commit a664ec9158 upstream.
We missed the window between the TIF flag update and the next reschedule.
Signed-off-by: Rodrigo Branco <bsdaemon@google.com>
Reviewed-by: Borislav Petkov (AMD) <bp@alien8.de>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Cc: <stable@vger.kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit d00dd2f264 upstream.
After
b3e34a47f9 ("x86/kexec: fix memory leak of elf header buffer"),
freeing image->elf_headers in the error path of crash_load_segments()
is not needed because kimage_file_post_load_cleanup() will take
care of that later. And not clearing it could result in a double-free.
Drop the superfluous vfree() call at the error path of
crash_load_segments().
Fixes: b3e34a47f9 ("x86/kexec: fix memory leak of elf header buffer")
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Signed-off-by: Borislav Petkov (AMD) <bp@alien8.de>
Acked-by: Baoquan He <bhe@redhat.com>
Acked-by: Vlastimil Babka <vbabka@suse.cz>
Cc: <stable@kernel.org>
Link: https://lore.kernel.org/r/20221122115122.13937-1-tiwai@suse.de
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 63dc6325ff upstream.
Since the CONFIG_RETHUNK and CONFIG_SLS will use INT3 for stopping
speculative execution after function return, kprobe jump optimization
always fails on the functions with such INT3 inside the function body.
(It already checks the INT3 padding between functions, but not inside
the function)
To avoid this issue, as same as kprobes, check whether the INT3 comes
from kgdb or not, and if so, stop decoding and make it fail. The other
INT3 will come from CONFIG_RETHUNK/CONFIG_SLS and those can be
treated as a one-byte instruction.
Fixes: e463a09af2 ("x86: Add straight-line-speculation mitigation")
Suggested-by: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Masami Hiramatsu (Google) <mhiramat@kernel.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: stable@vger.kernel.org
Link: https://lore.kernel.org/r/167146051929.1374301.7419382929328081706.stgit@devnote3
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 1993bf9799 upstream.
Since the CONFIG_RETHUNK and CONFIG_SLS will use INT3 for stopping
speculative execution after RET instruction, kprobes always failes to
check the probed instruction boundary by decoding the function body if
the probed address is after such sequence. (Note that some conditional
code blocks will be placed after function return, if compiler decides
it is not on the hot path.)
This is because kprobes expects kgdb puts the INT3 as a software
breakpoint and it will replace the original instruction.
But these INT3 are not such purpose, it doesn't need to recover the
original instruction.
To avoid this issue, kprobes checks whether the INT3 is owned by
kgdb or not, and if so, stop decoding and make it fail. The other
INT3 will come from CONFIG_RETHUNK/CONFIG_SLS and those can be
treated as a one-byte instruction.
Fixes: e463a09af2 ("x86: Add straight-line-speculation mitigation")
Suggested-by: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Masami Hiramatsu (Google) <mhiramat@kernel.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: stable@vger.kernel.org
Link: https://lore.kernel.org/r/167146051026.1374301.392728975473572291.stgit@devnote3
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit fd3dc56253 upstream.
After someone reported a bug report with a failed modification due to the
expected value not matching what was found, it came to my attention that
the ftrace_expected is no longer set when that happens. This makes for
debugging the issue a bit more difficult.
Set ftrace_expected to the expected code before calling ftrace_bug, so
that it shows what was expected and why it failed.
Link: https://lore.kernel.org/all/CA+wXwBQ-VhK+hpBtYtyZP-NiX4g8fqRRWithFOHQW-0coQ3vLg@mail.gmail.com/
Link: https://lore.kernel.org/linux-trace-kernel/20221209105247.01d4e51d@gandalf.local.home
Cc: Masami Hiramatsu <mhiramat@kernel.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: "x86@kernel.org" <x86@kernel.org>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: stable@vger.kernel.org
Fixes: 768ae4406a ("x86/ftrace: Use text_poke()")
Signed-off-by: Steven Rostedt (Google) <rostedt@goodmis.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit be1b670f61 upstream.
The retries in load_ucode_intel_ap() were in place to support systems
with mixed steppings. Mixed steppings are no longer supported and there is
only one microcode image at a time. Any retries will simply reattempt to
apply the same image over and over without making progress.
[ bp: Zap the circumstantial reasoning from the commit message. ]
Fixes: 06b8534cb7 ("x86/microcode: Rework microcode loading")
Signed-off-by: Ashok Raj <ashok.raj@intel.com>
Signed-off-by: Borislav Petkov (AMD) <bp@alien8.de>
Reviewed-by: Thomas Gleixner <tglx@linutronix.de>
Cc: stable@vger.kernel.org
Link: https://lore.kernel.org/r/20221129210832.107850-3-ashok.raj@intel.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 31de69f4ee upstream.
Set ENABLE_USR_WAIT_PAUSE in KVM's supported VMX MSR configuration if the
feature is supported in hardware and enabled in KVM's base, non-nested
configuration, i.e. expose ENABLE_USR_WAIT_PAUSE to L1 if it's supported.
This fixes a bug where saving/restoring, i.e. migrating, a vCPU will fail
if WAITPKG (the associated CPUID feature) is enabled for the vCPU, and
obviously allows L1 to enable the feature for L2.
KVM already effectively exposes ENABLE_USR_WAIT_PAUSE to L1 by stuffing
the allowed-1 control ina vCPU's virtual MSR_IA32_VMX_PROCBASED_CTLS2 when
updating secondary controls in response to KVM_SET_CPUID(2), but (a) that
depends on flawed code (KVM shouldn't touch VMX MSRs in response to CPUID
updates) and (b) runs afoul of vmx_restore_control_msr()'s restriction
that the guest value must be a strict subset of the supported host value.
Although no past commit explicitly enabled nested support for WAITPKG,
doing so is safe and functionally correct from an architectural
perspective as no additional KVM support is needed to virtualize TPAUSE,
UMONITOR, and UMWAIT for L2 relative to L1, and KVM already forwards
VM-Exits to L1 as necessary (commit bf653b78f9, "KVM: vmx: Introduce
handle_unexpected_vmexit and handle WAITPKG vmexit").
Note, KVM always keeps the hosts MSR_IA32_UMWAIT_CONTROL resident in
hardware, i.e. always runs both L1 and L2 with the host's power management
settings for TPAUSE and UMWAIT. See commit bf09fb6cba ("KVM: VMX: Stop
context switching MSR_IA32_UMWAIT_CONTROL") for more details.
Fixes: e69e72faa3 ("KVM: x86: Add support for user wait instructions")
Cc: stable@vger.kernel.org
Reported-by: Aaron Lewis <aaronlewis@google.com>
Reported-by: Yu Zhang <yu.c.zhang@linux.intel.com>
Signed-off-by: Sean Christopherson <seanjc@google.com>
Reviewed-by: Jim Mattson <jmattson@google.com>
Message-Id: <20221213062306.667649-2-seanjc@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit ef40757743 upstream.
When a VM reboots itself, the reset process will result in
an ioctl(KVM_SET_LAPIC, ...) to disable x2APIC mode and set
the xAPIC id of the vCPU to its default value, which is the
vCPU id.
That will be handled in KVM as follows:
kvm_vcpu_ioctl_set_lapic
kvm_apic_set_state
kvm_lapic_set_base => disable X2APIC mode
kvm_apic_state_fixup
kvm_lapic_xapic_id_updated
kvm_xapic_id(apic) != apic->vcpu->vcpu_id
kvm_set_apicv_inhibit(APICV_INHIBIT_REASON_APIC_ID_MODIFIED)
memcpy(vcpu->arch.apic->regs, s->regs, sizeof(*s)) => update APIC_ID
When kvm_apic_set_state invokes kvm_lapic_set_base to disable
x2APIC mode, the old 32-bit x2APIC id is still present rather
than the 8-bit xAPIC id. kvm_lapic_xapic_id_updated will set the
APICV_INHIBIT_REASON_APIC_ID_MODIFIED bit and disable APICv/x2AVIC.
Instead, kvm_lapic_xapic_id_updated must be called after APIC_ID is
changed.
In fact, this fixes another small issue in the code in that
potential changes to a vCPU's xAPIC ID need not be tracked for
KVM_GET_LAPIC.
Fixes: 3743c2f025 ("KVM: x86: inhibit APICv/AVIC on changes to APIC ID or APIC base")
Signed-off-by: Yuan ZhaoXiong <yuanzhaoxiong@baidu.com>
Message-Id: <1669984574-32692-1-git-send-email-yuanzhaoxiong@baidu.com>
Cc: stable@vger.kernel.org
Reported-by: Alejandro Jimenez <alejandro.j.jimenez@oracle.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 9cc409325d upstream.
Inject #GP for if VMXON is attempting with a CR0/CR4 that fails the
generic "is CRx valid" check, but passes the CR4.VMXE check, and do the
generic checks _after_ handling the post-VMXON VM-Fail.
The CR4.VMXE check, and all other #UD cases, are special pre-conditions
that are enforced prior to pivoting on the current VMX mode, i.e. occur
before interception if VMXON is attempted in VMX non-root mode.
All other CR0/CR4 checks generate #GP and effectively have lower priority
than the post-VMXON check.
Per the SDM:
IF (register operand) or (CR0.PE = 0) or (CR4.VMXE = 0) or ...
THEN #UD;
ELSIF not in VMX operation
THEN
IF (CPL > 0) or (in A20M mode) or
(the values of CR0 and CR4 are not supported in VMX operation)
THEN #GP(0);
ELSIF in VMX non-root operation
THEN VMexit;
ELSIF CPL > 0
THEN #GP(0);
ELSE VMfail("VMXON executed in VMX root operation");
FI;
which, if re-written without ELSIF, yields:
IF (register operand) or (CR0.PE = 0) or (CR4.VMXE = 0) or ...
THEN #UD
IF in VMX non-root operation
THEN VMexit;
IF CPL > 0
THEN #GP(0)
IF in VMX operation
THEN VMfail("VMXON executed in VMX root operation");
IF (in A20M mode) or
(the values of CR0 and CR4 are not supported in VMX operation)
THEN #GP(0);
Note, KVM unconditionally forwards VMXON VM-Exits that occur in L2 to L1,
i.e. there is no need to check the vCPU is not in VMX non-root mode. Add
a comment to explain why unconditionally forwarding such exits is
functionally correct.
Reported-by: Eric Li <ercli@ucdavis.edu>
Fixes: c7d855c2af ("KVM: nVMX: Inject #UD if VMXON is attempted with incompatible CR0/CR4")
Cc: stable@vger.kernel.org
Signed-off-by: Sean Christopherson <seanjc@google.com>
Link: https://lore.kernel.org/r/20221006001956.329314-1-seanjc@google.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit eb3992e833 upstream.
Resume the guest immediately when injecting a #GP on ECREATE due to an
invalid enclave size, i.e. don't attempt ECREATE in the host. The #GP is
a terminal fault, e.g. skipping the instruction if ECREATE is successful
would result in KVM injecting #GP on the instruction following ECREATE.
Fixes: 70210c044b ("KVM: VMX: Add SGX ENCLS[ECREATE] handler to enforce CPUID restrictions")
Cc: stable@vger.kernel.org
Cc: Kai Huang <kai.huang@intel.com>
Signed-off-by: Sean Christopherson <seanjc@google.com>
Reviewed-by: Kai Huang <kai.huang@intel.com>
Link: https://lore.kernel.org/r/20220930233132.1723330-1-seanjc@google.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit bc1b705b0e upstream.
AMD's MCA Thresholding feature counts errors of all severity levels, not
just correctable errors. If a deferred error causes the threshold limit
to be reached (it was the error that caused the overflow), then both a
deferred error interrupt and a thresholding interrupt will be triggered.
The order of the interrupts is not guaranteed. If the threshold
interrupt handler is executed first, then it will clear MCA_STATUS for
the error. It will not check or clear MCA_DESTAT which also holds a copy
of the deferred error. When the deferred error interrupt handler runs it
will not find an error in MCA_STATUS, but it will find the error in
MCA_DESTAT. This will cause two errors to be logged.
Check for deferred errors when handling a threshold interrupt. If a bank
contains a deferred error, then clear the bank's MCA_DESTAT register.
Define a new helper function to do the deferred error check and clearing
of MCA_DESTAT.
[ bp: Simplify, convert comment to passive voice. ]
Fixes: 37d43acfd7 ("x86/mce/AMD: Redo error logging from APIC LVT interrupt handlers")
Signed-off-by: Yazen Ghannam <yazen.ghannam@amd.com>
Signed-off-by: Borislav Petkov <bp@suse.de>
Cc: stable@vger.kernel.org
Link: https://lore.kernel.org/r/20220621155943.33623-1-yazen.ghannam@amd.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 48280042f2 upstream.
"XSAVE consistency problem" has been reported under Xen, but that's the extent
of my divination skills.
Modify XSTATE_WARN_ON() to force the caller to provide relevant diagnostic
information, and modify each caller suitably.
For check_xstate_against_struct(), this removes a double WARN() where one will
do perfectly fine.
CC stable as this has been wonky debugging for 7 years and it is good to
have there too.
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Signed-off-by: Borislav Petkov <bp@suse.de>
Cc: <stable@vger.kernel.org>
Link: https://lore.kernel.org/r/20220810221909.12768-1-andrew.cooper3@citrix.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 6532783310 upstream.
Current clear_attr_update procedure in pmu_set_mapping() sets attr_update
field in NULL that is not correct because intel_uncore_type pmu types can
contain several groups in attr_update field. For example, SPR platform
already has uncore_alias_group to update and then UPI topology group will
be added in next patches.
Fix current behavior and clear attr_update group related to mapping only.
Fixes: bb42b3d397 ("perf/x86/intel/uncore: Expose an Uncore unit to IIO PMON mapping")
Signed-off-by: Alexander Antonov <alexander.antonov@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Kan Liang <kan.liang@linux.intel.com>
Cc: stable@vger.kernel.org
Link: https://lore.kernel.org/r/20221117122833.3103580-4-alexander.antonov@linux.intel.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit efe062705d upstream.
Current implementation of I/O stacks to PMU mapping doesn't support ICX-D.
Detect ICX-D system to disable mapping.
Fixes: 10337e95e0 ("perf/x86/intel/uncore: Enable I/O stacks to IIO PMON mapping on ICX")
Signed-off-by: Alexander Antonov <alexander.antonov@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Kan Liang <kan.liang@linux.intel.com>
Cc: stable@vger.kernel.org
Link: https://lore.kernel.org/r/20221117122833.3103580-5-alexander.antonov@linux.intel.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
[ Upstream commit e3998434da ]
A kernel that was compiled without CONFIG_X86_X2APIC was unable to boot on
platforms that have x2APIC already enabled in the BIOS before starting the
kernel.
The kernel was supposed to panic with an approprite error message in
validate_x2apic() due to the missing X2APIC support.
However, validate_x2apic() was run too late in the boot cycle, and the
kernel tried to initialize the APIC nonetheless. This resulted in an
earlier panic in setup_local_APIC() because the APIC was not registered.
In my experiments, a panic message in setup_local_APIC() was not visible
in the graphical console, which resulted in a hang with no indication
what has gone wrong.
Instead of calling panic(), disable the APIC, which results in a somewhat
working system with the PIC only (and no SMP). This way the user is able to
diagnose the problem more easily.
Disabling X2APIC mode is not an option because it's impossible on systems
with locked x2APIC.
The proper place to disable the APIC in this case is in check_x2apic(),
which is called early from setup_arch(). Doing this in
__apic_intr_mode_select() is too late.
Make check_x2apic() unconditionally available and remove the empty stub.
Reported-by: Paul Menzel <pmenzel@molgen.mpg.de>
Reported-by: Robert Elliott (Servers) <elliott@hpe.com>
Signed-off-by: Mateusz Jończyk <mat.jonczyk@o2.pl>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Link: https://lore.kernel.org/lkml/d573ba1c-0dc4-3016-712a-cc23a8a33d42@molgen.mpg.de
Link: https://lore.kernel.org/lkml/20220911084711.13694-3-mat.jonczyk@o2.pl
Link: https://lore.kernel.org/all/20221129215008.7247-1-mat.jonczyk@o2.pl
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 32c97d980e ]
Hyper-V cleanup code comes under panic path where preemption and irq
is already disabled. So calling of unregister_syscore_ops might schedule
out the thread even for the case where mutex lock is free.
hyperv_cleanup
unregister_syscore_ops
mutex_lock(&syscore_ops_lock)
might_sleep
Here might_sleep might schedule out this thread, where voluntary preemption
config is on and this thread will never comes back. And also this was added
earlier to maintain the symmetry which is not required as this can comes
during crash shutdown path only.
To prevent the same, removing unregister_syscore_ops function call.
Signed-off-by: Gaurav Kohli <gauravkohli@linux.microsoft.com>
Reviewed-by: Michael Kelley <mikelley@microsoft.com>
Link: https://lore.kernel.org/r/1669443291-2575-1-git-send-email-gauravkohli@linux.microsoft.com
Signed-off-by: Wei Liu <wei.liu@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit f1e5250094 ]
When running as a Xen PV guest there is no need for setting up the
realmode trampoline, as realmode isn't supported in this environment.
Trying to setup the trampoline has been proven to be problematic in
some cases, especially when trying to debug early boot problems with
Xen requiring to keep the EFI boot-services memory mapped (some
firmware variants seem to claim basically all memory below 1Mb for boot
services).
Introduce new x86_platform_ops operations for that purpose, which can
be set to a NOP by the Xen PV specific kernel boot code.
[ bp: s/call_init_real_mode/do_init_real_mode/ ]
Fixes: 084ee1c641 ("x86, realmode: Relocator for realmode code")
Suggested-by: H. Peter Anvin <hpa@zytor.com>
Signed-off-by: Juergen Gross <jgross@suse.com>
Signed-off-by: Borislav Petkov <bp@suse.de>
Link: https://lore.kernel.org/r/20221123114523.3467-1-jgross@suse.com
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit ca84ce153d ]
In xen_init_lock_cpu(), the @name has allocated new string by kasprintf(),
if bind_ipi_to_irqhandler() fails, it should be freed, otherwise may lead
to a memory leak issue, fix it.
Fixes: 2d9e1e2f58 ("xen: implement Xen-specific spinlocks")
Signed-off-by: Xiu Jianfeng <xiujianfeng@huawei.com>
Reviewed-by: Juergen Gross <jgross@suse.com>
Link: https://lore.kernel.org/r/20221123155858.11382-3-xiujianfeng@huawei.com
Signed-off-by: Juergen Gross <jgross@suse.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 69143f6086 ]
These local variables @{resched|pmu|callfunc...}_name saves the new
string allocated by kasprintf(), and when bind_{v}ipi_to_irqhandler()
fails, it goes to the @fail tag, and calls xen_smp_intr_free{_pv}() to
free resource, however the new string is not saved, which cause a memory
leak issue. fix it.
Fixes: 9702785a74 ("i386: move xen")
Signed-off-by: Xiu Jianfeng <xiujianfeng@huawei.com>
Reviewed-by: Juergen Gross <jgross@suse.com>
Link: https://lore.kernel.org/r/20221123155858.11382-2-xiujianfeng@huawei.com
Signed-off-by: Juergen Gross <jgross@suse.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit cefa72129e ]
Intel ICC -hotpatch inserts 2-byte "0x66 0x90" NOP at the start of each
function to reserve extra space for hot-patching, and currently it is not
possible to probe these functions because branch_setup_xol_ops() wrongly
rejects NOP with REP prefix as it treats them like word-sized branch
instructions.
Fixes: 250bbd12c2 ("uprobes/x86: Refuse to attach uprobe to "word-sized" branch insns")
Reported-by: Seiji Nishikawa <snishika@redhat.com>
Suggested-by: Denys Vlasenko <dvlasenk@redhat.com>
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Acked-by: Masami Hiramatsu (Google) <mhiramat@kernel.org>
Link: https://lore.kernel.org/r/20221204173933.GA31544@redhat.com
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 17b8d847b9 ]
pci_get_device() will increase the reference count for the returned
pci_dev, so tgl_uncore_get_mc_dev() will return a pci_dev with its
reference count increased. We need to call pci_dev_put() to decrease the
reference count before exiting from __uncore_imc_init_box(). Add
pci_dev_put() for both normal and error path.
Fixes: fdb6482244 ("perf/x86: Add Intel Tiger Lake uncore support")
Signed-off-by: Xiongfeng Wang <wangxiongfeng2@huawei.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Kan Liang <kan.liang@linux.intel.com>
Link: https://lore.kernel.org/r/20221118063137.121512-5-wangxiongfeng2@huawei.com
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 8ebd16c11c ]
pci_get_device() will increase the reference count for the returned
pci_dev, so snr_uncore_get_mc_dev() will return a pci_dev with its
reference count increased. We need to call pci_dev_put() to decrease the
reference count. Let's add the missing pci_dev_put().
Fixes: ee49532b38 ("perf/x86/intel/uncore: Add IMC uncore support for Snow Ridge")
Signed-off-by: Xiongfeng Wang <wangxiongfeng2@huawei.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Kan Liang <kan.liang@linux.intel.com>
Link: https://lore.kernel.org/r/20221118063137.121512-4-wangxiongfeng2@huawei.com
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 1ff9dd6e70 ]
pci_get_device() will increase the reference count for the returned
'dev'. We need to call pci_dev_put() to decrease the reference count.
Since 'dev' is only used in pci_read_config_dword(), let's add
pci_dev_put() right after it.
Fixes: 9d480158ee ("perf/x86/intel/uncore: Remove uncore extra PCI dev HSWEP_PCI_PCU_3")
Signed-off-by: Xiongfeng Wang <wangxiongfeng2@huawei.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Kan Liang <kan.liang@linux.intel.com>
Link: https://lore.kernel.org/r/20221118063137.121512-3-wangxiongfeng2@huawei.com
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit c508eb042d ]
pci_get_device() will increase the reference count for the returned
pci_dev, and also decrease the reference count for the input parameter
*from* if it is not NULL.
If we break the loop in sad_cfg_iio_topology() with 'dev' not NULL. We
need to call pci_dev_put() to decrease the reference count. Since
pci_dev_put() can handle the NULL input parameter, we can just add one
pci_dev_put() right before 'return ret'.
Fixes: c1777be364 ("perf/x86/intel/uncore: Enable I/O stacks to IIO PMON mapping on SNR")
Signed-off-by: Xiongfeng Wang <wangxiongfeng2@huawei.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Kan Liang <kan.liang@linux.intel.com>
Link: https://lore.kernel.org/r/20221118063137.121512-2-wangxiongfeng2@huawei.com
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 727209376f ]
Commit b041b525da ("x86/split_lock: Make life miserable for split lockers")
changed the way the split lock detector works when in "warn" mode;
basically, it not only shows the warn message, but also intentionally
introduces a slowdown through sleeping plus serialization mechanism
on such task. Based on discussions in [0], seems the warning alone
wasn't enough motivation for userspace developers to fix their
applications.
This slowdown is enough to totally break some proprietary (aka.
unfixable) userspace[1].
Happens that originally the proposal in [0] was to add a new mode
which would warns + slowdown the "split locking" task, keeping the
old warn mode untouched. In the end, that idea was discarded and
the regular/default "warn" mode now slows down the applications. This
is quite aggressive with regards proprietary/legacy programs that
basically are unable to properly run in kernel with this change.
While it is understandable that a malicious application could DoS
by split locking, it seems unacceptable to regress old/proprietary
userspace programs through a default configuration that previously
worked. An example of such breakage was reported in [1].
Add a sysctl to allow controlling the "misery mode" behavior, as per
Thomas suggestion on [2]. This way, users running legacy and/or
proprietary software are allowed to still execute them with a decent
performance while still observing the warning messages on kernel log.
[0] https://lore.kernel.org/lkml/20220217012721.9694-1-tony.luck@intel.com/
[1] https://github.com/doitsujin/dxvk/issues/2938
[2] https://lore.kernel.org/lkml/87pmf4bter.ffs@tglx/
[ dhansen: minor changelog tweaks, including clarifying the actual
problem ]
Fixes: b041b525da ("x86/split_lock: Make life miserable for split lockers")
Suggested-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Guilherme G. Piccoli <gpiccoli@igalia.com>
Signed-off-by: Dave Hansen <dave.hansen@linux.intel.com>
Reviewed-by: Tony Luck <tony.luck@intel.com>
Tested-by: Andre Almeida <andrealmeid@igalia.com>
Link: https://lore.kernel.org/all/20221024200254.635256-1-gpiccoli%40igalia.com
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 7b72c823dd ]
commit 8795359e35 ("x86/sgx: Silence softlockup detection when
releasing large enclaves") introduced a cond_resched() during enclave
release where the EREMOVE instruction is applied to every 4k enclave
page. Giving other tasks an opportunity to run while tearing down a
large enclave placates the soft lockup detector but Iqbal found
that the fix causes a 25% performance degradation of a workload
run using Gramine.
Gramine maintains a 1:1 mapping between processes and SGX enclaves.
That means if a workload in an enclave creates a subprocess then
Gramine creates a duplicate enclave for that subprocess to run in.
The consequence is that the release of the enclave used to run
the subprocess can impact the performance of the workload that is
run in the original enclave, especially in large enclaves when
SGX2 is not in use.
The workload run by Iqbal behaves as follows:
Create enclave (enclave "A")
/* Initialize workload in enclave "A" */
Create enclave (enclave "B")
/* Run subprocess in enclave "B" and send result to enclave "A" */
Release enclave (enclave "B")
/* Run workload in enclave "A" */
Release enclave (enclave "A")
The performance impact of releasing enclave "B" in the above scenario
is amplified when there is a lot of SGX memory and the enclave size
matches the SGX memory. When there is 128GB SGX memory and an enclave
size of 128GB, from the time enclave "B" starts the 128GB SGX memory
is oversubscribed with a combined demand for 256GB from the two
enclaves.
Before commit 8795359e35 ("x86/sgx: Silence softlockup detection when
releasing large enclaves") enclave release was done in a tight loop
without giving other tasks a chance to run. Even though the system
experienced soft lockups the workload (run in enclave "A") obtained
good performance numbers because when the workload started running
there was no interference.
Commit 8795359e35 ("x86/sgx: Silence softlockup detection when
releasing large enclaves") gave other tasks opportunity to run while an
enclave is released. The impact of this in this scenario is that while
enclave "B" is released and needing to access each page that belongs
to it in order to run the SGX EREMOVE instruction on it, enclave "A"
is attempting to run the workload needing to access the enclave
pages that belong to it. This causes a lot of swapping due to the
demand for the oversubscribed SGX memory. Longer latencies are
experienced by the workload in enclave "A" while enclave "B" is
released.
Improve the performance of enclave release while still avoiding the
soft lockup detector with two enhancements:
- Only call cond_resched() after XA_CHECK_SCHED iterations.
- Use the xarray advanced API to keep the xarray locked for
XA_CHECK_SCHED iterations instead of locking and unlocking
at every iteration.
This batching solution is copied from sgx_encl_may_map() that
also iterates through all enclave pages using this technique.
With this enhancement the workload experiences a 5%
performance degradation when compared to a kernel without
commit 8795359e35 ("x86/sgx: Silence softlockup detection when
releasing large enclaves"), an improvement to the reported 25%
degradation, while still placating the soft lockup detector.
Scenarios with poor performance are still possible even with these
enhancements. For example, short workloads creating sub processes
while running in large enclaves. Further performance improvements
are pursued in user space through avoiding to create duplicate enclaves
for certain sub processes, and using SGX2 that will do lazy allocation
of pages as needed so enclaves created for sub processes start quickly
and release quickly.
Fixes: 8795359e35 ("x86/sgx: Silence softlockup detection when releasing large enclaves")
Reported-by: Md Iqbal Hossain <md.iqbal.hossain@intel.com>
Signed-off-by: Reinette Chatre <reinette.chatre@intel.com>
Signed-off-by: Dave Hansen <dave.hansen@linux.intel.com>
Tested-by: Md Iqbal Hossain <md.iqbal.hossain@intel.com>
Link: https://lore.kernel.org/all/00efa80dd9e35dc85753e1c5edb0344ac07bb1f0.1667236485.git.reinette.chatre%40intel.com
Signed-off-by: Sasha Levin <sashal@kernel.org>
commit 45be2ad007 upstream.
Recently, ld.lld moved from '--undefined-version' to
'--no-undefined-version' as the default, which breaks building the vDSO
when CONFIG_X86_SGX is not set:
ld.lld: error: version script assignment of 'LINUX_2.6' to symbol '__vdso_sgx_enter_enclave' failed: symbol not defined
__vdso_sgx_enter_enclave is only included in the vDSO when
CONFIG_X86_SGX is set. Only export it if it will be present in the final
object, which clears up the error.
Fixes: 8466436952 ("x86/vdso: Implement a vDSO for Intel SGX enclave call")
Signed-off-by: Nathan Chancellor <nathan@kernel.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Nick Desaulniers <ndesaulniers@google.com>
Link: https://github.com/ClangBuiltLinux/linux/issues/1756
Link: https://lore.kernel.org/r/20221109000306.1407357-1-nathan@kernel.org
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 6606515742 upstream.
The "force" argument to write_spec_ctrl_current() is currently ambiguous
as it does not guarantee the MSR write. This is due to the optimization
that writes to the MSR happen only when the new value differs from the
cached value.
This is fine in most cases, but breaks for S3 resume when the cached MSR
value gets out of sync with the hardware MSR value due to S3 resetting
it.
When x86_spec_ctrl_current is same as x86_spec_ctrl_base, the MSR write
is skipped. Which results in SPEC_CTRL mitigations not getting restored.
Move the MSR write from write_spec_ctrl_current() to a new function that
unconditionally writes to the MSR. Update the callers accordingly and
rename functions.
[ bp: Rework a bit. ]
Fixes: caa0ff24d5 ("x86/bugs: Keep a per-CPU IA32_SPEC_CTRL value")
Suggested-by: Borislav Petkov <bp@alien8.de>
Signed-off-by: Pawan Gupta <pawan.kumar.gupta@linux.intel.com>
Signed-off-by: Borislav Petkov (AMD) <bp@alien8.de>
Reviewed-by: Thomas Gleixner <tglx@linutronix.de>
Cc: <stable@kernel.org>
Link: https://lore.kernel.org/r/806d39b0bfec2fe8f50dc5446dff20f5bb24a959.1669821572.git.pawan.kumar.gupta@linux.intel.com
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 4dbd6a3e90 upstream.
Current code re-calculates the size after aligning the starting and
ending physical addresses on a page boundary. But the re-calculation
also embeds the masking of high order bits that exceed the size of
the physical address space (via PHYSICAL_PAGE_MASK). If the masking
removes any high order bits, the size calculation results in a huge
value that is likely to immediately fail.
Fix this by re-calculating the page-aligned size first. Then mask any
high order bits using PHYSICAL_PAGE_MASK.
Fixes: ffa71f33a8 ("x86, ioremap: Fix incorrect physical address handling in PAE mode")
Signed-off-by: Michael Kelley <mikelley@microsoft.com>
Signed-off-by: Borislav Petkov <bp@suse.de>
Acked-by: Dave Hansen <dave.hansen@linux.intel.com>
Cc: <stable@kernel.org>
Link: https://lore.kernel.org/r/1668624097-14884-2-git-send-email-mikelley@microsoft.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 50bcceb772 upstream.
pm_save_spec_msr() keeps a list of all the MSRs which _might_ need
to be saved and restored at hibernate and resume. However, it has
zero awareness of CPU support for these MSRs. It mostly works by
unconditionally attempting to manipulate these MSRs and relying on
rdmsrl_safe() being able to handle a #GP on CPUs where the support is
unavailable.
However, it's possible for reads (RDMSR) to be supported for a given MSR
while writes (WRMSR) are not. In this case, msr_build_context() sees
a successful read (RDMSR) and marks the MSR as valid. Then, later, a
write (WRMSR) fails, producing a nasty (but harmless) error message.
This causes restore_processor_state() to try and restore it, but writing
this MSR is not allowed on the Intel Atom N2600 leading to:
unchecked MSR access error: WRMSR to 0x122 (tried to write 0x0000000000000002) \
at rIP: 0xffffffff8b07a574 (native_write_msr+0x4/0x20)
Call Trace:
<TASK>
restore_processor_state
x86_acpi_suspend_lowlevel
acpi_suspend_enter
suspend_devices_and_enter
pm_suspend.cold
state_store
kernfs_fop_write_iter
vfs_write
ksys_write
do_syscall_64
? do_syscall_64
? up_read
? lock_is_held_type
? asm_exc_page_fault
? lockdep_hardirqs_on
entry_SYSCALL_64_after_hwframe
To fix this, add the corresponding X86_FEATURE bit for each MSR. Avoid
trying to manipulate the MSR when the feature bit is clear. This
required adding a X86_FEATURE bit for MSRs that do not have one already,
but it's a small price to pay.
[ bp: Move struct msr_enumeration inside the only function that uses it. ]
Fixes: 73924ec4d5 ("x86/pm: Save the MSR validity status at context setup")
Reported-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Pawan Gupta <pawan.kumar.gupta@linux.intel.com>
Signed-off-by: Borislav Petkov <bp@suse.de>
Reviewed-by: Dave Hansen <dave.hansen@linux.intel.com>
Acked-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Cc: <stable@kernel.org>
Link: https://lore.kernel.org/r/c24db75d69df6e66c0465e13676ad3f2837a2ed8.1668539735.git.pawan.kumar.gupta@linux.intel.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit aaa65d17ee upstream.
Support for the TSX control MSR is enumerated in MSR_IA32_ARCH_CAPABILITIES.
This is different from how other CPU features are enumerated i.e. via
CPUID. Currently, a call to tsx_ctrl_is_supported() is required for
enumerating the feature. In the absence of a feature bit for TSX control,
any code that relies on checking feature bits directly will not work.
In preparation for adding a feature bit check in MSR save/restore
during suspend/resume, set a new feature bit X86_FEATURE_TSX_CTRL when
MSR_IA32_TSX_CTRL is present. Also make tsx_ctrl_is_supported() use the
new feature bit to avoid any overhead of reading the MSR.
[ bp: Remove tsx_ctrl_is_supported(), add room for two more feature
bits in word 11 which are coming up in the next merge window. ]
Suggested-by: Andrew Cooper <andrew.cooper3@citrix.com>
Signed-off-by: Pawan Gupta <pawan.kumar.gupta@linux.intel.com>
Signed-off-by: Borislav Petkov <bp@suse.de>
Reviewed-by: Dave Hansen <dave.hansen@linux.intel.com>
Cc: <stable@kernel.org>
Link: https://lore.kernel.org/r/de619764e1d98afbb7a5fa58424f1278ede37b45.1668539735.git.pawan.kumar.gupta@linux.intel.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 05311ce954 upstream.
It is valid to receive external interrupt and have broken IDT entry,
which will lead to #GP with exit_int_into that will contain the index of
the IDT entry (e.g any value).
Other exceptions can happen as well, like #NP or #SS
(if stack switch fails).
Thus this warning can be user triggred and has very little value.
Cc: stable@vger.kernel.org
Signed-off-by: Maxim Levitsky <mlevitsk@redhat.com>
Message-Id: <20221103141351.50662-10-mlevitsk@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit f9697df251 upstream.
add kvm_leave_nested which wraps a call to nested_ops->leave_nested
into a function.
Cc: stable@vger.kernel.org
Signed-off-by: Maxim Levitsky <mlevitsk@redhat.com>
Message-Id: <20221103141351.50662-4-mlevitsk@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 16ae56d7e0 upstream.
Make sure that KVM uses vmcb01 before freeing nested state, and warn if
that is not the case.
This is a minimal fix for CVE-2022-3344 making the kernel print a warning
instead of a kernel panic.
Cc: stable@vger.kernel.org
Signed-off-by: Maxim Levitsky <mlevitsk@redhat.com>
Message-Id: <20221103141351.50662-3-mlevitsk@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit ed129ec905 upstream.
While not obivous, kvm_vcpu_reset() leaves the nested mode by clearing
'vcpu->arch.hflags' but it does so without all the required housekeeping.
On SVM, it is possible to have a vCPU reset while in guest mode because
unlike VMX, on SVM, INIT's are not latched in SVM non root mode and in
addition to that L1 doesn't have to intercept triple fault, which should
also trigger L1's reset if happens in L2 while L1 didn't intercept it.
If one of the above conditions happen, KVM will continue to use vmcb02
while not having in the guest mode.
Later the IA32_EFER will be cleared which will lead to freeing of the
nested guest state which will (correctly) free the vmcb02, but since
KVM still uses it (incorrectly) this will lead to a use after free
and kernel crash.
This issue is assigned CVE-2022-3344
Cc: stable@vger.kernel.org
Signed-off-by: Maxim Levitsky <mlevitsk@redhat.com>
Message-Id: <20221103141351.50662-5-mlevitsk@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 917401f26a upstream.
If the VM was terminated while nested, we free the nested state
while the vCPU still is in nested mode.
Soon a warning will be added for this condition.
Cc: stable@vger.kernel.org
Signed-off-by: Maxim Levitsky <mlevitsk@redhat.com>
Message-Id: <20221103141351.50662-2-mlevitsk@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 4ea9439fd5 upstream.
We shouldn't allow guests to poll on arbitrary port numbers off the end
of the event channel table.
Fixes: 1a65105a5a ("KVM: x86/xen: handle PV spinlocks slowpath")
[dwmw2: my bug though; the original version did check the validity as a
side-effect of an idr_find() which I ripped out in refactoring.]
Reported-by: Michal Luczaj <mhal@rbox.co>
Signed-off-by: David Woodhouse <dwmw@amazon.co.uk>
Reviewed-by: Sean Christopherson <seanjc@google.com>
Cc: stable@kernel.org
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit c2b8cdfaf3 upstream.
There are almost no hypercalls which are valid from CPL > 0, and definitely
none which are handled by the kernel.
Fixes: 2fd6df2f2b ("KVM: x86/xen: intercept EVTCHNOP_send from guests")
Reported-by: Michal Luczaj <mhal@rbox.co>
Signed-off-by: David Woodhouse <dwmw@amazon.co.uk>
Reviewed-by: Sean Christopherson <seanjc@google.com>
Cc: stable@kernel.org
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 47b0c2e4c2 upstream.
make_mmu_pages_available() must be called with mmu_lock held for write.
However, if the TDP MMU is used, it will be called with mmu_lock held for
read.
This function does nothing unless shadow pages are used, so there is no
race unless nested TDP is used.
Since nested TDP uses shadow pages, old shadow pages may be zapped by this
function even when the TDP MMU is enabled.
Since shadow pages are never allocated by kvm_tdp_mmu_map(), a race
condition can be avoided by not calling make_mmu_pages_available() if the
TDP MMU is currently in use.
I encountered this when repeatedly starting and stopping nested VM.
It can be artificially caused by allocating a large number of nested TDP
SPTEs.
For example, the following BUG and general protection fault are caused in
the host kernel.
pte_list_remove: 00000000cd54fc10 many->many
------------[ cut here ]------------
kernel BUG at arch/x86/kvm/mmu/mmu.c:963!
invalid opcode: 0000 [#1] PREEMPT SMP NOPTI
RIP: 0010:pte_list_remove.cold+0x16/0x48 [kvm]
Call Trace:
<TASK>
drop_spte+0xe0/0x180 [kvm]
mmu_page_zap_pte+0x4f/0x140 [kvm]
__kvm_mmu_prepare_zap_page+0x62/0x3e0 [kvm]
kvm_mmu_zap_oldest_mmu_pages+0x7d/0xf0 [kvm]
direct_page_fault+0x3cb/0x9b0 [kvm]
kvm_tdp_page_fault+0x2c/0xa0 [kvm]
kvm_mmu_page_fault+0x207/0x930 [kvm]
npf_interception+0x47/0xb0 [kvm_amd]
svm_invoke_exit_handler+0x13c/0x1a0 [kvm_amd]
svm_handle_exit+0xfc/0x2c0 [kvm_amd]
kvm_arch_vcpu_ioctl_run+0xa79/0x1780 [kvm]
kvm_vcpu_ioctl+0x29b/0x6f0 [kvm]
__x64_sys_ioctl+0x95/0xd0
do_syscall_64+0x5c/0x90
general protection fault, probably for non-canonical address
0xdead000000000122: 0000 [#1] PREEMPT SMP NOPTI
RIP: 0010:kvm_mmu_commit_zap_page.part.0+0x4b/0xe0 [kvm]
Call Trace:
<TASK>
kvm_mmu_zap_oldest_mmu_pages+0xae/0xf0 [kvm]
direct_page_fault+0x3cb/0x9b0 [kvm]
kvm_tdp_page_fault+0x2c/0xa0 [kvm]
kvm_mmu_page_fault+0x207/0x930 [kvm]
npf_interception+0x47/0xb0 [kvm_amd]
CVE: CVE-2022-45869
Fixes: a2855afc7e ("KVM: x86/mmu: Allow parallel page faults for the TDP MMU")
Signed-off-by: Kazuki Takiguchi <takiguchi.kazuki171@gmail.com>
Cc: stable@vger.kernel.org
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
[ Upstream commit ee68154163 ]
Commit e5d9b714fe ("x86/hyperv: fix root partition faults when writing
to VP assist page MSR") moved 'wrmsrl(HV_X64_MSR_VP_ASSIST_PAGE)' under
'if (*hvp)' condition. This works for root partition as hv_cpu_die()
does memunmap() and sets 'hv_vp_assist_page[cpu]' to NULL but breaks
non-root partitions as hv_cpu_die() doesn't free 'hv_vp_assist_page[cpu]'
for them. This causes VP assist page to remain unset after CPU
offline/online cycle:
$ rdmsr -p 24 0x40000073
10212f001
$ echo 0 > /sys/devices/system/cpu/cpu24/online
$ echo 1 > /sys/devices/system/cpu/cpu24/online
$ rdmsr -p 24 0x40000073
0
Fix the issue by always writing to HV_X64_MSR_VP_ASSIST_PAGE in
hv_cpu_init(). Note, checking 'if (!*hvp)', for root partition is
pointless as hv_cpu_die() always sets 'hv_vp_assist_page[cpu]' to
NULL (and it's also NULL initially).
Note: the fact that 'hv_vp_assist_page[cpu]' is reset to NULL may
present a (potential) issue for KVM. While Hyper-V uses
CPUHP_AP_ONLINE_DYN stage in CPU hotplug, KVM uses CPUHP_AP_KVM_STARTING
which comes earlier in CPU teardown sequence. It is theoretically
possible that Enlightened VMCS is still in use. It is unclear if the
issue is real and if using KVM with Hyper-V root partition is even
possible.
While on it, drop the unneeded smp_processor_id() call from hv_cpu_init().
Fixes: e5d9b714fe ("x86/hyperv: fix root partition faults when writing to VP assist page MSR")
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Reviewed-by: Michael Kelley <mikelley@microsoft.com>
Link: https://lore.kernel.org/r/20221103190601.399343-1-vkuznets@redhat.com
Signed-off-by: Wei Liu <wei.liu@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit baa014b954 ]
amd_pmu_enable_all() does:
if (!test_bit(idx, cpuc->active_mask))
continue;
amd_pmu_enable_event(cpuc->events[idx]);
A perf NMI of another event can come between these two steps. Perf NMI
handler internally disables and enables _all_ events, including the one
which nmi-intercepted amd_pmu_enable_all() was in process of enabling.
If that unintentionally enabled event has very low sampling period and
causes immediate successive NMI, causing the event to be throttled,
cpuc->events[idx] and cpuc->active_mask gets cleared by x86_pmu_stop().
This will result in amd_pmu_enable_event() getting called with event=NULL
when amd_pmu_enable_all() resumes after handling the NMIs. This causes a
kernel crash:
BUG: kernel NULL pointer dereference, address: 0000000000000198
#PF: supervisor read access in kernel mode
#PF: error_code(0x0000) - not-present page
[...]
Call Trace:
<TASK>
amd_pmu_enable_all+0x68/0xb0
ctx_resched+0xd9/0x150
event_function+0xb8/0x130
? hrtimer_start_range_ns+0x141/0x4a0
? perf_duration_warn+0x30/0x30
remote_function+0x4d/0x60
__flush_smp_call_function_queue+0xc4/0x500
flush_smp_call_function_queue+0x11d/0x1b0
do_idle+0x18f/0x2d0
cpu_startup_entry+0x19/0x20
start_secondary+0x121/0x160
secondary_startup_64_no_verify+0xe5/0xeb
</TASK>
amd_pmu_disable_all()/amd_pmu_enable_all() calls inside perf NMI handler
were recently added as part of BRS enablement but I'm not sure whether
we really need them. We can just disable BRS in the beginning and enable
it back while returning from NMI. This will solve the issue by not
enabling those events whose active_masks are set but are not yet enabled
in hw pmu.
Fixes: ada543459c ("perf/x86/amd: Add AMD Fam19h Branch Sampling support")
Reported-by: Linux Kernel Functional Testing <lkft@linaro.org>
Signed-off-by: Ravi Bangoria <ravi.bangoria@amd.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20221114044029.373-1-ravi.bangoria@amd.com
Signed-off-by: Sasha Levin <sashal@kernel.org>
commit ce0d998be9 upstream.
Deal with errata TGL052, ADL037 and RPL017 "Trace May Contain Incorrect
Data When Configured With Single Range Output Larger Than 4KB" by
disabling single range output whenever larger than 4KB.
Fixes: 670638477a ("perf/x86/intel/pt: Opportunistically use single range output mode")
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: stable@vger.kernel.org
Link: https://lkml.kernel.org/r/20221112151508.13768-1-adrian.hunter@intel.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 36b038791e upstream.
Mike Galbraith reported the following against an old fork of preempt-rt
but the same issue also applies to the current preempt-rt tree.
BUG: sleeping function called from invalid context at kernel/locking/spinlock_rt.c:46
in_atomic(): 1, irqs_disabled(): 0, non_block: 0, pid: 1, name: systemd
preempt_count: 1, expected: 0
RCU nest depth: 0, expected: 0
Preemption disabled at:
fpu_clone
CPU: 6 PID: 1 Comm: systemd Tainted: G E (unreleased)
Call Trace:
<TASK>
dump_stack_lvl
? fpu_clone
__might_resched
rt_spin_lock
fpu_clone
? copy_thread
? copy_process
? shmem_alloc_inode
? kmem_cache_alloc
? kernel_clone
? __do_sys_clone
? do_syscall_64
? __x64_sys_rt_sigprocmask
? syscall_exit_to_user_mode
? do_syscall_64
? syscall_exit_to_user_mode
? do_syscall_64
? syscall_exit_to_user_mode
? do_syscall_64
? exc_page_fault
? entry_SYSCALL_64_after_hwframe
</TASK>
Mike says:
The splat comes from fpu_inherit_perms() being called under fpregs_lock(),
and us reaching the spin_lock_irq() therein due to fpu_state_size_dynamic()
returning true despite static key __fpu_state_size_dynamic having never
been enabled.
Mike's assessment looks correct. fpregs_lock on a PREEMPT_RT kernel disables
preemption so calling spin_lock_irq() in fpu_inherit_perms() is unsafe. This
problem exists since commit
9e798e9aa1 ("x86/fpu: Prepare fpu_clone() for dynamically enabled features").
Even though the original bug report should not have enabled the paths at
all, the bug still exists.
fpregs_lock is necessary when editing the FPU registers or a task's FP
state but it is not necessary for fpu_inherit_perms(). The only write
of any FP state in fpu_inherit_perms() is for the new child which is
not running yet and cannot context switch or be borrowed by a kernel
thread yet. Hence, fpregs_lock is not protecting anything in the new
child until clone() completes and can be dropped earlier. The siglock
still needs to be acquired by fpu_inherit_perms() as the read of the
parent's permissions has to be serialised.
[ bp: Cleanup splat. ]
Fixes: 9e798e9aa1 ("x86/fpu: Prepare fpu_clone() for dynamically enabled features")
Reported-by: Mike Galbraith <efault@gmx.de>
Signed-off-by: Mel Gorman <mgorman@techsingularity.net>
Signed-off-by: Borislav Petkov <bp@suse.de>
Reviewed-by: Thomas Gleixner <tglx@linutronix.de>
Cc: <stable@vger.kernel.org>
Link: https://lore.kernel.org/r/20221110124400.zgymc2lnwqjukgfh@techsingularity.net
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
[ Upstream commit 9f2febf3f0 ]
Restoration of the host IA32_SPEC_CTRL value is probably too late
with respect to the return thunk training sequence.
With respect to the user/kernel boundary, AMD says, "If software chooses
to toggle STIBP (e.g., set STIBP on kernel entry, and clear it on kernel
exit), software should set STIBP to 1 before executing the return thunk
training sequence." I assume the same requirements apply to the guest/host
boundary. The return thunk training sequence is in vmenter.S, quite close
to the VM-exit. On hosts without V_SPEC_CTRL, however, the host's
IA32_SPEC_CTRL value is not restored until much later.
To avoid this, move the restoration of host SPEC_CTRL to assembly and,
for consistency, move the restoration of the guest SPEC_CTRL as well.
This is not particularly difficult, apart from some care to cover both
32- and 64-bit, and to share code between SEV-ES and normal vmentry.
Cc: stable@vger.kernel.org
Fixes: a149180fbc ("x86: Add magic AMD return-thunk")
Suggested-by: Jim Mattson <jmattson@google.com>
Reviewed-by: Sean Christopherson <seanjc@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit e287bd005a ]
Allow access to the percpu area via the GS segment base, which is
needed in order to access the saved host spec_ctrl value. In linux-next
FILL_RETURN_BUFFER also needs to access percpu data.
For simplicity, the physical address of the save area is added to struct
svm_cpu_data.
Cc: stable@vger.kernel.org
Fixes: a149180fbc ("x86: Add magic AMD return-thunk")
Reported-by: Nathan Chancellor <nathan@kernel.org>
Analyzed-by: Andrew Cooper <andrew.cooper3@citrix.com>
Tested-by: Nathan Chancellor <nathan@kernel.org>
Reviewed-by: Sean Christopherson <seanjc@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 73412dfeea ]
The svm_data percpu variable is a pointer, but it is allocated via
svm_hardware_setup() when KVM is loaded. Unlike hardware_enable()
this means that it is never NULL for the whole lifetime of KVM, and
static allocation does not waste any memory compared to the status quo.
It is also more efficient and more easily handled from assembly code,
so do it and don't look back.
Reviewed-by: Sean Christopherson <seanjc@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Stable-dep-of: e287bd005a ("KVM: SVM: restore host save area from assembly")
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 181d0fb0bb ]
The "cpu" field of struct svm_cpu_data has been write-only since commit
4b656b1202 ("KVM: SVM: force new asid on vcpu migration", 2009-08-05).
Remove it.
Reviewed-by: Sean Christopherson <seanjc@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Stable-dep-of: e287bd005a ("KVM: SVM: restore host save area from assembly")
Signed-off-by: Sasha Levin <sashal@kernel.org>
Rodrigo Branco