there's no initramfs for qemu target, while v86d is required to have uvesafb driver working. So current design is to have uvesafb as a module, and then later probed by a rcS script before psplash is started. We should remove such limitation to also support built-in driver.
Just to make this clear we need to address whatever the reason is uvesa has to be a module and allow it to be built in and still work.
The reason it fails when built-in is that the uvesafb driver tries and fails to start /sbin/v86d via call_usermode_helper() before the rootfs containing it has been mounted, and the probe that called it fails. We need to arrange for this to happen after the rootfs has been mounted.
According to the kernel documentation in Documentation/fb/uvesafb.txt: Note that the v86d userspace helper has to be available at all times in order for uvesafb to work properly. If you want to use uvesafb during early boot, you will have to include v86d into an initramfs image, and either compile it into the kernel or use it as an initrd. So apparently we'll need to have the qemu targets using initramfs for this to work compiled-in.
Moving to 1.2.
Reassiging to Anuj, as the meta-intel owner so effectively the best available for the qemux86 machines too. This is a very old bug that might be able to be closed once it's actually understood.
uvesa is now not loaded at all for qemux86 or genericx86.