currently, configurator will fail to generate config_summary.rst file if
user enable the "Cache Allocation Technology" because note function in rstcloth
is replaced by self.doc.note.
So this patch updates function and usage to fix this issue.
fixs: 9c2d0f8 ("config_tools: replace RstCloth library with class.")
Tracked-On: #8422
Signed-off-by: Kunhui-Li <kunhuix.li@intel.com>
Reviewed-by: Junjie Mao <junjie.mao@intel.com>
Set GUEST_FLAG_VHWP guest flag for VMs who own pCPUs. Windows does not
manage CPU frequency in virtualized environment, so this flag is not set
for waag. Platforms without HWP capability should not have this flag
either.
Tracked-On: #8414
Signed-off-by: Wu Zhou <wu.zhou@intel.com>
Reviewed-by: Junjie Mao <junjie.mao@intel.com>
When VHWP enabled, return 0 and px_cnt = 0 on ACRN_PMCMD_GET_PX_CNT,
so that DM will write _CPC to guests' ACPI.
Tracked-On: #8414
Signed-off-by: Wu Zhou <wu.zhou@intel.com>
Reviewed-by: Junjie Mao <junjie.mao@intel.com>
The optional object _CPC declares an interface that allows OSPM to
transition the processor into a performance state based on a continuous
range of allowable values.
It is associated with HWP on intel CPUs. Although Linux intel_pstate driver
can have its performance managing abilities without _CPC, it may still
need this _CPC table to implement some features such as providing the kernel
multi-core scheduler with core priority info.
As currently we are giving guests a vHWP interface for the multi-core
scheduler, this patch adds _CPC to the guest ACPI. _CPC is written only
when the hypervisor decides the guest should have vHWP, using the
existing pm hypercall ACRN_PMCMD_GET_PX_CNT. The idea is:
- If the VM supports vHWP, then the guest is having continuous p-state.
Thus it doesn't have a specific px_cnt. The hypercall returns success
and px_cnt = 0.
- If the VM's p-state is hidden or hv doesn't have its p-state info,
the hypercall returns fail.
Tracked-On: #8414
Signed-off-by: Wu Zhou <wu.zhou@intel.com>
Reviewed-by: Jian Jun Chen <jian.jun.chen@intel.com>
Changes made by this patch includes:
1. Emulate HWP and pstate MSRs/CPUIDs. Those are exposed to guest when
the GUEST_FLAG_VHWP is set:
- CPUID[6].EAX[7,9,10]: MSR_IA32_PM_ENABLE(enabled by hv, always read
1), MSR_IA32_HWP_CAPABILITIES, MSR_IA32_HWP_REQUEST,
MSR_IA32_HWP_STATUS,
- CPUID[6].ECX[0]: MSR_IA32_MPERF, MSR_IA32_APERF
- MSR_IA32_PERF_STATUS(read as base frequency when not owning pCPU)
- MSR_IA32_PERF_CTL(ignore writes)
2. Always hide HWP interrupt and package control MSRs/CPUIDs:
- CPUID[6].EAX[8]: MSR_IA32_HWP_INTERRUPT(currently ACRN is not able
to deliver thermal LVT virtual interrupt to guests)
- CPUID[6].EAX[11,22]: MSR_IA32_HWP_REQUEST_PKG, MSR_IA32_HWP_CTL
Tracked-On: #8414
Signed-off-by: Wu Zhou <wu.zhou@intel.com>
Reviewed-by: Junjie Mao <junjie.mao@intel.com>
Currently CPU frequency control is hidden to guests, and controlled
by hypervisor. While it is sufficient in most cases, some guest OS may
still need CPU performance info to make multi-core scheduling decisions.
This is seen on Linux kernel, which uses HWP highest performance level
as CPU core's priority in multi-core scheduling (CONFIG_SCHED_MC_PRIO).
Enabling this kernel feature could improve performance as single thread
workloads are scheduled on the highest performance cores. This is
significantly useful for guests with hybrid cores.
The concept is to expose performance interface to guest who exclusively
owns pCPU assigned to it. So that Linux guest can load intel_pstate
driver which will then provide the kernel with each core's schedule
priority.
Intel_pstate driver also relies on CONFIG_ACPI_CPPC_LIB to implement
this mechanic, this means we also need to provide ACPI _CPC in DM.
This patch sets up a guest flag GUEST_FLAG_VHWP to indicate whether
the guest can have VHWP feature.
Tracked-On: #8414
Signed-off-by: Wu Zhou <wu.zhou@intel.com>
Reviewed-by: Junjie Mao <junjie.mao@intel.com>
This patch refactors partition_epc() to make the code easier to
understand, also fixes the maybe-uninitialized warning for gcc-13.
Initializing 'vm_config' to get_vm_config(0) is okay here as scenario
validator ensures CONFIG_MAX_VM_NUM to be always larger than 0.
Tracked-On: #8413
Signed-off-by: Jiaqing Zhao <jiaqing.zhao@linux.intel.com>
This patch emulates the PLATFORM_INFO MSR in hypervisor to make it
only visible to Service VM, and only processor ratios (bit 15:8,
47:40 and 55:48) and sample part bit (27) are exponsed. This is
intended to prevent Service VM from changing processor parameters
like turbo ratio.
Tracked-On: #8406
Signed-off-by: Jiaqing Zhao <jiaqing.zhao@linux.intel.com>
add customized parameters in the launch script generation logic.
If user specify customized parameters in configurator, we will
append them to the dm parameters for launching VMs.
Tracked-On: #8408
Signed-off-by: Kunhui-Li <kunhuix.li@intel.com>
Reviewed-by: Junjie Mao <junjie.mao@intel.com>
`modprobe msr` will be invoked if no /dev/cpu/#/msr is found.
If the "msr" module is built-in that check will pass and
no `modprobe msr` should be invoked.
fixes: aeab5a2 ("config_tools: update loading msr driver logic")
Tracked-On: #8410
Signed-off-by: Kunhui-Li <kunhuix.li@intel.com>
Reviewed-by: Junjie Mao <junjie.mao@intel.com>
Linker option "-z,relro" will be automatically added to LDFLAGS when
building debian packages. Since x86_64-w64-mingw32-ld does not support
"-z" option, it causes failure in building the life_mngr windows exe.
And compared with ld, only "-z" option is not supported. This patch
fixes it by skipping the windows build of life_mngr if LDFLAGS contains
any "-z" option.
Tracked-On: #8399
Signed-off-by: Jiaqing Zhao <jiaqing.zhao@linux.intel.com>
Reviewed-by: Junjie Mao <junjie.mao@intel.com>
gvt_init_config() may perform out-of-range read on host_config, add
bound check before accessing it.
Tracked-On: #8382
Signed-off-by: Jiaqing Zhao <jiaqing.zhao@linux.intel.com>
Reviewed-by: Jian Jun Chen <jian.jun.chen@intel.com>
9c2d0f8858 ("config_tools: replace RstCloth library with class.")
removes all usage of rstcloth in code, but the rstcloth package is not
removed from acrn-configurator and it will still download dependencies
for rstcloth. This patch simply removes it.
Tracked-On: #8395
Signed-off-by: Jiaqing Zhao <jiaqing.zhao@linux.intel.com>
Reviewed-by: Junjie Mao <junjie.mao@intel.com>
This patch adds -v or --verbose option to build script to enable
verbose output by setting DH_VERBOSE environment variable. This is
useful when debugging build issues.
Tracked-On: #8399
Signed-off-by: Jiaqing Zhao <jiaqing.zhao@linux.intel.com>
Reviewed-by: Junjie Mao <junjie.mao@intel.com>
Currently, Configurator load fail because it needs to download
RstCloth related packages. This patch defines Doc class to replace
RstCloth library to fix this issue.
Tracked-On: #8395
Signed-off-by: Kunhui-Li <kunhuix.li@intel.com>
Reviewed-by: Junjie Mao <junjie.mao@intel.com>
If no TURBO_RATIO_LIMIT and TURBO_ACTIVATION_RATIO MSRs info
on target, board inspector will crash because of the IOError exception.
This patch captures the IOError exception to handle this error.
Tracked-On: #8380
Signed-off-by: Kunhui-Li <kunhuix.li@intel.com>
Reviewed-by: Junjie Mao <junjie.mao@intel.com>
If we use 5.15 kernel on rpl-p platform, some iomem info dumped with
non-ascii characters, then tool will raise a decode error.
So this patch filters non-ascii characters to handle this error.
Tracked-On: #8388
Signed-off-by: Kunhui-Li <kunhuix.li@intel.com>
Reviewed-by: Junjie Mao <junjie.mao@intel.com>
The new board xml have add a "module" node under the "processors/die"
which has cause an issue when we run the summary, this patch use "//"
to select all "cpu_id" under the "processors".
Tracked-On: #8385
Signed-off-by: Chenli Wei <chenli.wei@intel.com>
Starting from elementpath 4.0.0, the XPath2Parser class no longer has a
SYMBOLS member and completeness checking based on that are removed as
well. This patch updates the elementpath_overlay in the config tools to fix
the compilation issue when using recent elementpath versions.
Tracked-On: #8368
Signed-off-by: Junjie Mao <junjie.mao@intel.com>
LIST_FOREACH() doesn't allow var to be removed or freed within the
loop, but c_dev is freed inside the loop here. gcc 12 also reports
error on it. This patch uses list_foreach_safe() macro instead for
freeing var within the loop safely.
Tracked-On: #8382
Signed-off-by: Jiaqing Zhao <jiaqing.zhao@linux.intel.com>
Reviewed-by: Junjie Mao <junjie.mao@intel.com>
When assigning a physical interrupt to a Post-launched VM, if it has
been assigned to ServiceVM, we should remove that mapping first to reset
ioapic pin state and rte, and build new mapping for the Post-launched
VM.
Tracked-On: #8370
Signed-off-by: Qiang Zhang <qiang4.zhang@linux.intel.com>
Reviewed-by: Junjie Mao <junjie.mao@intel.com>
Acked-by: Eddie Dong <eddie.dong@intel.com>
ptirq_remapping_info records which physical interrupt is mapped to
the virtual interrupt in a VM. As we need to knonw whether a physical
sid has been mapped to a VM and whether a virtual sid in a VM has been
used, there should be two hash tables to link and iterate
ptirq_remapping_info:
- One is used to lookup from physical sid, linking phys_link.
- The other is used to lookup from virtual sid in a VM, linking virt_link
Without this patch, phys_link or virt_link from different
ptirq_remapping_info was linked by one hash list head if they got the
same hash value, as shown in following diagram.
When looking for a ptirq_remapping_info from physical sid, the original
code took all hash list node as phys_link and failed to get
ptirq_remapping_info linked with virt_link and later references
to its members are wrong.
The same problem also occurred when looking for a ptirq_remapping_info
from virtual sid and vm.
---------- <- hash table
|hlist_head| --actual ptirq_remapping_info address
---------- --------- / --used as ptirq_remapping_info
| ... | |phys_link| ___/
---------- --------- --------- / ---------
|hlist_head| -> |phys_link| <-> |virt_link| <-> |phys_link|
---------- --------- --------- ---------
| ... | |virt_link| | other | |virt_link|
---------- --------- | members | ---------
| other | --------- | other |
| members | | members |
--------- ---------
Tracked-On: #8370
Signed-off-by: Qiang Zhang <qiang4.zhang@linux.intel.com>
Reviewed-by: Junjie Mao <junjie.mao@intel.com>
Acked-by: Eddie Dong <eddie.dong@intel.com>
Sphinx version 5 generates <section> instead of <div> for
section headings. This throws off the custom CSS style for numbered
steps so the number icon wasn't showing up. Tweak the custom CSS to
look for both <section> and <div> with class=numbered-step
to handle Sphinx 4 and 5-generated content.
Signed-off-by: David B. Kinder <david.b.kinder@intel.com>
1. add kernel version to menuentry "Ubuntu-ACRN Board Inspector"
2. add kernel version and acrn version to menuentry "Ubuntu with ACRN hypervisor"
Tracked-On:#8359
Signed-off-by: Min Yang <minx.yang@intel.com>
delete building the ACRN kernel for the Service VM,
add some mention when building ACRN hypervisor,
and modify GRUB menuentry choice name.
Signed-off-by: Min Yang <minx.yang@intel.com>
1. update hypervisor and kernel version to release_3.2
2. modify debian package (of ACRN hypervisor and tools) build command
3. delete iasl command lines (substitute: acpica-tools generate iasl in GSG)
4. remove use of USB stick to copy files from development computer to
target (only use scp)
5. remove grub menu after reboot (it will boot automatically.)
Signed-off-by: Min Yang <minx.yang@intel.com>
1. add kernen version to menuentry "Ubuntu-ACRN Board Inspector"
2. add kernel version and acrn version to menuentry
"Ubuntu with ACRN hypervisor"
Please refer to #8357(master) and #8358(release_3.2) for code change.
Signed-off-by: Min Yang <minx.yang@intel.com>
The switch to Ubuntu 22.04 for ACRN v3.2 requires a few changes to the
sample application scripts:
- user _apt cannot access debian packages in user directories, so copy
them to /tmp to install
- HMI-VM image size needs to be bigger (running out of space during
image update)
Tracked-On: #8352
Signed-off-by: David B. Kinder <david.b.kinder@intel.com>
For STANDARD_VM 'own_pcpu' default value is 'n'. But RTVM always own pCPUs.
So check 'own_pcpu' of RTVM and make sure it's set to 'y'.
Tracked-On: #8253
Signed-off-by: Yuanyua Zhao <yuanyuan.zhao@linux.intel.com>
Reviewed-by: Junjie Mao <junjie.mao@intel.com>
* install all Debian packages from /tmp (not ~/acrn-work)
* remove use of USB stick to copy files between dev and target (use scp)
* fix kernel version numbers (5.15.44 to 5.15.77)
* update parameters for debian_build.sh
* remove display of GRUB menu (ACRN now takes over target and
automatically boots into hypervisor then Service VM
Signed-off-by: David B. Kinder <david.b.kinder@intel.com>
When searching for scenario XMLs that are saved under the same directory as
a board XML, debian/rules uses the wildcard `*` which includes other
non-XML files. That causes some non-XML files to be considered as scenario
XMLs as well and will cause build-time errors when the build system
attempts to parse them as XMLs.
Change the wildcard expression to `*.xml` to restrict the found files to be
XML.
Tracked-On: #8344
Signed-off-by: Junjie Mao <junjie.mao@intel.com>
The variables BOARDLIST and SCENARIOLIST serve as filters of XMLs that are
found under the user-given directories, and there is no need to assume any
filter if a user does not specify that explicitly.
Update the default filters to none so that all found XMLs will be used if a
user does not state otherwise.
Tracked-On: #8246
Signed-off-by: Junjie Mao <junjie.mao@intel.com>
Remove the common interface for sbuf setup, as it is not accept by
kernel side. Instead, use dedicate setup function for asyncio to init
its sbuf.
Tracked-On: #8209
Signed-off-by: Conghui <conghui.chen@intel.com>
Acked-by: Wang, Yu1 <yu1.wang@intel.com>
Update security advisory for v3.0.2 release.
Signed-off-by: Yonghua Huang <yonghua.huang@intel.com>
Signed-off-by: David B. Kinder <david.b.kinder@intel.com>
The division operator in XPATH is `div`, not `/`. Fix it in an error
message of the assertions.
Tracked-On: #6690
Signed-off-by: Junjie Mao <junjie.mao@intel.com>
Commit ce6500893 ("board_inspector: use executables found under system
paths") unintendedly changes how acpi.py invokes `rdmsr` and that impacts
the type of the execution results and causes incompatible method calls when
the results are parsed.
Convert the invocation back to shell-style.
Fixes: ce6500893 ("board_inspector: use executables found under system paths")
Tracked-On: #8315
Signed-off-by: Junjie Mao <junjie.mao@intel.com>
Kunhui-Li