With the following commit:
73d5e2b472 ("cpu/hotplug: detect SMT disabled by BIOS")
... the hotplug code attempted to detect when SMT was disabled by BIOS,
in which case it reported SMT as permanently disabled. However, that
code broke a virt hotplug scenario, where the guest is booted with only
primary CPU threads, and a sibling is brought online later.
The problem is that there doesn't seem to be a way to reliably
distinguish between the HW "SMT disabled by BIOS" case and the virt
"sibling not yet brought online" case. So the above-mentioned commit
was a bit misguided, as it permanently disabled SMT for both cases,
preventing future virt sibling hotplugs.
Going back and reviewing the original problems which were attempted to
be solved by that commit, when SMT was disabled in BIOS:
1) /sys/devices/system/cpu/smt/control showed "on" instead of
"notsupported"; and
2) vmx_vm_init() was incorrectly showing the L1TF_MSG_SMT warning.
I'd propose that we instead consider #1 above to not actually be a
problem. Because, at least in the virt case, it's possible that SMT
wasn't disabled by BIOS and a sibling thread could be brought online
later. So it makes sense to just always default the smt control to "on"
to allow for that possibility (assuming cpuid indicates that the CPU
supports SMT).
The real problem is #2, which has a simple fix: change vmx_vm_init() to
query the actual current SMT state -- i.e., whether any siblings are
currently online -- instead of looking at the SMT "control" sysfs value.
So fix it by:
a) reverting the original "fix" and its followup fix:
73d5e2b472 ("cpu/hotplug: detect SMT disabled by BIOS")
bc2d8d262c ("cpu/hotplug: Fix SMT supported evaluation")
and
b) changing vmx_vm_init() to query the actual current SMT state --
instead of the sysfs control value -- to determine whether the L1TF
warning is needed. This also requires the 'sched_smt_present'
variable to exported, instead of 'cpu_smt_control'.
Fixes: 73d5e2b472 ("cpu/hotplug: detect SMT disabled by BIOS")
Reported-by: Igor Mammedov <imammedo@redhat.com>
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Cc: Joe Mario <jmario@redhat.com>
Cc: Jiri Kosina <jikos@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: kvm@vger.kernel.org
Cc: stable@vger.kernel.org
Link: https://lkml.kernel.org/r/e3a85d585da28cc333ecbc1e78ee9216e6da9396.1548794349.git.jpoimboe@redhat.com
Pull x86 fixes from Thomas Gleixner:
"A set of fixes for x86:
- Fix the swapped outb() parameters in the KASLR code
- Fix the PKEY handling at fork which missed to preserve the pkey
state for the child. Comes with a test case to validate that.
- Fix the entry stack handling for XEN PV to respect that XEN PV
systems enter the function already on the current thread stack and
not on the trampoline.
- Fix kexec load failure caused by using a stale value when the
kexec_buf structure is reused for subsequent allocations.
- Fix a bogus sizeof() in the memory encryption code
- Enforce PCI dependency for the Intel Low Power Subsystem
- Enforce PCI_LOCKLESS_CONFIG when PCI is enabled"
* 'x86-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
x86/Kconfig: Select PCI_LOCKLESS_CONFIG if PCI is enabled
x86/entry/64/compat: Fix stack switching for XEN PV
x86/kexec: Fix a kexec_file_load() failure
x86/mm/mem_encrypt: Fix erroneous sizeof()
x86/selftests/pkeys: Fork() to check for state being preserved
x86/pkeys: Properly copy pkey state at fork()
x86/kaslr: Fix incorrect i8254 outb() parameters
x86/intel/lpss: Make PCI dependency explicit
Pull x86 timer fixes from Thomas Gleixner:
"Two commits which were missed to be sent during the merge window.
- The TSC calibration fix turns out to be more urgent as recent
Skylake-X systems seem to have massive trouble with calibration
disturbance. This should go back into stable for that reason and it
the risk of breakage is rather low.
- Drop an unused define"
* 'x86-timers-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
x86/hpet: Remove unused FSEC_PER_NSEC define
x86/tsc: Make calibration refinement more robust
and virtual APIC, 32-bit fixes, an important fix to restore operation on older
processors, and a bunch of hyper-v bugfixes. Several are marked stable.
There are also fixes for GCC warnings and for a GCC/objtool interaction.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.22 (GNU/Linux)
iQEcBAABAgAGBQJcTBqRAAoJEL/70l94x66DF8wH/0LKWQd4ay54OgpfdPRcUqXa
yW10qqnO5OZfpI19UU2rfv2QXrdSOnDafgN6xiQ2Dz7m3JsZB7gBDsCCYKzeUgCz
YB0UVcX1ZS+gr3igDDHIw3lWPBUqDIzKmEJO++9nAbDi4gOmWaPQ8vWrfORWAZcl
yx2nCjeljjbO65UdRTdr3TkUNbpFlJ2NEUrzzco8OgChNB9QoxLTSJHrZxeZ7dNn
J/ZDAaBwRxXN/aKH0A3+pwUFrP5nGuronT6nGo1048WWrlQzdMp7qh8fPtTBvWJ4
uqUrrYc7jY/EhfZ4k/aAUGkAdt4IZI1KyHjhqtmB9zf+hezphUJv66QYQGVTFts=
=yUth
-----END PGP SIGNATURE-----
Merge tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm
Pull KVM fixes from Paolo Bonzini:
"Quite a few fixes for x86: nested virtualization save/restore, AMD
nested virtualization and virtual APIC, 32-bit fixes, an important fix
to restore operation on older processors, and a bunch of hyper-v
bugfixes. Several are marked stable.
There are also fixes for GCC warnings and for a GCC/objtool interaction"
* tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm:
KVM: x86: Mark expected switch fall-throughs
KVM: x86: fix TRACE_INCLUDE_PATH and remove -I. header search paths
KVM: selftests: check returned evmcs version range
x86/kvm/hyper-v: nested_enable_evmcs() sets vmcs_version incorrectly
KVM: VMX: Move vmx_vcpu_run()'s VM-Enter asm blob to a helper function
kvm: selftests: Fix region overlap check in kvm_util
kvm: vmx: fix some -Wmissing-prototypes warnings
KVM: nSVM: clear events pending from svm_complete_interrupts() when exiting to L1
svm: Fix AVIC incomplete IPI emulation
svm: Add warning message for AVIC IPI invalid target
KVM: x86: WARN_ONCE if sending a PV IPI returns a fatal error
KVM: x86: Fix PV IPIs for 32-bit KVM host
x86/kvm/hyper-v: recommend using eVMCS only when it is enabled
x86/kvm/hyper-v: don't recommend doing reset via synthetic MSR
kvm: x86/vmx: Use kzalloc for cached_vmcs12
KVM: VMX: Use the correct field var when clearing VM_ENTRY_LOAD_IA32_PERF_GLOBAL_CTRL
KVM: x86: Fix single-step debugging
x86/kvm/hyper-v: don't announce GUEST IDLE MSR support
In preparation to enabling -Wimplicit-fallthrough, mark switch
cases where we are expecting to fall through.
This patch fixes the following warnings:
arch/x86/kvm/lapic.c:1037:27: warning: this statement may fall through [-Wimplicit-fallthrough=]
arch/x86/kvm/lapic.c:1876:3: warning: this statement may fall through [-Wimplicit-fallthrough=]
arch/x86/kvm/hyperv.c:1637:6: warning: this statement may fall through [-Wimplicit-fallthrough=]
arch/x86/kvm/svm.c:4396:6: warning: this statement may fall through [-Wimplicit-fallthrough=]
arch/x86/kvm/mmu.c:4372:36: warning: this statement may fall through [-Wimplicit-fallthrough=]
arch/x86/kvm/x86.c:3835:6: warning: this statement may fall through [-Wimplicit-fallthrough=]
arch/x86/kvm/x86.c:7938:23: warning: this statement may fall through [-Wimplicit-fallthrough=]
arch/x86/kvm/vmx/vmx.c:2015:6: warning: this statement may fall through [-Wimplicit-fallthrough=]
arch/x86/kvm/vmx/vmx.c:1773:6: warning: this statement may fall through [-Wimplicit-fallthrough=]
Warning level 3 was used: -Wimplicit-fallthrough=3
This patch is part of the ongoing efforts to enabling -Wimplicit-fallthrough.
Signed-off-by: Gustavo A. R. Silva <gustavo@embeddedor.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
The header search path -I. in kernel Makefiles is very suspicious;
it allows the compiler to search for headers in the top of $(srctree),
where obviously no header file exists.
The reason of having -I. here is to make the incorrectly set
TRACE_INCLUDE_PATH working.
As the comment block in include/trace/define_trace.h says,
TRACE_INCLUDE_PATH should be a relative path to the define_trace.h
Fix the TRACE_INCLUDE_PATH, and remove the iffy include paths.
Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Commit e2e871ab2f ("x86/kvm/hyper-v: Introduce nested_get_evmcs_version()
helper") broke EVMCS enablement: to set vmcs_version we now call
nested_get_evmcs_version() but this function checks
enlightened_vmcs_enabled flag which is not yet set so we end up returning
zero.
Fix the issue by re-arranging things in nested_enable_evmcs().
Fixes: e2e871ab2f ("x86/kvm/hyper-v: Introduce nested_get_evmcs_version() helper")
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
...along with the function's STACK_FRAME_NON_STANDARD tag. Moving the
asm blob results in a significantly smaller amount of code that is
marked with STACK_FRAME_NON_STANDARD, which makes it far less likely
that gcc will split the function and trigger a spurious objtool warning.
As a bonus, removing STACK_FRAME_NON_STANDARD from vmx_vcpu_run() allows
the bulk of code to be properly checked by objtool.
Because %rbp is not loaded via VMCS fields, vmx_vcpu_run() must manually
save/restore the host's RBP and load the guest's RBP prior to calling
vmx_vmenter(). Modifying %rbp triggers objtool's stack validation code,
and so vmx_vcpu_run() is tagged with STACK_FRAME_NON_STANDARD since it's
impossible to avoid modifying %rbp.
Unfortunately, vmx_vcpu_run() is also a gigantic function that gcc will
split into separate functions, e.g. so that pieces of the function can
be inlined. Splitting the function means that the compiled Elf file
will contain one or more vmx_vcpu_run.part.* functions in addition to
a vmx_vcpu_run function. Depending on where the function is split,
objtool may warn about a "call without frame pointer save/setup" in
vmx_vcpu_run.part.* since objtool's stack validation looks for exact
names when whitelisting functions tagged with STACK_FRAME_NON_STANDARD.
Up until recently, the undesirable function splitting was effectively
blocked because vmx_vcpu_run() was tagged with __noclone. At the time,
__noclone had an unintended side effect that put vmx_vcpu_run() into a
separate optimization unit, which in turn prevented gcc from inlining
the function (or any of its own function calls) and thus eliminated gcc's
motivation to split the function. Removing the __noclone attribute
allowed gcc to optimize vmx_vcpu_run(), exposing the objtool warning.
Kudos to Qian Cai for root causing that the fnsplit optimization is what
caused objtool to complain.
Fixes: 453eafbe65 ("KVM: VMX: Move VM-Enter + VM-Exit handling to non-inline sub-routines")
Tested-by: Qian Cai <cai@lca.pw>
Cc: Josh Poimboeuf <jpoimboe@redhat.com>
Reported-by: kbuild test robot <lkp@intel.com>
Signed-off-by: Sean Christopherson <sean.j.christopherson@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
We get some warnings when building kernel with W=1:
arch/x86/kvm/vmx/vmx.c:426:5: warning: no previous prototype for ‘kvm_fill_hv_flush_list_func’ [-Wmissing-prototypes]
arch/x86/kvm/vmx/nested.c:58:6: warning: no previous prototype for ‘init_vmcs_shadow_fields’ [-Wmissing-prototypes]
Make them static to fix this.
Signed-off-by: Yi Wang <wang.yi59@zte.com.cn>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
kvm-unit-tests' eventinj "NMI failing on IDT" test results in NMI being
delivered to the host (L1) when it's running nested. The problem seems to
be: svm_complete_interrupts() raises 'nmi_injected' flag but later we
decide to reflect EXIT_NPF to L1. The flag remains pending and we do NMI
injection upon entry so it got delivered to L1 instead of L2.
It seems that VMX code solves the same issue in prepare_vmcs12(), this was
introduced with code refactoring in commit 5f3d579997 ("KVM: nVMX: Rework
event injection and recovery").
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
In case of incomplete IPI with invalid interrupt type, the current
SVM driver does not properly emulate the IPI, and fails to boot
FreeBSD guests with multiple vcpus when enabling AVIC.
Fix this by update APIC ICR high/low registers, which also
emulate sending the IPI.
Signed-off-by: Suravee Suthikulpanit <suravee.suthikulpanit@amd.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Print warning message when IPI target ID is invalid due to one of
the following reasons:
* In logical mode: cluster > max_cluster (64)
* In physical mode: target > max_physical (512)
* Address is not present in the physical or logical ID tables
Signed-off-by: Suravee Suthikulpanit <suravee.suthikulpanit@amd.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
KVM hypercalls return a negative value error code in case of a fatal
error, e.g. when the hypercall isn't supported or was made with invalid
parameters. WARN_ONCE on fatal errors when sending PV IPIs as any such
error all but guarantees an SMP system will hang due to a missing IPI.
Fixes: aaffcfd1e8 ("KVM: X86: Implement PV IPIs in linux guest")
Cc: stable@vger.kernel.org
Cc: Wanpeng Li <wanpengli@tencent.com>
Signed-off-by: Sean Christopherson <sean.j.christopherson@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
The recognition of the KVM_HC_SEND_IPI hypercall was unintentionally
wrapped in "#ifdef CONFIG_X86_64", causing 32-bit KVM hosts to reject
any and all PV IPI requests despite advertising the feature. This
results in all KVM paravirtualized guests hanging during SMP boot due
to IPIs never being delivered.
Fixes: 4180bf1b65 ("KVM: X86: Implement "send IPI" hypercall")
Cc: stable@vger.kernel.org
Cc: Wanpeng Li <wanpengli@tencent.com>
Signed-off-by: Sean Christopherson <sean.j.christopherson@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
We shouldn't probably be suggesting using Enlightened VMCS when it's not
enabled (not supported from guest's point of view). Hyper-V on KVM seems
to be fine either way but let's be consistent.
Fixes: 2bc39970e9 ("x86/kvm/hyper-v: Introduce KVM_GET_SUPPORTED_HV_CPUID")
Reviewed-by: Liran Alon <liran.alon@oracle.com>
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
System reset through synthetic MSR is not recommended neither by genuine
Hyper-V nor my QEMU.
Fixes: 2bc39970e9 ("x86/kvm/hyper-v: Introduce KVM_GET_SUPPORTED_HV_CPUID")
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Reviewed-by: Liran Alon <liran.alon@oracle.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
This changes the allocation of cached_vmcs12 to use kzalloc instead of
kmalloc. This removes the information leak found by Syzkaller (see
Reported-by) in this case and prevents similar leaks from happening
based on cached_vmcs12.
It also changes vmx_get_nested_state to copy out the full 4k VMCS12_SIZE
in copy_to_user rather than only the size of the struct.
Tested: rebuilt against head, booted, and ran the syszkaller repro
https://syzkaller.appspot.com/text?tag=ReproC&x=174efca3400000 without
observing any problems.
Reported-by: syzbot+ded1696f6b50b615b630@syzkaller.appspotmail.com
Fixes: 8fcc4b5923
Cc: stable@vger.kernel.org
Signed-off-by: Tom Roeder <tmroeder@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Fix a recently introduced bug that results in the wrong VMCS control
field being updated when applying a IA32_PERF_GLOBAL_CTRL errata.
Fixes: c73da3fcab ("KVM: VMX: Properly handle dynamic VM Entry/Exit controls")
Reported-by: Harald Arnesen <harald@skogtun.org>
Tested-by: Harald Arnesen <harald@skogtun.org>
Signed-off-by: Sean Christopherson <sean.j.christopherson@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
The single-step debugging of KVM guests on x86 is broken: if we run
gdb 'stepi' command at the breakpoint when the guest interrupts are
enabled, RIP always jumps to native_apic_mem_write(). Then other
nasty effects follow.
Long investigation showed that on Jun 7, 2017 the
commit c8401dda2f ("KVM: x86: fix singlestepping over syscall")
introduced the kvm_run.debug corruption: kvm_vcpu_do_singlestep() can
be called without X86_EFLAGS_TF set.
Let's fix it. Please consider that for -stable.
Signed-off-by: Alexander Popov <alex.popov@linux.com>
Cc: stable@vger.kernel.org
Fixes: c8401dda2f ("KVM: x86: fix singlestepping over syscall")
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
HV_X64_MSR_GUEST_IDLE_AVAILABLE appeared in kvm_vcpu_ioctl_get_hv_cpuid()
by mistake: it announces support for HV_X64_MSR_GUEST_IDLE (0x400000F0)
which we don't support in KVM (yet).
Fixes: 2bc39970e9 ("x86/kvm/hyper-v: Introduce KVM_GET_SUPPORTED_HV_CPUID")
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
After commit
5d32a66541 ("PCI/ACPI: Allow ACPI to be built without CONFIG_PCI set")
dependencies on CONFIG_PCI that previously were satisfied implicitly
through dependencies on CONFIG_ACPI have to be specified directly.
PCI_LOCKLESS_CONFIG depends on PCI but this dependency has not been
mentioned in the Kconfig so add an explicit dependency here and fix
WARNING: unmet direct dependencies detected for PCI_LOCKLESS_CONFIG
Depends on [n]: PCI [=n]
Selected by [y]:
- X86 [=y]
Fixes: 5d32a66541 ("PCI/ACPI: Allow ACPI to be built without CONFIG_PCI set")
Signed-off-by: Sinan Kaya <okaya@kernel.org>
Signed-off-by: Borislav Petkov <bp@suse.de>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: linux-acpi@vger.kernel.org
Cc: x86-ml <x86@kernel.org>
Link: https://lkml.kernel.org/r/20190121231958.28255-2-okaya@kernel.org
Commit 594cc251fd ("make 'user_access_begin()' do 'access_ok()'")
makes the access_ok() check part of the user_access_begin() preceding a
series of 'unsafe' accesses. This has the desirable effect of ensuring
that all 'unsafe' accesses have been range-checked, without having to
pick through all of the callsites to verify whether the appropriate
checking has been made.
However, the consolidated range check does not inhibit speculation, so
it is still up to the caller to ensure that they are not susceptible to
any speculative side-channel attacks for user addresses that ultimately
fail the access_ok() check.
This is an oversight, so use __uaccess_begin_nospec() to ensure that
speculation is inhibited until the access_ok() check has passed.
Reported-by: Julien Thierry <julien.thierry@arm.com>
Signed-off-by: Will Deacon <will.deacon@arm.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
-----BEGIN PGP SIGNATURE-----
iHUEABYIAB0WIQRTLbB6QfY48x44uB6AXGG7T9hjvgUCXEGhYwAKCRCAXGG7T9hj
vu/qAP4ljV1QwAicTP2IosIpmSojJV4BkkJOs36dgF1u4DscPAEA7YhvTzGYcZr6
7rmBIbGve6Yb24Uh4aL3ir4w+LUvSAg=
=Jm9L
-----END PGP SIGNATURE-----
Merge tag 'for-linus-5.0-rc3-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/xen/tip
Pull xen fixes from Juergen Gross:
- Several fixes for the Xen pvcalls drivers (1 fix for the backend and
8 for the frontend).
- A fix for a rather longstanding bug in the Xen sched_clock()
interface which led to weird time jumps when migrating the system.
- A fix for avoiding accesses to x2apic MSRs in Xen PV guests.
* tag 'for-linus-5.0-rc3-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/xen/tip:
xen: Fix x86 sched_clock() interface for xen
pvcalls-front: fix potential null dereference
always clear the X2APIC_ENABLE bit for PV guest
pvcalls-front: Avoid get_free_pages(GFP_KERNEL) under spinlock
xen/pvcalls: remove set but not used variable 'intf'
pvcalls-back: set -ENOTCONN in pvcalls_conn_back_read
pvcalls-front: don't return error when the ring is full
pvcalls-front: properly allocate sk
pvcalls-front: don't try to free unallocated rings
pvcalls-front: read all data before closing the connection
While in the native case entry into the kernel happens on the trampoline
stack, PV Xen kernels get entered with the current thread stack right
away. Hence source and destination stacks are identical in that case,
and special care is needed.
Other than in sync_regs() the copying done on the INT80 path isn't
NMI / #MC safe, as either of these events occurring in the middle of the
stack copying would clobber data on the (source) stack.
There is similar code in interrupt_entry() and nmi(), but there is no fixup
required because those code paths are unreachable in XEN PV guests.
[ tglx: Sanitized subject, changelog, Fixes tag and stable mail address. Sigh ]
Fixes: 7f2590a110 ("x86/entry/64: Use a per-CPU trampoline stack for IDT entries")
Signed-off-by: Jan Beulich <jbeulich@suse.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Juergen Gross <jgross@suse.com>
Acked-by: Andy Lutomirski <luto@kernel.org>
Cc: Peter Anvin <hpa@zytor.com>
Cc: xen-devel@lists.xenproject.org>
Cc: Boris Ostrovsky <boris.ostrovsky@oracle.com>
Cc: stable@vger.kernel.org
Link: https://lkml.kernel.org/r/5C3E1128020000780020DFAD@prv1-mh.provo.novell.com
Commit f94c8d1169 ("sched/clock, x86/tsc: Rework the x86 'unstable'
sched_clock() interface") broke Xen guest time handling across
migration:
[ 187.249951] Freezing user space processes ... (elapsed 0.001 seconds) done.
[ 187.251137] OOM killer disabled.
[ 187.251137] Freezing remaining freezable tasks ... (elapsed 0.001 seconds) done.
[ 187.252299] suspending xenstore...
[ 187.266987] xen:grant_table: Grant tables using version 1 layout
[18446743811.706476] OOM killer enabled.
[18446743811.706478] Restarting tasks ... done.
[18446743811.720505] Setting capacity to 16777216
Fix that by setting xen_sched_clock_offset at resume time to ensure a
monotonic clock value.
[boris: replaced pr_info() with pr_info_once() in xen_callback_vector()
to avoid printing with incorrect timestamp during resume (as we
haven't re-adjusted the clock yet)]
Fixes: f94c8d1169 ("sched/clock, x86/tsc: Rework the x86 'unstable' sched_clock() interface")
Cc: <stable@vger.kernel.org> # 4.11
Reported-by: Hans van Kranenburg <hans.van.kranenburg@mendix.com>
Signed-off-by: Juergen Gross <jgross@suse.com>
Tested-by: Hans van Kranenburg <hans.van.kranenburg@mendix.com>
Signed-off-by: Boris Ostrovsky <boris.ostrovsky@oracle.com>
After commit 5d32a66541 (PCI/ACPI: Allow ACPI to be built without
CONFIG_PCI set) dependencies on CONFIG_PCI that previously were
satisfied implicitly through dependencies on CONFIG_ACPI have to be
specified directly.
LPSS code relies on PCI infrastructure but this dependency has not
been called out explicitly yet.
Fixes: 5d32a66541 ("PCI/ACPI: Allow ACPI to be built without CONFIG_PCI set")
Signed-off-by: Sinan Kaya <okaya@kernel.org>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Commit
b6664ba42f ("s390, kexec_file: drop arch_kexec_mem_walk()")
changed the behavior of kexec_locate_mem_hole(): it will try to allocate
free memory only when kbuf.mem is initialized to zero.
However, x86's kexec_file_load() implementation reuses a struct
kexec_buf allocated on the stack and its kbuf.mem member gets set by
each kexec_add_buffer() invocation.
The second kexec_add_buffer() will reuse the same kbuf but not
reinitialize kbuf.mem.
Therefore, explictily reset kbuf.mem each time in order for
kexec_locate_mem_hole() to locate a free memory region each time.
[ bp: massage commit message. ]
Fixes: b6664ba42f ("s390, kexec_file: drop arch_kexec_mem_walk()")
Signed-off-by: Dave Young <dyoung@redhat.com>
Signed-off-by: Borislav Petkov <bp@suse.de>
Acked-by: Baoquan He <bhe@redhat.com>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: AKASHI Takahiro <takahiro.akashi@linaro.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: Philipp Rudo <prudo@linux.vnet.ibm.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Vivek Goyal <vgoyal@redhat.com>
Cc: Yannik Sembritzki <yannik@sembritzki.me>
Cc: Yi Wang <wang.yi59@zte.com.cn>
Cc: kexec@lists.infradead.org
Cc: x86-ml <x86@kernel.org>
Link: https://lkml.kernel.org/r/20181228011247.GA9999@dhcp-128-65.nay.redhat.com
Using sizeof(pointer) for determining the size of a memset() only works
when the size of the pointer and the size of type to which it points are
the same. For pte_t this is only true for 64bit and 32bit-NONPAE. On 32bit
PAE systems this is wrong as the pointer size is 4 byte but the PTE entry
is 8 bytes. It's actually not a real world issue as this code depends on
64bit, but it's wrong nevertheless.
Use sizeof(*p) for correctness sake.
Fixes: aad983913d ("x86/mm/encrypt: Simplify sme_populate_pgd() and sme_populate_pgd_large()")
Signed-off-by: Peng Hao <peng.hao2@zte.com.cn>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Cc: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Cc: Tom Lendacky <thomas.lendacky@amd.com>
Cc: dave.hansen@linux.intel.com
Cc: peterz@infradead.org
Cc: luto@kernel.org
Link: https://lkml.kernel.org/r/1546065252-97996-1-git-send-email-peng.hao2@zte.com.cn
Memory protection key behavior should be the same in a child as it was
in the parent before a fork. But, there is a bug that resets the
state in the child at fork instead of preserving it.
The creation of new mm's is a bit convoluted. At fork(), the code
does:
1. memcpy() the parent mm to initialize child
2. mm_init() to initalize some select stuff stuff
3. dup_mmap() to create true copies that memcpy() did not do right
For pkeys two bits of state need to be preserved across a fork:
'execute_only_pkey' and 'pkey_allocation_map'.
Those are preserved by the memcpy(), but mm_init() invokes
init_new_context() which overwrites 'execute_only_pkey' and
'pkey_allocation_map' with "new" values.
The author of the code erroneously believed that init_new_context is *only*
called at execve()-time. But, alas, init_new_context() is used at execve()
and fork().
The result is that, after a fork(), the child's pkey state ends up looking
like it does after an execve(), which is totally wrong. pkeys that are
already allocated can be allocated again, for instance.
To fix this, add code called by dup_mmap() to copy the pkey state from
parent to child explicitly. Also add a comment above init_new_context() to
make it more clear to the next poor sod what this code is used for.
Fixes: e8c24d3a23 ("x86/pkeys: Allocation/free syscalls")
Signed-off-by: Dave Hansen <dave.hansen@linux.intel.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Thomas Gleixner <tglx@linutronix.de>
Cc: bp@alien8.de
Cc: hpa@zytor.com
Cc: peterz@infradead.org
Cc: mpe@ellerman.id.au
Cc: will.deacon@arm.com
Cc: luto@kernel.org
Cc: jroedel@suse.de
Cc: stable@vger.kernel.org
Cc: Borislav Petkov <bp@alien8.de>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: Will Deacon <will.deacon@arm.com>
Cc: Andy Lutomirski <luto@kernel.org>
Cc: Joerg Roedel <jroedel@suse.de>
Link: https://lkml.kernel.org/r/20190102215655.7A69518C@viggo.jf.intel.com
Commit e657fcc clears cpu capability bit instead of using fake cpuid
value, the EXTD should always be off for PV guest without depending
on cpuid value. So remove the cpuid check in xen_read_msr_safe() to
always clear the X2APIC_ENABLE bit.
Signed-off-by: Talons Lee <xin.li@citrix.com>
Reviewed-by: Juergen Gross <jgross@suse.com>
Signed-off-by: Boris Ostrovsky <boris.ostrovsky@oracle.com>
Minor fixes for new code, corner cases, and documentation.
Patches are isolated and sufficiently described by the shortlog.
-----BEGIN PGP SIGNATURE-----
iQEzBAABCAAdFiEEj/xJEkTHTJzZE/FJQP/qGw8qh+gFAlw52x0ACgkQQP/qGw8q
h+iejgf+NjD3jXCYAxLtGInXyQEwqw113Gr8941oC2E50CifTqrFhWNkRgl+GyCS
x92SCiNS1xpy+SnUZDj63MjsB90zOOogPG+RYsSE+BAN/i/9IiE2F5aYSXlen0rR
SOOfraQTDlS5rOp9auGMNlNWkmEKCO6YEsdmlNte/QVNuy4vrHfAJDvdfCUiC0qO
XJWMjIyZElsKJUtmZzKuTrCZniBvqn6yGH+/0kggrzDPPIHU+M7qSkMkAC7wqrFy
0zaeKNloArFLKlDQPQ3AykgxCrp6CJnXuR/Oo6R5rn9jF2kVdlzglx7pHBAc5OGp
wI+VDWRA2x2JevWcIdsN2Xp2vaCRGw==
=L1zP
-----END PGP SIGNATURE-----
Merge tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm
Pull KVM fixes from Radim Krčmář:
"Minor fixes for new code, corner cases, and documentation"
* tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm:
x86/kvm/nVMX: don't skip emulated instruction twice when vmptr address is not backed
Documentation/virtual/kvm: Update URL for AMD SEV API specification
KVM/VMX: Avoid return error when flush tlb successfully in the hv_remote_flush_tlb_with_range()
kvm: sev: Fail KVM_SEV_INIT if already initialized
KVM: validate userspace input in kvm_clear_dirty_log_protect()
KVM: x86: Fix bit shifting in update_intel_pt_cfg
After commit
5d32a66541 ("PCI/ACPI: Allow ACPI to be built without CONFIG_PCI set")
dependencies on CONFIG_PCI that previously were satisfied implicitly
through dependencies on CONFIG_ACPI have to be specified directly. LPSS
code relies on PCI infrastructure but this dependency has not been
explicitly called out so do that.
Fixes: 5d32a66541 ("PCI/ACPI: Allow ACPI to be built without CONFIG_PCI set")
Signed-off-by: Sinan Kaya <okaya@kernel.org>
Signed-off-by: Borislav Petkov <bp@suse.de>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: linux-acpi@vger.kernel.org
Cc: x86-ml <x86@kernel.org>
Link: https://lkml.kernel.org/r/20190102181038.4418-11-okaya@kernel.org
Since commit 09abb5e3e5 ("KVM: nVMX: call kvm_skip_emulated_instruction
in nested_vmx_{fail,succeed}") nested_vmx_failValid() results in
kvm_skip_emulated_instruction() so doing it again in handle_vmptrld() when
vmptr address is not backed is wrong, we end up advancing RIP twice.
Fixes: fca91f6d60 ("kvm: nVMX: Set VM instruction error for VMPTRLD of unbacked page")
Reported-by: Cornelia Huck <cohuck@redhat.com>
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Reviewed-by: Sean Christopherson <sean.j.christopherson@intel.com>
Reviewed-by: Cornelia Huck <cohuck@redhat.com>
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
The "ret" is initialized to be ENOTSUPP. The return value of
__hv_remote_flush_tlb_with_range() will be Or with "ret" when ept
table potiners are mismatched. This will cause return ENOTSUPP even if
flush tlb successfully. This patch is to fix the issue and set
"ret" to 0.
Fixes: a5c214dad1 ("KVM/VMX: Change hv flush logic when ept tables are mismatched.")
Signed-off-by: Lan Tianyu <Tianyu.Lan@microsoft.com>
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
By code inspection, it was found that multiple calls to KVM_SEV_INIT
could deplete asid bits and overwrite kvm_sev_info's regions_list.
Multiple calls to KVM_SVM_INIT is not likely to occur with QEMU, but this
should likely be fixed anyway.
This code is serialized by kvm->lock.
Fixes: 1654efcbc4 ("KVM: SVM: Add KVM_SEV_INIT command")
Reported-by: Cfir Cohen <cfir@google.com>
Signed-off-by: David Rientjes <rientjes@google.com>
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
ctl_bitmask in pt_desc is of type u64. When an integer like 0xf is
being left shifted more than 32 bits, the behavior is undefined.
Fix this by adding suffix ULL to integer 0xf.
Addresses-Coverity-ID: 1476095 ("Bad bit shift operation")
Fixes: 6c0f0bba85 ("KVM: x86: Introduce a function to initialize the PT configuration")
Signed-off-by: Gustavo A. R. Silva <gustavo@embeddedor.com>
Reviewed-by: Wei Yang <richardw.yang@linux.intel.com>
Reviewed-by: Luwei Kang <luwei.kang@intel.com>
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
CONFIG_RESCTRL is too generic. The final goal is to have a generic
option called like this which is selected by the arch-specific ones
CONFIG_X86_RESCTRL and CONFIG_ARM64_RESCTRL. The generic one will
cover the resctrl filesystem and other generic and shared bits of
functionality.
Signed-off-by: Borislav Petkov <bp@suse.de>
Suggested-by: Ingo Molnar <mingo@kernel.org>
Requested-by: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Babu Moger <babu.moger@amd.com>
Cc: Fenghua Yu <fenghua.yu@intel.com>
Cc: James Morse <james.morse@arm.com>
Cc: Reinette Chatre <reinette.chatre@intel.com>
Cc: Tony Luck <tony.luck@intel.com>
Cc: x86@kernel.org
Link: http://lkml.kernel.org/r/20190108171401.GC12235@zn.tnic
- improve boolinit.cocci and use_after_iter.cocci semantic patches
- fix alignment for kallsyms
- move 'asm goto' compiler test to Kconfig and clean up jump_label
CONFIG option
- generate asm-generic wrappers automatically if arch does not implement
mandatory UAPI headers
- remove redundant generic-y defines
- misc cleanups
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1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=I+r6
-----END PGP SIGNATURE-----
Merge tag 'kbuild-v4.21-3' of git://git.kernel.org/pub/scm/linux/kernel/git/masahiroy/linux-kbuild
Pull more Kbuild updates from Masahiro Yamada:
- improve boolinit.cocci and use_after_iter.cocci semantic patches
- fix alignment for kallsyms
- move 'asm goto' compiler test to Kconfig and clean up jump_label
CONFIG option
- generate asm-generic wrappers automatically if arch does not
implement mandatory UAPI headers
- remove redundant generic-y defines
- misc cleanups
* tag 'kbuild-v4.21-3' of git://git.kernel.org/pub/scm/linux/kernel/git/masahiroy/linux-kbuild:
kconfig: rename generated .*conf-cfg to *conf-cfg
kbuild: remove unnecessary stubs for archheader and archscripts
kbuild: use assignment instead of define ... endef for filechk_* rules
arch: remove redundant UAPI generic-y defines
kbuild: generate asm-generic wrappers if mandatory headers are missing
arch: remove stale comments "UAPI Header export list"
riscv: remove redundant kernel-space generic-y
kbuild: change filechk to surround the given command with { }
kbuild: remove redundant target cleaning on failure
kbuild: clean up rule_dtc_dt_yaml
kbuild: remove UIMAGE_IN and UIMAGE_OUT
jump_label: move 'asm goto' support test to Kconfig
kallsyms: lower alignment on ARM
scripts: coccinelle: boolinit: drop warnings on named constants
scripts: coccinelle: check for redeclaration
kconfig: remove unused "file" field of yylval union
nds32: remove redundant kernel-space generic-y
nios2: remove unneeded HAS_DMA define
Fix various regressions introduced in this cycles:
- fix dma-debug tracking for the map_page / map_single consolidatation
- properly stub out DMA mapping symbols for !HAS_DMA builds to avoid
link failures
- fix AMD Gart direct mappings
- setup the dma address for no kernel mappings using the remap
allocator
-----BEGIN PGP SIGNATURE-----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=H9J6
-----END PGP SIGNATURE-----
Merge tag 'dma-mapping-4.21-1' of git://git.infradead.org/users/hch/dma-mapping
Pull dma-mapping fixes from Christoph Hellwig:
"Fix various regressions introduced in this cycles:
- fix dma-debug tracking for the map_page / map_single
consolidatation
- properly stub out DMA mapping symbols for !HAS_DMA builds to avoid
link failures
- fix AMD Gart direct mappings
- setup the dma address for no kernel mappings using the remap
allocator"
* tag 'dma-mapping-4.21-1' of git://git.infradead.org/users/hch/dma-mapping:
dma-direct: fix DMA_ATTR_NO_KERNEL_MAPPING for remapped allocations
x86/amd_gart: fix unmapping of non-GART mappings
dma-mapping: remove a few unused exports
dma-mapping: properly stub out the DMA API for !CONFIG_HAS_DMA
dma-mapping: remove dmam_{declare,release}_coherent_memory
dma-mapping: implement dmam_alloc_coherent using dmam_alloc_attrs
dma-mapping: implement dma_map_single_attrs using dma_map_page_attrs
Now that Kbuild automatically creates asm-generic wrappers for missing
mandatory headers, it is redundant to list the same headers in
generic-y and mandatory-y.
Suggested-by: Sam Ravnborg <sam@ravnborg.org>
Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
Acked-by: Sam Ravnborg <sam@ravnborg.org>
These comments are leftovers of commit fcc8487d47 ("uapi: export all
headers under uapi directories").
Prior to that commit, exported headers must be explicitly added to
header-y. Now, all headers under the uapi/ directories are exported.
Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
Since commit 9c2af1c737 ("kbuild: add .DELETE_ON_ERROR special
target"), the target file is automatically deleted on failure.
The boilerplate code
... || { rm -f $@; false; }
is unneeded.
Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
Currently, CONFIG_JUMP_LABEL just means "I _want_ to use jump label".
The jump label is controlled by HAVE_JUMP_LABEL, which is defined
like this:
#if defined(CC_HAVE_ASM_GOTO) && defined(CONFIG_JUMP_LABEL)
# define HAVE_JUMP_LABEL
#endif
We can improve this by testing 'asm goto' support in Kconfig, then
make JUMP_LABEL depend on CC_HAS_ASM_GOTO.
Ugly #ifdef HAVE_JUMP_LABEL will go away, and CONFIG_JUMP_LABEL will
match to the real kernel capability.
Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
Acked-by: Michael Ellerman <mpe@ellerman.id.au> (powerpc)
Tested-by: Sedat Dilek <sedat.dilek@gmail.com>
Pull vfs mount API prep from Al Viro:
"Mount API prereqs.
Mostly that's LSM mount options cleanups. There are several minor
fixes in there, but nothing earth-shattering (leaks on failure exits,
mostly)"
* 'mount.part1' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs: (27 commits)
mount_fs: suppress MAC on MS_SUBMOUNT as well as MS_KERNMOUNT
smack: rewrite smack_sb_eat_lsm_opts()
smack: get rid of match_token()
smack: take the guts of smack_parse_opts_str() into a new helper
LSM: new method: ->sb_add_mnt_opt()
selinux: rewrite selinux_sb_eat_lsm_opts()
selinux: regularize Opt_... names a bit
selinux: switch away from match_token()
selinux: new helper - selinux_add_opt()
LSM: bury struct security_mnt_opts
smack: switch to private smack_mnt_opts
selinux: switch to private struct selinux_mnt_opts
LSM: hide struct security_mnt_opts from any generic code
selinux: kill selinux_sb_get_mnt_opts()
LSM: turn sb_eat_lsm_opts() into a method
nfs_remount(): don't leak, don't ignore LSM options quietly
btrfs: sanitize security_mnt_opts use
selinux; don't open-code a loop in sb_finish_set_opts()
LSM: split ->sb_set_mnt_opts() out of ->sb_kern_mount()
new helper: security_sb_eat_lsm_opts()
...
Merge more updates from Andrew Morton:
- procfs updates
- various misc bits
- lib/ updates
- epoll updates
- autofs
- fatfs
- a few more MM bits
* emailed patches from Andrew Morton <akpm@linux-foundation.org>: (58 commits)
mm/page_io.c: fix polled swap page in
checkpatch: add Co-developed-by to signature tags
docs: fix Co-Developed-by docs
drivers/base/platform.c: kmemleak ignore a known leak
fs: don't open code lru_to_page()
fs/: remove caller signal_pending branch predictions
mm/: remove caller signal_pending branch predictions
arch/arc/mm/fault.c: remove caller signal_pending_branch predictions
kernel/sched/: remove caller signal_pending branch predictions
kernel/locking/mutex.c: remove caller signal_pending branch predictions
mm: select HAVE_MOVE_PMD on x86 for faster mremap
mm: speed up mremap by 20x on large regions
mm: treewide: remove unused address argument from pte_alloc functions
initramfs: cleanup incomplete rootfs
scripts/gdb: fix lx-version string output
kernel/kcov.c: mark write_comp_data() as notrace
kernel/sysctl: add panic_print into sysctl
panic: add options to print system info when panic happens
bfs: extra sanity checking and static inode bitmap
exec: separate MM_ANONPAGES and RLIMIT_STACK accounting
...
In many cases we don't have to create a GART mapping at all, which
also means there is nothing to unmap. Fix the range check that was
incorrectly modified when removing the mapping_error method.
Fixes: 9e8aa6b546 ("x86/amd_gart: remove the mapping_error dma_map_ops method")
Reported-by: Michal Kubecek <mkubecek@suse.cz>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Tested-by: Michal Kubecek <mkubecek@suse.cz>
This has been broken forever, and nobody ever really noticed because
it's purely a performance issue.
Long long ago, in commit 6175ddf06b ("x86: Clean up mem*io functions")
Brian Gerst simplified the memory copies to and from iomem, since on
x86, the instructions to access iomem are exactly the same as the
regular instructions.
That is technically true, and things worked, and nobody said anything.
Besides, back then the regular memcpy was pretty simple and worked fine.
Nobody noticed except for David Laight, that is. David has a testing a
TLP monitor he was writing for an FPGA, and has been occasionally
complaining about how memcpy_toio() writes things one byte at a time.
Which is completely unacceptable from a performance standpoint, even if
it happens to technically work.
The reason it's writing one byte at a time is because while it's
technically true that accesses to iomem are the same as accesses to
regular memory on x86, the _granularity_ (and ordering) of accesses
matter to iomem in ways that they don't matter to regular cached memory.
In particular, when ERMS is set, we default to using "rep movsb" for
larger memory copies. That is indeed perfectly fine for real memory,
since the whole point is that the CPU is going to do cacheline
optimizations and executes the memory copy efficiently for cached
memory.
With iomem? Not so much. With iomem, "rep movsb" will indeed work, but
it will copy things one byte at a time. Slowly and ponderously.
Now, originally, back in 2010 when commit 6175ddf06b was done, we
didn't use ERMS, and this was much less noticeable.
Our normal memcpy() was simpler in other ways too.
Because in fact, it's not just about using the string instructions. Our
memcpy() these days does things like "read and write overlapping values"
to handle the last bytes of the copy. Again, for normal memory,
overlapping accesses isn't an issue. For iomem? It can be.
So this re-introduces the specialized memcpy_toio(), memcpy_fromio() and
memset_io() functions. It doesn't particularly optimize them, but it
tries to at least not be horrid, or do overlapping accesses. In fact,
this uses the existing __inline_memcpy() function that we still had
lying around that uses our very traditional "rep movsl" loop followed by
movsw/movsb for the final bytes.
Somebody may decide to try to improve on it, but if we've gone almost a
decade with only one person really ever noticing and complaining, maybe
it's not worth worrying about further, once it's not _completely_ broken?
Reported-by: David Laight <David.Laight@aculab.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
