| Commit message (Collapse) | Author | Age | Files | Lines |
|---|
| |
|
|
| |
Signed-off-by: Keir Fraser <keir.fraser@citrix.com>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
When creating an HVM domain, if e.g. another domain is created before
qemu allocates video memory, the extra 8MB memory ballooning is not
available any more, because it got consumed by the other domain.
This fixes it by taking video memory from the main memory:
- make hvmloader use e820_malloc to reserve some of the main memory
and notify ioemu of its address through the Xen platform PCI card.
- add XENMAPSPACE_mfn to the xen_add_to_physmap memory op, to allow
ioemu to move the MFNs between the original position and the PCI
mapping, when LFB acceleration is disabled/enabled
- add a remove_from_physmap memory op, to allow ioemu to unmap it
completely for the case of old guests with acceleration disabled.
- add xc_domain_memory_translate_gpfn_list to libxc to allow ioemu to
get the MFNs of the video memory.
- have xend save the PCI memory space instead of ioemu: if a memory
page is there, the guest can access it like usual memory, so xend
can safely be responsible to save it. The extra benefit is that
live migration will apply the logdirty optimization there too.
- handle old saved images, populating the video memory from ioemu if
really needed.
Signed-off-by: Samuel Thibault <samuel.thibault@eu.citrix.com>
|
| |
|
|
| |
Signed-off-by: Stefano Stabellini <stefano.stabellini@eu.citrix.com>
|
| |
|
|
|
|
|
|
|
|
| |
This adds a new HVM op that enables tracking dirty bits of a range of
video RAM. The idea is to optimize just for the most common case
(only one guest mapping, with sometimes some temporary other
mappings), which permits to keep the overhead on shadow as low as
possible.
Signed-off-by: Samuel Thibault <samuel.thibault@eu.citrix.com>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
- no need to call the colourdepth callback in text mode: the buffer
cannot be shared anyway;
- line size changes are currently undetected: since we added a
linesize parameter to the resize callback, we also need to detect
line size changes and call dpy_resize accordingly;
- 8 bit colour depth with the shared framebuffer is broken: in order
to fix it I disabled the shared buffer in vnc for the 8bit colour
depth case (it has to be done in software anyway..) and implemented
paletted colours in both opengl and sdl;
- opengl rendering is broken when there is padding in the framebuffer
lines: removing unnecessary GL_UNPACK_ALIGNMENT settings so that the
GL_UNPACK_ROW_LENGTH parameter can work properly.
Signed-off-by: Stefano Stabellini <stefano.stabellini@eu.citrix.com>
|
| |
|
|
| |
Signed-off-by: Christian Limpach <Christian.Limpach@xensource.com>
|
| |
|
|
|
| |
This makes the SSE-optimized vram_dirty() safe again.
Signed-off-by: Tim Deegan <Tim.Deegan@xensource.com>
|
| |
|
|
|
| |
Signed-off-by: Christian Limpach <Christian.Limpach@xensource.com>
|
| |
|
|
|
|
| |
Signed-off-by: Christian Limpach <Christian.Limpach@xensource.com>
|
| |
|
|
|
|
| |
Signed-off-by: Christian Limpach <Christian.Limpach@xensource.com>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
initial vga mode and linear framebuffer.
When creating device model, the vga memory is allocated through
xc_domain_memory_increase_reservation(), however, when switch to use
linear framebuffer, we didn't free that memory, and re-allocate the vga
memory through set_mm_mapping(), that caused memory leak. Now it is
changed to just using qemu_malloc when create device model, and free
that memory when switch to linear framebuffer.
After unset_vram_mapping(), the old memory pointer should be
unmapped; after set_vram_mapping(), the old memory should be freed.
Signed-off-by: Yunhong Jiang <yunhong.jiang@intel.com>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
OS was given write access to the VRAM the device model tracked all
VRAM changes by updating the entire screen on every output loop,
causing significant overhead (a CPU bound loop in a guest slows down
by about 35%) and significant mouse latency (VNC uses the same data
path for mouse events and video updates). With the shadow VRAM only
modified pages need to be updated and the comparison of the shadow
VRAM to the real VRAM only adds ~4% overhead while eliminating the
mouse latencies.
Signed-off-by: Don Dugger <donald.d.dugger@intel.com>
|
| |
|
|
|
|
|
|
| |
Signed-off-by: Yunhong Jiang <yunhong.jiang@intel.com>
Signed-off-by: Xin B Li <xin.b.li@intel.com>
Signed-off-by: Jun Nakajima <jun.nakajima@intel.com>
|
|
|
New qemu-based ioemu for fully virtualised guests.
Signed-off-by: Arun Sharma <arun.sharma@intel.com>
Signed-off-by: Keir Fraser <keir@xensource.com>
|
