Re: [RFC PATCH 0/3] KVM: x86: honor guest memory type

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On Wed, Feb 19, 2020 at 6:13 PM Tian, Kevin <kevin.tian@xxxxxxxxx> wrote:
> > > Curious... How is such slot exposed to the guest? A reserved memory
> > > region? Is it static or might be dynamically added?
> > The plan is for virtio-gpu device to reserve a huge memory region in
> > the guest.  Memslots may be added dynamically or statically to back
> > the region.
>
> so the region is marked as E820_RESERVED to prevent guest kernel
> from using it for other purpose and then virtio-gpu device will report
> virtio-gpu driver of the exact location of the region through its own
> interface?
The current plan is that the virtio-gpu device will have a bar for the
region, which is like the vram aperture on real GPUs.  The virtio-gpu
driver manages the region like managing vram.

When the guest userspace allocates from vram, the guest kernel
reserves an unused range from the region and tells the host the
offset.  The host allocates a real GPU buffer, maps the buffer, and
add a memslot with gpa==bar_base+offset (or mremap).  When the guest
userspace mmap, the guest kernel does a io_remap_pfn_range.



[Index of Archives]     [KVM ARM]     [KVM ia64]     [KVM ppc]     [Virtualization Tools]     [Spice Development]     [Libvirt]     [Libvirt Users]     [Linux USB Devel]     [Linux Audio Users]     [Yosemite Questions]     [Linux Kernel]     [Linux SCSI]     [XFree86]

  Powered by Linux