| Message ID | 6d0bd7f0e1aeec3ddb603ae4ff334c75c7d0d7b3.1678351495.git.viresh.kumar@linaro.org (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | qemu: vhost-user: Support Xen memory mapping quirks | expand |
On Thu, Mar 09, 2023 at 02:21:01PM +0530, Viresh Kumar wrote: > The current model of memory mapping at the back-end works fine where a > standard call to mmap() (for the respective file descriptor) is enough > before the front-end can start accessing the guest memory. > > There are other complex cases though where the back-end needs more > information and simple mmap() isn't enough. For example Xen, a type-1 > hypervisor, currently supports memory mapping via two different methods, > foreign-mapping (via /dev/privcmd) and grant-dev (via /dev/gntdev). In > both these cases, the back-end needs to call mmap() and ioctl(), with > extra information like the Xen domain-id of the guest whose memory we > are trying to map. > > Add a new protocol feature, 'VHOST_USER_PROTOCOL_F_XEN_MMAP', which lets > the back-end know about the additional memory mapping requirements. > When this feature is negotiated, the front-end will send the additional > information within the memory regions themselves. > > Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org> > --- > docs/interop/vhost-user.rst | 21 +++++++++++++++++++++ > 1 file changed, 21 insertions(+) Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Viresh Kumar <viresh.kumar@linaro.org> writes: > The current model of memory mapping at the back-end works fine where a > standard call to mmap() (for the respective file descriptor) is enough > before the front-end can start accessing the guest memory. > > There are other complex cases though where the back-end needs more > information and simple mmap() isn't enough. For example Xen, a type-1 > hypervisor, currently supports memory mapping via two different methods, > foreign-mapping (via /dev/privcmd) and grant-dev (via /dev/gntdev). In > both these cases, the back-end needs to call mmap() and ioctl(), with > extra information like the Xen domain-id of the guest whose memory we > are trying to map. > > Add a new protocol feature, 'VHOST_USER_PROTOCOL_F_XEN_MMAP', which lets > the back-end know about the additional memory mapping requirements. > When this feature is negotiated, the front-end will send the additional > information within the memory regions themselves. > > Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org> Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
diff --git a/docs/interop/vhost-user.rst b/docs/interop/vhost-user.rst index 1720d681264d..5a070adbc1aa 100644 --- a/docs/interop/vhost-user.rst +++ b/docs/interop/vhost-user.rst @@ -145,6 +145,26 @@ Memory region description :mmap offset: 64-bit offset where region starts in the mapped memory +When the ``VHOST_USER_PROTOCOL_F_XEN_MMAP`` protocol feature has been +successfully negotiated, the memory region description contains two extra +fields at the end. + ++---------------+------+--------------+-------------+----------------+-------+ +| guest address | size | user address | mmap offset | xen mmap flags | domid | ++---------------+------+--------------+-------------+----------------+-------+ + +:xen mmap flags: 32-bit bit field + +- Bit 0 is set for Xen foreign memory mapping. +- Bit 1 is set for Xen grant memory mapping. +- Bit 8 is set if the memory region can not be mapped in advance, and memory + areas within this region must be mapped / unmapped only when required by the + back-end. The back-end shouldn't try to map the entire region at once, as the + front-end may not allow it. The back-end should rather map only the required + amount of memory at once and unmap it after it is used. + +:domid: a 32-bit Xen hypervisor specific domain id. + Single memory region description ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ @@ -864,6 +884,7 @@ Protocol features #define VHOST_USER_PROTOCOL_F_INBAND_NOTIFICATIONS 14 #define VHOST_USER_PROTOCOL_F_CONFIGURE_MEM_SLOTS 15 #define VHOST_USER_PROTOCOL_F_STATUS 16 + #define VHOST_USER_PROTOCOL_F_XEN_MMAP 17 Front-end message types -----------------------
The current model of memory mapping at the back-end works fine where a standard call to mmap() (for the respective file descriptor) is enough before the front-end can start accessing the guest memory. There are other complex cases though where the back-end needs more information and simple mmap() isn't enough. For example Xen, a type-1 hypervisor, currently supports memory mapping via two different methods, foreign-mapping (via /dev/privcmd) and grant-dev (via /dev/gntdev). In both these cases, the back-end needs to call mmap() and ioctl(), with extra information like the Xen domain-id of the guest whose memory we are trying to map. Add a new protocol feature, 'VHOST_USER_PROTOCOL_F_XEN_MMAP', which lets the back-end know about the additional memory mapping requirements. When this feature is negotiated, the front-end will send the additional information within the memory regions themselves. Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org> --- docs/interop/vhost-user.rst | 21 +++++++++++++++++++++ 1 file changed, 21 insertions(+)