Splitting the definitions of a post-launched VM into two files, namely the
scenario XML and launch XML, introduces duplicated field in both files and
leads to a high probability of having inconsistencies between them (see
issue #7156 as an example). Further more, this split has also adds much
complexity to the configurator which has to either hide some of the items
from user interfaces or synchronize different fields upon changes.
The advantage of the split, however, is not widely adopted. Having a
separate XML capturing the VM definition tweakable in the service VM at
runtime seems to give users more flexibility to redefine a VM without
recompiling the hypervisor. But that is not a common practice in the
industry segment; instead it is preferred to have a static scenario
definition to make sure that all resources are allocated carefully in a
fixed manner in order for better determinism.
As a result, this patch merges the fields in launch XMLs into the schema of
scenario XMLs. Some fields are post-launched VM specific and thus added,
while the others have similar items in scenario XMLs today.
The launch script generator is also updated accordingly to generate launch
scripts from the new scenario XMLs (which now contain the same amount of
information as previous launch XMLs).
Tracked-On: #6690
Signed-off-by: Junjie Mao <junjie.mao@intel.com>
This patch adds the following annotations to the config items defined in
the schema of scenario XMLs:
- acrn:title, which defines the human-readable label of the corresponding
widgets in the configurator.
- acrn:views, which controls in which view(s) this item shows in the
configurator.
- acrn:applicable-vms, which specifies the kinds of VMs (pre-launched,
post-launched and/or service VM) this item applies to. An item not
applicable to a certain type of VM will not be shown in the
configurator, and will trigger validation error if that item exists for
a VM of that specific type.
v1 -> v2:
* Preserve the CPU affinity settings for service VMs. This will be needed
later when we want to restrict the available CPUs the service VM can use at
runtime (but not at initialization time).
Tracked-On: #6690
Signed-off-by: Junjie Mao <junjie.mao@intel.com>
A shared memory region can be provided either by the hypervisor or by the
device model. Before recent schema changes this is telled by the "hv:/" or
"dm:/" prefix.
This patch adds another node under an IVSHMEM region to represent the
provider, following the practice that information in the old-school
encodings is split and put as separate XML nodes.
Tracked-On: #6690
Signed-off-by: Junjie Mao <junjie.mao@intel.com>
This patch drops a few config items which are no longer needed, including:
- vm.os_config.name
- vm.UEFI_OS_LOADER_NAME
- vm.pci_dev_num
Tracked-On: #6690
Signed-off-by: Junjie Mao <junjie.mao@intel.com>
As is recommended by UX/DX reviews, the per-VM console virtual UART is now
limited to the following choices:
- Disabled
- a COM port from COM1 to COM4
- PCI based
This patch converts the schema of scenario XMLs to integrate this
recommendation and add logic in the scenario upgrader to migrate data from
old scenario XMLs.
v1 -> v2:
* Update the static allocators and C source transformers according to the
new console vUART config item.
Tracked-On: #6690
Signed-off-by: Junjie Mao <junjie.mao@intel.com>
Instead of using a Boolean variable indicating whether a build is for debug
or release, it is more intuitive to specify the build types as "debug" or
"release".
This patch converts the config item RELEASE to BUILD_TYPE which takes
"debug" or "release" as of now.
The generated header and makefile still uses RELEASE, and the command line
option RELEASE=<y or n> is also preserved.
Tracked-On: #6690
Signed-off-by: Junjie Mao <junjie.mao@intel.com>
With a bunch of dramatic improvement to the schema of scenario XML
recently, it is now non-trivial for users of previous releases to migrate
their scenario definitions to the next-gen schema, which will be a
major obstacle when upgrading to the next ACRN release.
In order to ease the upgrade of scenario XML, this patch introduces a
script that takes a scenario XML of previous releases and generates a new
one which satisfies the latest schema. The major data that can be migrated
include:
- Virtual UART and shared memory configurations
- VM types, load_order and guest flags
Other data are preserved as long as they are still needed according to the
schema. The script to print a list that summarizes the data that are
unintendedly discarded during the upgrade, so that users can double check
the results and edit the generated scenario using the configurator.
The upgrader share the same command-line interface of the default value
populator and scenario validator.
More upgrading logic will be added at the same time the schema is changed.
v1 -> v2:
* The upgrader now pretty-prints the upgraded XML.
* Apply the upgrader to the current scenario XMLs in the repo. The
ordering of nodes are adjusted to align with the schema, but no
essential changes are made.
* Set the log level to INFO.
Tracked-On: #6690
Signed-off-by: Junjie Mao <junjie.mao@intel.com>
We plan to add the following attributes to element definitions in the XML
schema:
- acrn:applicable-vms, which specify if an element applies to a
pre-launched VM, the service VM or a post-launched VM
- acrn:views, which specify if an element shall appear in the basic or
advanced tab in the configurator.
In order to reduce the attributes above to existing XML technologies, we
need to create new complex types that lists all config items that applies
to a pre-launched VM, the service VM or a post-launched VM, or that should
be shown in the basic or advanced view. Such types can then be used to
replace the original, all-in-one type during validation or configurator
rendering.
When unspecified, an element always applies under all possible
circumstances.
To realize this slicing mechanism, this patch adds a generic class
implementing the common part of slicing XML schema types and two
specific-purpose slicers according to the applicable VMs or views
attributes.
v2 -> v3:
* Update configdoc.xsl to recognize types in xs:alternative nodes.
Tracked-On: #6690
Signed-off-by: Junjie Mao <junjie.mao@intel.com>
Today the scripts that populate default values and validate scenarios have
different command-line interfaces: the former requires an XML schema as
input (which is cumbersome in most cases), while the latter always infer
where the XML schema is (which is inflexible).
This patch unifies the command line options of those scripts as follows:
- The scenario XML is always a required positional argument.
- The output file path (if any) is an optional positional argument.
- The schema XML file is an optional long option. When not specified, the
scripts will always use the one under the misc/config_tools/schema
directory.
Also, this patch makes the validator.py executable, as is done to other
executable scripts in the repo.
Tracked-On: #6690
Signed-off-by: Junjie Mao <junjie.mao@intel.com>
There is an increasing demand of composing different operations around XML
schemas and/or data in different ways for different purpose. Today we
already have:
- Validate XML data, which takes XML schemas and data (board and
scenario) as inputs.
- Fill in missing nodes in XML data with default values, which takes XML
schema and data (scenario only) as inputs.
In the near future we'll extend the operations around XMLs by introducing
XML schema preprocessing and XML data upgrading, adding more possibilities
to construct a larger operation by composing smaller ones.
In order for minimized code repetition and easier composition, this patch
introduces an infrasturcture that abstracts each operation as a pipeline
stage. Each stage defines its own inputs and outputs and can be composed
sequentially as a larger, single operation.
The existing operations listed above, along with XML file loaders, are then
refactored to provide pipeline stages. The main methods are also refined to
complete their tasks by constructing and invoking pipelines.
Tracked-On: #6690
Signed-off-by: Junjie Mao <junjie.mao@intel.com>
Today the default value populator will always create a node if it is
required by the schema but not provided in the given XML file. This could
hide issues in a given scenario XML if a node does not have default
values (i.e. require user's inputs) but accepts empty text.
This patch avoids the creation of nodes without a default value so that, at
validation stage, missing of essential data is always reported.
Tracked-On: #6690
Signed-off-by: Junjie Mao <junjie.mao@intel.com>
The "xs:alternative" nodes in XML schema enables specifying different types
for the same node according to the contents of that node. A typical usage
of this construct in the schema of scenario XMLs is to specify different
types for a `vm` node depending on whether it is a pre-launched,
post-launched or service VMs, so that each VM contains only configurations
supported in that kind of VM.
This patch adds support of xs:alternative nodes to the schema transformer
so that default values can be populated appropriately after we add
xs:alternative nodes to the schema.
Tracked-On: #6690
Signed-off-by: Junjie Mao <junjie.mao@intel.com>
Today the script default_populator.py fills in default values by visiting
an XML schema and an XML tree simultaneously and add a node to the latter
whenever one satisfying the schema does not exist. This visiting logic is
not only useful for filling in default values, but also for upgrading
XMLs to new schemas.
This patch abstracts the flow of the visiting above as a separate class,
just like tree visitors or transformers. The current default value
populator is then refactored to extend that class by inheritance.
Tracked-On: #6690
Signed-off-by: Junjie Mao <junjie.mao@intel.com>
Assertions in XML schema are placed under schema/checks as the central
place, in order to make it easier to enhance the error reporting
specifically on assertion failures in the near future.
This patch moves the RDT related assertions in the schema to a dedicate
file uner schema/checks.
Tracked-On: #6690
Signed-off-by: Junjie Mao <junjie.mao@intel.com>
HMAC_*, MD5_* and SHA256_* are deprecated since openssl3.0, replace them with the corresponding equivalents.
Tracked-On: #6743
Signed-off-by: Tw <wei.tan@intel.com>
Acked-by: Wang, Yu1 <yu1.wang@intel.com>
We have redesign the vuart and the UI for user, so the config tool
should change the schema and xform for the new xml, then change the
static_allocators to alloc irq for new connection.
This patch modify the xmls for vuart new design.
Tracked-On: #6690
Reviewed-by: Junjie Mao <junjie.mao@intel.com>
Signed-off-by: Chenli Wei <chenli.wei@linux.intel.com>
We have redesign the vuart and the UI for user, so the config tool
should change the schema and xform for the new xml, then change the
static_allocators to alloc irq and io_port for new connection.
This patch add a new vuart connection type and change the xforms to
adapter the new type.
Tracked-On: #6690
Reviewed-by: Junjie Mao <junjie.mao@intel.com>
Signed-off-by: Chenli Wei <chenli.wei@linux.intel.com>
Fix the issue that doesn't offline CPU in SCHED_NOOP mode.
Tracked-On: #7172
Signed-off-by: Kunhui-Li <kunhuix.li@intel.com>
Reviewed-by: Junjie Mao <junjie.mao@intel.com>
Since PRs #7115 and #6664 have landed, the ivshmem and hv_ram_size have
been refined, some related checks are obsolete in config tool.
So we remove these checks here.
Tracked-On: #7136
Signed-off-by: Kunhui-Li <kunhuix.li@intel.com>
Add user VM reboot command and related command handler in
lifecycle manager to support user VM reboot.
Libvirt will send user VM reboot command to lifecycle manager
of service VM through socket, this command is forwarded to the
specified user VM, user VM will execute reboot command to start
reboot itself.
v1-->v2:
Update some interfaces name to make it reable:
(1) enable_uart_channel_dev_resend -->
start_uart_channel_dev_resend
(2) enable_all_uart_channel_dev_resend -->
start_all_uart_channel_dev_resend
(3) disable_uart_channel_dev_resend -->
stop_uart_channel_dev_resend
(4) get_reboot_flag --> get_user_vm_reboot_flag
Tracked-On: #5921
Signed-off-by: Xiangyang Wu <xiangyang.wu@intel.com>
Acked-by: Wang, Yu1 <yu1.wang@intel.com>
The concept of guest_flags is hard to understand for users.
So turn guest_flags into several parameters in config tool
user interface, list as below:
GUEST_FLAG_LAPIC_PASSTHROUGH ---> lapic_passthrough
GUEST_FLAG_IO_COMPLETION_POLLING ---> io_completion_polling
GUEST_FLAG_VCAT_ENABLED ---> virtual_cat_support
GUEST_FLAG_SECURE_WORLD_ENABLED ---> secure_world_support
GUEST_FLAG_HIDE_MTRR ---> hide_mtrr_support
GUEST_FLAG_NVMX_ENABLED ---> nested_virtualization_support
GUEST_FLAG_SECURITY_VM ---> security_vm
GUEST_FLAG_RT ---> vm_type(RTVM)
GUEST_FLAG_TEE ---> vm_type(TEE_VM)
GUEST_FLAG_REE ---> vm_type(REE_VM)
In addition, HV global parameter NVMX_ENABLE is removed
from user interface, when config tool detects more than
one VM with nested_virtualization_support, NVMX_ENABLE is
assigned as 'y' automatically.
v1->v2:
*Rebase on the latest xml schema checking change
*Remove "all rights reserved" from the license header in guest_flags.py
v2->v3:
*Change the name of the new config items to CAPITAL_CASE style
*Combine guest flag policy to an XPATH in guest_flags.py
*Use count() to directly deduce NVMX_ENABLED in config_common.xsl and
update `boolean-by-key-value` to process 'true'
v3->v4:
*Change the name of the new config items to lower_case style
*Change guest_flag_node to allocation_vm_node in guest_flags.py
*Separate value case and key case for boolean-by-key-value
Tracked-On: #6690
Signed-off-by: hangliu1 <hang1.liu@linux.intel.com>
Reviewed-by: Junjie Mao <junjie.mao@intel.com>
Update the XPATH of apic_id to tolerate the variance of reported
intermediate levels.
Tracked-On: #7159
Signed-off-by: Kunhui-Li <kunhuix.li@intel.com>
Reviewed-by: Junjie Mao <junjie.mao@intel.com>
Remove the logic to parse SERIAL_CONSOLE and append to bootargs. Specify the console in bootargs directly.
Tracked-On: #7127
Signed-off-by: Yang,Yu-chu <yu-chu.yang@intel.com>
Reviewed-by: Junjie Mao <junjie.mao@intel.com>
Acked-by: Anthony Xu <anthony.xu@intel.com>
This patch includes:
1.add load_order(PRE_LAUNCHED_VM/SERVICE_VM/POST_LAUNCHED_VM) parameter
2.change vm_type parameter to values as RTVM, STANDARD_VM, TEE, REE
TEE and REE are hide in UI.
3.deduce vm severity in vm_configuration from vm_type and load_order
This patch not includes:
change for scenario_config and functions called by scenario_config about checking
v2->v3:
*Refine template load_order
v1->v2:
*Change variable name from vm_type to load_order
*Change LoadOptionType to LoadOrderType
*Change VMOptionsType to VMType
*Add TEE_VM/REE_VM description
*Refine acrn:is-pre-launched-vm
Tracked-On: #6690
Signed-off-by: hangliu1 <hang1.liu@linux.intel.com>
Reviewed-by: Junjie Mao <junjie.mao@intel.com>
Since PR #7113 has landed, we also remove PLATFORM_RAM_SIZE in the related
Python code, schema and all existing scenario XML files.
Tracked-On: #6690
Signed-off-by: Kunhui-Li <kunhuix.li@intel.com>
With the launch script generation script rewritten, some other files under
launch_config are no longer used. This patch removes these obsoleted files.
Tracked-On: #6690
Signed-off-by: Junjie Mao <junjie.mao@intel.com>
The launch script generator today (and the scripts that it generates) is
fundamentally built on the concept of PCI device classes, with the
restriction that at most one PCI function per class can be passed through
to a post-launched VM. This has put inproper constraint on the scenarios
users can set up, especially on server platforms or those with SR-IOV
capable devices.
As it is too tedious to change such deep-rooted concept, this patch
rewrites the launch script generator and refines the structure of the
generated scripts so that PCI functions are identified only by their
BDF.
This change serves as a mandatory step to align the way how passthrough
devices are configured for pre-launched and post-launched VMs, which
eventually allows us to present a unified view in the configurator for
assigning passthrough device.
v2 -> v3:
* Rename sos_id to service_vm_id and user_vmid to user_vm_id.
* Refine a couple of info messages in the launch script template.
v1 -> v2:
* Fix wording issues identified during review.
* Exit when the out_dir is an existing regular file.
Tracked-On: #6690
Signed-off-by: Junjie Mao <junjie.mao@intel.com>
Rewrite the configuration of ivshmem in scenario.xml according to
the format of new UI design.
Tracked-On: #6690
Signed-off-by: Yuanyuan Zhao <yuanyuan.zhao@linux.intel.com>
Reviewed-by: Junjie Mao <junjie.mao@intel.com>
According to the new design of DX, implement ivshmem configuration
and generate hv related files.
Tracked-On: #6690
Signed-off-by: Yuanyuan Zhao <yuanyuan.zhao@linux.intel.com>
Reviewed-by: Junjie Mao <junjie.mao@intel.com>
Libcjson package is needed by command monitor of DM, so this
should be installed in the target platform.
Tracked-On: #5921
Signed-off-by: Xiangyang Wu <xiangyang.wu@intel.com>
Since PR 7081 has updated the dm parameter`lapic_pt` usage,
according to the new design, we add the`--rtvm` parameter
for Hard RT in launch script generation logic.
Tracked-On: #6690
Signed-off-by: Kunhui-Li <kunhuix.li@intel.com>
category based on different log levels:
1) If the board inspector show CRITICAL error messages, it means that the
board inspector tool exit instantly and generate no file.
2) If the board inspector show ERROR messages, it means that the board
inspector will generate the board XML file successfully, but several conditions
may result in ACRN build or boot failure cannot continue execution due to the error.
3) If the board inspector show WARNING messages, this means that the board inspector
generate the board XML file successfully, but this board XML file is lack of
some feature which could still boot ACRN but loss some features.
4) If the board inspector show INFO messages, this means that board inspector
printed out some normal information.
5) If the board inspector show DEBUG messages, this means that it is used to
debug board inspector workflow.
v1->v2
1. Keep the assertions or exceptions under the acpiparser, inspectorlib
and smbiosparser directory.
2. Exit after all check_deps() completes instead of any unsatisfied dep
is identified.
3. As Amy's advice, Replace the message "{board_xml} saved successfully" with
the message "SUCCESS: Board configuration file <file name> generated successfully
and saved to <path>."
To do:
Print all messages using the colored font.
Tracked-On: #6689
Reviewed-by: Junjie Mao <junjie.mao@intel.com>
Signed-off-by: Kunhui-Li <kunhuix.li@intel.com>
Bring all the cores up in the beginning of the board_inspector.py
It is expected to run board_inspector in clean and native environment.
To resolve the issue that user may run other jobs which puts cores
offline, bringing up all cores online so tool can run cpuid to extract
all available cores information.
Tracked-On: #7119
Signed-off-by: Yang,Yu-chu <yu-chu.yang@intel.com>
Reviewed-by: Junjie Mao <junjie.mao@intel.com>
Acked-by: Anthony Xu <anthony.xu@intel.com>
In function mk_dir(), handler 'dir' returned
by opendir() is not released before function
return and results in memory leakage.
Tracked-On: #7098
Signed-off-by: Yonghua Huang <yonghua.huang@intel.com>
In 10-processors.py, level_types is an unordered dictionary from level type
encodings to human-readable terms, not a list. It is thus illegal to access the
highest level using the `[-1]` notation.
This patch fixes this by calculating the highest level in level_types and use
that level as the index instead.
Tracked-On: #6689
Fixes: b5b22bf98 ("board_inspector: avoid hard-coded topo level names")
Signed-off-by: Junjie Mao <junjie.mao@intel.com>
Manually specifying MAX_IOAPIC_NUM and MAX_IOAPIC_LINES in a scenario XML
is still usable as a workaround on boards without IOAPIC (e.g. emulated
ones) or when the board inspector does not collect IOAPIC information
correctly.
This patch takes user-specified values of those items if
provided. Otherwise the automatically calculated values will be used
instead, as is done today.
Tracked-On: #6986
Signed-off-by: Junjie Mao <junjie.mao@intel.com>
Today the config.h/config.mk generators do not put a newline after an entry
without a value, which causes the generated files ill-formed.
This patch fixes the issue by putting a newline at the end of an entry
unconditionally.
Tracked-On: #6355
Signed-off-by: Junjie Mao <junjie.mao@intel.com>
In XSL a string variable can be nul, the empty string or a non-empty
string. While `$var != ''` ensures that `var` is a non-empty string,
`$var = ''` is true only when `var` holds the empty string. In other words,
`$var = ''` and `$var != ''` can be both false if `var` evaluates to nul.
As a result, the config.h/config.mk generating scripts does not use the
given default value if the explicit value is nul.
This patch fixes the incorrect comparison in the XSL scripts that handles
default values, so that the default value behaves as expected.
Tracked-On: #6355
Signed-off-by: Junjie Mao <junjie.mao@intel.com>
The board configuration generator assumes all PCI BARs are programmed when
the board inspector runs, which is sometimes not the case. Unprogrammed
BARs will be reported using plain texts wrapped by angle brackets in the
output of `lspci`.
This patch gracefully skips the unprogrammed BARs and continues when
parsing the output of `lspci`, in order to avoid build-time failures due to
unprogrammed BARs.
Tracked-On: #6355
Signed-off-by: Junjie Mao <junjie.mao@intel.com>
While the terms to refer to CPU topology levels are originally borrowed
from MSR listings in SDMs (i.e. thread, core, die and package), we now
align the topology information with the CPUID topology enumeration
leaf (which does not have a "package" level) in the
implementation. However, some hard-coded level types are not updated
accordingly, leading to strangely organized nodes in the generated board
XML.
This patch refines the CPU extractor of the board inspector by removing
such hard code. Also the XPATH counting the total number of threads is
refined to tolerate the variance of reported intermediate levels.
Tracked-On: #6689
Signed-off-by: Junjie Mao <junjie.mao@intel.com>
