| Message ID | cover.1609231373.git.eafanasova@gmail.com (mailing list archive) |
|---|---|
| Headers | show |
| Series | Introduce MMIO/PIO dispatch file descriptors (ioregionfd) | expand |
On Tue, Dec 29, 2020 at 01:02:42PM +0300, Elena Afanasova wrote: > This patchset introduces a KVM dispatch mechanism which can be used > for handling MMIO/PIO accesses over file descriptors without returning > from ioctl(KVM_RUN). This allows device emulation to run in another task > separate from the vCPU task. > > This is achieved through KVM vm ioctl for registering MMIO/PIO regions and > a wire protocol that KVM uses to communicate with a task handling an > MMIO/PIO access. > > ioregionfd relies on kmemcg in order to limit the amount of kernel memory > that userspace can consume. Can NR_IOBUS_DEVS hardcoded limit be enforced > only in case kmemcg is disabled? Thanks for sharing this! Can you describe the todos? I noticed some in Patch 1 and highlighted them. In addition: * Signal handling when the vCPU thread is interrupted in kernel_read()/kernel_write() > Elena Afanasova (2): > KVM: add initial support for KVM_SET_IOREGION > KVM: add initial support for ioregionfd blocking read/write operations > > arch/x86/kvm/Kconfig | 1 + > arch/x86/kvm/Makefile | 1 + > arch/x86/kvm/x86.c | 1 + > include/linux/kvm_host.h | 17 ++ > include/uapi/linux/kvm.h | 23 +++ > virt/kvm/Kconfig | 3 + > virt/kvm/eventfd.c | 25 +++ > virt/kvm/eventfd.h | 14 ++ > virt/kvm/ioregion.c | 390 +++++++++++++++++++++++++++++++++++++++ > virt/kvm/ioregion.h | 15 ++ > virt/kvm/kvm_main.c | 20 +- > 11 files changed, 507 insertions(+), 3 deletions(-) > create mode 100644 virt/kvm/eventfd.h > create mode 100644 virt/kvm/ioregion.c > create mode 100644 virt/kvm/ioregion.h > > -- > 2.25.1 >
On Tue, 2020-12-29 at 12:06 +0000, Stefan Hajnoczi wrote: > On Tue, Dec 29, 2020 at 01:02:42PM +0300, Elena Afanasova wrote: > > This patchset introduces a KVM dispatch mechanism which can be > > used > > for handling MMIO/PIO accesses over file descriptors without > > returning > > from ioctl(KVM_RUN). This allows device emulation to run in another > > task > > separate from the vCPU task. > > > > This is achieved through KVM vm ioctl for registering MMIO/PIO > > regions and > > a wire protocol that KVM uses to communicate with a task handling > > an > > MMIO/PIO access. > > > > ioregionfd relies on kmemcg in order to limit the amount of kernel > > memory > > that userspace can consume. Can NR_IOBUS_DEVS hardcoded limit be > > enforced > > only in case kmemcg is disabled? > > Thanks for sharing this! Can you describe the todos? I noticed some > in > Patch 1 and highlighted them. In addition: > * Signal handling when the vCPU thread is interrupted in > kernel_read()/kernel_write() > TODOs: * Signal handling when the vCPU thread is interrupted in kernel_read()/kernel_write() * Add ioregionfd cmds/replies serialization * Implement KVM_EXIT_IOREGIONFD_FAILURE * Add non-x86 arch support * Add kvm-unittests > > Elena Afanasova (2): > > KVM: add initial support for KVM_SET_IOREGION > > KVM: add initial support for ioregionfd blocking read/write > > operations > > > > arch/x86/kvm/Kconfig | 1 + > > arch/x86/kvm/Makefile | 1 + > > arch/x86/kvm/x86.c | 1 + > > include/linux/kvm_host.h | 17 ++ > > include/uapi/linux/kvm.h | 23 +++ > > virt/kvm/Kconfig | 3 + > > virt/kvm/eventfd.c | 25 +++ > > virt/kvm/eventfd.h | 14 ++ > > virt/kvm/ioregion.c | 390 > > +++++++++++++++++++++++++++++++++++++++ > > virt/kvm/ioregion.h | 15 ++ > > virt/kvm/kvm_main.c | 20 +- > > 11 files changed, 507 insertions(+), 3 deletions(-) > > create mode 100644 virt/kvm/eventfd.h > > create mode 100644 virt/kvm/ioregion.c > > create mode 100644 virt/kvm/ioregion.h > > > > -- > > 2.25.1 > >