Currently vpid is not released in reset_vcpu() hence the vpid resource
could be exhausted easily if guests are re-launched.
This patch assigns vpid according to the fixed mapping of runtime vm_id
and vcpu_id to guarantee the uniqueness of vpid.
Tracked-On: #2700
Signed-off-by: Zide Chen <zide.chen@intel.com>
Signed-off-by: Sainath Grandhi <sainath.grandhi@intel.com>
Acked-by: Anthony Xu <anthony.xu@intel.com>
CRx 'guest/host mask' is widely used when talking
about CR0/CR4 virtualization in SDM spec,
rename them to 'VMX_CRx_GUEST_HOST_MASK'.
Tracked-On: #861
Signed-off-by: Yonghua Huang <yonghua.huang@intel.com>
Acked-by: Eddie Dong <eddie.dong@intel.com>
vmx.c should only take care host vmx operations, which should not
contain vcpu reference, so refine vmx on/off APIs, move out use
of vcpu by adding one per cpu vmcs_run pointer.
as now each pcpu only run on one vcpu, so just keep running vmcs
in per cpu vmcs_run pointer is enough.
Changes to be committed:
modified: arch/x86/cpu.c
modified: arch/x86/init.c
modified: arch/x86/pm.c
modified: arch/x86/vmcs.c
modified: arch/x86/vmx.c
modified: include/arch/x86/per_cpu.h
modified: include/arch/x86/vmx.h
Tracked-On: #1842
Signed-off-by: Jason Chen CJ <jason.cj.chen@intel.com>
Acked-by: Eddie Dong <eddie.dong@intel.com>
one part is pure vmx operations which keeps in vmx.c
the other part is vmcs operations which is vcpu related, move them
into vmcs.c
Changes to be committed:
modified: Makefile
copied: arch/x86/vmx.c -> arch/x86/vmcs.c
modified: arch/x86/vmx.c
modified: arch/x86/vmx_asm.S
modified: include/arch/x86/hv_arch.h
new file: include/arch/x86/vmcs.h
modified: include/arch/x86/vmx.h
Tracked-On: #1842
Signed-off-by: Jason Chen CJ <jason.cj.chen@intel.com>
Misra C reqires assembly code should comply with
the rules list below:
The assembly code's functionality should match the function's
name.If not,pls encapsulate the assembly code and give a suitable
name for describing the functionality.
V1->V2:
1.remove the dead code
2.update detail comment
V2->V3:
1.replace the macro name with upper case.
2.remove the typedef and rename the struct name
"_descriptor_table_" to "descriptor_table".
Tracked-On: #861
Signed-off-by: Huihuang Shi <huihuang.shi@intel.com>
Acked-by: Eddie Dong <eddie.dong@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>
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>
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>
to enable APICv Posted interrupt supported, following the
specifications defined in Intel SDM Section #29.6, Volume3.
Posted-interrupt processing is a feature by which a processor
processes the virtual interrupts by recording them as pending
on the virtual-APIC page.
Injecting interrupts to VCPU from remote CPU without causing
VM exit on the destination, following steps in SDM Section 29.6,volume3:
Tracked-On: #1447
Signed-off-by: Yonghua Huang <yonghua.huang@intel.com>
Acked-by: Eddie Dong <eddie.dong@intel.com>
MISRA-C required the suffix(U), such as:
(1U << 0) ---> (1U << 0U)
This patch will add the suffix(U) to come up MISRA-C.
Tracked-On: #1385
Signed-off-by: Wei Liu <weix.w.liu@intel.com>
Acked-by: Eddie Dong <eddie.dong@intel.com>
According to SDM 4.6 Access Rights definition, if SMAP/SMEP
is enabled, more conditions are used to decide the access rights
like SMAP/SMEP, EFLAGS.AC, R/W bit in every paging-structure entry
for page walking.
This patch add the check to guest page walking function and inject
page fault accordingly.
Tracked-On: #1207
Signed-off-by: Yin Fengwei <fengwei.yin@intel.com>
Acked-by: Anthony Xu <anthony.xu@intel.com>
- convert function like macros to inline functions based on MISRA-C
requirement
- remove some unused and duplicated macros
Tracked-On: #861
Signed-off-by: Shiqing Gao <shiqing.gao@intel.com>
Reviewed-by: Junjie Mao <junjie.mao@intel.com>
-- Change vmx_off/exec_vmxon/exec_vmclear/exec_vmptrld/
exec_vmxon_instr/init_vmcs to void type
-- for vmxon/vmclear/vmptrld, add pre-conditions to
guarantee sucessful execution.
Tracked-On: #861
Signed-off-by: Mingqiang Chi <mingqiang.chi@intel.com>
Reviewed-by: Junjie Mao <junjie.mao@intel.com>
Acked-by: Eddie Dong <eddie.dong@intel.com>
Check if the VMX capability is locked with incorrect value,
at the time when HV do the hardware capability detect.
Tracked-On: #861
Signed-off-by: Mingqiang Chi <mingqiang.chi@intel.com>
Reviewed-by: Junjie Mao <junjie.mao@intel.com>
Acked-by: Eddie Dong <eddie.dong@intel.com>
Since these APIs always return 0,
change them to void type,
vcpu_set_cr0()
vmx_write_cr0()
vcpu_set_cr4()
vmx_write_cr4()
kick_notification()
tsc_deadline_handler()
dmar_fault_handler()
ptdev_interrupt_handler()
Signed-off-by: Mingqiang Chi <mingqiang.chi@intel.com>
Reviewed-by: Junjie Mao <junjie.mao@intel.com>
- Clean up some code regarding to % and / operations since bit
operations are faster.
x % 64U ---> x & 0x3fU
x % 32U ---> x & 0x1fU
x % 16U ---> x & 0xfU
x % 8U ---> x & 0x7U
x % 4U ---> x & 0x3U
x % 2U ---> x & 0x1U
x / 64U ---> x >> 6U
x / 32U ---> x >> 5U
x / 16U ---> x >> 4U
x / 8U ---> x >> 3U
x / 4U ---> x >> 2U
x / 2U ---> x >> 1U
- Minor changes regarding to coding styles
Tracked-On: #861
Signed-off-by: Shiqing Gao <shiqing.gao@intel.com>
Acked-by: Eddie Dong <eddie.dong@intel.com>
V1:
In order to better comply with MISRA C,
add const qualifiers whereeven required.
In the patch, these are being added to pointers
which are normally used in "get" functions.
V2: Corrected the issues in the patch
pointed by Junjie in his review comments.
Moved the const qualifiers to the correct
places. Removed some changes which are not
needed.
V3: Updated patch comment.
This modifies a subset of all the functions
which might need constant qualifiers
for the their parameters.
This is not and exhaustive patch. This only
targets obvious places where we can use
the const qualifier. More changes will be
submitted in future patches, if required.
Signed-off-by: Arindam Roy <arindam.roy@intel.com>
there will be 3 types of vcpu runtime contexts:
- runtime contexts always saved/restored during VM exit/entry, which
include general registers rax/rcx/rdx/rbx/rbp/rsi/rdi/r8~r15, cr2 and
msr for spectre control (ia32_spec_ctrl)
- runtime contexts on-demand cached/updated during VM exit/entry, which
include frequently used registers rsp, rip, efer, rflags, cr0 and cr4
- runtime contexts always read/write from/to VMCS, which include left
registers not in above
this patch add get/set register APIs for vcpu runtime contexts, and unified
the save/restore method for them according to above description.
v3:
- update vcpu_get/set_cr0/4 as unified interface to get/set guest cr0/cr4,
use on-demand cache for reading, but always write to VMCS for writing.
v2:
- use reg_cached/reg_updated for on-demand runtime contexts
- always read/write cr3 from/to VMCS
Signed-off-by: Jason Chen CJ <jason.cj.chen@intel.com>
Acked-by: Eddie Dong <eddie.dong@intel.com>
Now, we let guest own most CR4 bit. Which means guest
handles whether the CR4 writting is invalid or not and
GP injection if it's invalid writing.
Two bits are exception here:
we filter VMX and PCID feature to guest (which means
they are supported on native).
So we can't depends on guest to inject GP for these bits.
Instead, we should explicitly trap these CR4 bits update
and inject GP to guest from HV.
Signed-off-by: Yin Fengwei <fengwei.yin@intel.com>
Acked-by: Anthony Xu <anthony.xu@intel.com>
According to SDM:
writing a nonzero value to 63:32 bits of CR0 and CR4 results #GP(0).
writing a nonzero value to reserved bit of CR4 results #GP(0).
We merge the check with always off mask of CR0 and CR4.
Signed-off-by: Yin Fengwei <fengwei.yin@intel.com>
Acked-by: Anthony Xu <anthony.xu@intel.com>
Added brackets for expression to make it easy to understand and
reduce the mistake of precedence. The rule is applied to the
mixed same level of prevedence opeartors, high level presedence
operators and logical expression.
Signed-off-by: Yang, Yu-chu <yu-chu.yang@intel.com>
Acked-by: Anthony Xu <anthony.xu@intel.com>
- explicitly declare the array size to fix the violation 'Array has no
bounds specified'
- minor changes for comments style
v1 -> v2:
* add the definition for exit reasons from 0x39 to 0x40 based on "SDM
APPENDIX C VMX BASIC EXIT REASONS"
Signed-off-by: Shiqing Gao <shiqing.gao@intel.com>
Acked-by: Anthony Xu <anthony.xu@intel.com>
- Remove those APIs without definition
- Minor changes to make sure the line length is less than 80
Signed-off-by: Shiqing Gao <shiqing.gao@intel.com>
Reviewed-by: Junjie Mao <junjie.mao@intel.com>
Acked-by: Eddie Dong <eddie.dong@intel.com>
In the hypervisor, VMCS fields include 16-bit fields,
32-bit fields, 64-bit fields and natural-width fields.
In the current implement, there are exec_vmread/exec_vmwrite
used for accessing 32-bit fields, 64-bit field and
natural-width fields. This usage will confue developer.
So there are many type casting for the return value and
parameters vmread/vmwrite operations.
Since exec_vmread/exec_vmwrite and exec_vmread64/exec_vmwrite64
are the same, update current exec_vmread/exec_vmwrite
implement into exec_vmread64/exec_vmwrite64 implement
and add MACRO define for exec_vmread/exec_vmwrite in
head file;
To access 64-bit fields in VMCS, callers use
exec_vmread64/exec_vmwrite64;
Update related variables type for vmread/vmwrite operations;
Update related caller according to VMCS fields size.
Note:Natural-width fields have 64 bits on processors
that support Intel 64 architecture.To access natural-width
fields in VMCS, callers still use exec_vmread/exec_vmwrite,
keep the current implementation.
V1--V2:
This is new part of this patch serial to only
update 64-bit vmread/vmread opertions and related
caller, for netural width fields, still use exec_vmread
or exec_vmwrite.
V2-->V3:
Fix few mistake updations for netural fields in VMCS,
just keep exec_vmread/exec_vmwrite to access them;
Fix few mistake updations for 64-bit fields in VMCS.
V3--V4:
Add "016ll" for 64-bit variable in log function;
Few updates for coding style;
Rename lssd32_idx as tr_sel in VMX module.
V4-->V5:
Use CPU_NATURAL_LAST in the vm_get_register and
vm_set_register to make condition statement more
understandable.
Signed-off-by: Xiangyang Wu <xiangyang.wu@intel.com>
Reviewed-by: Junjie Mao <junjie.mao@intel.com>
In the hypervisor, VMCS fields include 16-bit fields,
32-bit fields, 64-bit fields and natural-width fields.
In the current implement, no exec_vmread32/exec_vmwrite32
is for accessing 32-bit fields. So there are many type
casting for the return value and parameters vmread/vmwrite
operations.
Create exec_vmread32 and exec_vmwrite32 functions to
access 32-bit fields in VMCS;
Update related variables type for vmread/vmwrite operations;
Update related caller according to VMCS fields size.
V1--V2:
This is new part of this patch serial to only
update 32 bit vmread/vmread opertions and related
caller.
V2-->V3:
Update related variables type in data structure
for exec_vmread32/exec_vmwrite32.
Rename temp variable 'low' into 'value' for
exec_vmread32;
V3-->V4:
Remove useless type conversion.
Signed-off-by: Xiangyang Wu <xiangyang.wu@intel.com>
Reviewed-by: Junjie Mao <junjie.mao@intel.com>
In the hypervisor, VMCS fields include 16-bit fields,
32-bit fields, 64-bit fields and natural-width fields.
In the current implement, no exec_vmread16/exec_vmwrite16
is for accessing 16-bit fields. So there are many type
casting for the return value and parameters vmread/vmwrite
operations.
Create exec_vmread16 and exec_vmwrite16 functions to
access 16-bit fields in VMCS;
Update related variables type for vmread/vmwrite operations;
Update related caller according to VMCS fields size.
V1--V2:
This is new part of this patch serial to only
update 16-bit vmread/vmread opertions and related
caller.
V2--V3:
Add "hu" for uint16_t argument in log function;
Add comments for function get_vmcs_field;
Update related variables type for exec_vmread16;
Rename temp variable 'low' into 'value'.
V3-->V4:
Few updates for exec_vmread16.
V4-->V5:
Few updates for coding style;
Replace "hux" with "hu" in log function for 16-bit
variable.
V5-->V6:
CPU_REG_64BIT_LAST is used in the vm_get_register and
vm_set_register to make condition statement more
understandable.
Signed-off-by: Xiangyang Wu <xiangyang.wu@intel.com>
Reviewed-by: Junjie Mao <junjie.mao@intel.com>
To minimize explicit casts, this patch adjusts the types of function parameters
and structure fields related to vlapic and update the types in the internal
implementation accordingly.
Signed-off-by: Junjie Mao <junjie.mao@intel.com>
Acked-by: Dong Eddie <eddie.dong@intel.com>
For constants used in unsigned contexts, a 'U' suffix is required per MISRA C
standard.
Signed-off-by: Junjie Mao <junjie.mao@intel.com>
Acked-by: Dong Eddie <eddie.dong@intel.com>
There are some integer type conversions reported by static
analysis tool for vcpu id, number of created vcpus, and
vpid, to reduce these type conversions, redesign vcpu id,
number of created vcpus, and vpid type as uint16_t as per
their usage, related 16-bit atomic operations shall be
added in HV.
MISRA C requires that all unsigned constants should have the suffix 'U'
(e.g. 0xffU), but the assembler may not accept such C-style constants.
Add 16-bit atomic add/dec/store operations;
Update temporary variables type and parameters type of
related caller;
Update vpid type as uint16_t;
Replace Macro with constant value for CPU_PAGE_SIZE.
Note: According to SDM A.10, there are some bits defined
in the IA32_VMX_EPT_VPID_CAP MSR to support the INVVPID
instruction, these bits don't mean actual VPID, so
the vpid field in the data struct vmx_capability doesn't
be updated.
V1--V2:
update comments for assembly code as per coding style;
Signed-off-by: Xiangyang Wu <xiangyang.wu@intel.com>
In the current memory module, there are many constatn value
without U/UL suffix, it is reported as MISRA C violations by
static analysis tool.
Add 'U/UL' suffix for unsigned contant value in memory module
as needed.
Note:In the most case, CPU_PAGE_SIZE(0x1000) is used as
unsigned integer contant value, so CPU_PAGE_SIZE is defined
as unsigned integer contant value, and it is safety converted
into unsigned long type according to MISRA C standard.
Signed-off-by: Xiangyang Wu <xiangyang.wu@intel.com>
Acked-by: Eddie Dong <eddie.dong@intel.com>
This patch makes necessary integer narrowing and/or signedness conversion
explicit.
While some narrowing are expected behavior, the correctness of the others relies
on the specifications of some interfaces (e.g. the higher 32-bit of what
exec_vmread() returns is all 0s if the given field is 32-bit). Add a stub
for now to avoid missing them.
Signed-off-by: Junjie Mao <junjie.mao@intel.com>
This patch makes use of IA32_PAT MSR to emulate cache disabled behaviour
When the guest is requesting to set CR0.CD:
- Keep guest's CR0.CD and CR0.NW bits unchanged
- Write IA32_PAT MSR with all-UC entries to change the effective memory
type for all GPA to UC for the guest VCPU
- It depends on trapping wrmsr to IA32_PAT to prevent any entry in
IA32_PAT being changed to non UC type by the guest
When the guest is requesting to clear CR0.CD:
- restore the content of guest's IA32_PAT MSR
Signed-off-by: Zide Chen <zide.chen@intel.com>
Acked-by: Eddie Dong <eddie.dong@intel.com>
Preparing for emulating guest's CR0.CD and CR0.NW bits:
- Intercept both rdmsr and wrmsr for IA32_PAT
- Track guest's IA32_PAT MSR with vcpu.arch_vcpu.contexts.ia32_pat
Signed-off-by: Zide Chen <zide.chen@intel.com>
Acked-by: Eddie Dong <eddie.dong@intel.com>
MISRA C required function return type should be consistented.
Signed-off-by: Huihuang Shi <huihuang.shi@intel.com>
Acked-by: Eddie Dong <eddie.dong@intel.com>
In the hypervisor, physical cpu id is defined as "int" or "uint32_t"
type in the hypervisor. So there are some sign conversion issues
about physical cpu id (pcpu_id) reported by static analysis tool.
Sign conversion violates the rules of MISRA C:2012.
In this patch, define physical cpu id as "uint16_t" type for all
modules in the hypervisor and change related codes. The valid
range of pcpu_id is 0~65534, INVALID_PCPU_ID is defined to the
invalid pcpu_id for error detection, BROADCAST_PCPU_ID is
broadcast pcpu_id used to notify all valid pcpu.
The type of pcpu_id in the struct vcpu and vcpu_id is "int" type,
this will be fixed in another patch.
V1-->V2:
* Change the type of pcpu_id from uint32_t to uint16_t;
* Define INVALID_PCPU_ID for error detection;
* Define BROADCAST_PCPU_ID to notify all valid pcpu.
V2-->V3:
* Update comments for INVALID_PCPU_ID and BROADCAST_PCPU_ID;
* Update addtional pcpu_id;
* Convert hexadecimals to unsigned to meet the type of pcpu_id;
* Clean up for MIN_PCPU_ID and MAX_PCPU_ID, they will be
defined by configuration.
Note: fix bug in the init_lapic(), the pcpu_id shall be less than 8,
this is constraint by implement in the init_lapic().
Signed-off-by: Xiangyang Wu <xiangyang.wu@intel.com>
Acked-by: Eddie Dong <eddie.dong@intel.com>
To handle cpu down/up dynamically, arcn needs to support vmx off/on
dynamically. Following changes is introduced:
vmx_off will be used when down AP. It does:
- vmclear the mapped vcpu
- off vmx.
exec_vmxon_instr is updated to handle start and up AP both. It does
- if vmx was on on AP, load the vmxon_region saved. Otherwise,
allocate vmxon_region.
- if there is mapped vcpu, vmptrld mapped vcpu.
Signed-off-by: Zheng Gen <gen.zheng@intel.com>
Signed-off-by: Yin Fegnwei <fengwei.yin@intel.com>
Acked-by: Eddie Dong <Eddie.dong@intel.com>
Enable VMX vpid ctrl and assign an unique vpid to each vcpu
so that VMX transitions are not required to invalidate any
linear mappings or combined mappings.
SDM Vol 3 - 28.3.3.3
If EPT is in use, the logical processor associates all mappings
it creates with the value of bits 51:12 of current EPTP.
If a VMM uses different EPTP values for different guests, it may
use the same VPID for those guests. Doing so cannot result in one
guest using translations that pertain to the other.
In our UOS, the trusty world and normal world are using different
EPTP. So we can use the same VPID for it.
Signed-off-by: Li, Fei1 <fei1.li@intel.com>
Acked-by: Eddie Dong <eddie.dong@intel.com>
In current implemenation, cr0/cr4 host mask value are set
according to the value from fixed0/fixed1 values of cr0/cr4.
In fact, host mask can be set to the bits, which need to be trapped.
This patch, add code to support exiting long mode in CR0 write handling.
Add some check when modify CR0/CR4.
- CR0_PG, CR0_PE, CR0_WP, CR0_NE are trapped for CR0.
PG, PE are trapped to track vcpu mode switch.
WP is trapped for info of protection when paing walk.
NE is always on bit.
- CR4_PSE, CR4_PAE, CR4_VMXE are trapped for CR4.
PSE, PAE are trapped to track paging mode.
VMXE is always on bit.
- Reserved bits and always off bits are not allow to be set by guest.
If guest try to set these bits when vmexit, a #GP will be injected.
Signed-off-by: Binbin Wu <binbin.wu@intel.com>
Reviewed-by: Kevin Tian <kevin.tian@intel.com>
Reviewed-by: Jason Chen CJ <jason.cj.chen@intel.com>
Acked-by: Xu, Anthony <anthony.xu@intel.com>
Guest CR3 read/write operations are not trapped.
Remove CR3 handling in cr_access_vmexit_handler.
Also remove unused API vmx_read_cr3.
Signed-off-by: Binbin Wu <binbin.wu@intel.com>
Reviewed-by: Kevin Tian <kevin.tian@intel.com>
Move from vmexit.c to vmx.c
Declare the functions in vmx.h
Rename the functions' name with prefix vmx_.
Signed-off-by: Binbin Wu <binbin.wu@intel.com>
Reviewed-by: Kevin Tian <kevin.tian@intel.com>
Now one uint64_t type is used to obtain the corresponding descriptor_table
for GDT/IDT. This will cause the stack protect corruption under -O2.
So the descriptor_table struct is added to configure the GDT/IDT of VMCS.
V1->V2: Move the descriptor_table into vmx.h header file
And its type is renamed from dt_addr_t to descriptor_table.
Signed-off-by: Zhao Yakui <yakui.zhao@intel.com>
Signed-off-by: Zheng Gen <gen.zheng@intel.com>
Reviewed-by: Jason Chen CJ <jason.cj.chen@intel.com>
Acked-by: Anthony Xu <anthony.xu@intel.com>
Acked-by: Eddie Dong <eddie.dong@intel.com>
1. exception vector and other information
can be extracted from 'VM-Exit Interrupt-Information'
field of VMCS only if bit31 (Valid) is set.
-Intel SDM 24.9.2, Vol3
2. Rename 'exit-interrupt_info' to 'idt_vectoring_info'
in 'struct vcpu_arch', which is consistent with
SDM 24.9.3, Vol3
3. 'IDT-vectoring information' in VMCS is 32bit
-Intel SDM 24.9.3, Vol3
Update the type of 'idt_vectoring_info' in
'struct vcpu_arch'from 'uint32_t' to 'uint64_t'.
Signed-off-by: Yonghua Huang <yonghua.huang@intel.com>
