Currently, if board.xml and scenario.xml are specified,
the variable $(BOARD) and $(SCENARIO) are path, but the
parameters passed in to generate debian package should be
board and scenario name. So add a conversion function to
ensure that the function passed to the generated debian
package is board and scenario name.
Tracked-On: #6688
Signed-off-by: Hu Fenglin <fenglin.hu@intel.com>
When CPUID executes with EAX set to 02H, the processor returns information about cache and TLB information.
This information is percpu related, and should be obtained directly from the physical cpu.
Tracked-On: #6931
Signed-off-by: Tw <wei.tan@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>
When CPUID executes with EAX set to 1AH, the processor returns information about hybrid capabilities.
This information is percpu related, and should be obtained directly from the physical cpu.
Tracked-On: #6899
Signed-off-by: Wen <qian.wen@intel.com>
Signed-off-by: Tw <wei.tan@intel.com>
Rename get_sos_wakeup_reason to get_service_vm_wakeup_reason in acrn manager.
rename set_sos_timer to set_service_vm_timer.
rename sos_pci_devs to service_vm_pci_devs.
rename gen_sos_vm to gen_service_vm.
rename sos_used_bdf to service_vm_used_bdf.
rename is-sos-vm to is-service-vm.
rename sos_mmio_range to service_vm_mmio_range.
rename sos_guest_flags to service_vm_guest_flags.
rename get_sos_vmid to get_service_vmid.
rename sos_cpu_affinity to service_vm_cpu_affinity.
rename allocation_sos_vm_node to allocation_service_vm_node.
rename cpus_for_sos to cpus_for_service_vm.
rename sos_vm_cpus to service_vm_cpus.
rename get_sos_vuart_settings to get_service_vm_vuart_settings.
rename sos_extend_all_cpus to service_vm_extend_all_cpus.
rename sos_vm_id to servier_vm_id.
rename sos_vm_index to service_vm_index.
rename sos_vm_num to service_vm_num.
re-generate the board and scenario code by the new python and xml.
delete tdf8532 kernel module in python and shell script.
Tracked-On: #6744
Signed-off-by: Liu Long <long.liu@linux.intel.com>
Currently, in RTVM with multi vCPUs, lapic pass through is
configured, each vCPU works in x2apic mode. When one vCPU sends
IPI to all other vCPUs, writes ICR register with the value
0x00000000000c00f8, this ICR writting will trigger VM exit,
the hypervisor passes destination shorthand field 11B (All Excluding
Self) in the virtual ICR value into physical ICR value,
this IPI will be sent to each physical CPU core
in the platform according to 10.6.1 Interrupt Command Register (ICR),
Vol 3, SDM. This will cause one User VM with lapic pass through
configuration can send IPI with any vector (such as NMI or reboot
vector) to other VM, this interference may cause other VM hang.
In this patch, set "Destination Shorthand" field of the
ICR value as 00B (No Shorthand) since the emulation is done
through sending IPI to each VCPU in dmask one by one.
Tracked-On: #6908
Signed-off-by: Xiangyang Wu <xiangyang.wu@intel.com>
When a new thread is wakeup and added to runqueue, it will get the
smallest avt (svt) from runqueue to initiate its avt. But when all
threads are blocked, and removed from the runqueue, the svt is 0. The
new thread will get avt=0. avt=0 means very high priority, and can run
for a very long time until it catchup with other thread's avt in
runqueue.
To fix it, adjust the new thread's avt when there is another thread in
runqueue.
Tracked-On: #6888
Signed-off-by: Conghui <conghui.chen@intel.com>
The lifecycle manager in Service VM checks sync message every
5 seconds, so need add 5 seconds to sync delay for WaaG
lifecycle manager to avoid to unnecessary sync resending.
Tracked-On: #6652
Signed-off-by: Xiangyang Wu <xiangyang.wu@intel.com>
Currently, the lifecycle manager in Service VM will issue poweroff
command to shutdown service VM when User VM fails to shutdown. The
post-launched RTVM with virtio device may hang after Service VM
shutdown.
In this patch, the lifecycle manager in Service VM will stop to
shut down service VM when User VM fails to shutdown. Users have
chance to fix User VM shutdown failure (such as forcelly shutdown
User VM or execute poweroff command in User VM), after the failure
is fixed, user can trigger S5 in service VM to shutdown platform
gracefully.
Tracked-On: #6652
Signed-off-by: Xiangyang Wu <xiangyang.wu@intel.com>
If a file descriptor being monitored by epoll_wait is closed
in another thread, the result is unspecified. So add all mevents
removed in other threads to delete list. And drain the list in
the dispatch mevent thread.
Tracked-On: #6877
Signed-off-by: Yuanyuan Zhao <yuanyuan.zhao@linux.intel.com>
Acked-by: Wang, Yu1 <yu1.wang@intel.com>
Variable should not have a prefix of '_' per MISRA C standard. The patch
removes the prefix for _ld_ram_start and _ld_ram_end.
Tracked-On: #6885
Signed-off-by: Victor Sun <victor.sun@intel.com>
Acked-by: Anthony Xu <anthony.xu@intel.com>
In previous commit df7ffab441
the CONFIG_HV_RAM_SIZE was removed and the hv_ram_size was calculated in
link script by following formula:
ld_ram_size = _ld_ram_end - _ld_ram_start ;
but _ld_ram_start is a relative address in boot section whereas _ld_ram_end
is a absolute address in global, the mix operation cause hv_ram_size is
incorrect when HV binary is relocated.
The patch fix this issue by getting _ld_ram_start and _ld_ram_end respectively
and calculated at runtime.
Tracked-On: #6885
Signed-off-by: Victor Sun <victor.sun@intel.com>
Acked-by: Anthony Xu <anthony.xu@intel.com>
this patch validates input of virtio_console_control_tx()
function to avoid potential progream crash with malicious
input from guest.
Tracked-On: #6851
Signed-off-by: Yonghua Huang <yonghua.huang@intel.com>
This patch is to eliminate kernel error msgs:
'ACPI Error: AE_NOT_FOUND, Evaluating _PSS'
This is caused by missing of _PSS table in guest ACPI. It would
happen when pstate is not injected to the guest.
Kernel ACPI pstate driver first probes
_PPC(performance capabilites) and _PCT(performance control)
in ACPI. If they exist, then it loads the _PSS(performance state).
If _PPC/_PCT are presented while _PSS is missing, it prints
the error msg.
In acrn-dm, _PPC/_PCT are hard-coded to all vCPUs, while _PSS
are constructed with the pCPUs' pstate data. This is base on
assumption that all VMs can have pstate.
Now the pstate is given to VM only when the VM is not sharing
any CPU(and no RTVM is setup in the scenario).
When the VM doesn't have pstate, the hypercall will return px_cnt=0,
and the _PSS is not constructed. In this case, _PPC/PCT should not
be injected, too.
Tracked-On: #6848
Signed-off-by: Zhou, Wu <wu.zhou@intel.com>
Acked-by: Anthony Xu <anthony.xu@intel.com>
Acked-by: Eddie Dong <eddie.dong@Intel.com>
Avoid failing hypercalls by returning 0 for empty PX and CX tables on
HC_PM_GET_CPU_STATE/PMCMD_GET_PX_CNT and
HC_PM_GET_CPU_STATE/PMCMD_GET_CX_CNT.
Tracked-On: #6848
Signed-off-by: Helmut Buchsbaum <helmut.buchsbaum@opensource.tttech-industrial.com>
Acked-by: Anthony Xu <anthony.xu@intel.com>
Acked-by: Eddie Dong <eddie.dong@Intel.com>
replace MAX_VMNAME_LEN with MAX_VM_NAME_LEN
to keep sync with HV and devicemodel
Tracked-On: #6685
Signed-off-by: Mingqiang Chi <mingqiang.chi@intel.com>
-- refine get_vmid_by_name, return the first matching vm_id
-- the GUEST_FLAG_STATIC_VM is added to identify the static or
dynamic VM, the offline tool will set this flag for
all the pre-defined VMs.
-- only clear name field for dynamic VM instead of clear entire
vm_config
-- only set DM owned guest_flags in hcall_create_vm
-- add check dynamic flag in get_unused_vmid
Tracked-On: #6685
Signed-off-by: Mingqiang Chi <mingqiang.chi@intel.com>
The HV will be built failed with below compiler message:
common/efi_mmap.c: In function ‘init_efi_mmap_entries’:
common/efi_mmap.c:41:11: error: unused variable ‘efi_memdesc_nr’
[-Werror=unused-variable]
uint32_t efi_memdesc_nr = uefi_info->memmap_size / uefi_info->memdesc_size;
^~~~~~~~~~~~~~
cc1: all warnings being treated as errors
The root cause is ASSERT() api is for DEBUG only so efi_memdesc_nr is not used
in RELEASE code.
The patch fix this issue by removing efi_memdesc_nr declaration;
Tracked-On: #6834
Signed-off-by: Victor Sun <victor.sun@intel.com>
GUEST_FLAG_SECURITY_VM only for pre-launch vm, should not be
programmed by device model, so we remove it in DM_OWNED_GUEST_FLAG_MASK.
Tracked-On: #6366
Signed-off-by: Kunhui-Li <kunhuix.li@intel.com>
GET_PLATFORM_INFO interface is not supported from
hypervisor any more, remove this wrapper API in DM.
Tracked-On: #6724
Signed-off-by: Yonghua Huang <yonghua.huang@intel.com>
Acked-by: Wang, Yu1 <yu1.wang@intel.com>
- GET_PLATFORM_INFO will not be supported in ACRN.
- This patch removes the dependency on this interface
to build virtual ACPI table of post-launched VM:
Instead of calling 'vm_get_config()' to get local
APIC IDs, virtual ACPI builder calls new wrapper API
'lapcid_from_pcpuid()'.
Tracked-On: #6724
Signed-off-by: Yonghua Huang <yonghua.huang@intel.com>
Acked-by: Wang, Yu1 <yu1.wang@intel.com>
- GET_PLATFORM_INFO ioctl will not be supported by ACRN.
- This patch removes the dependency on above interface in
vRTCT ACPI table building:
1) Get cache hierarchy information (threads sharing shift)
through 'cpuid' instruction (leaf 0x04) instead of
call vm_get_config() API.
2) Get local APIC IDs of guest VM CPUs through new wrapper
API 'plapicid_from_pcpuid()'.
Tracked-On: #6724
Signed-off-by: Yonghua Huang <yonghua.huang@intel.com>
Acked-by: Wang, Yu1 <yu1.wang@intel.com>
GET_PLATFORM_INFO interface will be removed from ACRN hypervisor,
Service VM /ACRN-DM and other userland applications shall not use it any more.
For dm the lapic id was get from acrn_platform_info.
So in version 2.7, SOS madt table is parsed to get the lapic id for each pcpu.
There has an assumption. The SOS owned pcpu starts from physical cpu 0,
otherwise SOS doesn't know the mapping relationship between its vcpu0
and pcpu_id.
Tracked-On: #6724
Signed-off-by: Yuanyuan Zhao <yuanyuan.zhao@linux.intel.com>
Acked-by: Wang, Yu1 <yu1.wang@intel.com>
ACRN get messy default values for some CAT CBM MSRs,
these unexpected default value will result in TCC
Software SRAM initializes crash.
This patch resets above error default values before calling
CRL(cache reserve library) ABI initializaion function.
Tracked-On: #6780
Signed-off-by: Yonghua Huang <yonghua.huang@intel.com>
Rename function name to get-common-clos-count and refine the
calculation. Some platforms have odd clos counts. Using "floor" to get
integer numbers for macros.
Tracked-On: #6755
Signed-off-by: Yang,Yu-chu <yu-chu.yang@intel.com>
Commit cbf3825 "hv: Pass-through IA32_TSC_AUX MSR to L1 guest"
lets guest own the physical MSR IA32_TSC_AUX and does not handle this MSR
in the hypervisor.
If multiple vCPUs share the same pCPU, when one vCPU reads MSR IA32_TSC_AUX,
it may get the value set by other vCPUs.
To fix this issue, this patch does:
- initialize the MSR content to 0 for the given vCPU, which is consistent with
the value specified in SDM Vol3 "Table 9-1. IA-32 and Intel 64 Processor
States Following Power-up, Reset, or INIT"
- save/restore the MSR content for the given vCPU during context switch
v1 -> v2:
* According to Table 9-1, the content of IA32_TSC_AUX MSR is unchanged
following INIT, v2 updates the initialization logic so that the content for
vCPU is consistent with SDM.
Tracked-On: #6799
Signed-off-by: Shiqing Gao <shiqing.gao@intel.com>
Reviewed-by: Zide Chen <zide.chen@intel.com>
Acked-by: Eddie Dong <eddie.dong@intel.com>
1. ensure cpu affinity in launch xml is subset of its scenario settings.
2. cpu_affinity is a must have parameter for acrn_dm, if the user did not provide cpu affinity in launch xml, load it from scenario settings.
Tracked-On: #6724
Signed-off-by: Weiyi Feng <weiyix.feng@intel.com>
when we add two or more virtio block devices and rename them for a VM
in launch xml files, then generate launch scripts through the UI.
After ensuring the devices names are different, we repeat these operations
continuously. Finally we will get the duplicate PT_SLOT value for
different virtio block devices in a launch script. This is wrong.
After user click the "Generate Launch Script" button to generate the
launch scripts every time, we will reset PT_SLOT variate to fix the issue.
Tracked-On: #6767
Signed-off-by: Kunhui-Li <kunhuix.li@intel.com>
When some communication vuarts are configured in hypervisor
scenario files, serial.conf will be generated and should be
installed in the etc folder of service VM.
In this patch, update deb configuration file to install
serial.conf into etc folder of service VM.
Tracked-On: #6652
Signed-off-by: Xiangyang Wu <xiangyang.wu@intel.com>
COM2 in user VM is used by acrnctl to do user VM
shutdown.
In this patch series, COM2 in user VM is used as uart
channel device for new lifecycle manager. So acrnctl
can't be used to do user VM shutdown.
This patch provides a script to user to do guest
shutdown in service VM. Lifecycle manager in service
VM will send user VM shutdown command to user VM directly.
TODO: When user VM shutdown is supported in libvirt, this
script will be removed.
Tracked-On: #6652
Signed-off-by: Xiangyang Wu <xiangyang.wu@intel.com>
Reviewed-by: fei1.li@intel.com
The old lifecycle manager will not be used and remove it,
update Makefile to build and install new lifecycle manager.
v1-->v2:
Copy s5_trigger_win.py into build directory;
Add compiler parameter (-lwsock32) to build
life_mngr_win.c.
v3-->v4:
Update Makefile for s5_trigger_linux.py
v4-->v5:
Inlcude paths.make at the top-level folder.
Generate life_mngr.service according to bindir
variable.
Tracked-On: #6652
Signed-off-by: Xiangyang Wu <xiangyang.wu@intel.com>
Reviewed-by: fei1.li@intel.com
Refine lifecycle manager for windows VM since new commands
are introduced in the lifecycle manager.
v1-->v2:
Implement socket service to receive system shutdown
request command through local host socket.
Add one python script to trigger system shutdown
request in windows VM.
v2-->v3:
Update log message.
v3-->v4:
Support guest shutdown.
v4-->v5:
Update command name.
v5-->v7:
Add resend message logic.
Tracked-On: #6652
Signed-off-by: Xiangyang Wu <xiangyang.wu@intel.com>
Reviewed-by: fei1.li@intel.com
Provide a script to user to trigger system shutdown request
in service VM or user VM.
System shutdown logic have been integrated in the lifecycle
manager, only need to send system shutdown request command
through unix domain socket in this script.
v3-->v4:
Rewirte system shutdown trigger script using python
since phython script have better portability.
v4-->v5:
Update command name and copyright header.
Tracked-On: #6652
Signed-off-by: Xiangyang Wu <xiangyang.wu@intel.com>
Reviewed-by: fei1.li@intel.com
In the monitor module, the following functions are implemented:
-Implement main function of lifecycle manager, it will initialize
other module through invoking related interfaces, wait threads in
the uart module exit.
Todo:
Will refine the name of
init_socket_server_and_shutdown_commands,
init_uart_channel_devs_and_shutdown_commands.
v1-->v2:
Add comments in c file.
v2-->v3:
Update some log message.
v3-->v6:
Update monitior module for modulization.
v6-->v7:
Regiester command handler for ACK receiving timeout case
in user VM.
Tracked-On: #6652
Signed-off-by: Xiangyang Wu <xiangyang.wu@intel.com>
Reviewed-by: fei1.li@intel.com
In the log module, the following functions are implemented:
- open_log
Provide one interface to open log file;
-close_log
Provide one interface to close log file;
- LOG_PRINTF LOG_WRITE
Provide interfaces to write log message to log file.
v1-->v3:
Add timestamp in the log message.
Tracked-On: #6652
Signed-off-by: Xiangyang Wu <xiangyang.wu@intel.com>
Reviewed-by: fei1.li@intel.com
In the config module, the following functions are implemented:
- load_config
Provide one interface to load lifecycle manager configuration
from config file (life_mngr.conf), in the config file, user
can specify the VM type of VM which lifecycle manager will run
in, the VM name, the communication device name, and
the device name in service VM which is used to communicate with
the VM which is allowed to send system shutdown request.
- check_dir
Check folder exist or not, if not, create the folder
- get_allow_s5_config
Get the name of the device which is allowed to trigger
system shutdown.
v1-->v2:
Add comments in head file.
v2-->v3:
Update some log message.
v3-->v6:
Simply configuration item parsing logic.
Add get_allow_s5_config interface.
Tracked-On: #6652
Signed-off-by: Xiangyang Wu <xiangyang.wu@intel.com>
Reviewed-by: fei1.li@intel.com
In this module, the following functions are implemented:
For uart channel of service VM, implement handlers for sync command,
system shutdown request command, acked poweroff command, vm poweroff
timeout command.
The uart channel commands of service VM and related actions are
described below:
Command Actions
---------------------------------------------------------------
sync Send acked sync
command to uart in user VM
----------------------------------------------------------------
system shutdown If this request is valid, send
request poweroff command to each
connected user VM through uart,
Enable message reseding.
----------------------------------------------------------------
acked poweroff Remove uart of user VM from
command connection list, stop message polling,
if connection list is empty, will
wait for user VM status through check
ACRN DM process instance, if all user
VMs are shutdown, then shutdown
service VM
---------------------------------------------------------
ACK timeout If it is timeout of receiving poweroff ACK,
the action is similar to acked poweroff.
Ohterwise, just disable message reseding.
---------------------------------------------------------
For uart channel of user VM, implement handlers for acked sync
command, poweroff command, acked system shutdown request command,
ACK timeout command.
The uart channel commands of user VM and related actions are
described below:
Command Actions
--------------------------------------------------------
acked sync Print log message
--------------------------------------------------------
poweroff Disconnect uart channel,
exit message polling, close
unix domain socket, shutdown user VM
--------------------------------------------------------
user VM shutdown Exit message polling, close
unix domain socket, shutdown user VM
--------------------------------------------------------
acked system Print log message
shutdown request
-------------------------------------------------------
ACK timeout Disconnect uart channel,
exit message polling, close
unix domain socket
---------------------------------------------------------
For socket server in each VM, implement handler for system
shutdown request.
In user VM, forward this command to service VM through uart.
In service VM, send poweroff command to each connected user VM
through uart.
Implement handler for user VM shutdown request in socket server of
service VM, send user VM shutdown command to user VM which is
specified in the user VM shutdown request message.
v1-->v2:
Add comments in c file and head file.
v2-->v3:
Update commit message about allow s5 command and update
some log message.
v3-->v4:
Guest shutdown support.
v4-->v5:
Update command name.
v5-->v6:
Move uart channel operations into uart channel module.
v6-->v7:
Set resend requirement for some commands.
Add ACK receiving timeout handler for user VM.
v7-->v8:
Update message reseding enable, this interface will be
called before sending message.
v8-->v9:
Use strlen to calculate the length of string, this will be
the parameter of send_message_by_uart.
Tracked-On: #6652
Signed-off-by: Xiangyang Wu <xiangyang.wu@intel.com>
Reviewed-by: fei1.li@intel.com
