I have a question about the implementation of KVMGT. According to the
documentation, the virtual machine, send some requests to the gpu directly
(which are mainly related to memory access) and forward some other requests
to KVMGT module.
I read the code of KVMGT but i didn't understand how gpu requests are
trapped by KVMGT ? Are they sent by the VM to a certain shared memory that
is constantly read by KVMGT or there are other entities involved in this
(like KVM) ? and who decide which request to be directly sent to the gpu or
to be forwarded to kvmgt ? is it the graphics card driver of the VM (i915)
or KVMGT ?
Thanks a lot for your help.
Show replies by date