| Age | Commit message (Collapse) | Author |
|---|
|
|
|
commit aee9d30b9744d677509ef790f30f3a24c7841c3d upstream.
Commit
7825451fa4dc ("static_call: Add call depth tracking support")
failed to realize the problem fixed there is not specific to call depth
tracking but applies to all return-thunk uses.
Move the fix to the appropriate place and condition.
Fixes: ee88d363d156 ("x86,static_call: Use alternative RET encoding")
Reported-by: David Kaplan <David.Kaplan@amd.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Borislav Petkov (AMD) <bp@alien8.de>
Reviewed-by: Ingo Molnar <mingo@kernel.org>
Tested-by: Borislav Petkov (AMD) <bp@alien8.de>
Cc: <stable@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
commit 4ba89dd6ddeca2a733bdaed7c9a5cbe4e19d9124 upstream.
The following commit
095b8303f383 ("x86/alternative: Make custom return thunk unconditional")
made '__x86_return_thunk' a placeholder value. All code setting
X86_FEATURE_RETHUNK also changes the value of 'x86_return_thunk'. So
the optimization at the beginning of apply_returns() is dead code.
Also, before the above-mentioned commit, the optimization actually had a
bug It bypassed __static_call_fixup(), causing some raw returns to
remain unpatched in static call trampolines. Thus the 'Fixes' tag.
Fixes: d2408e043e72 ("x86/alternative: Optimize returns patching")
Signed-off-by: Josh Poimboeuf <jpoimboe@kernel.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Signed-off-by: Borislav Petkov (AMD) <bp@alien8.de>
Acked-by: Borislav Petkov (AMD) <bp@alien8.de>
Link: https://lore.kernel.org/r/16d19d2249d4485d8380fb215ffaae81e6b8119e.1693889988.git.jpoimboe@kernel.org
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
commit 5290e88ba2c742ca77c5f5b690e5af549cfd8591 upstream.
The UV code attempts to build a set of tables to allow it to do
bidirectional socket<=>node lookups.
But when nr_cpus is set to a smaller number than actually present, the
cpu_to_node() mapping information for unused CPUs is not available to
build_socket_tables(). This results in skipping some nodes or sockets
when creating the tables and leaving some -1's for later code to trip.
over, causing oopses.
The problem is that the socket<=>node lookups are created by doing a
loop over all CPUs, then looking up the CPU's APICID and socket. But
if a CPU is not present, there is no way to start this lookup.
Instead of looping over all CPUs, take CPUs out of the equation
entirely. Loop over all APICIDs which are mapped to a valid NUMA node.
Then just extract the socket-id from the APICID.
This avoid tripping over disabled CPUs.
Fixes: 8a50c5851927 ("x86/platform/uv: UV support for sub-NUMA clustering")
Signed-off-by: Steve Wahl <steve.wahl@hpe.com>
Signed-off-by: Dave Hansen <dave.hansen@linux.intel.com>
Cc: stable@vger.kernel.org
Link: https://lore.kernel.org/all/20230807141730.1117278-1-steve.wahl%40hpe.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
[ Upstream commit 75b2f7e4c9e0fd750a5a27ca9736d1daa7a3762a ]
-flto* implies -ffunction-sections. With LTO enabled, ld.lld generates
multiple .text sections for purgatory.ro:
$ readelf -S purgatory.ro | grep " .text"
[ 1] .text PROGBITS 0000000000000000 00000040
[ 7] .text.purgatory PROGBITS 0000000000000000 000020e0
[ 9] .text.warn PROGBITS 0000000000000000 000021c0
[13] .text.sha256_upda PROGBITS 0000000000000000 000022f0
[15] .text.sha224_upda PROGBITS 0000000000000000 00002be0
[17] .text.sha256_fina PROGBITS 0000000000000000 00002bf0
[19] .text.sha224_fina PROGBITS 0000000000000000 00002cc0
This causes WARNING from kexec_purgatory_setup_sechdrs():
WARNING: CPU: 26 PID: 110894 at kernel/kexec_file.c:919
kexec_load_purgatory+0x37f/0x390
Fix this by disabling LTO for purgatory.
[ AFAICT, x86 is the only arch that supports LTO and purgatory. ]
We could also fix this with an explicit linker script to rejoin .text.*
sections back into .text. However, given the benefit of LTOing purgatory
is small, simply disable the production of more .text.* sections for now.
Fixes: b33fff07e3e3 ("x86, build: allow LTO to be selected")
Signed-off-by: Song Liu <song@kernel.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Nick Desaulniers <ndesaulniers@google.com>
Reviewed-by: Sami Tolvanen <samitolvanen@google.com>
Link: https://lore.kernel.org/r/20230914170138.995606-1-song@kernel.org
Signed-off-by: Sasha Levin <sashal@kernel.org>
|
|
[ Upstream commit f530ee95b72e77b09c141c4b1a4b94d1199ffbd9 ]
The decompressor has a hard limit on the number of page tables it can
allocate. This limit is defined at compile-time and will cause boot
failure if it is reached.
The kernel is very strict and calculates the limit precisely for the
worst-case scenario based on the current configuration. However, it is
easy to forget to adjust the limit when a new use-case arises. The
worst-case scenario is rarely encountered during sanity checks.
In the case of enabling 5-level paging, a use-case was overlooked. The
limit needs to be increased by one to accommodate the additional level.
This oversight went unnoticed until Aaron attempted to run the kernel
via kexec with 5-level paging and unaccepted memory enabled.
Update wost-case calculations to include 5-level paging.
To address this issue, let's allocate some extra space for page tables.
128K should be sufficient for any use-case. The logic can be simplified
by using a single value for all kernel configurations.
[ Also add a warning, should this memory run low - by Dave Hansen. ]
Fixes: 34bbb0009f3b ("x86/boot/compressed: Enable 5-level paging during decompression stage")
Reported-by: Aaron Lu <aaron.lu@intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Link: https://lore.kernel.org/r/20230915070221.10266-1-kirill.shutemov@linux.intel.com
Signed-off-by: Sasha Levin <sashal@kernel.org>
|
|
[ Upstream commit 108af4b4bd3813610701379a58538e3339b162e4 ]
Commit 8f2d6c41e5a6 ("x86/sched: Rewrite topology setup") dropped the
SD_ASYM_PACKING flag in the DIE domain added in commit 044f0e27dec6
("x86/sched: Add the SD_ASYM_PACKING flag to the die domain of hybrid
processors"). Restore it on hybrid processors.
The die-level domain does not depend on any build configuration and now
x86_sched_itmt_flags() is always needed. Remove the build dependency on
CONFIG_SCHED_[SMT|CLUSTER|MC].
Fixes: 8f2d6c41e5a6 ("x86/sched: Rewrite topology setup")
Signed-off-by: Ricardo Neri <ricardo.neri-calderon@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Chen Yu <yu.c.chen@intel.com>
Tested-by: Caleb Callaway <caleb.callaway@intel.com>
Link: https://lkml.kernel.org/r/20230815035747.11529-1-ricardo.neri-calderon@linux.intel.com
Signed-off-by: Sasha Levin <sashal@kernel.org>
|
|
[ Upstream commit 7575e5a35267983dcbeb1e0d3a49d21ae3cf0b82 ]
Commit cb855971d717 ("x86/putuser: Provide room for padding") changed
__put_user_nocheck_*() into proper functions but failed to note that
SYM_FUNC_START() already provides ENDBR, rendering the explicit ENDBR
superfluous.
Fixes: cb855971d717 ("x86/putuser: Provide room for padding")
Reported-by: David Kaplan <David.Kaplan@amd.com>
Reviewed-by: Andrew Cooper <andrew.cooper3@citrix.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Link: https://lore.kernel.org/r/20230802110323.086971726@infradead.org
Signed-off-by: Sasha Levin <sashal@kernel.org>
|
|
[ Upstream commit 25e73b7e3f72a25aa30cbb2eecb49036e0acf066 ]
It was reported that under certain circumstances GCC emits ENDBR
instructions for _THIS_IP_ usage. Specifically, when it appears at the
start of a basic block -- but not elsewhere.
Since _THIS_IP_ is never used for control flow, these ENDBR
instructions are completely superfluous. Override the _THIS_IP_
definition for x86_64 to avoid this.
Less ENDBR instructions is better.
Fixes: 156ff4a544ae ("x86/ibt: Base IBT bits")
Reported-by: David Kaplan <David.Kaplan@amd.com>
Reviewed-by: Andrew Cooper <andrew.cooper3@citrix.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Link: https://lore.kernel.org/r/20230802110323.016197440@infradead.org
Signed-off-by: Sasha Levin <sashal@kernel.org>
|
|
[ Upstream commit ca96b162bfd21a5d55e3cd6099e4ee357a0eeb68 ]
Intel CPUs ship with ERMS for over a decade, but this is not true for
AMD. In particular one reasonably recent uarch (EPYC 7R13) does not
have it (or at least the bit is inactive when running on the Amazon EC2
cloud -- I found rather conflicting information about AMD CPUs vs the
extension).
Hand-rolled mov loops executing in this case are quite pessimal compared
to rep movsq for bigger sizes. While the upper limit depends on uarch,
everyone is well south of 1KB AFAICS and sizes bigger than that are
common.
While technically ancient CPUs may be suffering from rep usage, gcc has
been emitting it for years all over kernel code, so I don't think this
is a legitimate concern.
Sample result from read1_processes from will-it-scale (4KB reads/s):
before: 1507021
after: 1721828 (+14%)
Note that the cutoff point for rep usage is set to 64 bytes, which is
way too conservative but I'm sticking to what was done in 47ee3f1dd93b
("x86: re-introduce support for ERMS copies for user space accesses").
That is to say *some* copies will now go slower, which is fixable but
beyond the scope of this patch.
Signed-off-by: Mateusz Guzik <mjguzik@gmail.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
|
|
[ Upstream commit c64016609b6f66b753b5f37929a191477fa584c0 ]
Add new PCI Device IDs required to support AMD's new Family 1Ah-based
models 00h-1Fh, 20h and 40h-4Fh.
[ bp: Zap a useless sentence. ]
Co-developed-by: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Avadhut Naik <Avadhut.Naik@amd.com>
Signed-off-by: Borislav Petkov (AMD) <bp@alien8.de>
Link: https://lore.kernel.org/r/20230809035244.2722455-2-avadhut.naik@amd.com
Signed-off-by: Sasha Levin <sashal@kernel.org>
|
|
commit 1952e74da96fb3e48b72a2d0ece78c688a5848c1 upstream.
Skip initializing the VMSA physical address in the VMCB if the VMSA is
NULL, which occurs during intrahost migration as KVM initializes the VMCB
before copying over state from the source to the destination (including
the VMSA and its physical address).
In normal builds, __pa() is just math, so the bug isn't fatal, but with
CONFIG_DEBUG_VIRTUAL=y, the validity of the virtual address is verified
and passing in NULL will make the kernel unhappy.
Fixes: 6defa24d3b12 ("KVM: SEV: Init target VMCBs in sev_migrate_from")
Cc: stable@vger.kernel.org
Cc: Peter Gonda <pgonda@google.com>
Reviewed-by: Peter Gonda <pgonda@google.com>
Reviewed-by: Pankaj Gupta <pankaj.gupta@amd.com>
Link: https://lore.kernel.org/r/20230825022357.2852133-3-seanjc@google.com
Signed-off-by: Sean Christopherson <seanjc@google.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
commit f3cebc75e7425d6949d726bb8e937095b0aef025 upstream.
Update the target pCPU for IOMMU doorbells when updating IRTE routing if
KVM is actively running the associated vCPU. KVM currently only updates
the pCPU when loading the vCPU (via avic_vcpu_load()), and so doorbell
events will be delayed until the vCPU goes through a put+load cycle (which
might very well "never" happen for the lifetime of the VM).
To avoid inserting a stale pCPU, e.g. due to racing between updating IRTE
routing and vCPU load/put, get the pCPU information from the vCPU's
Physical APIC ID table entry (a.k.a. avic_physical_id_cache in KVM) and
update the IRTE while holding ir_list_lock. Add comments with --verbose
enabled to explain exactly what is and isn't protected by ir_list_lock.
Fixes: 411b44ba80ab ("svm: Implements update_pi_irte hook to setup posted interrupt")
Reported-by: dengqiao.joey <dengqiao.joey@bytedance.com>
Cc: stable@vger.kernel.org
Cc: Alejandro Jimenez <alejandro.j.jimenez@oracle.com>
Cc: Joao Martins <joao.m.martins@oracle.com>
Cc: Maxim Levitsky <mlevitsk@redhat.com>
Cc: Suravee Suthikulpanit <suravee.suthikulpanit@amd.com>
Tested-by: Alejandro Jimenez <alejandro.j.jimenez@oracle.com>
Reviewed-by: Joao Martins <joao.m.martins@oracle.com>
Link: https://lore.kernel.org/r/20230808233132.2499764-3-seanjc@google.com
Signed-off-by: Sean Christopherson <seanjc@google.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
commit 0c94e2468491cbf0754f49a5136ab51294a96b69 upstream.
When emulating nested VM-Exit, load L1's TSC multiplier if L1's desired
ratio doesn't match the current ratio, not if the ratio L1 is using for
L2 diverges from the default. Functionally, the end result is the same
as KVM will run L2 with L1's multiplier if L2's multiplier is the default,
i.e. checking that L1's multiplier is loaded is equivalent to checking if
L2 has a non-default multiplier.
However, the assertion that TSC scaling is exposed to L1 is flawed, as
userspace can trigger the WARN at will by writing the MSR and then
updating guest CPUID to hide the feature (modifying guest CPUID is
allowed anytime before KVM_RUN). E.g. hacking KVM's state_test
selftest to do
vcpu_set_msr(vcpu, MSR_AMD64_TSC_RATIO, 0);
vcpu_clear_cpuid_feature(vcpu, X86_FEATURE_TSCRATEMSR);
after restoring state in a new VM+vCPU yields an endless supply of:
------------[ cut here ]------------
WARNING: CPU: 10 PID: 206939 at arch/x86/kvm/svm/nested.c:1105
nested_svm_vmexit+0x6af/0x720 [kvm_amd]
Call Trace:
nested_svm_exit_handled+0x102/0x1f0 [kvm_amd]
svm_handle_exit+0xb9/0x180 [kvm_amd]
kvm_arch_vcpu_ioctl_run+0x1eab/0x2570 [kvm]
kvm_vcpu_ioctl+0x4c9/0x5b0 [kvm]
? trace_hardirqs_off+0x4d/0xa0
__se_sys_ioctl+0x7a/0xc0
__x64_sys_ioctl+0x21/0x30
do_syscall_64+0x41/0x90
entry_SYSCALL_64_after_hwframe+0x63/0xcd
Unlike the nested VMRUN path, hoisting the svm->tsc_scaling_enabled check
into the if-statement is wrong as KVM needs to ensure L1's multiplier is
loaded in the above scenario. Alternatively, the WARN_ON() could simply
be deleted, but that would make KVM's behavior even more subtle, e.g. it's
not immediately obvious why it's safe to write MSR_AMD64_TSC_RATIO when
checking only tsc_ratio_msr.
Fixes: 5228eb96a487 ("KVM: x86: nSVM: implement nested TSC scaling")
Cc: Maxim Levitsky <mlevitsk@redhat.com>
Cc: stable@vger.kernel.org
Link: https://lore.kernel.org/r/20230729011608.1065019-3-seanjc@google.com
Signed-off-by: Sean Christopherson <seanjc@google.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
commit 7cafe9b8e22bb3d77f130c461aedf6868c4aaf58 upstream.
Check for nested TSC scaling support on nested SVM VMRUN instead of
asserting that TSC scaling is exposed to L1 if L1's MSR_AMD64_TSC_RATIO
has diverged from KVM's default. Userspace can trigger the WARN at will
by writing the MSR and then updating guest CPUID to hide the feature
(modifying guest CPUID is allowed anytime before KVM_RUN). E.g. hacking
KVM's state_test selftest to do
vcpu_set_msr(vcpu, MSR_AMD64_TSC_RATIO, 0);
vcpu_clear_cpuid_feature(vcpu, X86_FEATURE_TSCRATEMSR);
after restoring state in a new VM+vCPU yields an endless supply of:
------------[ cut here ]------------
WARNING: CPU: 164 PID: 62565 at arch/x86/kvm/svm/nested.c:699
nested_vmcb02_prepare_control+0x3d6/0x3f0 [kvm_amd]
Call Trace:
<TASK>
enter_svm_guest_mode+0x114/0x560 [kvm_amd]
nested_svm_vmrun+0x260/0x330 [kvm_amd]
vmrun_interception+0x29/0x30 [kvm_amd]
svm_invoke_exit_handler+0x35/0x100 [kvm_amd]
svm_handle_exit+0xe7/0x180 [kvm_amd]
kvm_arch_vcpu_ioctl_run+0x1eab/0x2570 [kvm]
kvm_vcpu_ioctl+0x4c9/0x5b0 [kvm]
__se_sys_ioctl+0x7a/0xc0
__x64_sys_ioctl+0x21/0x30
do_syscall_64+0x41/0x90
entry_SYSCALL_64_after_hwframe+0x63/0xcd
RIP: 0033:0x45ca1b
Note, the nested #VMEXIT path has the same flaw, but needs a different
fix and will be handled separately.
Fixes: 5228eb96a487 ("KVM: x86: nSVM: implement nested TSC scaling")
Cc: Maxim Levitsky <mlevitsk@redhat.com>
Cc: stable@vger.kernel.org
Link: https://lore.kernel.org/r/20230729011608.1065019-2-seanjc@google.com
Signed-off-by: Sean Christopherson <seanjc@google.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
commit f1187ef24eb8f36e8ad8106d22615ceddeea6097 upstream.
Fix a goof where KVM tries to grab source vCPUs from the destination VM
when doing intrahost migration. Grabbing the wrong vCPU not only hoses
the guest, it also crashes the host due to the VMSA pointer being left
NULL.
BUG: unable to handle page fault for address: ffffe38687000000
#PF: supervisor read access in kernel mode
#PF: error_code(0x0000) - not-present page
PGD 0 P4D 0
Oops: 0000 [#1] SMP NOPTI
CPU: 39 PID: 17143 Comm: sev_migrate_tes Tainted: GO 6.5.0-smp--fff2e47e6c3b-next #151
Hardware name: Google, Inc. Arcadia_IT_80/Arcadia_IT_80, BIOS 34.28.0 07/10/2023
RIP: 0010:__free_pages+0x15/0xd0
RSP: 0018:ffff923fcf6e3c78 EFLAGS: 00010246
RAX: 0000000000000000 RBX: ffffe38687000000 RCX: 0000000000000100
RDX: 0000000000000100 RSI: 0000000000000000 RDI: ffffe38687000000
RBP: ffff923fcf6e3c88 R08: ffff923fcafb0000 R09: 0000000000000000
R10: 0000000000000000 R11: ffffffff83619b90 R12: ffff923fa9540000
R13: 0000000000080007 R14: ffff923f6d35d000 R15: 0000000000000000
FS: 0000000000000000(0000) GS:ffff929d0d7c0000(0000) knlGS:0000000000000000
CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: ffffe38687000000 CR3: 0000005224c34005 CR4: 0000000000770ee0
PKRU: 55555554
Call Trace:
<TASK>
sev_free_vcpu+0xcb/0x110 [kvm_amd]
svm_vcpu_free+0x75/0xf0 [kvm_amd]
kvm_arch_vcpu_destroy+0x36/0x140 [kvm]
kvm_destroy_vcpus+0x67/0x100 [kvm]
kvm_arch_destroy_vm+0x161/0x1d0 [kvm]
kvm_put_kvm+0x276/0x560 [kvm]
kvm_vm_release+0x25/0x30 [kvm]
__fput+0x106/0x280
____fput+0x12/0x20
task_work_run+0x86/0xb0
do_exit+0x2e3/0x9c0
do_group_exit+0xb1/0xc0
__x64_sys_exit_group+0x1b/0x20
do_syscall_64+0x41/0x90
entry_SYSCALL_64_after_hwframe+0x63/0xcd
</TASK>
CR2: ffffe38687000000
Fixes: 6defa24d3b12 ("KVM: SEV: Init target VMCBs in sev_migrate_from")
Cc: stable@vger.kernel.org
Cc: Peter Gonda <pgonda@google.com>
Reviewed-by: Peter Gonda <pgonda@google.com>
Reviewed-by: Pankaj Gupta <pankaj.gupta@amd.com>
Link: https://lore.kernel.org/r/20230825022357.2852133-2-seanjc@google.com
Signed-off-by: Sean Christopherson <seanjc@google.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
commit cb49631ad111570f1bad37702c11c2ae07fa2e3c upstream.
Don't inject a #UD if KVM attempts to "emulate" to skip an instruction
for an SEV guest, and instead resume the guest and hope that it can make
forward progress. When commit 04c40f344def ("KVM: SVM: Inject #UD on
attempted emulation for SEV guest w/o insn buffer") added the completely
arbitrary #UD behavior, there were no known scenarios where a well-behaved
guest would induce a VM-Exit that triggered emulation, i.e. it was thought
that injecting #UD would be helpful.
However, now that KVM (correctly) attempts to re-inject INT3/INTO, e.g. if
a #NPF is encountered when attempting to deliver the INT3/INTO, an SEV
guest can trigger emulation without a buffer, through no fault of its own.
Resuming the guest and retrying the INT3/INTO is architecturally wrong,
e.g. the vCPU will incorrectly re-hit code #DBs, but for SEV guests there
is literally no other option that has a chance of making forward progress.
Drop the #UD injection for all "skip" emulation, not just those related to
INT3/INTO, even though that means that the guest will likely end up in an
infinite loop instead of getting a #UD (the vCPU may also crash, e.g. if
KVM emulated everything about an instruction except for advancing RIP).
There's no evidence that suggests that an unexpected #UD is actually
better than hanging the vCPU, e.g. a soft-hung vCPU can still respond to
IRQs and NMIs to generate a backtrace.
Reported-by: Wu Zongyo <wuzongyo@mail.ustc.edu.cn>
Closes: https://lore.kernel.org/all/8eb933fd-2cf3-d7a9-32fe-2a1d82eac42a@mail.ustc.edu.cn
Fixes: 6ef88d6e36c2 ("KVM: SVM: Re-inject INT3/INTO instead of retrying the instruction")
Cc: stable@vger.kernel.org
Cc: Tom Lendacky <thomas.lendacky@amd.com>
Link: https://lore.kernel.org/r/20230825013621.2845700-2-seanjc@google.com
Signed-off-by: Sean Christopherson <seanjc@google.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
commit 4c08e737f056fec930b416a2bd37ed266d724f95 upstream.
Hoist the acquisition of ir_list_lock from avic_update_iommu_vcpu_affinity()
to its two callers, avic_vcpu_load() and avic_vcpu_put(), specifically to
encapsulate the write to the vCPU's entry in the AVIC Physical ID table.
This will allow a future fix to pull information from the Physical ID entry
when updating the IRTE, without potentially consuming stale information,
i.e. without racing with the vCPU being (un)loaded.
Add a comment to call out that ir_list_lock does NOT protect against
multiple writers, specifically that reading the Physical ID entry in
avic_vcpu_put() outside of the lock is safe.
To preserve some semblance of independence from ir_list_lock, keep the
READ_ONCE() in avic_vcpu_load() even though acuiring the spinlock
effectively ensures the load(s) will be generated after acquiring the
lock.
Cc: stable@vger.kernel.org
Tested-by: Alejandro Jimenez <alejandro.j.jimenez@oracle.com>
Reviewed-by: Joao Martins <joao.m.martins@oracle.com>
Link: https://lore.kernel.org/r/20230808233132.2499764-2-seanjc@google.com
Signed-off-by: Sean Christopherson <seanjc@google.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
commit 50011c2a245792993f2756e5b5b571512bfa409e upstream.
Reset the mask of available "registers" and refresh the IDT vectoring
info snapshot in vmx_vcpu_enter_exit(), before KVM potentially handles a
an NMI VM-Exit. One of the "registers" that KVM VMX lazily loads is the
vmcs.VM_EXIT_INTR_INFO field, which is holds the vector+type on "exception
or NMI" VM-Exits, i.e. is needed to identify NMIs. Clearing the available
registers bitmask after handling NMIs results in KVM querying info from
the last VM-Exit that read vmcs.VM_EXIT_INTR_INFO, and leads to both
missed NMIs and spurious NMIs in the host.
Opportunistically grab vmcs.IDT_VECTORING_INFO_FIELD early in the VM-Exit
path too, e.g. to guard against similar consumption of stale data. The
field is read on every "normal" VM-Exit, and there's no point in delaying
the inevitable.
Reported-by: Like Xu <like.xu.linux@gmail.com>
Fixes: 11df586d774f ("KVM: VMX: Handle NMI VM-Exits in noinstr region")
Cc: stable@vger.kernel.org
Link: https://lore.kernel.org/r/20230825014532.2846714-1-seanjc@google.com
Signed-off-by: Sean Christopherson <seanjc@google.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
[ Upstream commit 5df8ecfe3632d5879d1f154f7aa8de441b5d1c89 ]
Drop the explicit check on the extended CPUID level in cpu_has_svm(), the
kernel's cached CPUID info will leave the entire SVM leaf unset if said
leaf is not supported by hardware. Prior to using cached information,
the check was needed to avoid false positives due to Intel's rather crazy
CPUID behavior of returning the values of the maximum supported leaf if
the specified leaf is unsupported.
Fixes: 682a8108872f ("x86/kvm/svm: Simplify cpu_has_svm()")
Link: https://lore.kernel.org/r/20230721201859.2307736-13-seanjc@google.com
Signed-off-by: Sean Christopherson <seanjc@google.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
|
|
[ Upstream commit 389fbbec261b2842fd0e34b26a2b288b122cc406 ]
Immediately mark NMIs as unmasked in response to #VMGEXIT(NMI complete)
instead of setting awaiting_iret_completion and waiting until the *next*
VM-Exit to unmask NMIs. The whole point of "NMI complete" is that the
guest is responsible for telling the hypervisor when it's safe to inject
an NMI, i.e. there's no need to wait. And because there's no IRET to
single-step, the next VM-Exit could be a long time coming, i.e. KVM could
incorrectly hold an NMI pending for far longer than what is required and
expected.
Opportunistically fix a stale reference to HF_IRET_MASK.
Fixes: 916b54a7688b ("KVM: x86: Move HF_NMI_MASK and HF_IRET_MASK into "struct vcpu_svm"")
Fixes: 4444dfe4050b ("KVM: SVM: Add NMI support for an SEV-ES guest")
Cc: Tom Lendacky <thomas.lendacky@amd.com>
Link: https://lore.kernel.org/r/20230615063757.3039121-9-aik@amd.com
Signed-off-by: Sean Christopherson <seanjc@google.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
|
|
commit 6f7f984fa85b305799076a1bcec941b9377587de upstream.
Starting from SPR, the basic uncore PMON information is retrieved from
the discovery table (resides in an MMIO space populated by BIOS). It is
called the discovery method. The existing value of the type->num_boxes
is from the discovery table.
On some SPR variants, there is a firmware bug that makes the value from the
discovery table incorrect. We use the value from the
SPR_MSR_UNC_CBO_CONFIG MSR to replace the one from the discovery table:
38776cc45eb7 ("perf/x86/uncore: Correct the number of CHAs on SPR")
Unfortunately, the SPR_MSR_UNC_CBO_CONFIG isn't available for the EMR
XCC (Always returns 0), but the above firmware bug doesn't impact the
EMR XCC.
Don't let the value from the MSR replace the existing value from the
discovery table.
Fixes: 38776cc45eb7 ("perf/x86/uncore: Correct the number of CHAs on SPR")
Reported-by: Stephane Eranian <eranian@google.com>
Reported-by: Yunying Sun <yunying.sun@intel.com>
Signed-off-by: Kan Liang <kan.liang@linux.intel.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Tested-by: Yunying Sun <yunying.sun@intel.com>
Link: https://lore.kernel.org/r/20230905134248.496114-1-kan.liang@linux.intel.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
commit 65e710899fd19f435f40268f3a92dfaa11f14470 upstream.
With ":text =0xcccc", ld.lld fills unused text area with 0xcccc0000.
Example objdump -D output:
ffffffff82b04203: 00 00 add %al,(%rax)
ffffffff82b04205: cc int3
ffffffff82b04206: cc int3
ffffffff82b04207: 00 00 add %al,(%rax)
ffffffff82b04209: cc int3
ffffffff82b0420a: cc int3
Replace it with ":text =0xcccccccc", so we get the following instead:
ffffffff82b04203: cc int3
ffffffff82b04204: cc int3
ffffffff82b04205: cc int3
ffffffff82b04206: cc int3
ffffffff82b04207: cc int3
ffffffff82b04208: cc int3
gcc/ld doesn't seem to have the same issue. The generated code stays the
same for gcc/ld.
Signed-off-by: Song Liu <song@kernel.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Kees Cook <keescook@chromium.org>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Fixes: 7705dc855797 ("x86/vmlinux: Use INT3 instead of NOP for linker fill bytes")
Link: https://lore.kernel.org/r/20230906175215.2236033-1-song@kernel.org
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
commit 3d7d72a34e05b23e21bafc8bfb861e73c86b31f3 upstream.
On large enclaves we hit the softlockup warning with following call trace:
xa_erase()
sgx_vepc_release()
__fput()
task_work_run()
do_exit()
The latency issue is similar to the one fixed in:
8795359e35bc ("x86/sgx: Silence softlockup detection when releasing large enclaves")
The test system has 64GB of enclave memory, and all is assigned to a single VM.
Release of 'vepc' takes a longer time and causes long latencies, which triggers
the softlockup warning.
Add cond_resched() to give other tasks a chance to run and reduce
latencies, which also avoids the softlockup detector.
[ mingo: Rewrote the changelog. ]
Fixes: 540745ddbc70 ("x86/sgx: Introduce virtual EPC for use by KVM guests")
Reported-by: Yu Zhang <yu.zhang@ionos.com>
Signed-off-by: Jack Wang <jinpu.wang@ionos.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Tested-by: Yu Zhang <yu.zhang@ionos.com>
Reviewed-by: Jarkko Sakkinen <jarkko@kernel.org>
Reviewed-by: Kai Huang <kai.huang@intel.com>
Acked-by: Haitao Huang <haitao.huang@linux.intel.com>
Cc: stable@vger.kernel.org
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
commit 3f874c9b2aae8e30463efc1872bea4baa9ed25dc upstream.
Vasant reported that kexec() can hang or reset the machine when it tries to
park CPUs via INIT. This happens when the kernel is using extended APIC,
but the present mask has APIC IDs >= 0x100 enumerated.
As extended APIC can only handle 8 bit of APIC ID sending INIT to APIC ID
0x100 sends INIT to APIC ID 0x0. That's the boot CPU which is special on
x86 and INIT causes the system to hang or resets the machine.
Prevent this by sending INIT only to those CPUs which have been booted
once.
Fixes: 45e34c8af58f ("x86/smp: Put CPUs into INIT on shutdown if possible")
Reported-by: Dheeraj Kumar Srivastava <dheerajkumar.srivastava@amd.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Tested-by: Vasant Hegde <vasant.hegde@amd.com>
Link: https://lore.kernel.org/r/87cyzwjbff.ffs@tglx
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
commit 4240e2ebe67941ce2c4f5c866c3af4b5ac7a0c67 upstream.
The Instruction Fetch (IF) units on current AMD Zen-based systems do not
guarantee a synchronous #MC is delivered for poison consumption errors.
Therefore, MCG_STATUS[EIPV|RIPV] will not be set. However, the
microarchitecture does guarantee that the exception is delivered within
the same context. In other words, the exact rIP is not known, but the
context is known to not have changed.
There is no architecturally-defined method to determine this behavior.
The Code Segment (CS) register is always valid on such IF unit poison
errors regardless of the value of MCG_STATUS[EIPV|RIPV].
Add a quirk to save the CS register for poison consumption from the IF
unit banks.
This is needed to properly determine the context of the error.
Otherwise, the severity grading function will assume the context is
IN_KERNEL due to the m->cs value being 0 (the initialized value). This
leads to unnecessary kernel panics on data poison errors due to the
kernel believing the poison consumption occurred in kernel context.
Signed-off-by: Yazen Ghannam <yazen.ghannam@amd.com>
Signed-off-by: Borislav Petkov (AMD) <bp@alien8.de>
Cc: stable@vger.kernel.org
Link: https://lore.kernel.org/r/20230814200853.29258-1-yazen.ghannam@amd.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
commit ac3f9c9f1b37edaa7d1a9b908bc79d843955a1a2 upstream.
enc_dec_hypercall() accepted a page count instead of a size, which
forced its callers to round up. As a result, non-page aligned
vaddrs caused pages to be spuriously marked as decrypted via the
encryption status hypercall, which in turn caused consistent
corruption of pages during live migration. Live migration requires
accurate encryption status information to avoid migrating pages
from the wrong perspective.
Fixes: 064ce6c550a0 ("mm: x86: Invoke hypercall when page encryption status is changed")
Signed-off-by: Steve Rutherford <srutherford@google.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Tom Lendacky <thomas.lendacky@amd.com>
Reviewed-by: Pankaj Gupta <pankaj.gupta@amd.com>
Tested-by: Ben Hillier <bhillier@google.com>
Cc: stable@vger.kernel.org
Link: https://lore.kernel.org/r/20230824223731.2055016-1-srutherford@google.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
[ Upstream commit c9f4c45c8ec3f07f4f083f9750032a1ec3eab6b2 ]
The Gather Data Sampling (GDS) vulnerability is common to all Skylake
processors. However, the "client" Skylakes* are now in this list:
https://www.intel.com/content/www/us/en/support/articles/000022396/processors.html
which means they are no longer included for new vulnerabilities here:
https://www.intel.com/content/www/us/en/developer/topic-technology/software-security-guidance/processors-affected-consolidated-product-cpu-model.html
or in other GDS documentation. Thus, they were not included in the
original GDS mitigation patches.
Mark SKYLAKE and SKYLAKE_L as vulnerable to GDS to match all the
other Skylake CPUs (which include Kaby Lake). Also group the CPUs
so that the ones that share the exact same vulnerabilities are next
to each other.
Last, move SRBDS to the end of each line. This makes it clear at a
glance that SKYLAKE_X is unique. Of the five Skylakes, it is the
only "server" CPU and has a different implementation from the
clients of the "special register" hardware, making it immune to SRBDS.
This makes the diff much harder to read, but the resulting table is
worth it.
I very much appreciate the report from Michael Zhivich about this
issue. Despite what level of support a hardware vendor is providing,
the kernel very much needs an accurate and up-to-date list of
vulnerable CPUs. More reports like this are very welcome.
* Client Skylakes are CPUID 406E3/506E3 which is family 6, models
0x4E and 0x5E, aka INTEL_FAM6_SKYLAKE and INTEL_FAM6_SKYLAKE_L.
Reported-by: Michael Zhivich <mzhivich@akamai.com>
Fixes: 8974eb588283 ("x86/speculation: Add Gather Data Sampling mitigation")
Signed-off-by: Dave Hansen <dave.hansen@linux.intel.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Daniel Sneddon <daniel.sneddon@linux.intel.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
|
|
[ Upstream commit 4ba2909638a29630a346d6c4907a3105409bee7d ]
This source file already includes <linux/miscdevice.h>, which contains
the same macro. It doesn't need to be defined here again.
Fixes: 874bcd00f520 ("apm-emulation: move APM_MINOR_DEV to include/linux/miscdevice.h")
Signed-off-by: Randy Dunlap <rdunlap@infradead.org>
Cc: Jiri Kosina <jikos@kernel.org>
Cc: x86@kernel.org
Cc: Sohil Mehta <sohil.mehta@intel.com>
Cc: Corentin Labbe <clabbe.montjoie@gmail.com>
Reviewed-by: Sohil Mehta <sohil.mehta@intel.com>
Link: https://lore.kernel.org/r/20230728011120.759-1-rdunlap@infradead.org
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
|
|
[ Upstream commit 548cb932051fb6232ac983ed6673dae7bdf3cf4c ]
Visible glitches have been observed when running graphics applications on
Linux under Xen hypervisor. Those observations have been confirmed with
failures from kms_pwrite_crc Intel GPU test that verifies data coherency
of DRM frame buffer objects using hardware CRC checksums calculated by
display controllers, exposed to userspace via debugfs. Affected
processing paths have then been identified with new IGT test variants that
mmap the objects using different methods and caching modes [1].
When running as a Xen PV guest, Linux uses Xen provided PAT configuration
which is different from its native one. In particular, Xen specific PTE
encoding of write-combining caching, likely used by graphics applications,
differs from the Linux default one found among statically defined minimal
set of supported modes. Since Xen defines PTE encoding of the WC mode as
_PAGE_PAT, it no longer belongs to the minimal set, depends on correct
handling of _PAGE_PAT bit, and can be mismatched with write-back caching.
When a user calls mmap() for a DRM buffer object, DRM device specific
.mmap file operation, called from mmap_region(), takes care of setting PTE
encoding bits in a vm_page_prot field of an associated virtual memory area
structure. Unfortunately, _PAGE_PAT bit is not preserved when the vma's
.vm_flags are then applied to .vm_page_prot via vm_set_page_prot(). Bits
to be preserved are determined with _PAGE_CHG_MASK symbol that doesn't
cover _PAGE_PAT. As a consequence, WB caching is requested instead of WC
when running under Xen (also, WP is silently changed to WT, and UC
downgraded to UC_MINUS). When running on bare metal, WC is not affected,
but WP and WT extra modes are unintentionally replaced with WC and UC,
respectively.
WP and WT modes, encoded with _PAGE_PAT bit set, were introduced by commit
281d4078bec3 ("x86: Make page cache mode a real type"). Care was taken
to extend _PAGE_CACHE_MASK symbol with that additional bit, but that
symbol has never been used for identification of bits preserved when
applying page protection flags. Support for all cache modes under Xen,
including the problematic WC mode, was then introduced by commit
47591df50512 ("xen: Support Xen pv-domains using PAT").
The issue needs to be fixed by including _PAGE_PAT bit into a bitmask used
by pgprot_modify() for selecting bits to be preserved. We can do that
either internally to pgprot_modify() (as initially proposed), or by making
_PAGE_PAT a part of _PAGE_CHG_MASK. If we go for the latter then, since
_PAGE_PAT is the same as _PAGE_PSE, we need to note that _HPAGE_CHG_MASK
-- a huge pmds' counterpart of _PAGE_CHG_MASK, introduced by commit
c489f1257b8c ("thp: add pmd_modify"), defined as (_PAGE_CHG_MASK |
_PAGE_PSE) -- will no longer differ from _PAGE_CHG_MASK. If such
modification of _PAGE_CHG_MASK was irrelevant to its users then one might
wonder why that new _HPAGE_CHG_MASK symbol was introduced instead of
reusing the existing one with that otherwise irrelevant bit (_PAGE_PSE in
that case) added.
Add _PAGE_PAT to _PAGE_CHG_MASK and _PAGE_PAT_LARGE to _HPAGE_CHG_MASK for
symmetry. Split out common bits from both symbols to a common symbol for
clarity.
[ dhansen: tweak the solution changelog description ]
[1] https://gitlab.freedesktop.org/drm/igt-gpu-tools/-/commit/0f0754413f14
Fixes: 281d4078bec3 ("x86: Make page cache mode a real type")
Signed-off-by: Janusz Krzysztofik <janusz.krzysztofik@linux.intel.com>
Signed-off-by: Dave Hansen <dave.hansen@linux.intel.com>
Reviewed-by: Andi Shyti <andi.shyti@linux.intel.com>
Reviewed-by: Juergen Gross <jgross@suse.com>
Tested-by: Marek Marczykowski-Górecki <marmarek@invisiblethingslab.com>
Link: https://gitlab.freedesktop.org/drm/intel/-/issues/7648
Link: https://lore.kernel.org/all/20230710073613.8006-2-janusz.krzysztofik%40linux.intel.com
Signed-off-by: Sasha Levin <sashal@kernel.org>
|
|
[ Upstream commit 8d539b84f1e3478436f978ceaf55a0b6cab497b5 ]
The APIs that allow backtracing across CPUs have always had a way to
exclude the current CPU. This convenience means callers didn't need to
find a place to allocate a CPU mask just to handle the common case.
Let's extend the API to take a CPU ID to exclude instead of just a
boolean. This isn't any more complex for the API to handle and allows the
hardlockup detector to exclude a different CPU (the one it already did a
trace for) without needing to find space for a CPU mask.
Arguably, this new API also encourages safer behavior. Specifically if
the caller wants to avoid tracing the current CPU (maybe because they
already traced the current CPU) this makes it more obvious to the caller
that they need to make sure that the current CPU ID can't change.
[akpm@linux-foundation.org: fix trigger_allbutcpu_cpu_backtrace() stub]
Link: https://lkml.kernel.org/r/20230804065935.v4.1.Ia35521b91fc781368945161d7b28538f9996c182@changeid
Signed-off-by: Douglas Anderson <dianders@chromium.org>
Acked-by: Michal Hocko <mhocko@suse.com>
Cc: kernel test robot <lkp@intel.com>
Cc: Lecopzer Chen <lecopzer.chen@mediatek.com>
Cc: Petr Mladek <pmladek@suse.com>
Cc: Pingfan Liu <kernelfans@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Stable-dep-of: 1f38c86bb29f ("watchdog/hardlockup: avoid large stack frames in watchdog_hardlockup_check()")
Signed-off-by: Sasha Levin <sashal@kernel.org>
|
|
[ Upstream commit 264b82fdb4989cf6a44a2bcd0c6ea05e8026b2ac ]
The 4-to-5 level mode switch trampoline disables long mode and paging in
order to be able to flick the LA57 bit. According to section 3.4.1.1 of
the x86 architecture manual [0], 64-bit GPRs might not retain the upper
32 bits of their contents across such a mode switch.
Given that RBP, RBX and RSI are live at this point, preserve them on the
stack, along with the return address that might be above 4G as well.
[0] Intel® 64 and IA-32 Architectures Software Developer’s Manual, Volume 1: Basic Architecture
"Because the upper 32 bits of 64-bit general-purpose registers are
undefined in 32-bit modes, the upper 32 bits of any general-purpose
register are not preserved when switching from 64-bit mode to a 32-bit
mode (to protected mode or compatibility mode). Software must not
depend on these bits to maintain a value after a 64-bit to 32-bit
mode switch."
Fixes: 194a9749c73d650c ("x86/boot/compressed/64: Handle 5-level paging boot if kernel is above 4G")
Signed-off-by: Ard Biesheuvel <ardb@kernel.org>
Signed-off-by: Borislav Petkov (AMD) <bp@alien8.de>
Link: https://lore.kernel.org/r/20230807162720.545787-2-ardb@kernel.org
Signed-off-by: Sasha Levin <sashal@kernel.org>
|
|
[ Upstream commit d6b45484c130f4095313ae3edeb4aae662c12fb1 ]
Several architectures define arch_try_local_cmpxchg macro using
internal temporary variables named ___old, __old or _old. Remove
temporary varible in local_try_cmpxchg to avoid variable shadowing.
No functional change intended.
Fixes: d994f2c8e241 ("locking/arch: Wire up local_try_cmpxchg()")
Closes: https://lore.kernel.org/lkml/CAFGhKbyxtuk=LoW-E3yLXgcmR93m+Dfo5-u9oQA_YC5Fcy_t9g@mail.gmail.com/
Reported-by: Charlemagne Lasse <charlemagnelasse@gmail.com>
Signed-off-by: Uros Bizjak <ubizjak@gmail.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20230708090048.63046-1-ubizjak@gmail.com
Signed-off-by: Sasha Levin <sashal@kernel.org>
|
|
commit 0a6b58c5cd0dfd7961e725212f0fc8dfc5d96195 upstream.
On the parisc architecture, lockdep reports for all static objects which
are in the __initdata section (e.g. "setup_done" in devtmpfs,
"kthreadd_done" in init/main.c) this warning:
INFO: trying to register non-static key.
The warning itself is wrong, because those objects are in the __initdata
section, but the section itself is on parisc outside of range from
_stext to _end, which is why the static_obj() functions returns a wrong
answer.
While fixing this issue, I noticed that the whole existing check can
be simplified a lot.
Instead of checking against the _stext and _end symbols (which include
code areas too) just check for the .data and .bss segments (since we check a
data object). This can be done with the existing is_kernel_core_data()
macro.
In addition objects in the __initdata section can be checked with
init_section_contains(), and is_kernel_rodata() allows keys to be in the
_ro_after_init section.
This partly reverts and simplifies commit bac59d18c701 ("x86/setup: Fix static
memory detection").
Link: https://lkml.kernel.org/r/ZNqrLRaOi/3wPAdp@p100
Fixes: bac59d18c701 ("x86/setup: Fix static memory detection")
Signed-off-by: Helge Deller <deller@gmx.de>
Cc: Borislav Petkov <bp@suse.de>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Guenter Roeck <linux@roeck-us.net>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: "Rafael J. Wysocki" <rafael@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
0-Day found a 34.6% regression in stress-ng's 'af-alg' test case, and
bisected it to commit b81fac906a8f ("x86/fpu: Move FPU initialization into
arch_cpu_finalize_init()"), which optimizes the FPU init order, and moves
the CR4_OSXSAVE enabling into a later place:
arch_cpu_finalize_init
identify_boot_cpu
identify_cpu
generic_identify
get_cpu_cap --> setup cpu capability
...
fpu__init_cpu
fpu__init_cpu_xstate
cr4_set_bits(X86_CR4_OSXSAVE);
As the FPU is not yet initialized the CPU capability setup fails to set
X86_FEATURE_OSXSAVE. Many security module like 'camellia_aesni_avx_x86_64'
depend on this feature and therefore fail to load, causing the regression.
Cure this by setting X86_FEATURE_OSXSAVE feature right after OSXSAVE
enabling.
[ tglx: Moved it into the actual BSP FPU initialization code and added a comment ]
Fixes: b81fac906a8f ("x86/fpu: Move FPU initialization into arch_cpu_finalize_init()")
Reported-by: kernel test robot <oliver.sang@intel.com>
Signed-off-by: Feng Tang <feng.tang@intel.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Cc: stable@vger.kernel.org
Link: https://lore.kernel.org/lkml/202307192135.203ac24e-oliver.sang@intel.com
Link: https://lore.kernel.org/lkml/20230823065747.92257-1-feng.tang@intel.com
|
|
The thread flag TIF_NEED_FPU_LOAD indicates that the FPU saved state is
valid and should be reloaded when returning to userspace. However, the
kernel will skip doing this if the FPU registers are already valid as
determined by fpregs_state_valid(). The logic embedded there considers
the state valid if two cases are both true:
1: fpu_fpregs_owner_ctx points to the current tasks FPU state
2: the last CPU the registers were live in was the current CPU.
This is usually correct logic. A CPU’s fpu_fpregs_owner_ctx is set to
the current FPU during the fpregs_restore_userregs() operation, so it
indicates that the registers have been restored on this CPU. But this
alone doesn’t preclude that the task hasn’t been rescheduled to a
different CPU, where the registers were modified, and then back to the
current CPU. To verify that this was not the case the logic relies on the
second condition. So the assumption is that if the registers have been
restored, AND they haven’t had the chance to be modified (by being
loaded on another CPU), then they MUST be valid on the current CPU.
Besides the lazy FPU optimizations, the other cases where the FPU
registers might not be valid are when the kernel modifies the FPU register
state or the FPU saved buffer. In this case the operation modifying the
FPU state needs to let the kernel know the correspondence has been
broken. The comment in “arch/x86/kernel/fpu/context.h” has:
/*
...
* If the FPU register state is valid, the kernel can skip restoring the
* FPU state from memory.
*
* Any code that clobbers the FPU registers or updates the in-memory
* FPU state for a task MUST let the rest of the kernel know that the
* FPU registers are no longer valid for this task.
*
* Either one of these invalidation functions is enough. Invalidate
* a resource you control: CPU if using the CPU for something else
* (with preemption disabled), FPU for the current task, or a task that
* is prevented from running by the current task.
*/
However, this is not completely true. When the kernel modifies the
registers or saved FPU state, it can only rely on
__fpu_invalidate_fpregs_state(), which wipes the FPU’s last_cpu
tracking. The exec path instead relies on fpregs_deactivate(), which sets
the CPU’s FPU context to NULL. This was observed to fail to restore the
reset FPU state to the registers when returning to userspace in the
following scenario:
1. A task is executing in userspace on CPU0
- CPU0’s FPU context points to tasks
- fpu->last_cpu=CPU0
2. The task exec()’s
3. While in the kernel the task is preempted
- CPU0 gets a thread executing in the kernel (such that no other
FPU context is activated)
- Scheduler sets task’s fpu->last_cpu=CPU0 when scheduling out
4. Task is migrated to CPU1
5. Continuing the exec(), the task gets to
fpu_flush_thread()->fpu_reset_fpregs()
- Sets CPU1’s fpu context to NULL
- Copies the init state to the task’s FPU buffer
- Sets TIF_NEED_FPU_LOAD on the task
6. The task reschedules back to CPU0 before completing the exec() and
returning to userspace
- During the reschedule, scheduler finds TIF_NEED_FPU_LOAD is set
- Skips saving the registers and updating task’s fpu→last_cpu,
because TIF_NEED_FPU_LOAD is the canonical source.
7. Now CPU0’s FPU context is still pointing to the task’s, and
fpu->last_cpu is still CPU0. So fpregs_state_valid() returns true even
though the reset FPU state has not been restored.
So the root cause is that exec() is doing the wrong kind of invalidate. It
should reset fpu->last_cpu via __fpu_invalidate_fpregs_state(). Further,
fpu__drop() doesn't really seem appropriate as the task (and FPU) are not
going away, they are just getting reset as part of an exec. So switch to
__fpu_invalidate_fpregs_state().
Also, delete the misleading comment that says that either kind of
invalidate will be enough, because it’s not always the case.
Fixes: 33344368cb08 ("x86/fpu: Clean up the fpu__clear() variants")
Reported-by: Lei Wang <lei4.wang@intel.com>
Signed-off-by: Rick Edgecombe <rick.p.edgecombe@intel.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Tested-by: Lijun Pan <lijun.pan@intel.com>
Reviewed-by: Sohil Mehta <sohil.mehta@intel.com>
Acked-by: Lijun Pan <lijun.pan@intel.com>
Cc: stable@vger.kernel.org
Link: https://lore.kernel.org/r/20230818170305.502891-1-rick.p.edgecombe@intel.com
|
|
Specify how is SRSO mitigated when SMT is disabled. Also, correct the
SMT check for that.
Fixes: e9fbc47b818b ("x86/srso: Disable the mitigation on unaffected configurations")
Suggested-by: Josh Poimboeuf <jpoimboe@kernel.org>
Signed-off-by: Borislav Petkov (AMD) <bp@alien8.de>
Acked-by: Josh Poimboeuf <jpoimboe@kernel.org>
Link: https://lore.kernel.org/r/20230814200813.p5czl47zssuej7nv@treble
|
|
Christian reported spurious module load crashes after some of Song's
module memory layout patches.
Turns out that if the very last instruction on the very last page of the
module is a 'JMP __x86_return_thunk' then __static_call_fixup() will
trip a fault and die.
And while the module rework made this slightly more likely to happen,
it's always been possible.
Fixes: ee88d363d156 ("x86,static_call: Use alternative RET encoding")
Reported-by: Christian Bricart <christian@bricart.de>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: Josh Poimboeuf <jpoimboe@kernel.org>
Link: https://lkml.kernel.org/r/20230816104419.GA982867@hirez.programming.kicks-ass.net
|
|
The goal is to eventually have a proper documentation about all this.
Signed-off-by: Borislav Petkov (AMD) <bp@alien8.de>
Link: https://lore.kernel.org/r/20230814164447.GFZNpZ/64H4lENIe94@fat_crate.local
|
|
Similar to how it doesn't make sense to have UNTRAIN_RET have two
untrain calls, it also doesn't make sense for VMEXIT to have an extra
IBPB call.
This cures VMEXIT doing potentially unret+IBPB or double IBPB.
Also, the (SEV) VMEXIT case seems to have been overlooked.
Redefine the meaning of the synthetic IBPB flags to:
- ENTRY_IBPB -- issue IBPB on entry (was: entry + VMEXIT)
- IBPB_ON_VMEXIT -- issue IBPB on VMEXIT
And have 'retbleed=ibpb' set *BOTH* feature flags to ensure it retains
the previous behaviour and issues IBPB on entry+VMEXIT.
The new 'srso=ibpb_vmexit' option only sets IBPB_ON_VMEXIT.
Create UNTRAIN_RET_VM specifically for the VMEXIT case, and have that
check IBPB_ON_VMEXIT.
All this avoids having the VMEXIT case having to check both ENTRY_IBPB
and IBPB_ON_VMEXIT and simplifies the alternatives.
Fixes: fb3bd914b3ec ("x86/srso: Add a Speculative RAS Overflow mitigation")
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Borislav Petkov (AMD) <bp@alien8.de>
Link: https://lore.kernel.org/r/20230814121149.109557833@infradead.org
|
|
Since there can only be one active return_thunk, there only needs be
one (matching) untrain_ret. It fundamentally doesn't make sense to
allow multiple untrain_ret at the same time.
Fold all the 3 different untrain methods into a single (temporary)
helper stub.
Fixes: fb3bd914b3ec ("x86/srso: Add a Speculative RAS Overflow mitigation")
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Borislav Petkov (AMD) <bp@alien8.de>
Link: https://lore.kernel.org/r/20230814121149.042774962@infradead.org
|
|
For a more consistent namespace.
[ bp: Fixup names in the doc too. ]
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Borislav Petkov (AMD) <bp@alien8.de>
Link: https://lore.kernel.org/r/20230814121148.976236447@infradead.org
|
|
Rename the original retbleed return thunk and untrain_ret to
retbleed_return_thunk() and retbleed_untrain_ret().
No functional changes.
Suggested-by: Josh Poimboeuf <jpoimboe@kernel.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Borislav Petkov (AMD) <bp@alien8.de>
Link: https://lore.kernel.org/r/20230814121148.909378169@infradead.org
|
|
Use the existing configurable return thunk. There is absolute no
justification for having created this __x86_return_thunk alternative.
To clarify, the whole thing looks like:
Zen3/4 does:
srso_alias_untrain_ret:
nop2
lfence
jmp srso_alias_return_thunk
int3
srso_alias_safe_ret: // aliasses srso_alias_untrain_ret just so
add $8, %rsp
ret
int3
srso_alias_return_thunk:
call srso_alias_safe_ret
ud2
While Zen1/2 does:
srso_untrain_ret:
movabs $foo, %rax
lfence
call srso_safe_ret (jmp srso_return_thunk ?)
int3
srso_safe_ret: // embedded in movabs instruction
add $8,%rsp
ret
int3
srso_return_thunk:
call srso_safe_ret
ud2
While retbleed does:
zen_untrain_ret:
test $0xcc, %bl
lfence
jmp zen_return_thunk
int3
zen_return_thunk: // embedded in the test instruction
ret
int3
Where Zen1/2 flush the BTB entry using the instruction decoder trick
(test,movabs) Zen3/4 use BTB aliasing. SRSO adds a return sequence
(srso_safe_ret()) which forces the function return instruction to
speculate into a trap (UD2). This RET will then mispredict and
execution will continue at the return site read from the top of the
stack.
Pick one of three options at boot (evey function can only ever return
once).
[ bp: Fixup commit message uarch details and add them in a comment in
the code too. Add a comment about the srso_select_mitigation()
dependency on retbleed_select_mitigation(). Add moar ifdeffery for
32-bit builds. Add a dummy srso_untrain_ret_alias() definition for
32-bit alternatives needing the symbol. ]
Fixes: fb3bd914b3ec ("x86/srso: Add a Speculative RAS Overflow mitigation")
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Borislav Petkov (AMD) <bp@alien8.de>
Link: https://lore.kernel.org/r/20230814121148.842775684@infradead.org
|
|
There is infrastructure to rewrite return thunks to point to any
random thunk one desires, unwrap that from CALL_THUNKS, which up to
now was the sole user of that.
[ bp: Make the thunks visible on 32-bit and add ifdeffery for the
32-bit builds. ]
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Borislav Petkov (AMD) <bp@alien8.de>
Link: https://lore.kernel.org/r/20230814121148.775293785@infradead.org
|
|
vmlinux.o: warning: objtool: srso_untrain_ret() falls through to next function __x86_return_skl()
vmlinux.o: warning: objtool: __x86_return_thunk() falls through to next function __x86_return_skl()
This is because these functions (can) end with CALL, which objtool
does not consider a terminating instruction. Therefore, replace the
INT3 instruction (which is a non-fatal trap) with UD2 (which is a
fatal-trap).
This indicates execution will not continue past this point.
Fixes: fb3bd914b3ec ("x86/srso: Add a Speculative RAS Overflow mitigation")
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Borislav Petkov (AMD) <bp@alien8.de>
Link: https://lore.kernel.org/r/20230814121148.637802730@infradead.org
|
|
Commit
fb3bd914b3ec ("x86/srso: Add a Speculative RAS Overflow mitigation")
reimplemented __x86_return_thunk with a mix of SYM_FUNC_START and
SYM_CODE_END, this is not a sane combination.
Since nothing should ever actually 'CALL' this, make it consistently
CODE.
Fixes: fb3bd914b3ec ("x86/srso: Add a Speculative RAS Overflow mitigation")
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Borislav Petkov (AMD) <bp@alien8.de>
Link: https://lore.kernel.org/r/20230814121148.571027074@infradead.org
|
|
retpolines and IBT
The kprobes optimization check can_optimize() calls
insn_is_indirect_jump() to detect indirect jump instructions in
a target function. If any is found, creating an optprobe is disallowed
in the function because the jump could be from a jump table and could
potentially land in the middle of the target optprobe.
With retpolines, insn_is_indirect_jump() additionally looks for calls to
indirect thunks which the compiler potentially used to replace original
jumps. This extra check is however unnecessary because jump tables are
disabled when the kernel is built with retpolines. The same is currently
the case with IBT.
Based on this observation, remove the logic to look for calls to
indirect thunks and skip the check for indirect jumps altogether if the
kernel is built with retpolines or IBT. Remove subsequently the symbols
__indirect_thunk_start and __indirect_thunk_end which are no longer
needed.
Dropping this logic indirectly fixes a problem where the range
[__indirect_thunk_start, __indirect_thunk_end] wrongly included also the
return thunk. It caused that machines which used the return thunk as
a mitigation and didn't have it patched by any alternative ended up not
being able to use optprobes in any regular function.
Fixes: 0b53c374b9ef ("x86/retpoline: Use -mfunction-return")
Suggested-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Suggested-by: Masami Hiramatsu (Google) <mhiramat@kernel.org>
Signed-off-by: Petr Pavlu <petr.pavlu@suse.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Borislav Petkov (AMD) <bp@alien8.de>
Acked-by: Masami Hiramatsu (Google) <mhiramat@kernel.org>
Link: https://lore.kernel.org/r/20230711091952.27944-3-petr.pavlu@suse.com
|
|
The linker script arch/x86/kernel/vmlinux.lds.S matches the thunk
sections ".text.__x86.*" from arch/x86/lib/retpoline.S as follows:
.text {
[...]
TEXT_TEXT
[...]
__indirect_thunk_start = .;
*(.text.__x86.*)
__indirect_thunk_end = .;
[...]
}
Macro TEXT_TEXT references TEXT_MAIN which normally expands to only
".text". However, with CONFIG_LTO_CLANG, TEXT_MAIN becomes
".text .text.[0-9a-zA-Z_]*" which wrongly matches also the thunk
sections. The output layout is then different than expected. For
instance, the currently defined range [__indirect_thunk_start,
__indirect_thunk_end] becomes empty.
Prevent the problem by using ".." as the first separator, for example,
".text..__x86.indirect_thunk". This pattern is utilized by other
explicit section names which start with one of the standard prefixes,
such as ".text" or ".data", and that need to be individually selected in
the linker script.
[ nathan: Fix conflicts with SRSO and fold in fix issue brought up by
Andrew Cooper in post-review:
https://lore.kernel.org/20230803230323.1478869-1-andrew.cooper3@citrix.com ]
Fixes: dc5723b02e52 ("kbuild: add support for Clang LTO")
Signed-off-by: Petr Pavlu <petr.pavlu@suse.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Nathan Chancellor <nathan@kernel.org>
Signed-off-by: Borislav Petkov (AMD) <bp@alien8.de>
Link: https://lore.kernel.org/r/20230711091952.27944-2-petr.pavlu@suse.com
|
|
Skip the srso cmd line parsing which is not needed on Zen1/2 with SMT
disabled and with the proper microcode applied (latter should be the
case anyway) as those are not affected.
Fixes: 5a15d8348881 ("x86/srso: Tie SBPB bit setting to microcode patch detection")
Signed-off-by: Borislav Petkov (AMD) <bp@alien8.de>
Link: https://lore.kernel.org/r/20230813104517.3346-1-bp@alien8.de
|
Bruce Ashfield
Peter Zijlstra
Josh Poimboeuf
Steve Wahl
Song Liu
Kirill A. Shutemov
Ricardo Neri
Mateusz Guzik
Avadhut Naik
Sean Christopherson
Kan Liang
Jack Wang
Thomas Gleixner
Yazen Ghannam
Steve Rutherford
Dave Hansen
Randy Dunlap
Janusz Krzysztofik
Douglas Anderson
Ard Biesheuvel
Uros Bizjak
Helge Deller
Feng Tang
Rick Edgecombe
Borislav Petkov (AMD)
Petr Pavlu