirrus stuff. To clarify, that is a patch only for the kernel module enabling (CONFIG_DRM_BOCHS). Unifying drivers and cleanup still to do (once we are able to test the bochs). My apologies, but it is rather impolite to re-assign the task to someone else without asking them first, even if they did some work towards resolving it. isn't this fixed now ? https://git.openembedded.org/openembedded-core/commit/scripts?h=zeus&id=73cb104f3307736f4922f2e0c9648f9b2d3b3b6b also the kernal fragments went is too. https://git.openembedded.org/openembedded-core/commit/meta/recipes-kernel/linux?h=zeus&id=a777e0f34e106455f963bd58fd8728a16c588c4d https://git.openembedded.org/openembedded-core/commit/meta/recipes-kernel/linux?h=zeus&id=569d3f5d0454ed31f2f6df29f1703246a3dcd715 it in zeus for sure. This is up to Ross, I am not sure what else could be done really. Final piece: http://git.yoctoproject.org/cgit.cgi/poky/commit/?id=58e85c60cd15cf4c0b47cddcf507543461c1a328 Given the above commit, issue resolved. |