Compare commits

..

1 Commits

Author SHA1 Message Date
Fabiano Fidêncio
ca1e365a27 tests: arm64: Increase k8s tests timeout to 60 minutes
This allows adding different runners in case the powerful one goes down
for one reason or another.

Signed-off-by: Fabiano Fidêncio <ffidencio@nvidia.com>
2026-01-17 12:49:45 +01:00
19 changed files with 146 additions and 1785 deletions

View File

@@ -65,7 +65,7 @@ jobs:
run: bash tests/integration/kubernetes/gha-run.sh install-bats
- name: Run tests
timeout-minutes: 30
timeout-minutes: 60
run: bash tests/integration/kubernetes/gha-run.sh run-tests
- name: Report tests

View File

@@ -1 +1 @@
3.25.0
3.24.0

View File

@@ -133,17 +133,6 @@ PKGLIBEXECDIR := $(LIBEXECDIR)/$(PROJECT_DIR)
FIRMWAREPATH :=
FIRMWAREVOLUMEPATH :=
ROOTMEASURECONFIG ?= ""
KERNELTDXPARAMS += $(ROOTMEASURECONFIG)
# TDX
DEFSHAREDFS_QEMU_TDX_VIRTIOFS := none
FIRMWARETDXPATH := $(PREFIXDEPS)/share/ovmf/OVMF.inteltdx.fd
# SEV-SNP
FIRMWARE_SNP_PATH := $(PREFIXDEPS)/share/ovmf/AMDSEV.fd
FIRMWARE_VOLUME_SNP_PATH :=
##VAR DEFVCPUS=<number> Default number of vCPUs
DEFVCPUS := 1
##VAR DEFMAXVCPUS=<number> Default maximum number of vCPUs
@@ -187,7 +176,6 @@ DEFVIRTIOFSQUEUESIZE ?= 1024
# Make sure you quote args.
DEFVIRTIOFSEXTRAARGS ?= [\"--thread-pool-size=1\", \"-o\", \"announce_submounts\"]
DEFENABLEIOTHREADS := false
DEFINDEPIOTHREADS := 0
DEFENABLEVHOSTUSERSTORE := false
DEFVHOSTUSERSTOREPATH := $(PKGRUNDIR)/vhost-user
DEFVALIDVHOSTUSERSTOREPATHS := [\"$(DEFVHOSTUSERSTOREPATH)\"]
@@ -204,8 +192,6 @@ QEMUTDXQUOTEGENERATIONSERVICESOCKETPORT := 4050
DEFCREATECONTAINERTIMEOUT ?= 30
DEFCREATECONTAINERTIMEOUT_COCO ?= 60
DEFSTATICRESOURCEMGMT_COCO = true
DEFDISABLEIMAGENVDIMM ?= false
DEFPODRESOURCEAPISOCK := ""
SED = sed
CLI_DIR = cmd
@@ -306,30 +292,6 @@ ifneq (,$(QEMUCMD))
CONFIGS += $(CONFIG_QEMU)
CONFIG_FILE_QEMU_TDX = configuration-qemu-tdx-runtime-rs.toml
CONFIG_QEMU_TDX = config/$(CONFIG_FILE_QEMU_TDX)
CONFIG_QEMU_TDX_IN = $(CONFIG_QEMU_TDX).in
CONFIG_PATH_QEMU_TDX = $(abspath $(CONFDIR)/$(CONFIG_FILE_QEMU_TDX))
CONFIG_PATHS += $(CONFIG_PATH_QEMU_TDX)
SYSCONFIG_QEMU_TDX = $(abspath $(SYSCONFDIR)/$(CONFIG_FILE_QEMU_TDX))
SYSCONFIG_PATHS += $(SYSCONFIG_QEMU_TDX)
CONFIGS += $(CONFIG_QEMU_TDX)
CONFIG_FILE_QEMU_SNP = configuration-qemu-snp-runtime-rs.toml
CONFIG_QEMU_SNP = config/$(CONFIG_FILE_QEMU_SNP)
CONFIG_QEMU_SNP_IN = $(CONFIG_QEMU_SNP).in
CONFIG_PATH_QEMU_SNP = $(abspath $(CONFDIR)/$(CONFIG_FILE_QEMU_SNP))
CONFIG_PATHS += $(CONFIG_PATH_QEMU_SNP)
SYSCONFIG_QEMU_SNP = $(abspath $(SYSCONFDIR)/$(CONFIG_FILE_QEMU_SNP))
SYSCONFIG_PATHS += $(SYSCONFIG_QEMU_SNP)
CONFIGS += $(CONFIG_QEMU_SNP)
CONFIG_FILE_QEMU_SE = configuration-qemu-se-runtime-rs.toml
CONFIG_QEMU_SE = config/$(CONFIG_FILE_QEMU_SE)
CONFIG_QEMU_SE_IN = $(CONFIG_QEMU_SE).in
@@ -560,7 +522,6 @@ USER_VARS += DEFVIRTIOFSEXTRAARGS
USER_VARS += DEFENABLEANNOTATIONS
USER_VARS += DEFENABLEANNOTATIONS_COCO
USER_VARS += DEFENABLEIOTHREADS
USER_VARS += DEFINDEPIOTHREADS
USER_VARS += DEFSECCOMPSANDBOXPARAM
USER_VARS += DEFGUESTSELINUXLABEL
USER_VARS += DEFENABLEVHOSTUSERSTORE
@@ -581,7 +542,6 @@ USER_VARS += DEFSTATICRESOURCEMGMT_FC
USER_VARS += DEFSTATICRESOURCEMGMT_CLH
USER_VARS += DEFSTATICRESOURCEMGMT_QEMU
USER_VARS += DEFSTATICRESOURCEMGMT_COCO
USER_VARS += DEFDISABLEIMAGENVDIMM
USER_VARS += DEFBINDMOUNTS
USER_VARS += DEFVFIOMODE
USER_VARS += DEFVFIOMODE_SE
@@ -602,13 +562,6 @@ USER_VARS += DEFFORCEGUESTPULL
USER_VARS += QEMUTDXQUOTEGENERATIONSERVICESOCKETPORT
USER_VARS += DEFCREATECONTAINERTIMEOUT
USER_VARS += DEFCREATECONTAINERTIMEOUT_COCO
USER_VARS += QEMUTDXEXPERIMENTALCMD
USER_VARS += FIRMWARE_SNP_PATH
USER_VARS += FIRMWARE_VOLUME_SNP_PATH
USER_VARS += KERNELTDXPARAMS
USER_VARS += DEFSHAREDFS_QEMU_TDX_VIRTIOFS
USER_VARS += FIRMWARETDXPATH
USER_VARS += DEFPODRESOURCEAPISOCK
SOURCES := \
$(shell find . 2>&1 | grep -E '.*\.rs$$') \
@@ -646,8 +599,6 @@ GENERATED_VARS = \
VERSION \
CONFIG_DB_IN \
CONFIG_FC_IN \
CONFIG_QEMU_TDX_IN \
CONFIG_QEMU_SNP_IN \
$(USER_VARS)

View File

@@ -1,770 +0,0 @@
# Copyright (c) 2017-2019 Intel Corporation
# Copyright (c) 2021 Adobe Inc.
# Copyright (c) 2024 IBM Corp.
# Copyright (c) 2025-2026 Ant Group
#
# SPDX-License-Identifier: Apache-2.0
#
# XXX: WARNING: this file is auto-generated.
# XXX:
# XXX: Source file: "@CONFIG_QEMU_IN@"
# XXX: Project:
# XXX: Name: @PROJECT_NAME@
# XXX: Type: @PROJECT_TYPE@
[hypervisor.qemu]
path = "@QEMUPATH@"
kernel = "@KERNELPATH_COCO@"
initrd = "@INITRDCONFIDENTIALPATH@"
# image = "@IMAGECONFIDENTIALPATH@"
machine_type = "@MACHINETYPE@"
# Enable confidential guest support.
# Toggling that setting may trigger different hardware features, ranging
# from memory encryption to both memory and CPU-state encryption and integrity.
# The Kata Containers runtime dynamically detects the available feature set and
# aims at enabling the largest possible one, returning an error if none is
# available, or none is supported by the hypervisor.
#
# Known limitations:
# * Does not work by design:
# - CPU Hotplug
# - Memory Hotplug
# - NVDIMM devices
#
# Default false
confidential_guest = true
# Enable AMD SEV-SNP confidential guests
# In case of using confidential guests on AMD hardware that supports SEV-SNP,
# the following enables SEV-SNP guests. Default true
sev_snp_guest = true
# SNP 'ID Block' and 'ID Authentication Information Structure'.
# If one of snp_id_block or snp_id_auth is specified, the other must be specified, too.
# Notice that the default SNP policy of QEMU (0x30000) is used by Kata, if not explicitly
# set via 'snp_guest_policy' option. The IDBlock contains the guest policy as field, and
# it must match the value from 'snp_guest_policy' or, if unset, the QEMU default policy.
#
# 96-byte, base64-encoded blob to provide the ID Block structure for the
# SNP_LAUNCH_FINISH command defined in the SEV-SNP firmware ABI (QEMU default: all-zero)
snp_id_block = ""
# 4096-byte, base64-encoded blob to provide the ID Authentication Information Structure
# for the SNP_LAUNCH_FINISH command defined in the SEV-SNP firmware ABI (QEMU default: all-zero)
snp_id_auth = ""
# SNP Guest Policy, the POLICY parameter to the SNP_LAUNCH_START command.
# If unset, the QEMU default policy (0x30000) will be used.
# Notice that the guest policy is enforced at VM launch, and your pod VMs
# won't start at all if the policy denys it. This will be indicated by a
# 'SNP_LAUNCH_START' error.
snp_guest_policy = 196608
# rootfs filesystem type:
# - ext4 (default)
# - xfs
# - erofs
rootfs_type = @DEFROOTFSTYPE@
# Block storage driver to be used for the VM rootfs is backed
# by a block device. This is virtio-blk-pci, virtio-blk-mmio or nvdimm
vm_rootfs_driver = "virtio-blk-pci"
# Enable running QEMU VMM as a non-root user.
# By default QEMU VMM run as root. When this is set to true, QEMU VMM process runs as
# a non-root random user. See documentation for the limitations of this mode.
rootless = false
# List of valid annotation names for the hypervisor
# Each member of the list is a regular expression, which is the base name
# of the annotation, e.g. "path" for io.katacontainers.config.hypervisor.path"
enable_annotations = @DEFENABLEANNOTATIONS_COCO@
# List of valid annotations values for the hypervisor
# Each member of the list is a path pattern as described by glob(3).
# The default if not set is empty (all annotations rejected.)
# Your distribution recommends: @QEMUVALIDHYPERVISORPATHS@
valid_hypervisor_paths = @QEMUVALIDHYPERVISORPATHS@
# Optional space-separated list of options to pass to the guest kernel.
# For example, use `kernel_params = "vsyscall=emulate"` if you are having
# trouble running pre-2.15 glibc.
#
# WARNING: - any parameter specified here will take priority over the default
# parameter value of the same name used to start the virtual machine.
# Do not set values here unless you understand the impact of doing so as you
# may stop the virtual machine from booting.
# To see the list of default parameters, enable hypervisor debug, create a
# container and look for 'default-kernel-parameters' log entries.
kernel_params = "@KERNELPARAMS@"
# Path to the firmware.
# If you want that qemu uses the default firmware leave this option empty
firmware = "@FIRMWARE_SNP_PATH@"
# Path to the firmware volume.
# firmware TDVF or OVMF can be split into FIRMWARE_VARS.fd (UEFI variables
# as configuration) and FIRMWARE_CODE.fd (UEFI program image). UEFI variables
# can be customized per each user while UEFI code is kept same.
firmware_volume = "@FIRMWARE_VOLUME_SNP_PATH@"
# Machine accelerators
# comma-separated list of machine accelerators to pass to the hypervisor.
# For example, `machine_accelerators = "nosmm,nosmbus,nosata,nopit,static-prt,nofw"`
machine_accelerators = "@MACHINEACCELERATORS@"
# Qemu seccomp sandbox feature
# comma-separated list of seccomp sandbox features to control the syscall access.
# For example, `seccompsandbox= "on,obsolete=deny,spawn=deny,resourcecontrol=deny"`
# Note: "elevateprivileges=deny" doesn't work with daemonize option, so it's removed from the seccomp sandbox
# Another note: enabling this feature may reduce performance, you may enable
# /proc/sys/net/core/bpf_jit_enable to reduce the impact. see https://man7.org/linux/man-pages/man8/bpfc.8.html
# Recommended value when enabling: "on,obsolete=deny,spawn=deny,resourcecontrol=deny"
seccompsandbox = "@DEFSECCOMPSANDBOXPARAM@"
# CPU features
# comma-separated list of cpu features to pass to the cpu
# For example, `cpu_features = "pmu=off,vmx=off"
cpu_features = "@CPUFEATURES@"
# Default number of vCPUs per SB/VM:
# unspecified or 0 --> will be set to @DEFVCPUS@
# < 0 --> will be set to the actual number of physical cores
# > 0 <= number of physical cores --> will be set to the specified number
# > number of physical cores --> will be set to the actual number of physical cores
default_vcpus = @DEFVCPUS_QEMU@
# Default maximum number of vCPUs per SB/VM:
# unspecified or == 0 --> will be set to the actual number of physical cores or to the maximum number
# of vCPUs supported by KVM if that number is exceeded
# > 0 <= number of physical cores --> will be set to the specified number
# > number of physical cores --> will be set to the actual number of physical cores or to the maximum number
# of vCPUs supported by KVM if that number is exceeded
# WARNING: Depending of the architecture, the maximum number of vCPUs supported by KVM is used when
# the actual number of physical cores is greater than it.
# WARNING: Be aware that this value impacts the virtual machine's memory footprint and CPU
# the hotplug functionality. For example, `default_maxvcpus = 240` specifies that until 240 vCPUs
# can be added to a SB/VM, but the memory footprint will be big. Another example, with
# `default_maxvcpus = 8` the memory footprint will be small, but 8 will be the maximum number of
# vCPUs supported by the SB/VM. In general, we recommend that you do not edit this variable,
# unless you know what are you doing.
# NOTICE: on arm platform with gicv2 interrupt controller, set it to 8.
default_maxvcpus = @DEFMAXVCPUS_QEMU@
# Bridges can be used to hot plug devices.
# Limitations:
# * Currently only pci bridges are supported
# * Until 30 devices per bridge can be hot plugged.
# * Until 5 PCI bridges can be cold plugged per VM.
# This limitation could be a bug in qemu or in the kernel
# Default number of bridges per SB/VM:
# unspecified or 0 --> will be set to @DEFBRIDGES@
# > 1 <= 5 --> will be set to the specified number
# > 5 --> will be set to 5
default_bridges = @DEFBRIDGES@
# Default memory size in MiB for SB/VM.
# If unspecified then it will be set @DEFMEMSZ@ MiB.
default_memory = @DEFMEMSZ@
#
# Default memory slots per SB/VM.
# If unspecified then it will be set @DEFMEMSLOTS@.
# This is will determine the times that memory will be hotadded to sandbox/VM.
memory_slots = @DEFMEMSLOTS@
# Default maximum memory in MiB per SB / VM
# unspecified or == 0 --> will be set to the actual amount of physical RAM
# > 0 <= amount of physical RAM --> will be set to the specified number
# > amount of physical RAM --> will be set to the actual amount of physical RAM
default_maxmemory = @DEFMAXMEMSZ@
# The size in MiB will be plused to max memory of hypervisor.
# It is the memory address space for the NVDIMM device.
# If set block storage driver (block_device_driver) to "nvdimm",
# should set memory_offset to the size of block device.
# Default 0
memory_offset = 0
# Specifies virtio-mem will be enabled or not.
# Please note that this option should be used with the command
# "echo 1 > /proc/sys/vm/overcommit_memory".
# Default false
enable_virtio_mem = false
# Disable block device from being used for a container's rootfs.
# In case of a storage driver like devicemapper where a container's
# root file system is backed by a block device, the block device is passed
# directly to the hypervisor for performance reasons.
# This flag prevents the block device from being passed to the hypervisor,
# virtio-fs is used instead to pass the rootfs.
disable_block_device_use = @DEFDISABLEBLOCK@
# Shared file system type:
# - virtio-fs (default)
# - virtio-fs-nydus
# - none
shared_fs = "none"
# Path to vhost-user-fs daemon.
virtio_fs_daemon = "@DEFVIRTIOFSDAEMON@"
# List of valid annotations values for the virtiofs daemon
# The default if not set is empty (all annotations rejected.)
# Your distribution recommends: @DEFVALIDVIRTIOFSDAEMONPATHS@
valid_virtio_fs_daemon_paths = @DEFVALIDVIRTIOFSDAEMONPATHS@
# Default size of DAX cache in MiB
virtio_fs_cache_size = @DEFVIRTIOFSCACHESIZE@
# Default size of virtqueues
virtio_fs_queue_size = @DEFVIRTIOFSQUEUESIZE@
# Extra args for virtiofsd daemon
#
# Format example:
# ["-o", "arg1=xxx,arg2", "-o", "hello world", "--arg3=yyy"]
# Examples:
# Set virtiofsd log level to debug : ["-o", "log_level=debug"] or ["-d"]
#
# see `virtiofsd -h` for possible options.
virtio_fs_extra_args = @DEFVIRTIOFSEXTRAARGS@
# Cache mode:
#
# - never
# Metadata, data, and pathname lookup are not cached in guest. They are
# always fetched from host and any changes are immediately pushed to host.
#
# - metadata
# Metadata and pathname lookup are cached in guest and never expire.
# Data is never cached in guest.
#
# - auto
# Metadata and pathname lookup cache expires after a configured amount of
# time (default is 1 second). Data is cached while the file is open (close
# to open consistency).
#
# - always
# Metadata, data, and pathname lookup are cached in guest and never expire.
virtio_fs_cache = "@DEFVIRTIOFSCACHE@"
# Block storage driver to be used for the hypervisor in case the container
# rootfs is backed by a block device. This is virtio-scsi, virtio-blk
# or nvdimm.
block_device_driver = "@DEFBLOCKSTORAGEDRIVER_QEMU@"
# aio is the I/O mechanism used by qemu
# Options:
#
# - threads
# Pthread based disk I/O.
#
# - native
# Native Linux I/O.
#
# - io_uring
# Linux io_uring API. This provides the fastest I/O operations on Linux, requires kernel>5.1 and
# qemu >=5.0.
block_device_aio = "@DEFBLOCKDEVICEAIO_QEMU@"
# Specifies cache-related options will be set to block devices or not.
# Default false
block_device_cache_set = false
# Specifies cache-related options for block devices.
# Denotes whether use of O_DIRECT (bypass the host page cache) is enabled.
# Default false
block_device_cache_direct = false
# Specifies cache-related options for block devices.
# Denotes whether flush requests for the device are ignored.
# Default false
block_device_cache_noflush = false
# Enable iothreads (data-plane) to be used. This causes IO to be
# handled in a separate IO thread. This is currently only implemented
# for SCSI.
#
enable_iothreads = @DEFENABLEIOTHREADS@
# Independent IOThreads enables IO to be processed in a separate thread, it is
# for QEMU hotplug device attach to iothread, like virtio-blk.
indep_iothreads = @DEFINDEPIOTHREADS@
# Enable pre allocation of VM RAM, default false
# Enabling this will result in lower container density
# as all of the memory will be allocated and locked
# This is useful when you want to reserve all the memory
# upfront or in the cases where you want memory latencies
# to be very predictable
# Default false
enable_mem_prealloc = false
# Reclaim guest freed memory.
# Enabling this will result in the VM balloon device having f_reporting=on set.
# Then the hypervisor will use it to reclaim guest freed memory.
# This is useful for reducing the amount of memory used by a VM.
# Enabling this feature may sometimes reduce the speed of memory access in
# the VM.
#
# Default false
reclaim_guest_freed_memory = false
# Enable huge pages for VM RAM, default false
# Enabling this will result in the VM memory
# being allocated using huge pages.
# This is useful when you want to use vhost-user network
# stacks within the container. This will automatically
# result in memory pre allocation
enable_hugepages = false
# Enable vhost-user storage device, default false
# Enabling this will result in some Linux reserved block type
# major range 240-254 being chosen to represent vhost-user devices.
enable_vhost_user_store = @DEFENABLEVHOSTUSERSTORE@
# The base directory specifically used for vhost-user devices.
# Its sub-path "block" is used for block devices; "block/sockets" is
# where we expect vhost-user sockets to live; "block/devices" is where
# simulated block device nodes for vhost-user devices to live.
vhost_user_store_path = "@DEFVHOSTUSERSTOREPATH@"
# Enable vIOMMU, default false
# Enabling this will result in the VM having a vIOMMU device
# This will also add the following options to the kernel's
# command line: intel_iommu=on,iommu=pt
enable_iommu = false
# Enable IOMMU_PLATFORM, default false
# Enabling this will result in the VM device having iommu_platform=on set
enable_iommu_platform = false
# List of valid annotations values for the vhost user store path
# The default if not set is empty (all annotations rejected.)
# Your distribution recommends: @DEFVALIDVHOSTUSERSTOREPATHS@
valid_vhost_user_store_paths = @DEFVALIDVHOSTUSERSTOREPATHS@
# The timeout for reconnecting on non-server spdk sockets when the remote end goes away.
# qemu will delay this many seconds and then attempt to reconnect.
# Zero disables reconnecting, and the default is zero.
vhost_user_reconnect_timeout_sec = 0
# Enable file based guest memory support. The default is an empty string which
# will disable this feature. In the case of virtio-fs, this is enabled
# automatically and '/dev/shm' is used as the backing folder.
# This option will be ignored if VM templating is enabled.
file_mem_backend = ""
# List of valid annotations values for the file_mem_backend annotation
# The default if not set is empty (all annotations rejected.)
# Your distribution recommends: @DEFVALIDFILEMEMBACKENDS@
valid_file_mem_backends = @DEFVALIDFILEMEMBACKENDS@
# -pflash can add image file to VM. The arguments of it should be in format
# of ["/path/to/flash0.img", "/path/to/flash1.img"]
pflashes = []
# This option changes the default hypervisor and kernel parameters
# to enable debug output where available. And Debug also enable the hmp socket.
#
# Default false
enable_debug = false
# Disable the customizations done in the runtime when it detects
# that it is running on top a VMM. This will result in the runtime
# behaving as it would when running on bare metal.
#
disable_nesting_checks = true
# If false and nvdimm is supported, use nvdimm device to plug guest image.
# Otherwise virtio-block device is used.
#
# nvdimm is not supported when `confidential_guest = true`.
disable_image_nvdimm = @DEFDISABLEIMAGENVDIMM@
# Before hot plugging a PCIe device, you need to add a pcie_root_port device.
# Use this parameter when using some large PCI bar devices, such as Nvidia GPU
# The value means the number of pcie_root_port
# Default 0
pcie_root_port = 0
# If vhost-net backend for virtio-net is not desired, set to true. Default is false, which trades off
# security (vhost-net runs ring0) for network I/O performance.
disable_vhost_net = false
# This option allows to add an extra HMP or QMP socket when `enable_debug = true`
#
# WARNING: Anyone with access to the extra socket can take full control of
# Qemu. This is for debugging purpose only and must *NEVER* be used in
# production.
#
# Valid values are :
# - "hmp"
# - "qmp"
# - "qmp-pretty" (same as "qmp" with pretty json formatting)
#
# If set to the empty string "", no extra monitor socket is added. This is
# the default.
#extra_monitor_socket = "hmp"
#
# Default entropy source.
# The path to a host source of entropy (including a real hardware RNG)
# /dev/urandom and /dev/random are two main options.
# Be aware that /dev/random is a blocking source of entropy. If the host
# runs out of entropy, the VMs boot time will increase leading to get startup
# timeouts.
# The source of entropy /dev/urandom is non-blocking and provides a
# generally acceptable source of entropy. It should work well for pretty much
# all practical purposes.
entropy_source = "@DEFENTROPYSOURCE@"
# List of valid annotations values for entropy_source
# The default if not set is empty (all annotations rejected.)
# Your distribution recommends: @DEFVALIDENTROPYSOURCES@
valid_entropy_sources = @DEFVALIDENTROPYSOURCES@
# Path to OCI hook binaries in the *guest rootfs*.
# This does not affect host-side hooks which must instead be added to
# the OCI spec passed to the runtime.
#
# You can create a rootfs with hooks by customizing the osbuilder scripts:
# https://github.com/kata-containers/kata-containers/tree/main/tools/osbuilder
#
# Hooks must be stored in a subdirectory of guest_hook_path according to their
# hook type, i.e. "guest_hook_path/{prestart,poststart,poststop}".
# The agent will scan these directories for executable files and add them, in
# lexicographical order, to the lifecycle of the guest container.
# Hooks are executed in the runtime namespace of the guest. See the official documentation:
# https://github.com/opencontainers/runtime-spec/blob/v1.0.1/config.md#posix-platform-hooks
# Warnings will be logged if any error is encountered while scanning for hooks,
# but it will not abort container execution.
# Recommended value when enabling: "/usr/share/oci/hooks"
guest_hook_path = ""
#
# Use rx Rate Limiter to control network I/O inbound bandwidth(size in bits/sec for SB/VM).
# In Qemu, we use classful qdiscs HTB(Hierarchy Token Bucket) to discipline traffic.
# Default 0-sized value means unlimited rate.
rx_rate_limiter_max_rate = 0
# Use tx Rate Limiter to control network I/O outbound bandwidth(size in bits/sec for SB/VM).
# In Qemu, we use classful qdiscs HTB(Hierarchy Token Bucket) and ifb(Intermediate Functional Block)
# to discipline traffic.
# Default 0-sized value means unlimited rate.
tx_rate_limiter_max_rate = 0
# Set where to save the guest memory dump file.
# If set, when GUEST_PANICKED event occurred,
# guest memeory will be dumped to host filesystem under guest_memory_dump_path,
# This directory will be created automatically if it does not exist.
#
# The dumped file(also called vmcore) can be processed with crash or gdb.
#
# WARNING:
# Dump guest's memory can take very long depending on the amount of guest memory
# and use much disk space.
# Recommended value when enabling: "/var/crash/kata"
guest_memory_dump_path = ""
# If enable paging.
# Basically, if you want to use "gdb" rather than "crash",
# or need the guest-virtual addresses in the ELF vmcore,
# then you should enable paging.
#
# See: https://www.qemu.org/docs/master/qemu-qmp-ref.html#Dump-guest-memory for details
guest_memory_dump_paging = false
# Enable swap in the guest. Default false.
# When enable_guest_swap is enabled, insert a raw file to the guest as the swap device
# if the swappiness of a container (set by annotation "io.katacontainers.container.resource.swappiness")
# is bigger than 0.
# The size of the swap device should be
# swap_in_bytes (set by annotation "io.katacontainers.container.resource.swap_in_bytes") - memory_limit_in_bytes.
# If swap_in_bytes is not set, the size should be memory_limit_in_bytes.
# If swap_in_bytes and memory_limit_in_bytes is not set, the size should
# be default_memory.
enable_guest_swap = false
# use legacy serial for guest console if available and implemented for architecture. Default false
use_legacy_serial = false
# disable applying SELinux on the VMM process (default false)
disable_selinux = @DEFDISABLESELINUX@
# disable applying SELinux on the container process
# If set to false, the type `container_t` is applied to the container process by default.
# Note: To enable guest SELinux, the guest rootfs must be CentOS that is created and built
# with `SELINUX=yes`.
# (default: true)
disable_guest_selinux = @DEFDISABLEGUESTSELINUX@
[factory]
# VM templating support. Once enabled, new VMs are created from template
# using vm cloning. They will share the same initial kernel, initramfs and
# agent memory by mapping it readonly. It helps speeding up new container
# creation and saves a lot of memory if there are many kata containers running
# on the same host.
#
# When disabled, new VMs are created from scratch.
#
# Note: Requires "initrd=" to be set ("image=" is not supported).
#
# Default false
enable_template = false
# Specifies the path of template.
#
# Default "/run/vc/vm/template"
template_path = "/run/vc/vm/template"
# The number of caches of VMCache:
# unspecified or == 0 --> VMCache is disabled
# > 0 --> will be set to the specified number
#
# VMCache is a function that creates VMs as caches before using it.
# It helps speed up new container creation.
# The function consists of a server and some clients communicating
# through Unix socket. The protocol is gRPC in protocols/cache/cache.proto.
# The VMCache server will create some VMs and cache them by factory cache.
# It will convert the VM to gRPC format and transport it when gets
# requestion from clients.
# Factory grpccache is the VMCache client. It will request gRPC format
# VM and convert it back to a VM. If VMCache function is enabled,
# kata-runtime will request VM from factory grpccache when it creates
# a new sandbox.
#
# Default 0
vm_cache_number = 0
# Specify the address of the Unix socket that is used by VMCache.
#
# Default /var/run/kata-containers/cache.sock
vm_cache_endpoint = "/var/run/kata-containers/cache.sock"
[agent.@PROJECT_TYPE@]
# If enabled, make the agent display debug-level messages.
# (default: disabled)
enable_debug = false
# Enable agent tracing.
#
# If enabled, the agent will generate OpenTelemetry trace spans.
#
# Notes:
#
# - If the runtime also has tracing enabled, the agent spans will be
# associated with the appropriate runtime parent span.
# - If enabled, the runtime will wait for the container to shutdown,
# increasing the container shutdown time slightly.
#
# (default: disabled)
enable_tracing = false
# Comma separated list of kernel modules and their parameters.
# These modules will be loaded in the guest kernel using modprobe(8).
# The following example can be used to load two kernel modules with parameters
# - kernel_modules=["e1000e InterruptThrottleRate=3000,3000,3000 EEE=1", "i915 enable_ppgtt=0"]
# The first word is considered as the module name and the rest as its parameters.
# Container will not be started when:
# * A kernel module is specified and the modprobe command is not installed in the guest
# or it fails loading the module.
# * The module is not available in the guest or it doesn't met the guest kernel
# requirements, like architecture and version.
#
kernel_modules = []
# Enable debug console.
# If enabled, user can connect guest OS running inside hypervisor
# through "kata-runtime exec <sandbox-id>" command
debug_console_enabled = false
# Agent dial timeout in millisecond.
# (default: 10)
dial_timeout_ms = 10
# Agent reconnect timeout in millisecond.
# Retry times = reconnect_timeout_ms / dial_timeout_ms (default: 300)
# If you find pod cannot connect to the agent when starting, please
# consider increasing this value to increase the retry times.
# You'd better not change the value of dial_timeout_ms, unless you have an
# idea of what you are doing.
# (default: 3000)
reconnect_timeout_ms = 3000
# Create Container Request Timeout
# This timeout value is used to set the maximum duration for the agent to process a CreateContainerRequest.
# It's also used to ensure that workloads, especially those involving large image pulls within the guest,
# have sufficient time to complete.
#
# Effective Timeout Determination:
# The effective timeout for a CreateContainerRequest is determined by taking the minimum of the following two values:
# - create_container_timeout: The timeout value configured for creating containers (default: 30,000 milliseconds).
# - runtime-request-timeout: The timeout value specified in the Kubelet configuration described as the link below:
# (https://kubernetes.io/docs/reference/command-line-tools-reference/kubelet/#:~:text=runtime%2Drequest%2Dtimeout)
# Defaults to @DEFCREATECONTAINERTIMEOUT_COCO@ second(s)
create_container_timeout = @DEFCREATECONTAINERTIMEOUT_COCO@
[runtime]
# If enabled, the runtime will log additional debug messages to the
# system log
# (default: disabled)
enable_debug = false
#
# Internetworking model
# Determines how the VM should be connected to the
# the container network interface
# Options:
#
# - macvtap
# Used when the Container network interface can be bridged using
# macvtap.
#
# - none
# Used when customize network. Only creates a tap device. No veth pair.
#
# - tcfilter
# Uses tc filter rules to redirect traffic from the network interface
# provided by plugin to a tap interface connected to the VM.
#
internetworking_model="@DEFNETWORKMODEL_QEMU@"
name="@RUNTIMENAME@"
hypervisor_name="@HYPERVISOR_QEMU@"
agent_name="@PROJECT_TYPE@"
# disable guest seccomp
# Determines whether container seccomp profiles are passed to the virtual
# machine and applied by the kata agent. If set to true, seccomp is not applied
# within the guest
# (default: true)
disable_guest_seccomp = @DEFDISABLEGUESTSECCOMP@
# vCPUs pinning settings
# if enabled, each vCPU thread will be scheduled to a fixed CPU
# qualified condition: num(vCPU threads) == num(CPUs in sandbox's CPUSet)
enable_vcpus_pinning = false
# Apply a custom SELinux security policy to the container process inside the VM.
# This is used when you want to apply a type other than the default `container_t`,
# so general users should not uncomment and apply it.
# (format: "user:role:type")
# Note: You cannot specify MCS policy with the label because the sensitivity levels and
# categories are determined automatically by high-level container runtimes such as containerd.
guest_selinux_label = "@DEFGUESTSELINUXLABEL@"
# If enabled, the runtime will create opentracing.io traces and spans.
# (See https://www.jaegertracing.io/docs/getting-started).
# (default: disabled)
enable_tracing = false
# Set the full url to the Jaeger HTTP Thrift collector.
# The default if not set will be "http://localhost:14268/api/traces"
jaeger_endpoint = ""
# Sets the username to be used if basic auth is required for Jaeger.
jaeger_user = ""
# Sets the password to be used if basic auth is required for Jaeger.
jaeger_password = ""
# If enabled, the runtime will not create a network namespace for shim and hypervisor processes.
# This option may have some potential impacts to your host. It should only be used when you know what you're doing.
# `disable_new_netns` conflicts with `internetworking_model=tcfilter` and `internetworking_model=macvtap`. It works only
# with `internetworking_model=none`. The tap device will be in the host network namespace and can connect to a bridge
# (like OVS) directly.
# (default: false)
disable_new_netns = false
# if enabled, the runtime will add all the kata processes inside one dedicated cgroup.
# The container cgroups in the host are not created, just one single cgroup per sandbox.
# The runtime caller is free to restrict or collect cgroup stats of the overall Kata sandbox.
# The sandbox cgroup path is the parent cgroup of a container with the PodSandbox annotation.
# The sandbox cgroup is constrained if there is no container type annotation.
# See: https://pkg.go.dev/github.com/kata-containers/kata-containers/src/runtime/virtcontainers#ContainerType
sandbox_cgroup_only = @DEFSANDBOXCGROUPONLY_QEMU@
# If enabled, the runtime will attempt to determine appropriate sandbox size (memory, CPU) before booting the virtual machine. In
# this case, the runtime will not dynamically update the amount of memory and CPU in the virtual machine. This is generally helpful
# when a hardware architecture or hypervisor solutions is utilized which does not support CPU and/or memory hotplug.
# Compatibility for determining appropriate sandbox (VM) size:
# - When running with pods, sandbox sizing information will only be available if using Kubernetes >= 1.23 and containerd >= 1.6. CRI-O
# does not yet support sandbox sizing annotations.
# - When running single containers using a tool like ctr, container sizing information will be available.
static_sandbox_resource_mgmt = @DEFSTATICRESOURCEMGMT_COCO@
# If specified, sandbox_bind_mounts identifieds host paths to be mounted (ro) into the sandboxes shared path.
# This is only valid if filesystem sharing is utilized. The provided path(s) will be bindmounted into the shared fs directory.
# If defaults are utilized, these mounts should be available in the guest at `/run/kata-containers/shared/containers/sandbox-mounts`
# These will not be exposed to the container workloads, and are only provided for potential guest services.
sandbox_bind_mounts = @DEFBINDMOUNTS@
# VFIO Mode
# Determines how VFIO devices should be be presented to the container.
# Options:
#
# - vfio
# Matches behaviour of OCI runtimes (e.g. runc) as much as
# possible. VFIO devices will appear in the container as VFIO
# character devices under /dev/vfio. The exact names may differ
# from the host (they need to match the VM's IOMMU group numbers
# rather than the host's)
#
# - guest-kernel
# This is a Kata-specific behaviour that's useful in certain cases.
# The VFIO device is managed by whatever driver in the VM kernel
# claims it. This means it will appear as one or more device nodes
# or network interfaces depending on the nature of the device.
# Using this mode requires specially built workloads that know how
# to locate the relevant device interfaces within the VM.
#
vfio_mode = "@DEFVFIOMODE@"
# If enabled, the runtime will not create Kubernetes emptyDir mounts on the guest filesystem. Instead, emptyDir mounts will
# be created on the host and shared via virtio-fs. This is potentially slower, but allows sharing of files from host to guest.
disable_guest_empty_dir = @DEFDISABLEGUESTEMPTYDIR@
# Enabled experimental feature list, format: ["a", "b"].
# Experimental features are features not stable enough for production,
# they may break compatibility, and are prepared for a big version bump.
# Supported experimental features:
# for example:
# experimental=["force_guest_pull"]
# which is for enable force_guest_pull mode in CoCo scenarios.
# (default: [])
experimental = @DEFAULTEXPFEATURES@
# If enabled, user can run pprof tools with shim v2 process through kata-monitor.
# (default: false)
enable_pprof = false
# Base directory of directly attachable network config.
# Network devices for VM-based containers are allowed to be placed in the
# host netns to eliminate as many hops as possible, which is what we
# called a "Directly Attachable Network". The config, set by special CNI
# plugins, is used to tell the Kata containers what devices are attached
# to the hypervisor.
# (default: /run/kata-containers/dans)
dan_conf = "@DEFDANCONF@"
# pod_resource_api_sock specifies the unix socket for the Kubelet's
# PodResource API endpoint. If empty, kubernetes based cold plug
# will not be attempted. In order for this feature to work, the
# KubeletPodResourcesGet featureGate must be enabled in Kubelet,
# if using Kubelet older than 1.34.
#
# The pod resource API's socket is relative to the Kubelet's root-dir,
# which is defined by the cluster admin, and its location is:
# ${KubeletRootDir}/pod-resources/kubelet.sock
#
# cold_plug_vfio(see hypervisor config) acts as a feature gate:
# cold_plug_vfio = no_port (default) => no cold plug
# cold_plug_vfio != no_port AND pod_resource_api_sock = "" => need
# explicit CDI annotation for cold plug (applies mainly
# to non-k8s cases)
# cold_plug_vfio != no_port AND pod_resource_api_sock != "" => kubelet
# based cold plug.
pod_resource_api_sock = "@DEFPODRESOURCEAPISOCK@"

View File

@@ -1,746 +0,0 @@
# Copyright (c) 2017-2019 Intel Corporation
# Copyright (c) 2021 Adobe Inc.
# Copyright (c) 2025-2026 Ant Group
#
# SPDX-License-Identifier: Apache-2.0
#
# XXX: WARNING: this file is auto-generated.
# XXX:
# XXX: Source file: "@CONFIG_QEMU_IN@"
# XXX: Project:
# XXX: Name: @PROJECT_NAME@
# XXX: Type: @PROJECT_TYPE@
[hypervisor.qemu]
path = "@QEMUPATH@"
kernel = "@KERNELPATH_COCO@"
image = "@IMAGECONFIDENTIALPATH@"
# initrd = "@INITRDPATH@"
machine_type = "@MACHINETYPE@"
tdx_quote_generation_service_socket_port = @QEMUTDXQUOTEGENERATIONSERVICESOCKETPORT@
# rootfs filesystem type:
# - ext4 (default)
# - xfs
# - erofs
rootfs_type = @DEFROOTFSTYPE@
# Block storage driver to be used for the VM rootfs is backed
# by a block device. This is virtio-blk-pci, virtio-blk-mmio or nvdimm
vm_rootfs_driver = "virtio-blk-pci"
# Enable confidential guest support.
# Toggling that setting may trigger different hardware features, ranging
# from memory encryption to both memory and CPU-state encryption and integrity.
# The Kata Containers runtime dynamically detects the available feature set and
# aims at enabling the largest possible one, returning an error if none is
# available, or none is supported by the hypervisor.
#
# Known limitations:
# * Does not work by design:
# - CPU Hotplug
# - Memory Hotplug
# - NVDIMM devices
#
# Default false
confidential_guest = true
# Enable running QEMU VMM as a non-root user.
# By default QEMU VMM run as root. When this is set to true, QEMU VMM process runs as
# a non-root random user. See documentation for the limitations of this mode.
rootless = false
# List of valid annotation names for the hypervisor
# Each member of the list is a regular expression, which is the base name
# of the annotation, e.g. "path" for io.katacontainers.config.hypervisor.path"
enable_annotations = @DEFENABLEANNOTATIONS_COCO@
# List of valid annotations values for the hypervisor
# Each member of the list is a path pattern as described by glob(3).
# The default if not set is empty (all annotations rejected.)
# Your distribution recommends: @QEMUVALIDHYPERVISORPATHS@
valid_hypervisor_paths = @QEMUVALIDHYPERVISORPATHS@
# Optional space-separated list of options to pass to the guest kernel.
# For example, use `kernel_params = "vsyscall=emulate"` if you are having
# trouble running pre-2.15 glibc.
#
# WARNING: - any parameter specified here will take priority over the default
# parameter value of the same name used to start the virtual machine.
# Do not set values here unless you understand the impact of doing so as you
# may stop the virtual machine from booting.
# To see the list of default parameters, enable hypervisor debug, create a
# container and look for 'default-kernel-parameters' log entries.
kernel_params = "@KERNELTDXPARAMS@"
# Path to the firmware.
# If you want that qemu uses the default firmware leave this option empty
firmware = "@FIRMWARETDXPATH@"
# Path to the firmware volume.
# firmware TDVF or OVMF can be split into FIRMWARE_VARS.fd (UEFI variables
# as configuration) and FIRMWARE_CODE.fd (UEFI program image). UEFI variables
# can be customized per each user while UEFI code is kept same.
firmware_volume = "@FIRMWAREVOLUMEPATH@"
# Machine accelerators
# comma-separated list of machine accelerators to pass to the hypervisor.
# For example, `machine_accelerators = "nosmm,nosmbus,nosata,nopit,static-prt,nofw"`
machine_accelerators = "@MACHINEACCELERATORS@"
# Qemu seccomp sandbox feature
# comma-separated list of seccomp sandbox features to control the syscall access.
# For example, `seccompsandbox= "on,obsolete=deny,spawn=deny,resourcecontrol=deny"`
# Note: "elevateprivileges=deny" doesn't work with daemonize option, so it's removed from the seccomp sandbox
# Another note: enabling this feature may reduce performance, you may enable
# /proc/sys/net/core/bpf_jit_enable to reduce the impact. see https://man7.org/linux/man-pages/man8/bpfc.8.html
# Recommended value when enabling: "on,obsolete=deny,spawn=deny,resourcecontrol=deny"
seccompsandbox = "@DEFSECCOMPSANDBOXPARAM@"
# CPU features
# comma-separated list of cpu features to pass to the cpu
# For example, `cpu_features = "pmu=off,vmx=off"
cpu_features = "@CPUFEATURES@"
# Default number of vCPUs per SB/VM:
# unspecified or 0 --> will be set to @DEFVCPUS@
# < 0 --> will be set to the actual number of physical cores
# > 0 <= number of physical cores --> will be set to the specified number
# > number of physical cores --> will be set to the actual number of physical cores
default_vcpus = 1
# Default maximum number of vCPUs per SB/VM:
# unspecified or == 0 --> will be set to the actual number of physical cores or to the maximum number
# of vCPUs supported by KVM if that number is exceeded
# > 0 <= number of physical cores --> will be set to the specified number
# > number of physical cores --> will be set to the actual number of physical cores or to the maximum number
# of vCPUs supported by KVM if that number is exceeded
# WARNING: Depending of the architecture, the maximum number of vCPUs supported by KVM is used when
# the actual number of physical cores is greater than it.
# WARNING: Be aware that this value impacts the virtual machine's memory footprint and CPU
# the hotplug functionality. For example, `default_maxvcpus = 240` specifies that until 240 vCPUs
# can be added to a SB/VM, but the memory footprint will be big. Another example, with
# `default_maxvcpus = 8` the memory footprint will be small, but 8 will be the maximum number of
# vCPUs supported by the SB/VM. In general, we recommend that you do not edit this variable,
# unless you know what are you doing.
# NOTICE: on arm platform with gicv2 interrupt controller, set it to 8.
default_maxvcpus = @DEFMAXVCPUS@
# Bridges can be used to hot plug devices.
# Limitations:
# * Currently only pci bridges are supported
# * Until 30 devices per bridge can be hot plugged.
# * Until 5 PCI bridges can be cold plugged per VM.
# This limitation could be a bug in qemu or in the kernel
# Default number of bridges per SB/VM:
# unspecified or 0 --> will be set to @DEFBRIDGES@
# > 1 <= 5 --> will be set to the specified number
# > 5 --> will be set to 5
default_bridges = @DEFBRIDGES@
# Default memory size in MiB for SB/VM.
# If unspecified then it will be set @DEFMEMSZ@ MiB.
default_memory = @DEFMEMSZ@
#
# Default memory slots per SB/VM.
# If unspecified then it will be set @DEFMEMSLOTS@.
# This is will determine the times that memory will be hotadded to sandbox/VM.
memory_slots = @DEFMEMSLOTS@
# Default maximum memory in MiB per SB / VM
# unspecified or == 0 --> will be set to the actual amount of physical RAM
# > 0 <= amount of physical RAM --> will be set to the specified number
# > amount of physical RAM --> will be set to the actual amount of physical RAM
default_maxmemory = @DEFMAXMEMSZ@
# The size in MiB will be plused to max memory of hypervisor.
# It is the memory address space for the NVDIMM device.
# If set block storage driver (block_device_driver) to "nvdimm",
# should set memory_offset to the size of block device.
# Default 0
memory_offset = 0
# Specifies virtio-mem will be enabled or not.
# Please note that this option should be used with the command
# "echo 1 > /proc/sys/vm/overcommit_memory".
# Default false
enable_virtio_mem = false
# Disable block device from being used for a container's rootfs.
# In case of a storage driver like devicemapper where a container's
# root file system is backed by a block device, the block device is passed
# directly to the hypervisor for performance reasons.
# This flag prevents the block device from being passed to the hypervisor,
# virtio-fs is used instead to pass the rootfs.
disable_block_device_use = @DEFDISABLEBLOCK@
# Shared file system type:
# - virtio-fs (default)
# - virtio-fs-nydus
# - none
shared_fs = "@DEFSHAREDFS_QEMU_TDX_VIRTIOFS@"
# Path to vhost-user-fs daemon.
virtio_fs_daemon = "@DEFVIRTIOFSDAEMON@"
# List of valid annotations values for the virtiofs daemon
# The default if not set is empty (all annotations rejected.)
# Your distribution recommends: @DEFVALIDVIRTIOFSDAEMONPATHS@
valid_virtio_fs_daemon_paths = @DEFVALIDVIRTIOFSDAEMONPATHS@
# Default size of DAX cache in MiB
virtio_fs_cache_size = @DEFVIRTIOFSCACHESIZE@
# Default size of virtqueues
virtio_fs_queue_size = @DEFVIRTIOFSQUEUESIZE@
# Extra args for virtiofsd daemon
#
# Format example:
# ["-o", "arg1=xxx,arg2", "-o", "hello world", "--arg3=yyy"]
# Examples:
# Set virtiofsd log level to debug : ["-o", "log_level=debug"] or ["-d"]
#
# see `virtiofsd -h` for possible options.
virtio_fs_extra_args = @DEFVIRTIOFSEXTRAARGS@
# Cache mode:
#
# - never
# Metadata, data, and pathname lookup are not cached in guest. They are
# always fetched from host and any changes are immediately pushed to host.
#
# - metadata
# Metadata and pathname lookup are cached in guest and never expire.
# Data is never cached in guest.
#
# - auto
# Metadata and pathname lookup cache expires after a configured amount of
# time (default is 1 second). Data is cached while the file is open (close
# to open consistency).
#
# - always
# Metadata, data, and pathname lookup are cached in guest and never expire.
virtio_fs_cache = "@DEFVIRTIOFSCACHE@"
# Block storage driver to be used for the hypervisor in case the container
# rootfs is backed by a block device. This is virtio-scsi, virtio-blk
# or nvdimm.
block_device_driver = "@DEFBLOCKSTORAGEDRIVER_QEMU@"
# aio is the I/O mechanism used by qemu
# Options:
#
# - threads
# Pthread based disk I/O.
#
# - native
# Native Linux I/O.
#
# - io_uring
# Linux io_uring API. This provides the fastest I/O operations on Linux, requires kernel>5.1 and
# qemu >=5.0.
block_device_aio = "@DEFBLOCKDEVICEAIO_QEMU@"
# Specifies cache-related options will be set to block devices or not.
# Default false
block_device_cache_set = false
# Specifies cache-related options for block devices.
# Denotes whether use of O_DIRECT (bypass the host page cache) is enabled.
# Default false
block_device_cache_direct = false
# Specifies cache-related options for block devices.
# Denotes whether flush requests for the device are ignored.
# Default false
block_device_cache_noflush = false
# Enable iothreads (data-plane) to be used. This causes IO to be
# handled in a separate IO thread. This is currently implemented
# for virtio-scsi and virtio-blk.
#
enable_iothreads = @DEFENABLEIOTHREADS@
# Independent IOThreads enables IO to be processed in a separate thread, it is
# for QEMU hotplug device attach to iothread, like virtio-blk.
indep_iothreads = @DEFINDEPIOTHREADS@
# Enable pre allocation of VM RAM, default false
# Enabling this will result in lower container density
# as all of the memory will be allocated and locked
# This is useful when you want to reserve all the memory
# upfront or in the cases where you want memory latencies
# to be very predictable
# Default false
enable_mem_prealloc = false
# Reclaim guest freed memory.
# Enabling this will result in the VM balloon device having f_reporting=on set.
# Then the hypervisor will use it to reclaim guest freed memory.
# This is useful for reducing the amount of memory used by a VM.
# Enabling this feature may sometimes reduce the speed of memory access in
# the VM.
#
# Default false
reclaim_guest_freed_memory = false
# Enable huge pages for VM RAM, default false
# Enabling this will result in the VM memory
# being allocated using huge pages.
# This is useful when you want to use vhost-user network
# stacks within the container. This will automatically
# result in memory pre allocation
enable_hugepages = false
# Enable vhost-user storage device, default false
# Enabling this will result in some Linux reserved block type
# major range 240-254 being chosen to represent vhost-user devices.
enable_vhost_user_store = @DEFENABLEVHOSTUSERSTORE@
# The base directory specifically used for vhost-user devices.
# Its sub-path "block" is used for block devices; "block/sockets" is
# where we expect vhost-user sockets to live; "block/devices" is where
# simulated block device nodes for vhost-user devices to live.
vhost_user_store_path = "@DEFVHOSTUSERSTOREPATH@"
# Enable vIOMMU, default false
# Enabling this will result in the VM having a vIOMMU device
# This will also add the following options to the kernel's
# command line: intel_iommu=on,iommu=pt
enable_iommu = false
# Enable IOMMU_PLATFORM, default false
# Enabling this will result in the VM device having iommu_platform=on set
enable_iommu_platform = false
# List of valid annotations values for the vhost user store path
# The default if not set is empty (all annotations rejected.)
# Your distribution recommends: @DEFVALIDVHOSTUSERSTOREPATHS@
valid_vhost_user_store_paths = @DEFVALIDVHOSTUSERSTOREPATHS@
# The timeout for reconnecting on non-server spdk sockets when the remote end goes away.
# qemu will delay this many seconds and then attempt to reconnect.
# Zero disables reconnecting, and the default is zero.
vhost_user_reconnect_timeout_sec = 0
# Enable file based guest memory support. The default is an empty string which
# will disable this feature. In the case of virtio-fs, this is enabled
# automatically and '/dev/shm' is used as the backing folder.
# This option will be ignored if VM templating is enabled.
file_mem_backend = "@DEFFILEMEMBACKEND@"
# List of valid annotations values for the file_mem_backend annotation
# The default if not set is empty (all annotations rejected.)
# Your distribution recommends: @DEFVALIDFILEMEMBACKENDS@
valid_file_mem_backends = @DEFVALIDFILEMEMBACKENDS@
# -pflash can add image file to VM. The arguments of it should be in format
# of ["/path/to/flash0.img", "/path/to/flash1.img"]
pflashes = []
# This option changes the default hypervisor and kernel parameters
# to enable debug output where available. And Debug also enable the hmp socket.
#
# Default false
enable_debug = false
# This option allows to add an extra HMP or QMP socket when `enable_debug = true`
#
# WARNING: Anyone with access to the extra socket can take full control of
# Qemu. This is for debugging purpose only and must *NEVER* be used in
# production.
#
# Valid values are :
# - "hmp"
# - "qmp"
# - "qmp-pretty" (same as "qmp" with pretty json formatting)
#
# If set to the empty string "", no extra monitor socket is added. This is
# the default.
extra_monitor_socket = ""
# Disable the customizations done in the runtime when it detects
# that it is running on top a VMM. This will result in the runtime
# behaving as it would when running on bare metal.
#
disable_nesting_checks = false
# If false and nvdimm is supported, use nvdimm device to plug guest image.
# Otherwise virtio-block device is used.
#
# nvdimm is not supported when `confidential_guest = true`.
disable_image_nvdimm = @DEFDISABLEIMAGENVDIMM@
# Before hot plugging a PCIe device, you need to add a pcie_root_port device.
# Use this parameter when using some large PCI bar devices, such as Nvidia GPU
# The value means the number of pcie_root_port
# Default 0
pcie_root_port = 0
# If vhost-net backend for virtio-net is not desired, set to true. Default is false, which trades off
# security (vhost-net runs ring0) for network I/O performance.
disable_vhost_net = false
#
# Default entropy source.
# The path to a host source of entropy (including a real hardware RNG)
# /dev/urandom and /dev/random are two main options.
# Be aware that /dev/random is a blocking source of entropy. If the host
# runs out of entropy, the VMs boot time will increase leading to get startup
# timeouts.
# The source of entropy /dev/urandom is non-blocking and provides a
# generally acceptable source of entropy. It should work well for pretty much
# all practical purposes.
entropy_source = "@DEFENTROPYSOURCE@"
# List of valid annotations values for entropy_source
# The default if not set is empty (all annotations rejected.)
# Your distribution recommends: @DEFVALIDENTROPYSOURCES@
valid_entropy_sources = @DEFVALIDENTROPYSOURCES@
# Path to OCI hook binaries in the *guest rootfs*.
# This does not affect host-side hooks which must instead be added to
# the OCI spec passed to the runtime.
#
# You can create a rootfs with hooks by customizing the osbuilder scripts:
# https://github.com/kata-containers/kata-containers/tree/main/tools/osbuilder
#
# Hooks must be stored in a subdirectory of guest_hook_path according to their
# hook type, i.e. "guest_hook_path/{prestart,poststart,poststop}".
# The agent will scan these directories for executable files and add them, in
# lexicographical order, to the lifecycle of the guest container.
# Hooks are executed in the runtime namespace of the guest. See the official documentation:
# https://github.com/opencontainers/runtime-spec/blob/v1.0.1/config.md#posix-platform-hooks
# Warnings will be logged if any error is encountered while scanning for hooks,
# but it will not abort container execution.
# Recommended value when enabling: "/usr/share/oci/hooks"
guest_hook_path = ""
#
# Use rx Rate Limiter to control network I/O inbound bandwidth(size in bits/sec for SB/VM).
# In Qemu, we use classful qdiscs HTB(Hierarchy Token Bucket) to discipline traffic.
# Default 0-sized value means unlimited rate.
rx_rate_limiter_max_rate = 0
# Use tx Rate Limiter to control network I/O outbound bandwidth(size in bits/sec for SB/VM).
# In Qemu, we use classful qdiscs HTB(Hierarchy Token Bucket) and ifb(Intermediate Functional Block)
# to discipline traffic.
# Default 0-sized value means unlimited rate.
tx_rate_limiter_max_rate = 0
# Set where to save the guest memory dump file.
# If set, when GUEST_PANICKED event occurred,
# guest memeory will be dumped to host filesystem under guest_memory_dump_path,
# This directory will be created automatically if it does not exist.
#
# The dumped file(also called vmcore) can be processed with crash or gdb.
#
# WARNING:
# Dump guest's memory can take very long depending on the amount of guest memory
# and use much disk space.
# Recommended value when enabling: "/var/crash/kata"
guest_memory_dump_path = ""
# If enable paging.
# Basically, if you want to use "gdb" rather than "crash",
# or need the guest-virtual addresses in the ELF vmcore,
# then you should enable paging.
#
# See: https://www.qemu.org/docs/master/qemu-qmp-ref.html#Dump-guest-memory for details
guest_memory_dump_paging = false
# Enable swap in the guest. Default false.
# When enable_guest_swap is enabled, insert a raw file to the guest as the swap device
# if the swappiness of a container (set by annotation "io.katacontainers.container.resource.swappiness")
# is bigger than 0.
# The size of the swap device should be
# swap_in_bytes (set by annotation "io.katacontainers.container.resource.swap_in_bytes") - memory_limit_in_bytes.
# If swap_in_bytes is not set, the size should be memory_limit_in_bytes.
# If swap_in_bytes and memory_limit_in_bytes is not set, the size should
# be default_memory.
enable_guest_swap = false
# use legacy serial for guest console if available and implemented for architecture. Default false
use_legacy_serial = false
# disable applying SELinux on the VMM process (default false)
disable_selinux = @DEFDISABLESELINUX@
# disable applying SELinux on the container process
# If set to false, the type `container_t` is applied to the container process by default.
# Note: To enable guest SELinux, the guest rootfs must be CentOS that is created and built
# with `SELINUX=yes`.
# (default: true)
disable_guest_selinux = @DEFDISABLEGUESTSELINUX@
[factory]
# VM templating support. Once enabled, new VMs are created from template
# using vm cloning. They will share the same initial kernel, initramfs and
# agent memory by mapping it readonly. It helps speeding up new container
# creation and saves a lot of memory if there are many kata containers running
# on the same host.
#
# When disabled, new VMs are created from scratch.
#
# Note: Requires "initrd=" to be set ("image=" is not supported).
#
# Default false
enable_template = false
# Specifies the path of template.
#
# Default "/run/vc/vm/template"
template_path = "/run/vc/vm/template"
# The number of caches of VMCache:
# unspecified or == 0 --> VMCache is disabled
# > 0 --> will be set to the specified number
#
# VMCache is a function that creates VMs as caches before using it.
# It helps speed up new container creation.
# The function consists of a server and some clients communicating
# through Unix socket. The protocol is gRPC in protocols/cache/cache.proto.
# The VMCache server will create some VMs and cache them by factory cache.
# It will convert the VM to gRPC format and transport it when gets
# requestion from clients.
# Factory grpccache is the VMCache client. It will request gRPC format
# VM and convert it back to a VM. If VMCache function is enabled,
# kata-runtime will request VM from factory grpccache when it creates
# a new sandbox.
#
# Default 0
vm_cache_number = 0
# Specify the address of the Unix socket that is used by VMCache.
#
# Default /var/run/kata-containers/cache.sock
vm_cache_endpoint = "/var/run/kata-containers/cache.sock"
[agent.@PROJECT_TYPE@]
# If enabled, make the agent display debug-level messages.
# (default: disabled)
enable_debug = false
# Enable agent tracing.
#
# If enabled, the agent will generate OpenTelemetry trace spans.
#
# Notes:
#
# - If the runtime also has tracing enabled, the agent spans will be
# associated with the appropriate runtime parent span.
# - If enabled, the runtime will wait for the container to shutdown,
# increasing the container shutdown time slightly.
#
# (default: disabled)
enable_tracing = false
# Comma separated list of kernel modules and their parameters.
# These modules will be loaded in the guest kernel using modprobe(8).
# The following example can be used to load two kernel modules with parameters
# - kernel_modules=["e1000e InterruptThrottleRate=3000,3000,3000 EEE=1", "i915 enable_ppgtt=0"]
# The first word is considered as the module name and the rest as its parameters.
# Container will not be started when:
# * A kernel module is specified and the modprobe command is not installed in the guest
# or it fails loading the module.
# * The module is not available in the guest or it doesn't met the guest kernel
# requirements, like architecture and version.
#
kernel_modules = []
# Enable debug console.
# If enabled, user can connect guest OS running inside hypervisor
# through "kata-runtime exec <sandbox-id>" command
debug_console_enabled = false
# Agent dial timeout in millisecond.
# (default: 10)
dial_timeout_ms = 10
# Agent reconnect timeout in millisecond.
# Retry times = reconnect_timeout_ms / dial_timeout_ms (default: 300)
# If you find pod cannot connect to the agent when starting, please
# consider increasing this value to increase the retry times.
# You'd better not change the value of dial_timeout_ms, unless you have an
# idea of what you are doing.
# (default: 3000)
reconnect_timeout_ms = 3000
# Create Container Request Timeout
# This timeout value is used to set the maximum duration for the agent to process a CreateContainerRequest.
# It's also used to ensure that workloads, especially those involving large image pulls within the guest,
# have sufficient time to complete.
#
# Effective Timeout Determination:
# The effective timeout for a CreateContainerRequest is determined by taking the minimum of the following two values:
# - create_container_timeout: The timeout value configured for creating containers (default: 30,000 milliseconds).
# - runtime-request-timeout: The timeout value specified in the Kubelet configuration described as the link below:
# (https://kubernetes.io/docs/reference/command-line-tools-reference/kubelet/#:~:text=runtime%2Drequest%2Dtimeout)
# Defaults to @DEFCREATECONTAINERTIMEOUT_COCO@ second(s)
create_container_timeout = @DEFCREATECONTAINERTIMEOUT_COCO@
[runtime]
# If enabled, the runtime will log additional debug messages to the
# system log
# (default: disabled)
enable_debug = false
#
# Internetworking model
# Determines how the VM should be connected to the
# the container network interface
# Options:
#
# - macvtap
# Used when the Container network interface can be bridged using
# macvtap.
#
# - none
# Used when customize network. Only creates a tap device. No veth pair.
#
# - tcfilter
# Uses tc filter rules to redirect traffic from the network interface
# provided by plugin to a tap interface connected to the VM.
#
internetworking_model = "@DEFNETWORKMODEL_QEMU@"
name="@RUNTIMENAME@"
hypervisor_name="@HYPERVISOR_QEMU@"
agent_name="@PROJECT_TYPE@"
# disable guest seccomp
# Determines whether container seccomp profiles are passed to the virtual
# machine and applied by the kata agent. If set to true, seccomp is not applied
# within the guest
# (default: true)
disable_guest_seccomp = @DEFDISABLEGUESTSECCOMP@
# vCPUs pinning settings
# if enabled, each vCPU thread will be scheduled to a fixed CPU
# qualified condition: num(vCPU threads) == num(CPUs in sandbox's CPUSet)
enable_vcpus_pinning = false
# Apply a custom SELinux security policy to the container process inside the VM.
# This is used when you want to apply a type other than the default `container_t`,
# so general users should not uncomment and apply it.
# (format: "user:role:type")
# Note: You cannot specify MCS policy with the label because the sensitivity levels and
# categories are determined automatically by high-level container runtimes such as containerd.
# Example value when enabling: "system_u:system_r:container_t"
guest_selinux_label = "@DEFGUESTSELINUXLABEL@"
# If enabled, the runtime will create opentracing.io traces and spans.
# (See https://www.jaegertracing.io/docs/getting-started).
# (default: disabled)
enable_tracing = false
# Set the full url to the Jaeger HTTP Thrift collector.
# The default if not set will be "http://localhost:14268/api/traces"
jaeger_endpoint = ""
# Sets the username to be used if basic auth is required for Jaeger.
jaeger_user = ""
# Sets the password to be used if basic auth is required for Jaeger.
jaeger_password = ""
# If enabled, the runtime will not create a network namespace for shim and hypervisor processes.
# This option may have some potential impacts to your host. It should only be used when you know what you're doing.
# `disable_new_netns` conflicts with `internetworking_model=tcfilter` and `internetworking_model=macvtap`. It works only
# with `internetworking_model=none`. The tap device will be in the host network namespace and can connect to a bridge
# (like OVS) directly.
# (default: false)
disable_new_netns = false
# if enabled, the runtime will add all the kata processes inside one dedicated cgroup.
# The container cgroups in the host are not created, just one single cgroup per sandbox.
# The runtime caller is free to restrict or collect cgroup stats of the overall Kata sandbox.
# The sandbox cgroup path is the parent cgroup of a container with the PodSandbox annotation.
# The sandbox cgroup is constrained if there is no container type annotation.
# See: https://pkg.go.dev/github.com/kata-containers/kata-containers/src/runtime/virtcontainers#ContainerType
sandbox_cgroup_only = @DEFSANDBOXCGROUPONLY_QEMU@
# If enabled, the runtime will attempt to determine appropriate sandbox size (memory, CPU) before booting the virtual machine. In
# this case, the runtime will not dynamically update the amount of memory and CPU in the virtual machine. This is generally helpful
# when a hardware architecture or hypervisor solutions is utilized which does not support CPU and/or memory hotplug.
# Compatibility for determining appropriate sandbox (VM) size:
# - When running with pods, sandbox sizing information will only be available if using Kubernetes >= 1.23 and containerd >= 1.6. CRI-O
# does not yet support sandbox sizing annotations.
# - When running single containers using a tool like ctr, container sizing information will be available.
static_sandbox_resource_mgmt = @DEFSTATICRESOURCEMGMT_COCO@
# If specified, sandbox_bind_mounts identifieds host paths to be mounted (ro) into the sandboxes shared path.
# This is only valid if filesystem sharing is utilized. The provided path(s) will be bindmounted into the shared fs directory.
# If defaults are utilized, these mounts should be available in the guest at `/run/kata-containers/shared/containers/sandbox-mounts`
# These will not be exposed to the container workloads, and are only provided for potential guest services.
sandbox_bind_mounts = @DEFBINDMOUNTS@
# VFIO Mode
# Determines how VFIO devices should be be presented to the container.
# Options:
#
# - vfio
# Matches behaviour of OCI runtimes (e.g. runc) as much as
# possible. VFIO devices will appear in the container as VFIO
# character devices under /dev/vfio. The exact names may differ
# from the host (they need to match the VM's IOMMU group numbers
# rather than the host's)
#
# - guest-kernel
# This is a Kata-specific behaviour that's useful in certain cases.
# The VFIO device is managed by whatever driver in the VM kernel
# claims it. This means it will appear as one or more device nodes
# or network interfaces depending on the nature of the device.
# Using this mode requires specially built workloads that know how
# to locate the relevant device interfaces within the VM.
#
vfio_mode = "@DEFVFIOMODE@"
# If enabled, the runtime will not create Kubernetes emptyDir mounts on the guest filesystem. Instead, emptyDir mounts will
# be created on the host and shared via virtio-fs. This is potentially slower, but allows sharing of files from host to guest.
disable_guest_empty_dir = @DEFDISABLEGUESTEMPTYDIR@
# Enabled experimental feature list, format: ["a", "b"].
# Experimental features are features not stable enough for production,
# they may break compatibility, and are prepared for a big version bump.
# Supported experimental features:
# for example:
# experimental=["force_guest_pull"]
# which is for enable force_guest_pull mode in CoCo scenarios.
# (default: [])
experimental = @DEFAULTEXPFEATURES@
# If enabled, user can run pprof tools with shim v2 process through kata-monitor.
# (default: false)
enable_pprof = false
# Base directory of directly attachable network config.
# Network devices for VM-based containers are allowed to be placed in the
# host netns to eliminate as many hops as possible, which is what we
# called a "Directly Attachable Network". The config, set by special CNI
# plugins, is used to tell the Kata containers what devices are attached
# to the hypervisor.
# (default: /run/kata-containers/dans)
dan_conf = "@DEFDANCONF@"
# pod_resource_api_sock specifies the unix socket for the Kubelet's
# PodResource API endpoint. If empty, kubernetes based cold plug
# will not be attempted. In order for this feature to work, the
# KubeletPodResourcesGet featureGate must be enabled in Kubelet,
# if using Kubelet older than 1.34.
#
# The pod resource API's socket is relative to the Kubelet's root-dir,
# which is defined by the cluster admin, and its location is:
# ${KubeletRootDir}/pod-resources/kubelet.sock
#
# cold_plug_vfio(see hypervisor config) acts as a feature gate:
# cold_plug_vfio = no_port (default) => no cold plug
# cold_plug_vfio != no_port AND pod_resource_api_sock = "" => need
# explicit CDI annotation for cold plug (applies mainly
# to non-k8s cases)
# cold_plug_vfio != no_port AND pod_resource_api_sock != "" => kubelet
# based cold plug.
pod_resource_api_sock = "@DEFPODRESOURCEAPISOCK@"

View File

@@ -74,21 +74,43 @@ impl KernelParams {
pub(crate) fn new_rootfs_kernel_params(rootfs_driver: &str, rootfs_type: &str) -> Result<Self> {
let mut params = vec![];
// DAX is disabled on aarch64 due to kernel panic in dax_disassociate_entry
// with virtio-pmem on kernel 6.18.x
#[cfg(target_arch = "aarch64")]
let use_dax = false;
#[cfg(not(target_arch = "aarch64"))]
let use_dax = true;
match rootfs_driver {
VM_ROOTFS_DRIVER_PMEM => {
params.push(Param::new("root", VM_ROOTFS_ROOT_PMEM));
match rootfs_type {
VM_ROOTFS_FILESYSTEM_EXT4 => {
params.push(Param::new(
"rootflags",
"dax,data=ordered,errors=remount-ro ro",
));
if use_dax {
params.push(Param::new(
"rootflags",
"dax,data=ordered,errors=remount-ro ro",
));
} else {
params.push(Param::new(
"rootflags",
"data=ordered,errors=remount-ro ro",
));
}
}
VM_ROOTFS_FILESYSTEM_XFS => {
params.push(Param::new("rootflags", "dax ro"));
if use_dax {
params.push(Param::new("rootflags", "dax ro"));
} else {
params.push(Param::new("rootflags", "ro"));
}
}
VM_ROOTFS_FILESYSTEM_EROFS => {
params.push(Param::new("rootflags", "dax ro"));
if use_dax {
params.push(Param::new("rootflags", "dax ro"));
} else {
params.push(Param::new("rootflags", "ro"));
}
}
_ => {
return Err(anyhow!("Unsupported rootfs type {}", rootfs_type));
@@ -233,6 +255,22 @@ mod tests {
#[test]
fn test_rootfs_kernel_params() {
// DAX is disabled on aarch64
#[cfg(target_arch = "aarch64")]
let ext4_pmem_rootflags = "data=ordered,errors=remount-ro ro";
#[cfg(not(target_arch = "aarch64"))]
let ext4_pmem_rootflags = "dax,data=ordered,errors=remount-ro ro";
#[cfg(target_arch = "aarch64")]
let xfs_pmem_rootflags = "ro";
#[cfg(not(target_arch = "aarch64"))]
let xfs_pmem_rootflags = "dax ro";
#[cfg(target_arch = "aarch64")]
let erofs_pmem_rootflags = "ro";
#[cfg(not(target_arch = "aarch64"))]
let erofs_pmem_rootflags = "dax ro";
let tests = &[
// EXT4
TestData {
@@ -241,7 +279,7 @@ mod tests {
expect_params: KernelParams {
params: [
Param::new("root", VM_ROOTFS_ROOT_PMEM),
Param::new("rootflags", "dax,data=ordered,errors=remount-ro ro"),
Param::new("rootflags", ext4_pmem_rootflags),
Param::new("rootfstype", VM_ROOTFS_FILESYSTEM_EXT4),
]
.to_vec(),
@@ -268,7 +306,7 @@ mod tests {
expect_params: KernelParams {
params: [
Param::new("root", VM_ROOTFS_ROOT_PMEM),
Param::new("rootflags", "dax ro"),
Param::new("rootflags", xfs_pmem_rootflags),
Param::new("rootfstype", VM_ROOTFS_FILESYSTEM_XFS),
]
.to_vec(),
@@ -295,7 +333,7 @@ mod tests {
expect_params: KernelParams {
params: [
Param::new("root", VM_ROOTFS_ROOT_PMEM),
Param::new("rootflags", "dax ro"),
Param::new("rootflags", erofs_pmem_rootflags),
Param::new("rootfstype", VM_ROOTFS_FILESYSTEM_EROFS),
]
.to_vec(),

View File

@@ -585,7 +585,9 @@ func (clh *cloudHypervisor) CreateVM(ctx context.Context, id string, network Net
clh.vmconfig.Cpus = chclient.NewCpusConfig(int32(clh.config.NumVCPUs()), int32(clh.config.DefaultMaxVCPUs))
disableNvdimm := (clh.config.DisableImageNvdimm || clh.config.ConfidentialGuest)
enableDax := !disableNvdimm
// DAX is disabled on aarch64 due to kernel panic in dax_disassociate_entry
// with virtio-pmem on kernel 6.18.x
enableDax := !disableNvdimm && runtime.GOARCH != "arm64"
params, err := getNonUserDefinedKernelParams(hypervisorConfig.RootfsType, disableNvdimm, enableDax, clh.config.Debug, clh.config.ConfidentialGuest, clh.config.IOMMU)
if err != nil {

View File

@@ -69,9 +69,11 @@ func newQemuArch(config HypervisorConfig) (qemuArch, error) {
kernelParamsDebug: kernelParamsDebug,
kernelParams: kernelParams,
disableNvdimm: config.DisableImageNvdimm,
dax: true,
protection: noneProtection,
legacySerial: config.LegacySerial,
// DAX is disabled on aarch64 due to kernel panic in dax_disassociate_entry
// with virtio-pmem on kernel 6.18.x
dax: false,
protection: noneProtection,
legacySerial: config.LegacySerial,
},
measurementAlgo: config.MeasurementAlgo,
}

View File

@@ -1415,13 +1415,6 @@ func (s *Sandbox) startVM(ctx context.Context, prestartHookFunc func(context.Con
if err != nil {
return err
}
// If we want the network, scan the netns again to update the network
// configuration after the prestart hooks have run.
if !s.config.NetworkConfig.DisableNewNetwork {
if _, err := s.network.AddEndpoints(ctx, s, nil, false); err != nil {
return err
}
}
}
if err := s.network.Run(ctx, func() error {

View File

@@ -74,6 +74,8 @@ Extra environment variables:
AGENT_BIN: Use it to change the expected agent binary name
AGENT_INIT: Use kata agent as init process
BLOCK_SIZE: Use to specify the size of blocks in bytes. DEFAULT: 4096
DAX_DISABLE: If set to "yes", skip DAX metadata header (for kernels without FS_DAX support).
DEFAULT: not set
IMAGE_REGISTRY: Hostname for the image registry used to pull down the rootfs build image.
NSDAX_BIN: Use to specify path to pre-compiled 'nsdax' tool.
USE_DOCKER: If set will build image in a Docker Container (requries docker)
@@ -169,6 +171,7 @@ build_with_container() {
--env BLOCK_SIZE="${block_size}" \
--env ROOT_FREE_SPACE="${root_free_space}" \
--env NSDAX_BIN="${nsdax_bin}" \
--env DAX_DISABLE="${DAX_DISABLE:-no}" \
--env MEASURED_ROOTFS="${MEASURED_ROOTFS}" \
--env SELINUX="${SELINUX}" \
--env DEBUG="${DEBUG}" \
@@ -304,8 +307,12 @@ calculate_img_size() {
local fs_type="$3"
local block_size="$4"
# rootfs start + DAX header size + rootfs end
local reserved_size_mb=$((rootfs_start + dax_header_sz + rootfs_end))
# rootfs start + DAX header size (if enabled) + rootfs end
local dax_sz=0
if [ "${DAX_DISABLE:-no}" != "yes" ]; then
dax_sz="${dax_header_sz}"
fi
local reserved_size_mb=$((rootfs_start + dax_sz + rootfs_end))
disk_size="$(calculate_required_disk_size "${rootfs}" "${fs_type}" "${block_size}")"
@@ -624,25 +631,35 @@ main() {
die "Invalid rootfs"
fi
# Determine DAX header size based on DAX_DISABLE setting
local dax_sz=0
if [ "${DAX_DISABLE:-no}" != "yes" ]; then
dax_sz="${dax_header_sz}"
fi
if [ "${fs_type}" == 'erofs' ]; then
# mkfs.erofs accepts an src root dir directory as an input
# rather than some device, so no need to guess the device dest size first.
create_erofs_rootfs_image "${rootfs}" "${image}" \
"${block_size}" "${agent_bin}"
rootfs_img_size=$?
img_size=$((rootfs_img_size + dax_header_sz))
img_size=$((rootfs_img_size + dax_sz))
else
img_size=$(calculate_img_size "${rootfs}" "${root_free_space}" \
"${fs_type}" "${block_size}")
# the first 2M are for the first MBR + NVDIMM metadata and were already
# consider in calculate_img_size
rootfs_img_size=$((img_size - dax_header_sz))
# consider in calculate_img_size (if DAX is enabled)
rootfs_img_size=$((img_size - dax_sz))
create_rootfs_image "${rootfs}" "${image}" "${rootfs_img_size}" \
"${fs_type}" "${block_size}" "${agent_bin}"
fi
# insert at the beginning of the image the MBR + DAX header
set_dax_header "${image}" "${img_size}" "${fs_type}" "${nsdax_bin}"
if [ "${DAX_DISABLE:-no}" != "yes" ]; then
set_dax_header "${image}" "${img_size}" "${fs_type}" "${nsdax_bin}"
else
info "Skipping DAX header (DAX_DISABLE=yes)"
fi
chown "${USER}:${GROUP}" "${image}"
}

View File

@@ -18,9 +18,12 @@ die() {
exit 1
}
arch_target=$1
nvidia_gpu_stack="$2"
base_os="$3"
run_file_name=$2
run_fm_file_name=$3
arch_target=$4
nvidia_gpu_stack="$5"
driver_version=""
base_os="noble"
APT_INSTALL="apt -o Dpkg::Options::='--force-confdef' -o Dpkg::Options::='--force-confold' -yqq --no-install-recommends install"
@@ -31,6 +34,31 @@ is_feature_enabled() {
[[ ",${nvidia_gpu_stack}," == *",${feature},"* ]]
}
set_driver_version() {
# Extract the driver=XXX part first, then get the value
if [[ "${nvidia_gpu_stack}" =~ driver=([^,]+) ]]; then
driver_version="${BASH_REMATCH[1]}"
fi
echo "chroot: driver_version: ${driver_version}"
echo "chroot: TODO remove with new NVRC"
cat <<-CHROOT_EOF > "/supported-gpu.devids"
0x230E
0x2321
0x2322
0x2324
0x2329
0x232C
0x2330
0x2331
0x2335
0x2339
0x233A
0x233B
0x2342
0x2348
CHROOT_EOF
}
install_nvidia_ctk() {
echo "chroot: Installing NVIDIA GPU container runtime"
# Base gives a nvidia-ctk and the nvidia-container-runtime
@@ -48,19 +76,13 @@ install_nvidia_fabricmanager() {
}
install_userspace_components() {
# Extract the driver=XXX part first, then get the value
if [[ "${nvidia_gpu_stack}" =~ driver=([^,]+) ]]; then
driver_version="${BASH_REMATCH[1]}"
fi
echo "chroot: driver_version: ${driver_version}"
eval "${APT_INSTALL}" nvidia-driver-pinning-"${driver_version}"
eval "${APT_INSTALL}" nvidia-imex nvidia-firmware \
eval "${APT_INSTALL}" nvidia-imex nvidia-firmware \
libnvidia-cfg1 libnvidia-gl libnvidia-extra \
libnvidia-decode libnvidia-fbc1 libnvidia-encode \
libnvidia-nscq
apt-mark hold nvidia-imex nvidia-firmware \
apt-mark hold nvidia-imex nvidia-firmware \
libnvidia-cfg1 libnvidia-gl libnvidia-extra \
libnvidia-decode libnvidia-fbc1 libnvidia-encode \
libnvidia-nscq
@@ -89,13 +111,12 @@ setup_apt_repositories() {
rm -f /etc/apt/sources.list.d/*
key="/usr/share/keyrings/ubuntu-archive-keyring.gpg"
comp="main restricted universe multiverse"
cat <<-CHROOT_EOF > /etc/apt/sources.list.d/"${base_os}".list
deb [arch=${deb_arch} signed-by=${key}] http://${mirror} ${base_os} ${comp}
deb [arch=${deb_arch} signed-by=${key}] http://${mirror} ${base_os}-updates ${comp}
deb [arch=${deb_arch} signed-by=${key}] http://${mirror} ${base_os}-security ${comp}
deb [arch=${deb_arch} signed-by=${key}] http://${mirror} ${base_os}-backports ${comp}
deb [arch=${deb_arch} signed-by=${key}] http://${mirror} ${base_os} main restricted universe multiverse
deb [arch=${deb_arch} signed-by=${key}] http://${mirror} ${base_os}-updates main restricted universe multiverse
deb [arch=${deb_arch} signed-by=${key}] http://${mirror} ${base_os}-security main restricted universe multiverse
deb [arch=${deb_arch} signed-by=${key}] http://${mirror} ${base_os}-backports main restricted universe multiverse
CHROOT_EOF
local arch="${arch_target}"
@@ -108,7 +129,7 @@ setup_apt_repositories() {
curl -fsSL -O "https://developer.download.nvidia.com/compute/cuda/repos/${osver}/${arch}/${keyring}"
dpkg -i "${keyring}" && rm -f "${keyring}"
# Set priorities: CUDA repos highest, Ubuntu non-driver next, Ubuntu blocked for driver packages
# Set priorities: Ubuntu repos highest, NVIDIA Container Toolkit next, CUDA repo blocked for driver packages
cat <<-CHROOT_EOF > /etc/apt/preferences.d/nvidia-priority
Package: *
Pin: $(dirname "${mirror}")
@@ -160,6 +181,7 @@ cleanup_rootfs() {
# Start of script
echo "chroot: Setup NVIDIA GPU rootfs stage one"
set_driver_version
setup_apt_repositories
install_userspace_components
install_nvidia_fabricmanager

View File

@@ -94,11 +94,25 @@ setup_nvidia_gpu_rootfs_stage_one() {
mkdir -p ./lib/modules/
tar --zstd -xvf "${BUILD_DIR}"/kata-static-kernel-nvidia-gpu"${appendix}"-modules.tar.zst -C ./lib/modules/
# If we find a local downloaded run file build the kernel modules
# with it, otherwise use the distribution packages. Run files may have
# more recent drivers available then the distribution packages.
local run_file_name="nvidia-driver.run"
if [[ -f ${BUILD_DIR}/${run_file_name} ]]; then
cp -L "${BUILD_DIR}"/"${run_file_name}" ./"${run_file_name}"
fi
local run_fm_file_name="nvidia-fabricmanager.run"
if [[ -f ${BUILD_DIR}/${run_fm_file_name} ]]; then
cp -L "${BUILD_DIR}"/"${run_fm_file_name}" ./"${run_fm_file_name}"
fi
mount --rbind /dev ./dev
mount --make-rslave ./dev
mount -t proc /proc ./proc
chroot . /bin/bash -c "/nvidia_chroot.sh ${machine_arch} ${NVIDIA_GPU_STACK} noble"
chroot . /bin/bash -c "/nvidia_chroot.sh $(uname -r) ${run_file_name} \
${run_fm_file_name} ${machine_arch} ${NVIDIA_GPU_STACK}"
umount -R ./dev
umount ./proc
@@ -245,6 +259,7 @@ chisseled_init() {
cp -a "${stage_one}"/etc/kata-opa etc/.
fi
cp -a "${stage_one}"/etc/resolv.conf etc/.
cp -a "${stage_one}"/supported-gpu.devids .
cp -a "${stage_one}"/lib/firmware/nvidia lib/firmware/.
cp -a "${stage_one}"/sbin/ldconfig.real sbin/ldconfig

View File

@@ -15,13 +15,13 @@ type: application
# This is the chart version. This version number should be incremented each time you make changes
# to the chart and its templates, including the app version.
# Versions are expected to follow Semantic Versioning (https://semver.org/)
version: "3.25.0"
version: "3.24.0"
# This is the version number of the application being deployed. This version number should be
# incremented each time you make changes to the application. Versions are not expected to
# follow Semantic Versioning. They should reflect the version the application is using.
# It is recommended to use it with quotes.
appVersion: "3.25.0"
appVersion: "3.24.0"
dependencies:
- name: node-feature-discovery

View File

@@ -114,6 +114,7 @@ RUNTIME_CHOICE="${RUNTIME_CHOICE:-both}"
IMAGE_SIZE_ALIGNMENT_MB=${IMAGE_SIZE_ALIGNMENT_MB:-}
KERNEL_DEBUG_ENABLED="${KERNEL_DEBUG_ENABLED:-}"
INIT_DATA="${INIT_DATA:-yes}"
DAX_DISABLE="${DAX_DISABLE:-no}"
docker run \
-v $HOME/.docker:/root/.docker \
@@ -160,6 +161,7 @@ docker run \
--env CROSS_BUILD="${CROSS_BUILD}" \
--env TARGET_ARCH="${TARGET_ARCH}" \
--env ARCH="${ARCH}" \
--env DAX_DISABLE="${DAX_DISABLE}" \
--rm \
-w ${script_dir} \
build-kata-deploy "${kata_deploy_create}" "$@"

View File

@@ -468,6 +468,12 @@ install_image() {
export REPO_COMPONENTS
fi
# Disable DAX for ARM64 due to kernel panic in dax_disassociate_entry
# with virtio-pmem on kernel 6.18.x
if [ "${ARCH}" == "aarch64" ]; then
export DAX_DISABLE=yes
fi
"${rootfs_builder}" --osname="${os_name}" --osversion="${os_version}" --imagetype=image --prefix="${prefix}" --destdir="${destdir}" --image_initrd_suffix="${variant}"
}

View File

@@ -1 +1 @@
176
175

View File

@@ -1,170 +0,0 @@
From mboxrd@z Thu Jan 1 00:00:00 1970
Received: from foss.arm.com (foss.arm.com [217.140.110.172])
by smtp.subspace.kernel.org (Postfix) with ESMTP id BCE6D2FFDD5
for <nvdimm@lists.linux.dev>; Wed, 14 Jan 2026 17:49:59 +0000 (UTC)
Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=217.140.110.172
ARC-Seal:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116;
t=1768413001; cv=none; b=mYidLRrJZn5ooS7h+lyKLsbA8/GKL/ZqDOCHo66hKab86eV5cgpwbWPeudpYGPiMp/QhczPxwDzq1J9qP57FU8xy5AmIFwF6cAn6FPN0BzBWxBQUeT/pxDwfOkvh33RigAd/HACiTa+9waLWNn94H1tPpUOn1SUKYcC2anb/EMA=
ARC-Message-Signature:i=1; a=rsa-sha256; d=subspace.kernel.org;
s=arc-20240116; t=1768413001; c=relaxed/simple;
bh=/AHMpb4+3MfzIhYBPESp8KGt8HTeUy14LUGKGDaY7Jc=;
h=Message-ID:Date:MIME-Version:To:Cc:From:Subject:Content-Type; b=nYCvtwNmfNPRI2kMt6FzMOpG8Xv2GrytpiVJh3K4jRBpJFo3NO0icdYoGz0yjvq1G4vQvh/VrhrLhOVAEHdNkuGQz1yL6qHm/9KniwafY98ihbvaadCAZtdiNtjhZ646/irNi48nnwxquGqUkiPk2n9PqYSVLR9VYf60nr/nAOA=
ARC-Authentication-Results:i=1; smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=arm.com; spf=pass smtp.mailfrom=arm.com; arc=none smtp.client-ip=217.140.110.172
Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=arm.com
Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=arm.com
Received: from usa-sjc-imap-foss1.foss.arm.com (unknown [10.121.207.14])
by usa-sjc-mx-foss1.foss.arm.com (Postfix) with ESMTP id 4F3F21515;
Wed, 14 Jan 2026 09:49:52 -0800 (PST)
Received: from [10.1.37.132] (unknown [10.1.37.132])
by usa-sjc-imap-foss1.foss.arm.com (Postfix) with ESMTPSA id 0CA253F59E;
Wed, 14 Jan 2026 09:49:57 -0800 (PST)
Message-ID: <18af3213-6c46-4611-ba75-da5be5a1c9b0@arm.com>
Date: Wed, 14 Jan 2026 17:49:30 +0000
Precedence: bulk
X-Mailing-List: nvdimm@lists.linux.dev
List-Id: <nvdimm.lists.linux.dev>
List-Subscribe: <mailto:nvdimm+subscribe@lists.linux.dev>
List-Unsubscribe: <mailto:nvdimm+unsubscribe@lists.linux.dev>
MIME-Version: 1.0
User-Agent: Mozilla Thunderbird
Content-Language: en-US
To: linux-kernel@vger.kernel.org
Cc: linux-fsdevel@vger.kernel.org, nvdimm@lists.linux.dev,
dan.j.williams@intel.com, willy@infradead.org, jack@suse.cz,
Nick.Connolly@arm.com, ffidencio@nvidia.com
From: Seunguk Shin <seunguk.shin@arm.com>
Subject: [PATCH] fs/dax: check zero or empty entry before converting xarray
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 8bit
Trying to convert zero or empty xarray entry causes kernel panic.
[ 0.737679] EXT4-fs (pmem0p1): mounted filesystem
79676804-7c8b-491a-b2a6-9bae3c72af70 ro with ordered data mode. Quota
mode: disabled.
[ 0.737891] VFS: Mounted root (ext4 filesystem) readonly on device 259:1.
[ 0.739119] devtmpfs: mounted
[ 0.739476] Freeing unused kernel memory: 1920K
[ 0.740156] Run /sbin/init as init process
[ 0.740229] with arguments:
[ 0.740286] /sbin/init
[ 0.740321] with environment:
[ 0.740369] HOME=/
[ 0.740400] TERM=linux
[ 0.743162] Unable to handle kernel paging request at virtual address
fffffdffbf000008
[ 0.743285] Mem abort info:
[ 0.743316] ESR = 0x0000000096000006
[ 0.743371] EC = 0x25: DABT (current EL), IL = 32 bits
[ 0.743444] SET = 0, FnV = 0
[ 0.743489] EA = 0, S1PTW = 0
[ 0.743545] FSC = 0x06: level 2 translation fault
[ 0.743610] Data abort info:
[ 0.743656] ISV = 0, ISS = 0x00000006, ISS2 = 0x00000000
[ 0.743720] CM = 0, WnR = 0, TnD = 0, TagAccess = 0
[ 0.743785] GCS = 0, Overlay = 0, DirtyBit = 0, Xs = 0
[ 0.743848] swapper pgtable: 4k pages, 48-bit VAs, pgdp=00000000b9d17000
[ 0.743931] [fffffdffbf000008] pgd=10000000bfa3d403,
p4d=10000000bfa3d403, pud=1000000040bfe403, pmd=0000000000000000
[ 0.744070] Internal error: Oops: 0000000096000006 [#1] SMP
[ 0.748888] CPU: 0 UID: 0 PID: 1 Comm: init Not tainted 6.18.4 #1 NONE
[ 0.749421] pstate: 004000c5 (nzcv daIF +PAN -UAO -TCO -DIT -SSBS
BTYPE=--)
[ 0.749969] pc : dax_disassociate_entry.constprop.0+0x20/0x50
[ 0.750444] lr : dax_insert_entry+0xcc/0x408
[ 0.750802] sp : ffff80008000b9e0
[ 0.751083] x29: ffff80008000b9e0 x28: 0000000000000000 x27:
0000000000000000
[ 0.751682] x26: 0000000001963d01 x25: ffff0000004f7d90 x24:
0000000000000000
[ 0.752264] x23: 0000000000000000 x22: ffff80008000bcc8 x21:
0000000000000011
[ 0.752836] x20: ffff80008000ba90 x19: 0000000001963d01 x18:
0000000000000000
[ 0.753407] x17: 0000000000000000 x16: 0000000000000000 x15:
0000000000000000
[ 0.753970] x14: ffffbf3154b9ae70 x13: 0000000000000000 x12:
ffffbf3154b9ae70
[ 0.754548] x11: ffffffffffffffff x10: 0000000000000000 x9 :
0000000000000000
[ 0.755122] x8 : 000000000000000d x7 : 000000000000001f x6 :
0000000000000000
[ 0.755707] x5 : 0000000000000000 x4 : 0000000000000000 x3 :
fffffdffc0000000
[ 0.756287] x2 : 0000000000000008 x1 : 0000000040000000 x0 :
fffffdffbf000000
[ 0.756871] Call trace:
[ 0.757107] dax_disassociate_entry.constprop.0+0x20/0x50 (P)
[ 0.757592] dax_iomap_pte_fault+0x4fc/0x808
[ 0.757951] dax_iomap_fault+0x28/0x30
[ 0.758258] ext4_dax_huge_fault+0x80/0x2dc
[ 0.758594] ext4_dax_fault+0x10/0x3c
[ 0.758892] __do_fault+0x38/0x12c
[ 0.759175] __handle_mm_fault+0x530/0xcf0
[ 0.759518] handle_mm_fault+0xe4/0x230
[ 0.759833] do_page_fault+0x17c/0x4dc
[ 0.760144] do_translation_fault+0x30/0x38
[ 0.760483] do_mem_abort+0x40/0x8c
[ 0.760771] el0_ia+0x4c/0x170
[ 0.761032] el0t_64_sync_handler+0xd8/0xdc
[ 0.761371] el0t_64_sync+0x168/0x16c
[ 0.761677] Code: f9453021 f2dfbfe3 cb813080 8b001860 (f9400401)
[ 0.762168] ---[ end trace 0000000000000000 ]---
[ 0.762550] note: init[1] exited with irqs disabled
[ 0.762631] Kernel panic - not syncing: Attempted to kill init!
exitcode=0x0000000b
This patch just reorders checking and converting.
Signed-off-by: Seunguk Shin <seunguk.shin@arm.com>
---
fs/dax.c | 9 ++++++---
1 file changed, 6 insertions(+), 3 deletions(-)
diff --git a/fs/dax.c b/fs/dax.c
index 516f995a9..69ac2562c 100644
--- a/fs/dax.c
+++ b/fs/dax.c
@@ -443,11 +443,12 @@ static void dax_associate_entry(void *entry, struct address_space *mapping,
unsigned long address, bool shared)
{
unsigned long size = dax_entry_size(entry), index;
- struct folio *folio = dax_to_folio(entry);
+ struct folio *folio;
if (dax_is_zero_entry(entry) || dax_is_empty_entry(entry))
return;
+ folio = dax_to_folio(entry);
index = linear_page_index(vma, address & ~(size - 1));
if (shared && (folio->mapping || dax_folio_is_shared(folio))) {
if (folio->mapping)
@@ -468,21 +469,23 @@ static void dax_associate_entry(void *entry, struct address_space *mapping,
static void dax_disassociate_entry(void *entry, struct address_space *mapping,
bool trunc)
{
- struct folio *folio = dax_to_folio(entry);
+ struct folio *folio;
if (dax_is_zero_entry(entry) || dax_is_empty_entry(entry))
return;
+ folio = dax_to_folio(entry);
dax_folio_put(folio);
}
static struct page *dax_busy_page(void *entry)
{
- struct folio *folio = dax_to_folio(entry);
+ struct folio *folio;
if (dax_is_zero_entry(entry) || dax_is_empty_entry(entry))
return NULL;
+ folio = dax_to_folio(entry);
if (folio_ref_count(folio) - folio_mapcount(folio))
return &folio->page;
else
--
2.43.0

View File

@@ -40,7 +40,6 @@ function _check_required_env_var()
KATA_STATIC_TARBALL) env_var="${KATA_STATIC_TARBALL}" ;;
KATA_DEPLOY_IMAGE_TAGS) env_var="${KATA_DEPLOY_IMAGE_TAGS}" ;;
KATA_DEPLOY_REGISTRIES) env_var="${KATA_DEPLOY_REGISTRIES}" ;;
KATA_TOOLS_STATIC_TARBALL) env_var="${KATA_TOOLS_STATIC_TARBALL}" ;;
*) >&2 _die "Invalid environment variable \"${1}\"" ;;
esac