add 2 MACROs: DEFAULT_PCI_MMCFG_START_BUS and DEFAULT_PCI_MMCFG_END_BUS
in platform_acpi_info.h.
Tracked-On: #5233
Signed-off-by: Shixiong Zhang <shixiongx.zhang@intel.com>
Acked-by: Victor Sun <victor.sun@intel.com>
Per PCI Firmware Specification Revision 3.0, 4.1.2. MCFG Table Description:
Memory Mapped Enhanced Configuration Space Base Address Allocation Structure
assign the Start Bus Number and the End Bus Number which could decoded by the
Host Bridge. We should not access the PCI device which bus number outside of
the range of [Start Bus Number, End Bus Number).
For ACRN, we should:
1. Don't detect PCI device which bus number outside the range of
[Start Bus Number, End Bus Number) of MCFG ACPI Table.
2. Only trap the ECAM MMIO size: [MMCFG_BASE_ADDRESS, MMCFG_BASE_ADDRESS +
(End Bus Number - Start Bus Number + 1) * 0x100000) for SOS.
Tracked-On: #5233
Signed-off-by: Li Fei1 <fei1.li@intel.com>
Acked-by: Eddie Dong <eddie.dong@intel.com>
compile ACPI tables for pre-launched VMs to one binary when pre-build
hypervisor.
Tracked-On: #5266
Signed-off-by: Shuang Zheng <shuang.zheng@intel.com>
Acked-by: Victor Sun <victor.sun@intel.com>
use offline tool to generate one binary of ACPI tables for pre-launched
VMs, then load the ACPI binary into guest physical memory as grub module.
Users can configure the resources or devices like TPM2 for the
pre-launched VM from sceanrio XMLs or UI, and the offline tool will
generate ASL code of the ACPI tables with the configured resources or
devices, then compile the ASL code to one binary when building ACRN.
Tracked-On: #5266
Signed-off-by: Shuang Zheng <shuang.zheng@intel.com>
Acked-by: Victor Sun <victor.sun@intel.com>
The old method of build pre-launched VM vacpi by HV source code is deprecated,
so remove related source code;
Tracked-On: #5266
Signed-off-by: Victor Sun <victor.sun@intel.com>
Acked-by: Eddie Dong <eddie.dong@intel.com>
Previously we use a pre-defined structure as vACPI table for pre-launched
VM, the structure is initialized by HV code. Now change the method to use a
pre-loaded multiboot module instead. The module file will be generated by
acrn-config tool and loaded to GPA 0x7ff00000, a hardcoded RSDP table at
GPA 0x000f2400 will point to the XSDT table which at GPA 0x7ff00080;
Tracked-On: #5266
Signed-off-by: Victor Sun <victor.sun@intel.com>
Signed-off-by: Shuang Zheng <shuang.zheng@intel.com>
Acked-by: Eddie Dong <eddie.dong@intel.com>
Previously the ACPI table was stored in F segment which might not be big
enough for a customized ACPI table, hence reserve 1MB space in pre-launched
VM e820 table to store the ACPI related data:
0x7ff00000 ~ 0x7ffeffff : ACPI Reclaim memory
0x7fff0000 ~ 0x7fffffff : ACPI NVS memory
Tracked-On: #5266
Signed-off-by: Victor Sun <victor.sun@intel.com>
Acked-by: Eddie Dong <eddie.dong@intel.com>
EHL PSE TSN GbE driver is default set to use 32bit of dma addressing.
net: stmmac: configure PSE Gbe to 32bit dma addressing
https://github.com/intel/linux-intel-lts/commit/011c8f
When VM has more than 4GB physical memory, Linux kernel uses the bounce
buffers (swiotlb) to translate kernel data in 64bit memory to 32bit
range for the sake of the DMA because iommu is not available. The
default swiotlb value 32768 is insufficient to support two PSE TSN GbEs
at the same time. Increase the value to 131072 otherwise two GbEs can't
link up.
Tracked-On: #5243
Signed-off-by: Toshiki Nishioka <toshiki.nishioka@intel.com>
Acked-by: Victor Sun <victor.sun@intel.com>
When creating a new launch setting, config app can't fine the
scenario config which caused the app failed to load the new
launch setting.
Tracked-On: #5282
Signed-off-by: Shuang Zheng <shuang.zheng@intel.com>
Reviewed-by: Victor Sun <victor.sun@intel.com>
Build and install 'acrnctl' for regardless of the RELEASE value. Kata Containers
depends on 'acrnctl' in order to work correctly and we therefore need to make
sure that it is built and installed correctly regardless of whether this is a
debug or a release build.
Tracked-On: #4940
Signed-off-by: Geoffroy Van Cutsem <geoffroy.vancutsem@intel.com>
Add a new scenario xml file for EHL which is derived from hybrid_rt for
validation of certain passthru devices in prelaunched RTVM. Because the
configuration requires to disable GPIO support for SOS VM, it should
not be merged into the standard hybrid_rt scenario. According to this
change, remove the SCI passthru setting from existing hybrid_rt since
from now on hybrid_rt_fusa should be used for SCI passthru test.
Tracked-On: #5278
Signed-off-by: Toshiki Nishioka <toshiki.nishioka@intel.com>
Reviewed-by: Junjie Mao <junjie.mao@intel.com>
Acked-by: Victor Sun <victor.sun@intel.com>
hv: vpci: Add 0x45, which is the high-byte of device id of EHL,
to the enumeration array in vhostbridge.c. This is to fix the
problem that PCIe extended capabilities like SR-IOV cannot be
used on EHL.
Tracked-On: #5256
Signed-off-by: Qian Wang <qian1.wang@intel.com>
minor_ver should be le 2 when major_ver == 2.
Tracked-On: #5276
Signed-off-by: Shuang Zheng <shuang.zheng@intel.com>
Acked-by: Victor Sun <victor.sun@intel.com>
Previously the min load_addr for HV image is hard coded to 0x10000000 when
CONFIG_RELOC is enabled, now use CONFIG_HV_RAM_START as its prefer minimum
address like setting of CONFIG_PHYSICAL_START do in Linux kernel.
With this patch, we can offload the CONFIG_HV_RAM_START algorithm to
acrn-config or manually set it in scenario XML on some special boards.
Tracked-On: #5275
Signed-off-by: Victor Sun <victor.sun@intel.com>
update the size of ivshmem memory region to [2MB, 512MB].
Tracked-On: #4853
Signed-off-by: Shuang Zheng <shuang.zheng@intel.com>
Reviewed-by: Yuan Liu <yuan1.liu@intel.com>
When HV pass through the P2SB MMIO device to pre-launched VM, vgpio
device model traps MMIO access to the GPIO registers within P2SB so
that it can expose virtual IOAPIC pins to the VM in accordance with
the programmed mappings between gsi and vgsi.
Tracked-On: #5246
Signed-off-by: Toshiki Nishioka <toshiki.nishioka@intel.com>
Reviewed-by: Junjie Mao <junjie.mao@intel.com>
Acked-by: Eddie Dong <eddie.dong@intel.com>
Always use P2SB_ as a prefix for all macro definitions related to P2SB
MMIO passthru. And introduce the new P2SB_VGPIO_DM_ENABLED macro to
indicate the presence of the pre-launched VM which requires the feature.
This macro intends to be used to enclose source files with ifdef where
macros defined by config-tool being used which are available only when
the feature is enabled. It is required to avoid causing compilation
errors when users build HV without enabling the feature.
Tracked-On: #5246
Signed-off-by: Toshiki Nishioka <toshiki.nishioka@intel.com>
Reviewed-by: Junjie Mao <junjie.mao@intel.com>
Acked-by: Eddie Dong <eddie.dong@intel.com>
Add the capability of forwarding specified physical IOAPIC interrupt
lines to pre-launched VMs as virtual IOAPIC interrupts. This is for the
sake of the certain MMIO pass-thru devices on EHL CRB which can support
only INTx interrupts.
Tracked-On: #5245
Signed-off-by: Toshiki Nishioka <toshiki.nishioka@intel.com>
Reviewed-by: Junjie Mao <junjie.mao@intel.com>
Acked-by: Eddie Dong <eddie.dong@intel.com>
Kernel driver and ACPI ASL may access a platform hidden device
thru PIO, e.g., Intel ICH LPC driver. If the access is originated
in SOS or Pre-launched OS, vpci_pio_cfgdata_write/read should support
it.
This commit also reworks vpci_write_cfg/vpci_read_cfg to do the access
check and elimiates the access from post-launched VM (that should be
handled by DM).
Tracked-On: #5257
Signed-off-by: Stanley Chang <stanley.chang@intel.com>
Reviewed-by: Li Fei <fei1.li@intel.com>
Acked-by: Eddie Dong <eddie.dong@intel.com>
Add Null pointer check in Set TR Dequeue command handle function
to avoid Null pointer exception. Fllow xHCI spec 4.6.10, for Set
TR Dequeue pointer command, when the slot is not enabled the
completion code should be Slot Not Enabled Error.
Tracked-On: #5263
Signed-off-by: Long Liu <long.liu@intel.com>
Acked-by: Yu Wang yu1.wang@intel.com
According to xHCI Spec 4.10.1 Transfer TRBS chapter, there have three
scenarios shall generate a Transfer Event TRB when transfer TRBS:
1. If upon transfer completion of a TRB the Interrupt On Completion(IOC)
flag is set, the xHC shall generate a Transfer Event TRB.
2. A Short Packet will trigger the generation of a Transfer Event TRB on
the Event Ring if the Interrupt-on-Short (ISP) or Interrupt On Completion
(IOC) flags are set in the TRB that the Short Packet was detected on.
3. The detection of an error during a transfer shall always generate a Transfer
Event, irrespective of whether the Interrupt-on-Short or Interrupt On Completion
(IOC) flags are set in the Transfer TRB.
When an error condition is encountered which requires an endpoint to halt; the xHC
shall stop on the TRB in error, the endpoint shall be halted, and
software shall use a Set TR Dequeue Pointer Command to advance
the Transfer Ring to the next TD.
Tracked-On: #5263
Signed-off-by: Long Liu <long.liu@intel.com>
Acked-by: Yu Wang yu1.wang@intel.com
Two PSE-GPIO controllers of EHL CRB require MSI-X emulation for
pass-thru because it uses multiple MSI vectors. Currently acrn-config
enables MSI-X emulation for only TSN devices. Enable MSI-X emulation
for PSE-GPIOs, too.
Tracked-On: #5242
Signed-off-by: Toshiki Nishioka <toshiki.nishioka@intel.com>
Acked-by: Victor Sun <victor.sun@intel.com>
A previous update changed "slave" to "secondary", but the code comments
were changed to use "client", so update the documentation to match.
Signed-off-by: David B. Kinder <david.b.kinder@intel.com>
After below commit in https://github.com/zephyrproject-rtos/zephyr
commit d0126a037d23484feebba00d2c0eac27e6393fef
Author: Zide Chen <zide.chen@intel.com>
Date: Wed Feb 5 08:32:00 2020 -0800
boards/x86/acrn: build it in x86_64 mode and switch to X2APIC
The zephyr image for acrn would be built in x86_64 mode by default, then the
load/entry address for pre-launched Zephyr image should be changed from
0x100000 to 0x8000 accordingly per below definition in zephyr .ld file:
zephyrproject_src/zephyr/include/arch/x86/intel64/linker.ld
SECTIONS
{
/*
* The "locore" must be in the 64K of RAM, so that 16-bit code (with
* segment registers == 0x0000) and 32/64-bit code agree on addresses.
* ... there is no 16-bit code yet, but there will be when we add SMP.
*/
.locore 0x8000 : ALIGN(16)
{
_locore_start = .;
The commit in zephyrproject is merged before zephyr v2.2 release, so from v2.2
on, HV need this fix to boot Zephyr as pre-launched VM.
Tracked-On: #5259
Signed-off-by: Victor Sun <victor.sun@intel.com>
ACRN 2.1 supports two virtual boot modes, deprivilege boot mode and
direct boot mode. The deprivilege boot mode’s main purpose is to support
booting Clear Linux Service VM with UEFI service support, but this
brings scalability problems when porting ACRN to new Intel platforms.
For the 2.2 release, deprivilege mode is removed, and only direct boot
is supported, and with this we've removed support for Clear Linux as the
service VM, which impacts over 50 ACRN documents. This PR removes
documents we don't intend to update, and fixes broken links that would
occur from references to these deleted docs.
Signed-off-by: David B. Kinder <david.b.kinder@intel.com>
Replace white/black master/slave terms with alternatives. We're not
changing "master" when used in the context of GitHub branches. GitHub
advises they have a plan to help this transition. In the text body we
rever to the "master" branch as the "main" branch, but leave any urls or
code-block commands still using "master".
Signed-off-by: David B. Kinder <david.b.kinder@intel.com>
Convert SERIAL_PCI_BDF string value to hex value.
Tracked-On: #4937
Signed-off-by: Wei Liu <weix.w.liu@intel.com>
Tested-by: Shuo A Liu <shuo.a.liu@intel.com>
BDF string can be parsed by the configuration tool. A 16bit WORD value with
format (B:8, D:5, F:3) can be passed from configuration to the
hypervisor directly to save some BDF string parse code.
Tracked-On: #4937
Signed-off-by: Shuo A Liu <shuo.a.liu@intel.com>
Acked-by: Eddie Dong <eddie.dong@intel.com>
Users can add one or more ivshmem shm regions for uos when the shm
regions are configured from scenario setting.
Tracked-On: #4853
Signed-off-by: Shuang Zheng <shuang.zheng@intel.com>
Acked-by: Victor Sun <victor.sun@intel.com>
add shm_region config in default launch XMLs to configure Inter-
VM communication for post-launched VMs.
Tracked-On: #4853
Signed-off-by: Shuang Zheng <shuang.zheng@intel.com>
Acked-by: Victor Sun <victor.sun@intel.com>
This patch is to expose GPIO chassis interrupts as INTx to safety VM for
EHL. User can configure this per-VM attribute in scenario xml using the
following format:
<pt_intx desc="pt intx mapping.">
(phys_gsi0, virt_gsi0), (phys_gsi1, virt_gsi1), (phys_gsiN, virt_gsiN)
</pt_intx>
The physical and virtual interrupt gsi in each pair are separated by a
comma and enclosed in parentheses. If an integer begins with 0x or 0X,
it is hexadecimal, otherwise, it is assumed to be decimal. Example:
<pt_intx desc="pt intx mapping.">
(1, 0), (0x3, 1), (0x4, 2), (5, 6), (89, 0x12)
</pt_intx>
Tracked-On: #5241
Signed-off-by: dongshen <dongsheng.x.zhang@intel.com>
On EHL platform, we need to expose GPIO chassis interrupt to pre-launched VM
as INTx. Add related data structures so that they can be used in subsequent
commits.
Tracked-On: #5241
Signed-off-by: dongshen <dongsheng.x.zhang@intel.com>
This patch is to support direct assignment of P2SB bridge to one pre-launched
VM for EHL. User can configure this per-VM attribute in scenario xml:
<mmio_resources desc="MMIO resources.">
<p2sb>y</p2sb>
</mmio_resources>
Set p2sb to y to passthru P2SB bridge to VM, and n otherwise.
Tracked-On: #5221
Signed-off-by: dongshen <dongsheng.x.zhang@intel.com>
On EHL platform, we need to pass through P2SB bridge to pre-launched VM.
Use pt_p2sb_bar to indicate whether to passthru p2sb bridge to pre-launched VM
or not.
Tracked-On: #5221
Signed-off-by: dongshen <dongsheng.x.zhang@intel.com>
Add missing IVSHMEM tag in mrb board xml file to fix build issue
Correct misspelled function name
Use better error messages
Tracked-On: #5221
Signed-off-by: dongshen <dongsheng.x.zhang@intel.com>
Because ivshmem memory uses hv memory, if the ivshmem feature is
enabled, HV_RAM_SIZE will include IVSHMEM_SHM_SIZE.
Tracked-On: #4853
Signed-off-by: Shuang Zheng <shuang.zheng@intel.com>
Acked-by: Victor Sun <victor.sun@intel.com>
When trying to passthru a DHRD-ignored PCI device,
iommu_attach_device shall report success. Otherwise,
the assign_vdev_pt_iommu_domain will result in HV panic.
Same for iommu_detach_device case.
Tracked-On: #5240
Signed-off-by: Stanley Chang <stanley.chang@intel.com>
Acked-by: Eddie Dong <eddie.dong@intel.com>
add hybrid_rt scenario for the ElkhartLake CRB board so that user can
launch Yocto Linux as pre-launched VM.
Tracked-On: #5238
Signed-off-by: "Nishioka, Toshiki" <toshiki.nishioka@intel.com>
Acked-by: Victor Sun <victor.sun@intel.com>
Some hypercalls to a target VM are only acceptable in some certain
states, else it impacts target VM. Add some restrictive status checks to
avoid that.
Tracked-On: #5208
Signed-off-by: Shuo A Liu <shuo.a.liu@intel.com>
Acked-by: Eddie Dong <eddie.dong@intel.com>
Virtual interrupts injection and memory mapping operations can impact
target VM. By design, these type of operations from lower severity VM
to higher severity VM should be blocked by the hypervisor.
While the hypercalls are the interface between SOS VM and the
hypervisor, severity checks can be implemented at the beginning of
hypercalls needed.
Added severity checks in below hypercalls:
* hcall_set_vm_memory_regions()
* hcall_notify_ioreq_finish()
* hcall_set_irqline()
* hcall_inject_msi()
* hcall_write_protect_page()
Tracked-On: #5208
Signed-off-by: Shuo A Liu <shuo.a.liu@intel.com>
Acked-by: Eddie Dong <eddie.dong@intel.com>
This patch is used to create and destroy an ivshmem device which
is emulated in hypervisor.
Tracked-On: #4853
Signed-off-by: Yuan Liu <yuan1.liu@intel.com>
Acked-by: Wang, Yu1 <yu1.wang@intel.com>
add create_ivshmem_from_dm and destroy_ivshmem_from_dm for
ivshmem device creation and destruction in dm-land
Tracked-On: #4853
Signed-off-by: Yuan Liu <yuan1.liu@intel.com>
Acked-by: Wang, Yu1 <yu1.wang@intel.com>
Add IC_CREATE_DEVICE and IC_DESTROY_DEVICE ioctls to create and
destroy an emulated device in hypervisor
v3: change IC_CREATE_DEVICE and IC_DESTROY_DEVICE to IC_CREATE_HV_VDEV
and IC_DESTROY_HV_VDEV
Tracked-On: #4853
Signed-off-by: Yuan Liu <yuan1.liu@intel.com>
Acked-by: Wang, Yu1 <yu1.wang@intel.com>
if device configuration vbdf is unassigned, then the corresponding
vdev will not be initialized, instead, the vdev will be initialized
by device model through hypercall.
Tracked-On: #4853
Signed-off-by: Yuan Liu <yuan1.liu@intel.com>
Acked-by: Eddie Dong <eddie.dong@intel.com>
For ivshmem vdev creation, the vdev vBDF, vBARs, shared memory region
name and size are set by device model. The shared memory name and size
must be same as the corresponding device configuration which is configured
by offline tool.
v3: add a comment to the vbar_base member of the acrn_vm_pci_dev_config
structure that vbar_base is power-on default value
Tracked-On: #4853
Signed-off-by: Yuan Liu <yuan1.liu@intel.com>
Acked-by: Eddie Dong <eddie.dong@intel.com>