mirror of
https://github.com/hanzoai/visor.git
synced 2026-09-24 17:04:14 +00:00
Tianyu Zhou
fc1afc9de5
SlimVM is a lightweight VMX-based platform for gVisor that accelerates sentry system calls by eliminating unnecessary user/kernel context switches. In a traditional gVisor + KVM setup, every guest syscall triggers a full VM exit to userspace, then re-enters the kernel via a regular syscall: gr0 -> hr0 (vm_exit) -> hr3 (ioctl return) -> hr0 (re-invoke syscall) SlimVM eliminates this overhead. When the guest executes a VMCALL, the VM exits into VMX root mode where SlimVM directly invokes the host kernel's syscall handler, then immediately resumes the guest: gr0 -> hr0 (vmcall + direct function call) SlimVM requires a companion kernel module that provides the /dev/slimvm device interface. See https://github.com/antgroup/slimvm for the kernel module and full documentation. A sentry->host getpid microbenchmark (Xeon 6982P-C, host kernel 5.10.134) shows the effect: │ ns/op │ vs base │ base 82.5 1.0x slimvm 420.9 5.1x kvm 3566.1 43.2x slimvm is ~8.5x faster than the kvm platform on this path. See the issue for details. Updates: https://github.com/google/gvisor/issues/13330 Co-developed-by: Chenggang <[email protected]> Co-developed-by: Robin Luk <[email protected]> Co-developed-by: Tiwei Bie <[email protected]> Co-developed-by: Jianfeng Tan <[email protected]> Co-developed-by: liushi.ls <[email protected]> Co-developed-by: Yong He <[email protected]> Co-developed-by: Aaron Lu <[email protected]> Co-developed-by: chris.zn <[email protected]> Co-developed-by: Lai Jiangshan <[email protected]> Co-developed-by: Lingfu <[email protected]> Co-developed-by: Min Le <[email protected]> Co-developed-by: Quan Xu <[email protected]> Co-developed-by: Zhang Haoyu <[email protected]> Co-developed-by: Dawei Shen <[email protected]> Signed-off-by: Tianyu Zhou <[email protected]>
91 lines
2.5 KiB
Go
91 lines
2.5 KiB
Go
// Copyright 2026 The gVisor Authors.
|
|
//
|
|
// Licensed under the Apache License, Version 2.0 (the "License");
|
|
// you may not use this file except in compliance with the License.
|
|
// You may obtain a copy of the License at
|
|
//
|
|
// http://www.apache.org/licenses/LICENSE-2.0
|
|
//
|
|
// Unless required by applicable law or agreed to in writing, software
|
|
// distributed under the License is distributed on an "AS IS" BASIS,
|
|
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
|
// See the License for the specific language governing permissions and
|
|
// limitations under the License.
|
|
|
|
//go:build amd64
|
|
// +build amd64
|
|
|
|
package slimvm
|
|
|
|
import (
|
|
"unsafe"
|
|
|
|
"golang.org/x/sys/unix"
|
|
"gvisor.dev/gvisor/pkg/abi/linux"
|
|
"gvisor.dev/gvisor/pkg/hostsyscall"
|
|
)
|
|
|
|
// loadSegments copies the current segments.
|
|
//
|
|
// This may be called from within the signal context and throws on error.
|
|
//
|
|
//go:nosplit
|
|
func (c *vCPU) loadSegments(tid uint64) {
|
|
if errno := hostsyscall.RawSyscallErrno(
|
|
unix.SYS_ARCH_PRCTL,
|
|
linux.ARCH_GET_FS,
|
|
uintptr(unsafe.Pointer(&c.CPU.Registers().Fs_base)),
|
|
0); errno != 0 {
|
|
throw("getting FS segment")
|
|
}
|
|
if errno := hostsyscall.RawSyscallErrno(
|
|
unix.SYS_ARCH_PRCTL,
|
|
linux.ARCH_GET_GS,
|
|
uintptr(unsafe.Pointer(&c.CPU.Registers().Gs_base)),
|
|
0); errno != 0 {
|
|
throw("getting GS segment")
|
|
}
|
|
c.tid.Store(tid)
|
|
}
|
|
|
|
// setUserRegisters sets user registers in the vCPU.
|
|
func (c *vCPU) setUserRegisters(uregs *userRegs) error {
|
|
c.vmxConfig.userRegs = *uregs
|
|
return nil
|
|
}
|
|
|
|
// setSystemRegisters sets system registers.
|
|
func (c *vCPU) setSystemRegisters(sregs *systemRegs) error {
|
|
c.vmxConfig.sysRegs = *sregs
|
|
return nil
|
|
}
|
|
|
|
// setCPUID sets the CPUID to be used by the guest.
|
|
//
|
|
// In SlimVM, guest CPU features are controlled via CPUID faulting and emulated
|
|
// by the sentry, so no per-vCPU CPUID is needed.
|
|
func (c *vCPU) setCPUID() error {
|
|
return nil
|
|
}
|
|
|
|
// setSystemTime sets the TSC for the vCPU.
|
|
//
|
|
// In SlimVM, the guest shares the host TSC directly, so no TSC
|
|
// synchronization is needed and there is no host/guest TSC offset.
|
|
func (c *vCPU) setSystemTime() error {
|
|
return nil
|
|
}
|
|
|
|
// setSignalMask sets the vCPU signal mask.
|
|
//
|
|
// This must be called prior to running the vCPU.
|
|
//
|
|
// In KVM, this uses KVM_SET_SIGNAL_MASK ioctl to configure which
|
|
// signals are blocked while the vCPU is running. In SlimVM, the
|
|
// kernel module manages the signal mask internally during
|
|
// _SLIMVM_RUN (blocking all signals except SIGKILL, SIGSTOP,
|
|
// SIG_BOUNCE, and SIGPROF), so no userspace configuration is needed.
|
|
func (c *vCPU) setSignalMask() error {
|
|
return nil
|
|
}
|