Enclaves encounter exceptions for lots of reasons: everything from enclave
page faults to NULL pointer dereferences, to system calls that must be
“proxied” to the kernel from outside the enclave.
In addition to the code contained inside an enclave, there is also
supporting code outside the enclave called an “SGX runtime”, which is
virtually always implemented inside a shared library. The runtime helps
build the enclave and handles things like *re*building the enclave if it
got destroyed by something like a suspend/resume cycle.
The rebuilding has traditionally been handled in SIGSEGV handlers,
registered by the library. But, being process-wide, shared state, signal
handling and shared libraries do not mix well.
Introduce a vDSO function call that wraps the enclave entry functions
(EENTER/ERESUME functions of the ENCLU instruciton) and returns information
about any exceptions to the caller in the SGX runtime.
Instead of generating a signal, the kernel places exception information in
RDI, RSI and RDX. The kernel-provided userspace portion of the vDSO handler
will place this information in a user-provided buffer or trigger a
user-provided callback at the time of the exception.
The vDSO function calling convention uses the standard RDI RSI, RDX, RCX,
R8 and R9 registers. This makes it possible to declare the vDSO as a C
prototype, but other than that there is no specific support for SystemV
ABI. Things like storing XSAVE are the responsibility of the enclave and
the runtime.
[ bp: Change vsgx.o build dependency to CONFIG_X86_SGX. ]
Suggested-by: Andy Lutomirski <luto@amacapital.net>
Signed-off-by: Sean Christopherson <sean.j.christopherson@intel.com>
Co-developed-by: Cedric Xing <cedric.xing@intel.com>
Signed-off-by: Cedric Xing <cedric.xing@intel.com>
Co-developed-by: Jarkko Sakkinen <jarkko@kernel.org>
Signed-off-by: Jarkko Sakkinen <jarkko@kernel.org>
Signed-off-by: Borislav Petkov <bp@suse.de>
Tested-by: Jethro Beekman <jethro@fortanix.com>
Link: https://lkml.kernel.org/r/20201112220135.165028-20-jarkko@kernel.org
Signals are a horrid little mechanism. They are especially nasty in
multi-threaded environments because signal state like handlers is global
across the entire process. But, signals are basically the only way that
userspace can “gracefully” handle and recover from exceptions.
The kernel generally does not like exceptions to occur during execution.
But, exceptions are a fact of life and must be handled in some
circumstances. The kernel handles them by keeping a list of individual
instructions which may cause exceptions. Instead of truly handling the
exception and returning to the instruction that caused it, the kernel
instead restarts execution at a *different* instruction. This makes it
obvious to that thread of execution that the exception occurred and lets
*that* code handle the exception instead of the handler.
This is not dissimilar to the try/catch exceptions mechanisms that some
programming languages have, but applied *very* surgically to single
instructions. It effectively changes the visible architecture of the
instruction.
Problem
=======
SGX generates a lot of signals, and the code to enter and exit enclaves and
muck with signal handling is truly horrid. At the same time, an approach
like kernel exception fixup can not be easily applied to userspace
instructions because it changes the visible instruction architecture.
Solution
========
The vDSO is a special page of kernel-provided instructions that run in
userspace. Any userspace calling into the vDSO knows that it is special.
This allows the kernel a place to legitimately rewrite the user/kernel
contract and change instruction behavior.
Add support for fixing up exceptions that occur while executing in the
vDSO. This replaces what could traditionally only be done with signal
handling.
This new mechanism will be used to replace previously direct use of SGX
instructions by userspace.
Just introduce the vDSO infrastructure. Later patches will actually
replace signal generation with vDSO exception fixup.
Suggested-by: Andy Lutomirski <luto@amacapital.net>
Signed-off-by: Sean Christopherson <sean.j.christopherson@intel.com>
Signed-off-by: Jarkko Sakkinen <jarkko@kernel.org>
Signed-off-by: Borislav Petkov <bp@suse.de>
Acked-by: Jethro Beekman <jethro@fortanix.com>
Link: https://lkml.kernel.org/r/20201112220135.165028-17-jarkko@kernel.org
The x86 architecture has a set of page fault error codes. These indicate
things like whether the fault occurred from a write, or whether it
originated in userspace.
The SGX hardware architecture has its own per-page memory management
metadata (EPCM) [*] and hardware which is separate from the normal x86 MMU.
The architecture has a new page fault error code: PF_SGX. This new error
code bit is set whenever a page fault occurs as the result of the SGX MMU.
These faults occur for a variety of reasons. For instance, an access
attempt to enclave memory from outside the enclave causes a PF_SGX fault.
PF_SGX would also be set for permission conflicts, such as if a write to an
enclave page occurs and the page is marked read-write in the x86 page
tables but is read-only in the EPCM.
These faults do not always indicate errors, though. SGX pages are
encrypted with a key that is destroyed at hardware reset, including
suspend. Throwing a SIGSEGV allows user space software to react and recover
when these events occur.
Include PF_SGX in the PF error codes list and throw SIGSEGV when it is
encountered.
[*] Intel SDM: 36.5.1 Enclave Page Cache Map (EPCM)
[ bp: Add bit 15 to the comment above enum x86_pf_error_code too. ]
Signed-off-by: Sean Christopherson <sean.j.christopherson@intel.com>
Signed-off-by: Jarkko Sakkinen <jarkko@kernel.org>
Signed-off-by: Borislav Petkov <bp@suse.de>
Acked-by: Jethro Beekman <jethro@fortanix.com>
Link: https://lkml.kernel.org/r/20201112220135.165028-7-jarkko@kernel.org
The SGX Launch Control hardware helps restrict which enclaves the
hardware will run. Launch control is intended to restrict what software
can run with enclave protections, which helps protect the overall system
from bad enclaves.
For the kernel's purposes, there are effectively two modes in which the
launch control hardware can operate: rigid and flexible. In its rigid
mode, an entity other than the kernel has ultimate authority over which
enclaves can be run (firmware, Intel, etc...). In its flexible mode, the
kernel has ultimate authority over which enclaves can run.
Enable X86_FEATURE_SGX_LC to enumerate when the CPU supports SGX Launch
Control in general.
Add MSR_IA32_SGXLEPUBKEYHASH{0, 1, 2, 3}, which when combined contain a
SHA256 hash of a 3072-bit RSA public key. The hardware allows SGX enclaves
signed with this public key to initialize and run [*]. Enclaves not signed
with this key can not initialize and run.
Add FEAT_CTL_SGX_LC_ENABLED, which informs whether the SGXLEPUBKEYHASH MSRs
can be written by the kernel.
If the MSRs do not exist or are read-only, the launch control hardware is
operating in rigid mode. Linux does not and will not support creating
enclaves when hardware is configured in rigid mode because it takes away
the authority for launch decisions from the kernel. Note, this does not
preclude KVM from virtualizing/exposing SGX to a KVM guest when launch
control hardware is operating in rigid mode.
[*] Intel SDM: 38.1.4 Intel SGX Launch Control Configuration
Signed-off-by: Sean Christopherson <sean.j.christopherson@intel.com>
Co-developed-by: Jarkko Sakkinen <jarkko@kernel.org>
Signed-off-by: Jarkko Sakkinen <jarkko@kernel.org>
Signed-off-by: Borislav Petkov <bp@suse.de>
Acked-by: Jethro Beekman <jethro@fortanix.com>
Link: https://lkml.kernel.org/r/20201112220135.165028-5-jarkko@kernel.org
Populate X86_FEATURE_SGX feature from CPUID and tie it to the Kconfig
option with disabled-features.h.
IA32_FEATURE_CONTROL.SGX_ENABLE must be examined in addition to the CPUID
bits to enable full SGX support. The BIOS must both set this bit and lock
IA32_FEATURE_CONTROL for SGX to be supported (Intel SDM section 36.7.1).
The setting or clearing of this bit has no impact on the CPUID bits above,
which is why it needs to be detected separately.
Signed-off-by: Sean Christopherson <sean.j.christopherson@intel.com>
Co-developed-by: Jarkko Sakkinen <jarkko@kernel.org>
Signed-off-by: Jarkko Sakkinen <jarkko@kernel.org>
Signed-off-by: Borislav Petkov <bp@suse.de>
Acked-by: Jethro Beekman <jethro@fortanix.com>
Link: https://lkml.kernel.org/r/20201112220135.165028-4-jarkko@kernel.org
Pull kvm fixes from Paolo Bonzini:
"Fixes for ARM and x86, the latter especially for old processors
without two-dimensional paging (EPT/NPT)"
* tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm:
kvm: mmu: fix is_tdp_mmu_check when the TDP MMU is not in use
KVM: SVM: Update cr3_lm_rsvd_bits for AMD SEV guests
KVM: x86: Introduce cr3_lm_rsvd_bits in kvm_vcpu_arch
KVM: x86: clflushopt should be treated as a no-op by emulation
KVM: arm64: Handle SCXTNUM_ELx traps
KVM: arm64: Unify trap handlers injecting an UNDEF
KVM: arm64: Allow setting of ID_AA64PFR0_EL1.CSV2 from userspace
Pull x86 fixes from Thomas Gleixner:
"A small set of fixes for x86:
- Cure the fallout from the MSI irqdomain overhaul which missed that
the Intel IOMMU does not register virtual function devices and
therefore never reaches the point where the MSI interrupt domain is
assigned. This made the VF devices use the non-remapped MSI domain
which is trapped by the IOMMU/remap unit
- Remove an extra space in the SGI_UV architecture type procfs output
for UV5
- Remove a unused function which was missed when removing the UV BAU
TLB shootdown handler"
* tag 'x86-urgent-2020-11-15' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
iommu/vt-d: Cure VF irqdomain hickup
x86/platform/uv: Fix copied UV5 output archtype
x86/platform/uv: Drop last traces of uv_flush_tlb_others
This patch is heavily based on previous work from Lei Cao
<lei.cao@stratus.com> and Paolo Bonzini <pbonzini@redhat.com>. [1]
KVM currently uses large bitmaps to track dirty memory. These bitmaps
are copied to userspace when userspace queries KVM for its dirty page
information. The use of bitmaps is mostly sufficient for live
migration, as large parts of memory are be dirtied from one log-dirty
pass to another. However, in a checkpointing system, the number of
dirty pages is small and in fact it is often bounded---the VM is
paused when it has dirtied a pre-defined number of pages. Traversing a
large, sparsely populated bitmap to find set bits is time-consuming,
as is copying the bitmap to user-space.
A similar issue will be there for live migration when the guest memory
is huge while the page dirty procedure is trivial. In that case for
each dirty sync we need to pull the whole dirty bitmap to userspace
and analyse every bit even if it's mostly zeros.
The preferred data structure for above scenarios is a dense list of
guest frame numbers (GFN). This patch series stores the dirty list in
kernel memory that can be memory mapped into userspace to allow speedy
harvesting.
This patch enables dirty ring for X86 only. However it should be
easily extended to other archs as well.
[1] https://patchwork.kernel.org/patch/10471409/
Signed-off-by: Lei Cao <lei.cao@stratus.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Peter Xu <peterx@redhat.com>
Message-Id: <20201001012222.5767-1-peterx@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Originally, we have three code paths that can dirty a page without
vcpu context for X86:
- init_rmode_identity_map
- init_rmode_tss
- kvmgt_rw_gpa
init_rmode_identity_map and init_rmode_tss will be setup on
destination VM no matter what (and the guest cannot even see them), so
it does not make sense to track them at all.
To do this, allow __x86_set_memory_region() to return the userspace
address that just allocated to the caller. Then in both of the
functions we directly write to the userspace address instead of
calling kvm_write_*() APIs.
Another trivial change is that we don't need to explicitly clear the
identity page table root in init_rmode_identity_map() because no
matter what we'll write to the whole page with 4M huge page entries.
Suggested-by: Paolo Bonzini <pbonzini@redhat.com>
Reviewed-by: Sean Christopherson <sean.j.christopherson@intel.com>
Signed-off-by: Peter Xu <peterx@redhat.com>
Message-Id: <20201001012044.5151-4-peterx@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Background: We have a lightweight HV, it needs INIT-VMExit and
SIPI-VMExit to wake-up APs for guests since it do not monitor
the Local APIC. But currently virtual wait-for-SIPI(WFS) state
is not supported in nVMX, so when running on top of KVM, the L1
HV cannot receive the INIT-VMExit and SIPI-VMExit which cause
the L2 guest cannot wake up the APs.
According to Intel SDM Chapter 25.2 Other Causes of VM Exits,
SIPIs cause VM exits when a logical processor is in
wait-for-SIPI state.
In this patch:
1. introduce SIPI exit reason,
2. introduce wait-for-SIPI state for nVMX,
3. advertise wait-for-SIPI support to guest.
When L1 hypervisor is not monitoring Local APIC, L0 need to emulate
INIT-VMExit and SIPI-VMExit to L1 to emulate INIT-SIPI-SIPI for
L2. L2 LAPIC write would be traped by L0 Hypervisor(KVM), L0 should
emulate the INIT/SIPI vmexit to L1 hypervisor to set proper state
for L2's vcpu state.
Handle procdure:
Source vCPU:
L2 write LAPIC.ICR(INIT).
L0 trap LAPIC.ICR write(INIT): inject a latched INIT event to target
vCPU.
Target vCPU:
L0 emulate an INIT VMExit to L1 if is guest mode.
L1 set guest VMCS, guest_activity_state=WAIT_SIPI, vmresume.
L0 set vcpu.mp_state to INIT_RECEIVED if (vmcs12.guest_activity_state
== WAIT_SIPI).
Source vCPU:
L2 write LAPIC.ICR(SIPI).
L0 trap LAPIC.ICR write(INIT): inject a latched SIPI event to traget
vCPU.
Target vCPU:
L0 emulate an SIPI VMExit to L1 if (vcpu.mp_state == INIT_RECEIVED).
L1 set CS:IP, guest_activity_state=ACTIVE, vmresume.
L0 resume to L2.
L2 start-up.
Signed-off-by: Yadong Qi <yadong.qi@intel.com>
Message-Id: <20200922052343.84388-1-yadong.qi@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Message-Id: <20201106065122.403183-1-yadong.qi@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Split out VMX's checks on CR4.VMXE to a dedicated hook, .is_valid_cr4(),
and invoke the new hook from kvm_valid_cr4(). This fixes an issue where
KVM_SET_SREGS would return success while failing to actually set CR4.
Fixing the issue by explicitly checking kvm_x86_ops.set_cr4()'s return
in __set_sregs() is not a viable option as KVM has already stuffed a
variety of vCPU state.
Note, kvm_valid_cr4() and is_valid_cr4() have different return types and
inverted semantics. This will be remedied in a future patch.
Fixes: 5e1746d620 ("KVM: nVMX: Allow setting the VMXE bit in CR4")
Signed-off-by: Sean Christopherson <sean.j.christopherson@intel.com>
Message-Id: <20201007014417.29276-5-sean.j.christopherson@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
When CONFIG_HAVE_DYNAMIC_FTRACE_WITH_ARGS is available, the ftrace call
will be able to set the ip of the calling function. This will improve the
performance of live kernel patching where it does not need all the regs to
be stored just to change the instruction pointer.
If all archs that support live kernel patching also support
HAVE_DYNAMIC_FTRACE_WITH_ARGS, then the architecture specific function
klp_arch_set_pc() could be made generic.
It is possible that an arch can support HAVE_DYNAMIC_FTRACE_WITH_ARGS but
not HAVE_DYNAMIC_FTRACE_WITH_REGS and then have access to live patching.
Cc: Josh Poimboeuf <jpoimboe@redhat.com>
Cc: Jiri Kosina <jikos@kernel.org>
Cc: live-patching@vger.kernel.org
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: Miroslav Benes <mbenes@suse.cz>
Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
Currently, the only way to get access to the registers of a function via a
ftrace callback is to set the "FL_SAVE_REGS" bit in the ftrace_ops. But as this
saves all regs as if a breakpoint were to trigger (for use with kprobes), it
is expensive.
The regs are already saved on the stack for the default ftrace callbacks, as
that is required otherwise a function being traced will get the wrong
arguments and possibly crash. And on x86, the arguments are already stored
where they would be on a pt_regs structure to use that code for both the
regs version of a callback, it makes sense to pass that information always
to all functions.
If an architecture does this (as x86_64 now does), it is to set
HAVE_DYNAMIC_FTRACE_WITH_ARGS, and this will let the generic code that it
could have access to arguments without having to set the flags.
This also includes having the stack pointer being saved, which could be used
for accessing arguments on the stack, as well as having the function graph
tracer not require its own trampoline!
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
SEV guests fail to boot on a system that supports the PCID feature.
While emulating the RSM instruction, KVM reads the guest CR3
and calls kvm_set_cr3(). If the vCPU is in the long mode,
kvm_set_cr3() does a sanity check for the CR3 value. In this case,
it validates whether the value has any reserved bits set. The
reserved bit range is 63:cpuid_maxphysaddr(). When AMD memory
encryption is enabled, the memory encryption bit is set in the CR3
value. The memory encryption bit may fall within the KVM reserved
bit range, causing the KVM emulation failure.
Introduce a new field cr3_lm_rsvd_bits in kvm_vcpu_arch which will
cache the reserved bits in the CR3 value. This will be initialized
to rsvd_bits(cpuid_maxphyaddr(vcpu), 63).
If the architecture has any special bits(like AMD SEV encryption bit)
that needs to be masked from the reserved bits, should be cleared
in vendor specific kvm_x86_ops.vcpu_after_set_cpuid handler.
Fixes: a780a3ea62 ("KVM: X86: Fix reserved bits check for MOV to CR3")
Signed-off-by: Babu Moger <babu.moger@amd.com>
Message-Id: <160521947657.32054.3264016688005356563.stgit@bmoger-ubuntu>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Non RT kernels need to protect FPU against preemption and bottom half
processing. This is achieved by disabling bottom halfs via
local_bh_disable() which implictly disables preemption.
On RT kernels this protection mechanism is not sufficient because
local_bh_disable() does not disable preemption. It serializes bottom half
related processing via a CPU local lock.
As bottom halfs are running always in thread context on RT kernels
disabling preemption is the proper choice as it implicitly prevents bottom
half processing.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Link: https://lore.kernel.org/r/20201027101349.588965083@linutronix.de
Commit 39297dde73 ("x86/platform/uv: Remove UV BAU TLB Shootdown
Handler") removed uv_flush_tlb_others. Its declaration was removed also
from asm/uv/uv.h. But only for the CONFIG_X86_UV=y case. The inline
definition (!X86_UV case) is still in place.
So remove this implementation with everything what was added to support
uv_flush_tlb_others:
* include of asm/tlbflush.h
* forward declarations of struct cpumask, mm_struct, and flush_tlb_info
Signed-off-by: Jiri Slaby <jslaby@suse.cz>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Acked-by: Mike Travis <mike.travis@hpe.com>
Acked-by: Steve Wahl <steve.wahl@hpe.com>
Link: https://lore.kernel.org/r/20201109093653.2042-1-jslaby@suse.cz
Enable AMD Fam17h RAPL support for the power capping framework.
The support is as per AMD Fam17h Model31h (Zen2) and model 00-ffh
(Zen1) PPR.
Tested by comparing the results of following two sysfs entries and the
values directly read from corresponding MSRs via /dev/cpu/[x]/msr:
/sys/class/powercap/intel-rapl/intel-rapl:0/energy_uj
/sys/class/powercap/intel-rapl/intel-rapl:0/intel-rapl:0:0/energy_uj
Signed-off-by: Victor Ding <victording@google.com>
Acked-by: Kim Phillips <kim.phillips@amd.com>
[ rjw: Changelog edits ]
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Starting with Arch Perfmon v5, the anythread filter on generic counters may be
deprecated. The current kernel was exporting the any filter without checking.
On Icelake, it means you could do cpu/event=0x3c,any/ even though the filter
does not exist. This patch corrects the problem by relying on the CPUID 0xa leaf
function to determine if anythread is supported or not as described in the
Intel SDM Vol3b 18.2.5.1 AnyThread Deprecation section.
Signed-off-by: Stephane Eranian <eranian@google.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20201028194247.3160610-1-eranian@google.com
Conflicts:
include/asm-generic/atomic-instrumented.h
kernel/kprobes.c
Use the upstream atomic-instrumented.h checksum, and pick
the kprobes version of kernel/kprobes.c, which effectively
reverts this upstream workaround:
645f224e7b: ("kprobes: Tell lockdep about kprobe nesting")
Since the new code *should* be fine without nesting.
Knock on wood ...
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Move the x86 IMA arch code into security/integrity/ima/ima_efi.c,
so that we will be able to wire it up for arm64 in a future patch.
Co-developed-by: Chester Lin <clin@suse.com>
Signed-off-by: Chester Lin <clin@suse.com>
Acked-by: Mimi Zohar <zohar@linux.ibm.com>
Signed-off-by: Ard Biesheuvel <ardb@kernel.org>
Lockdep state handling on NMI enter and exit is nothing specific to X86. It's
not any different on other architectures. Also the extra state type is not
necessary, irqentry_state_t can carry the necessary information as well.
Move it to common code and extend irqentry_state_t to carry lockdep state.
[ Ira: Make exit_rcu and lockdep a union as they are mutually exclusive
between the IRQ and NMI exceptions, and add kernel documentation for
struct irqentry_state_t ]
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ira Weiny <ira.weiny@intel.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Link: https://lore.kernel.org/r/20201102205320.1458656-7-ira.weiny@intel.com
When a Linux VM runs on Hyper-V, if the VM has CPUs with >255 APIC IDs,
the CPUs can't be the destination of IOAPIC interrupts, because the
IOAPIC RTE's Dest Field has only 8 bits. Currently the hackery driver
drivers/iommu/hyperv-iommu.c is used to ensure IOAPIC interrupts are
only routed to CPUs that don't have >255 APIC IDs. However, there is
an issue with kdump, because the kdump kernel can run on any CPU, and
hence IOAPIC interrupts can't work if the kdump kernel run on a CPU
with a >255 APIC ID.
The kdump issue can be fixed by the Extended Dest ID, which is introduced
recently by David Woodhouse (for IOAPIC, see the field virt_destid_8_14 in
struct IO_APIC_route_entry). Of course, the Extended Dest ID needs the
support of the underlying hypervisor. The latest Hyper-V has added the
support recently: with this commit, on such a Hyper-V host, Linux VM
does not use hyperv-iommu.c because hyperv_prepare_irq_remapping()
returns -ENODEV; instead, Linux kernel's generic support of Extended Dest
ID from David is used, meaning that Linux VM is able to support up to
32K CPUs, and IOAPIC interrupts can be routed to all the CPUs.
On an old Hyper-V host that doesn't support the Extended Dest ID, nothing
changes with this commit: Linux VM is still able to bring up the CPUs with
> 255 APIC IDs with the help of hyperv-iommu.c, but IOAPIC interrupts still
can not go to such CPUs, and the kdump kernel still can not work properly
on such CPUs.
[ tglx: Updated comment as suggested by David ]
Signed-off-by: Dexuan Cui <decui@microsoft.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Acked-by: David Woodhouse <dwmw@amazon.co.uk>
Link: https://lore.kernel.org/r/20201103011136.59108-1-decui@microsoft.com
The Xeon versions of Sandy Bridge, Ivy Bridge and Haswell support an
optional additional error logging mode which is enabled by an MSR.
Previously, this mode was enabled from the mcelog(8) tool via /dev/cpu,
but userspace should not be poking at MSRs. So move the enabling into
the kernel.
[ bp: Correct the explanation why this is done. ]
Suggested-by: Boris Petkov <bp@alien8.de>
Signed-off-by: Tony Luck <tony.luck@intel.com>
Signed-off-by: Borislav Petkov <bp@suse.de>
Link: https://lkml.kernel.org/r/20201030190807.GA13884@agluck-desk2.amr.corp.intel.com
Commit
b4e0409a36 ("x86: check vmlinux limits, 64-bit")
added a check that the size of the 64-bit kernel is less than
KERNEL_IMAGE_SIZE.
The check uses (_end - _text), but this is not enough. The initial
PMD used in startup_64() (level2_kernel_pgt) can only map upto
KERNEL_IMAGE_SIZE from __START_KERNEL_map, not from _text, and the
modules area (MODULES_VADDR) starts at KERNEL_IMAGE_SIZE.
The correct check is what is currently done for 32-bit, since
LOAD_OFFSET is defined appropriately for the two architectures. Just
check (_end - LOAD_OFFSET) against KERNEL_IMAGE_SIZE unconditionally.
Note that on 32-bit, the limit is not strict: KERNEL_IMAGE_SIZE is not
really used by the main kernel. The higher the kernel is located, the
less the space available for the vmalloc area. However, it is used by
KASLR in the compressed stub to limit the maximum address of the kernel
to a safe value.
Clean up various comments to clarify that despite the name,
KERNEL_IMAGE_SIZE is not a limit on the size of the kernel image, but a
limit on the maximum virtual address that the image can occupy.
Signed-off-by: Arvind Sankar <nivedita@alum.mit.edu>
Signed-off-by: Borislav Petkov <bp@suse.de>
Link: https://lkml.kernel.org/r/20201029161903.2553528-1-nivedita@alum.mit.edu
Some hypervisors can allow the guest to use the Extended Destination ID
field in the MSI address to address up to 32768 CPUs.
This applies to all downstream devices which generate MSI cycles,
including HPET, I/O-APIC and PCI MSI.
HPET and PCI MSI use the same __irq_msi_compose_msg() function, while
I/O-APIC generates its own and had support for the extended bits added in
a previous commit.
Signed-off-by: David Woodhouse <dwmw@amazon.co.uk>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Link: https://lore.kernel.org/r/20201024213535.443185-33-dwmw2@infradead.org
Bits 63-48 of the I/OAPIC Redirection Table Entry map directly to bits 19-4
of the address used in the resulting MSI cycle.
Historically, the x86 MSI format only used the top 8 of those 16 bits as
the destination APIC ID, and the "Extended Destination ID" in the lower 8
bits was unused.
With interrupt remapping, the lowest bit of the Extended Destination ID
(bit 48 of RTE, bit 4 of MSI address) is now used to indicate a remappable
format MSI.
A hypervisor can use the other 7 bits of the Extended Destination ID to
permit guests to address up to 15 bits of APIC IDs, thus allowing 32768
vCPUs before having to expose a vIOMMU and interrupt remapping to the
guest.
No behavioural change in this patch, since nothing yet permits APIC IDs
above 255 to be used with the non-IR I/OAPIC domain.
[ tglx: Converted it to the cleaned up entry/msi_msg format and added
commentry ]
Signed-off-by: David Woodhouse <dwmw@amazon.co.uk>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Link: https://lore.kernel.org/r/20201024213535.443185-32-dwmw2@infradead.org
The I/O-APIC generates an MSI cycle with address/data bits taken from its
Redirection Table Entry in some combination which used to make sense, but
now is just a bunch of bits which get passed through in some seemingly
arbitrary order.
Instead of making IRQ remapping drivers directly frob the I/OA-PIC RTE, let
them just do their job and generate an MSI message. The bit swizzling to
turn that MSI message into the I/O-APIC's RTE is the same in all cases,
since it's a function of the I/O-APIC hardware. The IRQ remappers have no
real need to get involved with that.
The only slight caveat is that the I/OAPIC is interpreting some of those
fields too, and it does want the 'vector' field to be unique to make EOI
work. The AMD IOMMU happens to put its IRTE index in the bits that the
I/O-APIC thinks are the vector field, and accommodates this requirement by
reserving the first 32 indices for the I/O-APIC. The Intel IOMMU doesn't
actually use the bits that the I/O-APIC thinks are the vector field, so it
fills in the 'pin' value there instead.
[ tglx: Replaced the unreadably macro maze with the cleaned up RTE/msi_msg
bitfields and added commentry to explain the mapping magic ]
Signed-off-by: David Woodhouse <dwmw@amazon.co.uk>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Link: https://lore.kernel.org/r/20201024213535.443185-22-dwmw2@infradead.org
Having two seperate structs for the I/O-APIC RTE entries (non-remapped and
DMAR remapped) requires type casts and makes it hard to map.
Combine them in IO_APIC_routing_entry by defining a union of two 64bit
bitfields. Use naming which reflects which bits are shared and which bits
are actually different for the operating modes.
[dwmw2: Fix it up and finish the job, pulling the 32-bit w1,w2 words for
register access into the same union and eliminating a few more
places where bits were accessed through masks and shifts.]
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: David Woodhouse <dwmw@amazon.co.uk>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Link: https://lore.kernel.org/r/20201024213535.443185-21-dwmw2@infradead.org
'trigger' and 'polarity' are used throughout the I/O-APIC code for handling
the trigger type (edge/level) and the active low/high configuration. While
there are defines for initializing these variables and struct members, they
are not used consequently and the meaning of 'trigger' and 'polarity' is
opaque and confusing at best.
Rename them to 'is_level' and 'active_low' and make them boolean in various
structs so it's entirely clear what the meaning is.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: David Woodhouse <dwmw@amazon.co.uk>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Link: https://lore.kernel.org/r/20201024213535.443185-20-dwmw2@infradead.org
Create shadow structs with named bitfields for msi_msg data, address_lo and
address_hi and use them in the MSI message composer.
Provide a function to retrieve the destination ID. This could be inline,
but that'd create a circular header dependency.
[dwmw2: fix bitfields not all to be a union]
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: David Woodhouse <dwmw@amazon.co.uk>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Link: https://lore.kernel.org/r/20201024213535.443185-13-dwmw2@infradead.org
struct apic has two members which store information about the destination
mode: dest_logical and irq_dest_mode.
dest_logical contains a mask which was historically used to set the
destination mode in IPI messages. Over time the usage was reduced and the
logical/physical functions were seperated.
There are only a few places which still use 'dest_logical' but they can
use 'irq_dest_mode' instead.
irq_dest_mode is actually a boolean where 0 means physical destination mode
and 1 means logical destination mode. Of course the name does not reflect
the functionality. This will be cleaned up in a subsequent change.
Remove apic::dest_logical and fixup the remaining users.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: David Woodhouse <dwmw@amazon.co.uk>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Link: https://lore.kernel.org/r/20201024213535.443185-8-dwmw2@infradead.org
Currently, Linux as a hypervisor guest will enable x2apic only if there are
no CPUs present at boot time with an APIC ID above 255.
Hotplugging a CPU later with a higher APIC ID would result in a CPU which
cannot be targeted by external interrupts.
Add a filter in x2apic_apic_id_valid() which can be used to prevent such
CPUs from coming online, and allow x2apic to be enabled even if they are
present at boot time.
Fixes: ce69a78450 ("x86/apic: Enable x2APIC without interrupt remapping under KVM")
Signed-off-by: David Woodhouse <dwmw@amazon.co.uk>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Link: https://lore.kernel.org/r/20201024213535.443185-2-dwmw2@infradead.org
