This patch introduce a new mode of IO request completion, polling mode.
Now, the sketch of ioreq process can be,
A. UOS vcpu0 generate PIO/MMIO ->
B. pcpu1(vcpu0 of UOS) trap into HV ->
C. pcpu1 build ioreq, send IPI and enter idle ->
D.1 pcpu0(vcpu0 of SOS) response IPI,
D.2 pcpu0 handle the ioreq in HV, kernel, DM,
D.3 pcpu0 mark ioreq as complete,
D.4 pcpu0 hypercall to enter HV ->
E.1 pcpu0 send IPI to wake pcpu1 up
E.2 UOS vcpu0 continue running
With this change, it skips D.4, E.1 steps. In step C, pcpu1 will enter a
polling ioreq state idle after send out the IPI.
It can save about ~5000 cpu cycles.
In polling mode, we do the polling in idle instead of pause cpu all the
time. It will consume more power. A better way is to use monitor/mwait
instructions which can put cpu into a sleep state with monitoring a
memory address. Unfortunately, APL has bug with monitor. We can gather
all ioreqs state into one monitorable memory and take advantage of
monitor/mwait for future platform.
The way polling or notification is per VM. We can config VMs in
different mode. By default, IO request completion will use notification
mode for all VMs. We can switch it by Kconfig.
Tracked-On: #1821
Signed-off-by: Shuo Liu <shuo.a.liu@intel.com>
Reviewed-by: Eddie Dong <eddie.dong@intel.com>
Acked-by: Anthony Xu <anthony.xu@intel.com>
In current code, DMA remapping is enabled at very early stage, when there is no
device added.
Devices are added to VT-d context table when creating VMs.
There is potential risk that there may be device DMA transactions when DMA remapping is
enabled, but the device is not added to the VT-d context table. (Eg. this could happen in UEFI
enviroment when USB keyboard use in efi intenal shell.)
This patch defers DMA remapping enabling until vm creation.
Tracked-On: #1855
Signed-off-by: Binbin Wu <binbin.wu@intel.com>
Acked-by: Anthony Xu <anthony.xu@intel.com>
remove the usage of HV_DEBUG in hypercall.c and vmcall.c
TO-DO:
Enhance Makefile to compile debug/release into 2 libraries
Tracked-On: #861
Signed-off-by: Shiqing Gao <shiqing.gao@intel.com>
Acked-by: Eddie Dong <eddie.dong@intel.com>
ioreqs acrossing VM reset is meaningless. So we reset their status when
VM reset.
Please note, device model and service os need to handle various ioreqs
pending status in emergency reset condition carefully. Otherwises, the
post processing of such ioreqs might overwrite this reset.
Tracked-On: projectacrn#1821
Signed-off-by: Shuo Liu <shuo.a.liu@intel.com>
Acked-by: Eddie Dong <eddie.dong@intel.com>
Even in partition mode, we also need to clear vmid in vmid_bitmap to
indicate the VM is not present.
Tracked-On: projectacrn#1821
Signed-off-by: Shuo Liu <shuo.a.liu@intel.com>
Acked-by: Anthony Xu <anthony.xu@intel.com>
MISRA-C requires that the function call in which the returned
value is discarded shall be clearly indicated using (void).
This patch fixes the violations related to the following
function calls.
- instr_check_gva
- vlapic_set_local_intr
- prepare_vm
- enter_s3
- emulate_instruction
- ptdev_intx_pin_remap
- register_mmio_emulation_handler
v1 -> v2:
* discard the return value of enter_s3
Tracked-On: #861
Signed-off-by: Shiqing Gao <shiqing.gao@intel.com>
fix the following integer violations:
1. Signed/unsigned conversion without cast
2. Literal value requires a U suffix
3. Implicit conversion of underlying type
v3 -> v4:
* change the type of npk_loglevel/mem_loglevel/console_loglevel
from uint32_t to uint16_t
v2 -> v3:
* discard the return value of update_ept
* discard changes related to npk loglevel
v1 -> v2:
* remove the unnecessary changes related to the false positive
issues caused by scanning tool
* change the type of the local variable 'vlapic_id' from uint8_t
to uint32_t in function 'vlapic_build_id'
* change the type of the struct member 'flags' in shared_buf from
uint64_t to uint32_t
Tracked-On: #861
Signed-off-by: Shiqing Gao <shiqing.gao@intel.com>
Acked-by: Eddie Dong <eddie.dong@intel.com>
clean up mmio_read/mmio_write in instr_emul.c
- make these two functions inline
- discard the return value and add the pre condition
- rename these two functions to vie_mmio_read/vie_mmio_write
Tracked-On: #861
Signed-off-by: Shiqing Gao <shiqing.gao@intel.com>
Acked-by: Eddie Dong <eddie.dong@intel.com>
MISRA-C requires that the function call in which the returned
value is discarded shall be clearly indicated using (void).
This patch fixes the violations related to the following
function calls.
- vlapic_set_intr
- vlapic_intr_edge
Tracked-On: #861
Signed-off-by: Shiqing Gao <shiqing.gao@intel.com>
to decrease the value of 'create_vcpus' in fail case.
Tracked-On: #861
Signed-off-by: Yonghua Huang <yonghua.huang@intel.com>
Acked-by: Eddie Dong <eddie.dong@intel.com>
this patch is to avoid posisble inconsisent states for
VMs and vCPUs in configuration that SOS is SMP based
and Device Model may be multi-threads based.
Tracked-On: #861
Signed-off-by: Yonghua Huang <yonghua.huang@intel.com>
Acked-by: Eddie Dong <eddie.dong@intel.com>
-- Config MAX_EMULATED_MMIO_REGIONS 16 in Kconfig
-- Add emulated mmio array and emulated mmio regions
in vm structure
-- Remove mmio list in vm structure
-- Remove unregister_mmio_emulation_handler and
vioapic_cleanup APIs
Tracked-On: #861
Signed-off-by: Mingqiang Chi <mingqiang.chi@intel.com>
Acked-by: Eddie Dong <eddie.dong@intel.com>
-- Add emulated port io index
-- Add emulated pio array in vm structure
-- Remove port list in vm structure
-- Remove free_io_emulation_resource/register_io_handler/
create_io_handler APIs
v2-->v3:
-- not add 'is_emulated', check len == 0U
-- Check if io_read/io_write handler is NULL before calling
-- Replace ENUM with MACRO for emulated pio index to avoid
MISRA-C violations
v1-->v2:
-- Remove EMUL_PIO_NUM in Kconfig, add emulated pio index
for PIC/PCI/UART/RTC/PM
Tracked-On: #861
Signed-off-by: Mingqiang Chi <mingqiang.chi@intel.com>
Reviewed-by: Jason Chen CJ <jason.cj.chen@intel.com>
Acked-by: Eddie Dong <eddie.dong@intel.com>
MISRA-C requires that the controlling expression of an if statement or
an iteration-statement shall be Boolean type.
v1 -> v2:
* update the U suffix of constant to let the type of operands
to bit operations (&) is the same.
* update the type of 'i' from uint64_t to uint16_t in
'profiling_set_control' since it is limited by phys_cpu_num,
which is uint16_t.
Tracked-On: #861
Signed-off-by: Shiqing Gao <shiqing.gao@intel.com>
Acked-by: Eddie Dong <eddie.dong@intel.com>
It is not necessary to use atomic operation to
set the initial value for 'created_vcpu' in vm structure.
Tracked-On: #861
Signed-off-by: Mingqiang Chi <mingqiang.chi@intel.com>
- Fix the integer violations related to the following rules:
1. The operands to shift operations (<<, >>) shall be unsigned
integers.
2. The operands to bit operations (&, |, ~) shall be unsigned
integers.
- Replace 12U with CPU_PAGE_SHIFT when it is address shift case.
v1 -> v2:
* use existed MACRO to get bus/slot/func values
* update PCI_SLOT MACRO to make it more straightforward
* remove the incorrect replacement of 12U with CPU_PAGE_SHIFT
dmar_fault_msi_write
Tracked-On: #861
Signed-off-by: Shiqing Gao <shiqing.gao@intel.com>
Acked-by: Eddie Dong <eddie.dong@intel.com>
For data structure types "struct vm", its name is identical
with variable name in the same scope. This is a MISRA C violation.
Naming convention rule:If the data structure type is used by multi
modules, its corresponding logic resource is exposed to external
components (such as SOS, UOS), and its name meaning is simplistic
(such as vcpu, vm), its name needs prefix "acrn_".
The following udpates are made:
struct vm *vm-->struct acrn_vm *vm
Tracked-On: #861
Signed-off-by: Xiangyang Wu <xiangyang.wu@linux.intel.com>
For data structure types "struct vcpu_arch", its name
shall follow Naming convention.
Naming convention rule:If the data structure type is
used by multi modules, its corresponding logic resource
is exposed to external components (such as SOS, UOS),
and its name meaning is simplistic (such as vcpu, vm),
its name needs prefix "acrn_". Variable name can be
shortened from its data structure type name.
The following udpates are made:
struct vcpu_arch arch_vcpu-->struct acrn_vcpu_arch arch
Tracked-On: #861
Signed-off-by: Xiangyang Wu <xiangyang.wu@linux.intel.com>
For data structure types "struct vcpu", its name is identical
with variable name in the same scope. This is a MISRA C violation.
Naming convention rule:If the data structure type is used by multi
modules, its corresponding logic resource is exposed to external
components (such as SOS, UOS), and its name meaning is simplistic
(such as vcpu, vm), its name needs prefix "acrn_".
The following udpates are made:
struct vcpu *vcpu-->struct acrn_vcpu *vcpu
Tracked-On: #861
Signed-off-by: Xiangyang Wu <xiangyang.wu@linux.intel.com>
ACRN, in partition mode, supports LAPIC pass-thru to guests. Guest needs
to use x2APIC mode of LAPIC for pass-thru to be enabled.
ACRN also needs the user to configure lapic_pt to true in vm_desc
for the VM.
Interrupt Command Register (ICR) is the only APIC register that is
intercepted. Reference code in partition/vm_description.c enables
LAPIC pass-thru for vm2.
Tracked-On: #1626
Signed-off-by: Sainath Grandhi <sainath.grandhi@intel.com>
Reviewed-by: Xu Anthony <anthony.xu@intel.com>
This patch adds support for self-IPI virtualization when guest
uses vLAPIC in x2APIC mode.
Tracked-On: #1626
Signed-off-by: Sainath Grandhi <sainath.grandhi@intel.com>
Reviewed-by: Xu Anthony <anthony.xu@intel.com>
When guest switches from xAPIC mode to x2APIC mode of vLAPIC operation,
MSRs are used to access vLAPIC. This patch adds APICv support for
MSR accesses to vLAPIC. Switching from xAPIC to x2APIC is supported via
APIC BASE MSR. Other modifications like disabling and switching back to
xAPIC are not supported.
Tracked-On: #1626
Signed-off-by: Sainath Grandhi <sainath.grandhi@intel.com>
Reviewed-by: Xu Anthony <anthony.xu@intel.com>
This patch adds support to return APIC ID if guest uses vLAPIC
in x2APIC mode.
Tracked-On: #1626
Signed-off-by: Sainath Grandhi <sainath.grandhi@intel.com>
Reviewed-by: Xu Anthony <anthony.xu@intel.com>
This patch converts x2APIC MSR accesses to corresponding LAPIC MMIO offset to
utitlize vlapic_write/read APIs to virtualize LAPIC. Also adds support to inject
GP fault when read-only registers are attempted to be written to or vice versa.
Tracked-On: #1626
Signed-off-by: Sainath Grandhi <sainath.grandhi@intel.com>
Reviewed-by: Xu Anthony <anthony.xu@intel.com>
When guest uses vLAPIC in x2APIC mode, ICR write is a single MSR write.
Also, the destination field for device interrupts and IPIs should not be
handled in the same way as xAPIC mode. This patch adds support for x2APIC
mode operation of guest vLAPIC.
Tracked-On: #1626
Signed-off-by: Sainath Grandhi <sainath.grandhi@intel.com>
Reviewed-by: Xu Anthony <anthony.xu@intel.com>
x2APIC ID and LDR are in different format compared to xAPIC mode of operation.
This patch adds code to build ID and LDR when guest uses vLAPIC in x2APIC mode.
Tracked-On: #1626
Signed-off-by: Sainath Grandhi <sainath.grandhi@intel.com>
Reviewed-by: Xu Anthony <anthony.xu@intel.com>
Extending enable_msr_interception to accept mode as input. Mode specifies
if the API user wants ACRN to intercept on read-only or write-only or both
read and write or disable MSR interception altogether.
Tracked-On: #1626
Signed-off-by: Sainath Grandhi <sainath.grandhi@intel.com>
Reviewed-by: Xu Anthony <anthony.xu@intel.com>
Accessing x2APIC MSRs in xAPIC mode should result in GP exception according
to SDM section 10.12.2. Adding support to ACRN to inject GP into guests for
the same reason.
Tracked-On: #1626
Signed-off-by: Sainath Grandhi <sainath.grandhi@intel.com>
Reviewed-by: Xu Anthony <anthony.xu@intel.com>
Replace dynamic memory allocation in memory management with static memory allocation.
Since the static memory allocation can guarantee the allocation never failed, so
split_large_page and construct_pgentry don't need to return a errno any more.
Besides, the destroy_ept don't need to free page any more. Instead, it would memset
the eptp to 0 to make sure we can't walk this paging table again.
Tracked-On: #861
Signed-off-by: Li, Fei1 <fei1.li@intel.com>
Add static paging table allocation API for hypervisor.
Note: must configure PLATFORM_RAM_SIZE and PLATFORM_MMIO_SIZE exactly as the platform.
Rename RAM_START/RAM_SIZE to HV_RAM_START/HV_RAM_SIZE for HV.
Tracked-On: #861
Signed-off-by: Li, Fei1 <fei1.li@intel.com>
add size check for other hypervisor console command;
they could be overflow for shell log buffer output.
Tracked-On: #1587
Signed-off-by: Minggui Cao <minggui.cao@intel.com>
Reviewed-by: Yin Fengwei <fengwei.yin@intel.com>
Acked-by: Eddie Dong <eddie.dong@intel.com>
After enabling vPCI in hypervisor for vm0, UOS may not able be launched
successfully. Consider this scenario (take MSI-X for example):
- DM makes hypercall to hypervisor to do MSI-X remapping on behalf of
UOS guests.
- After the hypercall, VHM module in SOS kernel updates the physical
MSI-X table with the physical Message Data/Addr.
- These MMIO write requests are intercepted by hypervisor, which will
call ptdev_msix_remap() to do MSI-S remapping.
It may fail due to 2 possible reasons:
1) wrong target VM because:
hypervisor thinks it's a VM0 MSI-X device but they have been registered
as UOS guests through HC_SET_PTDEV_INTR_INFO hypercall.
2) wrong ptdev_msi_info->vmsi_data because:
The virtual MSI-X table is supposed to hold virtual Message data/addr
but the SOS VHM writes the physical ones to it.
This patch resolves these problems by ignoring the HC_VM_PCI_MSIX_REMAP
hypercall, so virtual and physical Message Data are the same from SOS'
perspective and it won't mess up the virtual PCI device in HV.
Also in HC_SET_PTDEV_INTR_INFO handler, vpci updates the target VM
when the PCI devices are assigned to different VMs.
The UOS' MSI/MSI-X remapping is triggered by hypervisor when SOS (either
DM or VHM) updates the Message Data/Addr.
Tracked-On: #1568
Signed-off-by: dongshen <dongsheng.x.zhang@intel.com>
Signed-off-by: Zide Chen <zide.chen@intel.com>
Acked-by: Anthony Xu <anthony.xu@intel.com>
For SOS BSP, we reuse native saved cs.limit
For UOS BSP, we set cs.limit in DM
For AP, we use initialized data from realmode_init_regs.
Tracked-On: #1231
Signed-off-by: Yin Fengwei <fengwei.yin@intel.com>
Acked-by: Eddie Dong <Eddie.dong@intel.com>
- flush L1 cache before VM entry only on platform
affected by L1TF
- flush operation is configurable by below MACRO:
--CONFIG_L1D_FLUSH_VMENTRY_ENABLED
Tracked-On: #1672
Signed-off-by: Yonghua Huang <yonghua.huang@intel.com>
Reviewed-by: Kevin Tian <kevin.tian@intel.com>
The operands to shift operations (<<, >>) shall be unsigned integers.
v1 -> v2:
* replace 12U with CPU_PAGE_SHIFT when it is address shift case.
* replace 6UL with 0x6UL
Tracked-On: #861
Signed-off-by: Shiqing Gao <shiqing.gao@intel.com>
Acked-by: Eddie Dong <eddie.dong@intel.com>
Guest OS can use vLAPIC in x2APIC mode. Upon vlapic_reset, vLAPIC should
be set to xAPIC mode.
Tracked-On: #1626
Signed-off-by: Sainath Grandhi <sainath.grandhi@intel.com>
Functions shall not call themselves, either directly or indirectly.
V1->V2:
add pre-assumption.
V2->V3:
when vector < 16,call vlapic_set_error
Tracked-On: #861
Signed-off-by: Huihuang Shi <huihuang.shi@intel.com>
Acked-by: Eddie Dong <eddie.dong@intel.com>
Currently we compare vcpu_id with phys_cpu_num,
vcpu_id is not related with phys_cpu_num,
this patch cleanup them.
Tracked-On: #861
Signed-off-by: Mingqiang Chi <mingqiang.chi@intel.com>
Reviewed-by: Li, Fei1 <fei1.li@intel.com>
Acked-by: Eddie Dong <eddie.dong@intel.com>
This patch adds support to sep/socwatch profiling
Adds 2 new files include/arch/x86/profiling.h and arch/x86/profiling.c
which contains most of the implementation for profiling,most of the functions
in profiling.c have dummy implementation and will be implemented in next patches
a. cpu.c, Initial profiling setup is done as part of bsp_boot_post
and cpu_secondary_post flow
b. vmcall.c, New ioctl is added for performing profiling related
operations in vmcall_vmexit_handler
ioctl - HC_PROFILING_OPS
function - hcall_profiling_ops()
c. common/hypercall.c, hcall_profiling_ops() implementation.
d. hv_main.c, In vcpu_thread calling profiling related functions
to save vm context
e. acrn_hv_defs.h, list all the profiling command types
Tracked-On: projectacrn#1409
Acked-by: Eddie Dong <eddie.dong@intel.com>
Signed-off-by: Chinthapally, Manisha <manisha.chinthapally@intel.com>
Fix violations whose parameter can be read-only.
This patch only fix the parameter whose name is vcpu.
Tracked-On: #861
Signed-off-by: Huihuang Shi <huihuang.shi@intel.com>
Acked-by: Anthony Xu <anthony.xu@intel.com>
Fix violations for function whose parameter can be read-only.
Tracked-On: #861
Signed-off-by: Huihuang Shi <huihuang.shi@intel.com>
Acked-by: Anthony Xu <anthony.xu@intel.com>
guest software loading is per VM instead of vcpu. So we move it
from prepare_vcpu to prepare_vm. And make sure it's called for
all VMs for partition mode.
Tracked-On: #1565
Signed-off-by: Yin Fengwei <fengwei.yin@intel.com>
Acked-by: Anthony Xu <anthony.xu@intel.com>
All the platforms supported by ACRN supports x2APIC. So enabled
x2APIC for ACRN hv. Removed any code that is needed for xAPIC mode
of operation.
Tracked-On: #1455
Signed-off-by: Sainath Grandhi <sainath.grandhi@intel.com>
Reviewed-by: Eddie Dong <eddie.dong@intel.com>
Reviewed by: Yonghua Huang <yonghua.huang@intel.com>
