build: Update for v0.14.1 bug fix release

Signed-off-by: Rob Bradford <robert.bradford@intel.com>
vmm: Add missing syscall for vCPU unplug
2026-08-05 02:19:16 +00:00 · 2021-03-31 16:56:05 +01:00 · 2021-03-31 16:56:05 +01:00 · 2021-03-31 16:56:05 +01:00 · 2021-03-31 16:56:05 +01:00 · 2021-03-31 16:56:05 +01:00
6 changed files with 30 additions and 9 deletions
--- a/Cargo.lock
+++ b/Cargo.lock
@@ -212,7 +212,7 @@ dependencies = [

 [[package]]
 name = "cloud-hypervisor"
-version = "0.14.0"
+version = "0.14.1"
 dependencies = [
 "anyhow",
 "api_client",
--- a/Cargo.toml
+++ b/Cargo.toml
@@ -1,6 +1,6 @@
 [package]
 name = "cloud-hypervisor"
-version = "0.14.0"
+version = "0.14.1"
 authors = ["The Cloud Hypervisor Authors"]
 edition = "2018"
 default-run = "cloud-hypervisor"
--- a/release-notes.md
+++ b/release-notes.md
@@ -124,6 +124,17 @@
    - [Unit testing](#unit-testing)
    - [Integration tests parallelization](#integration-tests-parallelization)

+# v0.14.1
+
+Bug fix release branched off the v0.14.0 release. The following bugs were fixed
+in this release:
+
+* CPU hotplug on Windows failed due to misreported CPU state information and
+  the lack of HyperV CPUID bit enabled (#2437, #2449, #2436)
+* A seccomp rule was missing that was triggered on CPU unplug (#2455)
+* A bounds check in VIRTIO queue validation was erroneously generating
+  DescriptorChainTooShort errors in certain circumstances (#2450, #2424)
+
 # v0.14.0

 This release has been tracked through the [0.14.0 project](https://github.com/cloud-hypervisor/cloud-hypervisor/projects/17).
--- a/vm-virtio/src/queue.rs
+++ b/vm-virtio/src/queue.rs
@@ -265,10 +265,7 @@ impl<'a> DescriptorChain<'a> {
    }

    fn is_valid(&self) -> bool {
-        !(self
-            .mem
-            .checked_offset(self.addr, self.len as usize)
-            .is_none()
+        !(!self.mem.check_range(self.addr, self.len as usize)
            || (self.has_next() && self.next >= self.table_size))
    }

--- a/vmm/src/cpu.rs
+++ b/vmm/src/cpu.rs
@@ -433,12 +433,16 @@ const CPU_SELECTION_OFFSET: u64 = 0;

 impl BusDevice for CpuManager {
    fn read(&mut self, _base: u64, offset: u64, data: &mut [u8]) {
+        // The Linux kernel, quite reasonably, doesn't zero the memory it gives us.
+        data.copy_from_slice(&[0; 8][0..data.len()]);
+
        match offset {
+            CPU_SELECTION_OFFSET => {
+                data[0] = self.selected_cpu;
+            }
            CPU_STATUS_OFFSET => {
                if self.selected_cpu < self.present_vcpus() {
                    let state = &self.vcpu_states[usize::from(self.selected_cpu)];
-                    // The Linux kernel, quite reasonably, doesn't zero the memory it gives us.
-                    data.copy_from_slice(&[0; 8][0..data.len()]);
                    if state.active() {
                        data[0] |= 1 << CPU_ENABLE_FLAG;
                    }
@@ -742,10 +746,18 @@ impl CpuManager {
                       | 1 << 2 // AccessSynicRegs
                       | 1 << 3 // AccessSyntheticTimerRegs
                       | 1 << 9, // AccessPartitionReferenceTsc
+                    edx: 1 << 3, // CPU dynamic partitioning
                    ..Default::default()
                })
                .map_err(Error::CpuidKvmHyperV)?;
-            for i in 0x4000_0004..=0x4000_000a {
+            cpuid
+                .push(CpuIdEntry {
+                    function: 0x4000_0004,
+                    eax: 1 << 5, // Recommend relaxed timing
+                    ..Default::default()
+                })
+                .map_err(Error::CpuidKvmHyperV)?;
+            for i in 0x4000_0005..=0x4000_000a {
                cpuid
                    .push(CpuIdEntry {
                        function: i,
--- a/vmm/src/seccomp_filters.rs
+++ b/vmm/src/seccomp_filters.rs
@@ -418,6 +418,7 @@ fn vcpu_thread_rules() -> Result<Vec<SyscallRuleSet>, Error> {
    Ok(vec![
        allow_syscall(libc::SYS_brk),
        allow_syscall(libc::SYS_clock_gettime),
+        allow_syscall(libc::SYS_clock_nanosleep),
        allow_syscall(libc::SYS_close),
        allow_syscall(libc::SYS_dup),
        allow_syscall(libc::SYS_exit),
Author	SHA1	Message	Date
Rob Bradford	bb02c28b62	build: Update for v0.14.1 bug fix release Signed-off-by: Rob Bradford <robert.bradford@intel.com>	2021-03-31 16:56:05 +01:00
Sebastien Boeuf	b2c96dea24	vmm: Add missing syscall for vCPU unplug clock_nanosleep() is triggered when hot-unplugging a vCPU. Signed-off-by: Sebastien Boeuf <sebastien.boeuf@intel.com> (cherry picked from commit `46f96f27a4`) Signed-off-by: Rob Bradford <robert.bradford@intel.com>	2021-03-31 16:56:05 +01:00
Rob Bradford	2c81312433	vm-virtio: queue: Fix descriptor chain validation DescriptorChain::is_valid() wrongly used .checked_offset() to attempt to validate that the descriptor's data is in valid memory. This works in all cases except where the guest has placed the data at the very end of the guest memory as the offset + offset will be outside the range (as the combined offset will be the next byte and as such out of the guest memory). Instead use the function .check_range() takes an offset and a length to validate This fixes issues see with error messages featuring the DescriptorChainTooShort error. Fixes: #2424 Signed-off-by: Rob Bradford <robert.bradford@intel.com> (cherry picked from commit `1eb4ebe3d1`) Signed-off-by: Rob Bradford <robert.bradford@intel.com>	2021-03-31 16:56:05 +01:00
Anatol Belski	e8ee29b4fe	CpuManager: Fix MMIO read handling There are two parts: - Unconditionally zero the output area. The length of the incoming vector has been seen from 1 to 4 bytes, even though just the first byte might need to be handled. But also, this ensures any possibly unhandled offset will return zeroed result to the caller. The former implementation used an I/O port which seems to behave differently from MMIO and wouldn't require explicit output zeroing. - An access with zero offset still takes place and needs to be handled. Fixes #2437. Signed-off-by: Anatol Belski <anbelski@linux.microsoft.com> (cherry picked from commit `9e9aba7c0b`) Signed-off-by: Rob Bradford <robert.bradford@intel.com>	2021-03-31 16:56:05 +01:00
Anatol Belski	498939c297	hyperv: Fix CPU hotadd The following is from the Hyper-V specification v6.0b. Cpuid leaf 0x40000003 EDX: Bit 3: Support for physical CPU dynamic partitioning events is available. When Windows determines to be running under a hypervisor, it will require this cpuid bit to be set to support dynamic CPU operations. Cpuid leaf 0x40000004 EAX: Bit 5: Recommend using relaxed timing for this partition. If used, the VM should disable any watchdog timeouts that rely on the timely delivery of external interrupts. This bit has been figured out as required after seeing guest BSOD when CPU hotplug bit is enabled. Race conditions seem to arise after a hotplug operation, when a system watchdog has expired. Closes #1799. Signed-off-by: Anatol Belski <anbelski@linux.microsoft.com> (cherry picked from commit `5b168f54a6`) Signed-off-by: Rob Bradford <robert.bradford@intel.com>	2021-03-31 16:56:05 +01:00