| Age | Commit message (Collapse) | Author |
|---|
|
svm_recalc_instruction_intercepts() is always called at least once
before the vCPU is started, so the setting or clearing of the RDTSCP
intercept can be dropped from the TSC_AUX virtualization support.
Extracted from a patch by Tom Lendacky.
Cc: stable@vger.kernel.org
Fixes: 296d5a17e793 ("KVM: SEV-ES: Use V_TSC_AUX if available instead of RDTSC/MSR_TSC_AUX intercepts")
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
|
|
Stop zapping invalidate TDP MMU roots via work queue now that KVM
preserves TDP MMU roots until they are explicitly invalidated. Zapping
roots asynchronously was effectively a workaround to avoid stalling a vCPU
for an extended during if a vCPU unloaded a root, which at the time
happened whenever the guest toggled CR0.WP (a frequent operation for some
guest kernels).
While a clever hack, zapping roots via an unbound worker had subtle,
unintended consequences on host scheduling, especially when zapping
multiple roots, e.g. as part of a memslot. Because the work of zapping a
root is no longer bound to the task that initiated the zap, things like
the CPU affinity and priority of the original task get lost. Losing the
affinity and priority can be especially problematic if unbound workqueues
aren't affined to a small number of CPUs, as zapping multiple roots can
cause KVM to heavily utilize the majority of CPUs in the system, *beyond*
the CPUs KVM is already using to run vCPUs.
When deleting a memslot via KVM_SET_USER_MEMORY_REGION, the async root
zap can result in KVM occupying all logical CPUs for ~8ms, and result in
high priority tasks not being scheduled in in a timely manner. In v5.15,
which doesn't preserve unloaded roots, the issues were even more noticeable
as KVM would zap roots more frequently and could occupy all CPUs for 50ms+.
Consuming all CPUs for an extended duration can lead to significant jitter
throughout the system, e.g. on ChromeOS with virtio-gpu, deleting memslots
is a semi-frequent operation as memslots are deleted and recreated with
different host virtual addresses to react to host GPU drivers allocating
and freeing GPU blobs. On ChromeOS, the jitter manifests as audio blips
during games due to the audio server's tasks not getting scheduled in
promptly, despite the tasks having a high realtime priority.
Deleting memslots isn't exactly a fast path and should be avoided when
possible, and ChromeOS is working towards utilizing MAP_FIXED to avoid the
memslot shenanigans, but KVM is squarely in the wrong. Not to mention
that removing the async zapping eliminates a non-trivial amount of
complexity.
Note, one of the subtle behaviors hidden behind the async zapping is that
KVM would zap invalidated roots only once (ignoring partial zaps from
things like mmu_notifier events). Preserve this behavior by adding a flag
to identify roots that are scheduled to be zapped versus roots that have
already been zapped but not yet freed.
Add a comment calling out why kvm_tdp_mmu_invalidate_all_roots() can
encounter invalid roots, as it's not at all obvious why zapping
invalidated roots shouldn't simply zap all invalid roots.
Reported-by: Pattara Teerapong <pteerapong@google.com>
Cc: David Stevens <stevensd@google.com>
Cc: Yiwei Zhang<zzyiwei@google.com>
Cc: Paul Hsia <paulhsia@google.com>
Cc: stable@vger.kernel.org
Signed-off-by: Sean Christopherson <seanjc@google.com>
Message-Id: <20230916003916.2545000-4-seanjc@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
|
|
All callers except the MMU notifier want to process all address spaces.
Remove the address space ID argument of for_each_tdp_mmu_root_yield_safe()
and switch the MMU notifier to use __for_each_tdp_mmu_root_yield_safe().
Extracted out of a patch by Sean Christopherson <seanjc@google.com>
Cc: stable@vger.kernel.org
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull x86 rethunk fixes from Borislav Petkov:
"Fix the patching ordering between static calls and return thunks"
* tag 'x86_urgent_for_v6.6-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
x86,static_call: Fix static-call vs return-thunk
x86/alternatives: Remove faulty optimization
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull misc x86 fixes from Ingo Molnar:
- Fix a kexec bug
- Fix an UML build bug
- Fix a handful of SRSO related bugs
- Fix a shadow stacks handling bug & robustify related code
* tag 'x86-urgent-2023-09-22' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
x86/shstk: Add warning for shadow stack double unmap
x86/shstk: Remove useless clone error handling
x86/shstk: Handle vfork clone failure correctly
x86/srso: Fix SBPB enablement for spec_rstack_overflow=off
x86/srso: Don't probe microcode in a guest
x86/srso: Set CPUID feature bits independently of bug or mitigation status
x86/srso: Fix srso_show_state() side effect
x86/asm: Fix build of UML with KASAN
x86/mm, kexec, ima: Use memblock_free_late() from ima_free_kexec_buffer()
|
|
Add "#define pr_fmt()" in hv_init.c to use "Hyper-V:" as common
print prefix for all pr_*() statements in this file.
Remove the "Hyper-V:" already prefixed in couple of prints.
Signed-off-by: Saurabh Sengar <ssengar@linux.microsoft.com>
Signed-off-by: Wei Liu <wei.liu@kernel.org>
Link: https://lore.kernel.org/r/1695123361-8877-1-git-send-email-ssengar@linux.microsoft.com
|
|
There has been cases reported where HYPERV_VTL_MODE is enabled by mistake,
on a non Hyper-V platforms. This causes the hv_vtl_early_init function to
be called in an non Hyper-V/VTL platforms which results the memory
corruption.
Remove the early_initcall for hv_vtl_early_init and call it at the end of
hyperv_init to make sure it is never called in a non Hyper-V platform by
mistake.
Reported-by: Mathias Krause <minipli@grsecurity.net>
Closes: https://lore.kernel.org/lkml/40467722-f4ab-19a5-4989-308225b1f9f0@grsecurity.net/
Signed-off-by: Saurabh Sengar <ssengar@linux.microsoft.com>
Acked-by: Mathias Krause <minipli@grsecurity.net>
Signed-off-by: Wei Liu <wei.liu@kernel.org>
Link: https://lore.kernel.org/r/1695358720-27681-1-git-send-email-ssengar@linux.microsoft.com
|
|
When Linux runs in a non-default VTL (CONFIG_HYPERV_VTL_MODE=y),
get_vtl() must never fail as its return value is used in negotiations
with the host. In the more generic case, (CONFIG_HYPERV_VTL_MODE=n) the
VTL is always zero so there's no need to do the hypercall.
Make get_vtl() BUG() in case of failure and put the implementation under
"if IS_ENABLED(CONFIG_HYPERV_VTL_MODE)" to avoid the call altogether in
the most generic use case.
Signed-off-by: Saurabh Sengar <ssengar@linux.microsoft.com>
Reviewed-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: Wei Liu <wei.liu@kernel.org>
Link: https://lore.kernel.org/r/1695182675-13405-1-git-send-email-ssengar@linux.microsoft.com
|
|
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>
|
|
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
|
|
Inspired by Salvatore Mesoraca's earlier[1] efforts to provide some
in-tree guidance for kernel hardening Kconfig options, add a new fragment
named "hardening-basic.config" (along with some arch-specific fragments)
that enable a basic set of kernel hardening options that have the least
(or no) performance impact and remove a reasonable set of legacy APIs.
Using this fragment is as simple as running "make hardening.config".
More extreme fragments can be added[2] in the future to cover all the
recognized hardening options, and more per-architecture files can be
added too.
For now, document the fragments directly via comments. Perhaps .rst
documentation can be generated from them in the future (rather than the
other way around).
[1] https://lore.kernel.org/kernel-hardening/1536516257-30871-1-git-send-email-s.mesoraca16@gmail.com/
[2] https://github.com/KSPP/linux/issues/14
Cc: Salvatore Mesoraca <s.mesoraca16@gmail.com>
Cc: x86@kernel.org
Cc: linux-arm-kernel@lists.infradead.org
Cc: linux-doc@vger.kernel.org
Cc: linux-kbuild@vger.kernel.org
Signed-off-by: Kees Cook <keescook@chromium.org>
|
|
Kernels older than v5.19 do not support PerfMonV2 and the PMI handler
does not clear the overflow bits of the PerfCntrGlobalStatus register.
Because of this, loading a recent kernel using kexec from an older
kernel can result in inconsistent register states on Zen 4 systems.
The PMI handler of the new kernel gets confused and shows a warning when
an overflow occurs because some of the overflow bits are set even if the
corresponding counters are inactive. These are remnants from overflows
that were handled by the older kernel.
During CPU hotplug, the PerfCntrGlobalCtl and PerfCntrGlobalStatus
registers should always be cleared for PerfMonV2-capable processors.
However, a condition used for NB event constaints applicable only to
older processors currently prevents this from happening. Move the reset
sequence to an appropriate place and also clear the LBR Freeze bit.
Fixes: 21d59e3e2c40 ("perf/x86/amd/core: Detect PerfMonV2 support")
Signed-off-by: Sandipan Das <sandipan.das@amd.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Link: https://lore.kernel.org/r/882a87511af40792ba69bb0e9026f19a2e71e8a3.1694696888.git.sandipan.das@amd.com
|
|
.discard.retpoline_safe sections do not have the SHF_ALLOC flag. These
sections referencing text sections' STT_SECTION symbols with PC-relative
relocations like R_386_PC32 [0] is conceptually not suitable. Newer
LLD will report warnings for REL relocations even for relocatable links [1]:
ld.lld: warning: vmlinux.a(drivers/i2c/busses/i2c-i801.o):(.discard.retpoline_safe+0x120): has non-ABS relocation R_386_PC32 against symbol ''
Switch to absolute relocations instead, which indicate link-time
addresses. In a relocatable link, these addresses are also output
section offsets, used by checks in tools/objtool/check.c. When linking
vmlinux, these .discard.* sections will be discarded, therefore it is
not a problem that R_X86_64_32 cannot represent a kernel address.
Alternatively, we could set the SHF_ALLOC flag for .discard.* sections,
but I think non-SHF_ALLOC for sections to be discarded makes more sense.
Note: if we decide to never support REL architectures (e.g. arm, i386),
we can utilize R_*_NONE relocations (.reloc ., BFD_RELOC_NONE, sym),
making .discard.* sections zero-sized. That said, the section content
waste is 4 bytes per entry, much smaller than sizeof(Elf{32,64}_Rel).
[0] commit 1c0c1faf5692 ("objtool: Use relative pointers for annotations")
[1] https://github.com/ClangBuiltLinux/linux/issues/1937
Signed-off-by: Fangrui Song <maskray@google.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Josh Poimboeuf <jpoimboe@redhat.com>
Link: https://lore.kernel.org/r/20230920001728.1439947-1-maskray@google.com
|
|
When SVM is disabled by BIOS, one cannot use KVM but the
SVM feature is still shown in the output of /proc/cpuinfo.
On Intel machines, VMX is cleared by init_ia32_feat_ctl(),
so do the same on AMD and Hygon processors.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Borislav Petkov (AMD) <bp@alien8.de>
Link: https://lore.kernel.org/r/20230921114940.957141-1-pbonzini@redhat.com
|
|
Header cleanups in the fast-headers tree highlighted that we have an
unused assembly implementation for __sw_hweight64():
WARNING: modpost: EXPORT symbol "__sw_hweight64" [vmlinux] version ...
__arch_hweight64() on x86-32 is defined in the
arch/x86/include/asm/arch_hweight.h header as an inline, using
__arch_hweight32():
#ifdef CONFIG_X86_32
static inline unsigned long __arch_hweight64(__u64 w)
{
return __arch_hweight32((u32)w) +
__arch_hweight32((u32)(w >> 32));
}
*But* there's also a __sw_hweight64() assembly implementation:
arch/x86/lib/hweight.S
SYM_FUNC_START(__sw_hweight64)
#ifdef CONFIG_X86_64
...
#else /* CONFIG_X86_32 */
/* We're getting an u64 arg in (%eax,%edx): unsigned long hweight64(__u64 w) */
pushl %ecx
call __sw_hweight32
movl %eax, %ecx # stash away result
movl %edx, %eax # second part of input
call __sw_hweight32
addl %ecx, %eax # result
popl %ecx
ret
#endif
But this __sw_hweight64 assembly implementation is unused - and it's
essentially doing the same thing that the inline wrapper does.
Remove the assembly version and add a comment about it.
Reported-by: Nathan Chancellor <nathan@kernel.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: linux-kernel@vger.kernel.org
|
|
from <asm/processor.h> to <asm/pgtable.h>
<linux/mm.h> relies on these definitions being included first,
which is true currently due to historic header spaghetti,
but in the future <asm/processor.h> will not guaranteed to be
included by the MM code.
Move these definitions over into a suitable MM header.
This is a preparatory patch for x86 header dependency simplifications
and reductions.
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Cc: linux-kernel@vger.kernel.org
|
|
Cross-merge networking fixes after downstream PR.
No conflicts.
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
|
|
Finish off the 'simple' futex2 syscall group by adding
sys_futex_requeue(). Unlike sys_futex_{wait,wake}() its arguments are
too numerous to fit into a regular syscall. As such, use struct
futex_waitv to pass the 'source' and 'destination' futexes to the
syscall.
This syscall implements what was previously known as FUTEX_CMP_REQUEUE
and uses {val, uaddr, flags} for source and {uaddr, flags} for
destination.
This design explicitly allows requeueing between different types of
futex by having a different flags word per uaddr.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Thomas Gleixner <tglx@linutronix.de>
Acked-by: Geert Uytterhoeven <geert@linux-m68k.org>
Link: https://lore.kernel.org/r/20230921105248.511860556@noisy.programming.kicks-ass.net
|
|
To complement sys_futex_waitv()/wake(), add sys_futex_wait(). This
syscall implements what was previously known as FUTEX_WAIT_BITSET
except it uses 'unsigned long' for the value and bitmask arguments,
takes timespec and clockid_t arguments for the absolute timeout and
uses FUTEX2 flags.
The 'unsigned long' allows FUTEX2_SIZE_U64 on 64bit platforms.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Thomas Gleixner <tglx@linutronix.de>
Acked-by: Geert Uytterhoeven <geert@linux-m68k.org>
Link: https://lore.kernel.org/r/20230921105248.164324363@noisy.programming.kicks-ass.net
|
|
To complement sys_futex_waitv() add sys_futex_wake(). This syscall
implements what was previously known as FUTEX_WAKE_BITSET except it
uses 'unsigned long' for the bitmask and takes FUTEX2 flags.
The 'unsigned long' allows FUTEX2_SIZE_U64 on 64bit platforms.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Thomas Gleixner <tglx@linutronix.de>
Acked-by: Geert Uytterhoeven <geert@linux-m68k.org>
Link: https://lore.kernel.org/r/20230921105247.936205525@noisy.programming.kicks-ass.net
|
|
The mmu_notifier path is a bit of a special snowflake, e.g. it zaps only a
single address space (because it's per-slot), and can't always yield.
Because of this, it calls kvm_tdp_mmu_zap_leafs() in ways that no one
else does.
Iterate manually over the leafs in response to an mmu_notifier
invalidation, instead of invoking kvm_tdp_mmu_zap_leafs(). Drop the
@can_yield param from kvm_tdp_mmu_zap_leafs() as its sole remaining
caller unconditionally passes "true".
Cc: stable@vger.kernel.org
Signed-off-by: Sean Christopherson <seanjc@google.com>
Message-Id: <20230916003916.2545000-2-seanjc@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
|
|
Signed-off-by: Yang Li <yang.lee@linux.alibaba.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Link: https://lore.kernel.org/r/20230816003842.116574-1-yang.lee@linux.alibaba.com
|
|
The fallback alternative uses %rsi register to manually load pointer
to the percpu variable before the call to the emulation function.
This is unoptimal, because the load is hidden from the compiler.
Move the load of %rsi outside inline asm, so the compiler can
reuse the value. The code in slub.o improves from:
55ac: 49 8b 3c 24 mov (%r12),%rdi
55b0: 48 8d 4a 40 lea 0x40(%rdx),%rcx
55b4: 49 8b 1c 07 mov (%r15,%rax,1),%rbx
55b8: 4c 89 f8 mov %r15,%rax
55bb: 48 8d 37 lea (%rdi),%rsi
55be: e8 00 00 00 00 callq 55c3 <...>
55bf: R_X86_64_PLT32 this_cpu_cmpxchg16b_emu-0x4
55c3: 75 a3 jne 5568 <...>
55c5: ...
0000000000000000 <.altinstr_replacement>:
5: 65 48 0f c7 0f cmpxchg16b %gs:(%rdi)
to:
55ac: 49 8b 34 24 mov (%r12),%rsi
55b0: 48 8d 4a 40 lea 0x40(%rdx),%rcx
55b4: 49 8b 1c 07 mov (%r15,%rax,1),%rbx
55b8: 4c 89 f8 mov %r15,%rax
55bb: e8 00 00 00 00 callq 55c0 <...>
55bc: R_X86_64_PLT32 this_cpu_cmpxchg16b_emu-0x4
55c0: 75 a6 jne 5568 <...>
55c2: ...
Where the alternative replacement instruction now uses %rsi:
0000000000000000 <.altinstr_replacement>:
5: 65 48 0f c7 0e cmpxchg16b %gs:(%rsi)
The instruction (effectively a reg-reg move) at 55bb: in the original
assembly is removed. Also, both the CALL and replacement CMPXCHG16B
are 5 bytes long, removing the need for NOPs in the asm code.
Suggested-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Uros Bizjak <ubizjak@gmail.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Link: https://lore.kernel.org/r/20230918151452.62344-1-ubizjak@gmail.com
|
|
The 'mid' pointer is being initialized with a value that is never read,
it is being re-assigned and used inside a for-loop. Remove the
redundant initialization.
Cleans up clang scan build warning:
arch/x86/kernel/unwind_orc.c:88:7: warning: Value stored to 'mid' during its initialization is never read [deadcode.DeadStores]
Signed-off-by: Colin Ian King <colin.i.king@gmail.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Cc: Josh Poimboeuf <jpoimboe@redhat.com>
Link: https://lore.kernel.org/r/20230920114141.118919-1-colin.i.king@gmail.com
|
|
x86 optimized crypto modules are built as modules rather than build-in and
they are not loaded when the crypto API is initialized, resulting in the
generic builtin module (sha1-generic) being used instead.
It was discovered when creating a sha1/sha256 checksum of a 2Gb file by
using kcapi-tools because it would take significantly longer than creating
a sha512 checksum of the same file. trace-cmd showed that for sha1/256 the
generic module was used, whereas for sha512 the optimized module was used
instead.
Add module aliases() for these x86 optimized crypto modules based on CPU
feature bits so udev gets a chance to load them later in the boot
process. This resulted in ~3x decrease in the real-time execution of
kcapi-dsg.
Fix is inspired from commit
aa031b8f702e ("crypto: x86/sha512 - load based on CPU features")
where a similar fix was done for sha512.
Cc: stable@vger.kernel.org # 5.15+
Suggested-by: Dimitri John Ledkov <dimitri.ledkov@canonical.com>
Suggested-by: Julian Andres Klode <julian.klode@canonical.com>
Signed-off-by: Roxana Nicolescu <roxana.nicolescu@canonical.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
|
|
Remove the repeated word "if" in comments.
Signed-off-by: Bo Liu <liubo03@inspur.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
|
|
There are several ways a thread's shadow stacks can get unmapped. This
can happen on exit or exec, as well as error handling in exec or clone.
The task struct already keeps track of the thread's shadow stack. Use the
size variable to keep track of if the shadow stack has already been freed.
When an attempt to double unmap the thread shadow stack is caught, warn
about it and abort the operation.
Signed-off-by: Rick Edgecombe <rick.p.edgecombe@intel.com>
Signed-off-by: Dave Hansen <dave.hansen@linux.intel.com>
Tested-by: H.J. Lu <hjl.tools@gmail.com>
Link: https://lore.kernel.org/all/20230908203655.543765-4-rick.p.edgecombe%40intel.com
|
|
When clone fails after the shadow stack is allocated, any allocated shadow
stack is cleaned up in exit_thread() in copy_process(). So the logic in
copy_thread() is unneeded, and also will not handle failures that happen
outside of copy_thread().
In addition, since there is a second attempt to unmap the same shadow
stack, there is a race where an newly mapped region could get unmapped.
So remove the logic in copy_thread() and rely on exit_thread() to handle
clone failure.
Fixes: b2926a36b97a ("x86/shstk: Handle thread shadow stack")
Signed-off-by: Rick Edgecombe <rick.p.edgecombe@intel.com>
Signed-off-by: Dave Hansen <dave.hansen@linux.intel.com>
Tested-by: H.J. Lu <hjl.tools@gmail.com>
Link: https://lore.kernel.org/all/20230908203655.543765-3-rick.p.edgecombe%40intel.com
|
|
Shadow stacks are allocated automatically and freed on exit, depending
on the clone flags. The two cases where new shadow stacks are not
allocated are !CLONE_VM (fork()) and CLONE_VFORK (vfork()). For
!CLONE_VM, although a new stack is not allocated, it can be freed normally
because it will happen in the child's copy of the VM.
However, for CLONE_VFORK the parent and the child are actually using the
same shadow stack. So the kernel doesn't need to allocate *or* free a
shadow stack for a CLONE_VFORK child. CLONE_VFORK children already need
special tracking to avoid returning to userspace until the child exits or
execs. Shadow stack uses this same tracking to avoid freeing CLONE_VFORK
shadow stacks.
However, the tracking is not setup until the clone has succeeded
(internally). Which means, if a CLONE_VFORK fails, the existing logic will
not know it is a CLONE_VFORK and proceed to unmap the parents shadow stack.
This error handling cleanup logic runs via exit_thread() in the
bad_fork_cleanup_thread label in copy_process(). The issue was seen in
the glibc test "posix/tst-spawn3-pidfd" while running with shadow stack
using currently out-of-tree glibc patches.
Fix it by not unmapping the vfork shadow stack in the error case as well.
Since clone is implemented in core code, it is not ideal to pass the clone
flags along the error path in order to have shadow stack code have
symmetric logic in the freeing half of the thread shadow stack handling.
Instead use the existing state for thread shadow stacks to track whether
the thread is managing its own shadow stack. For CLONE_VFORK, simply set
shstk->base and shstk->size to 0, and have it mean the thread is not
managing a shadow stack and so should skip cleanup work. Implement this
by breaking up the CLONE_VFORK and !CLONE_VM cases in
shstk_alloc_thread_stack() to separate conditionals since, the logic is
now different between them. In the case of CLONE_VFORK && !CLONE_VM, the
existing behavior is to not clean up the shadow stack in the child (which
should go away quickly with either be exit or exec), so maintain that
behavior by handling the CLONE_VFORK case first in the allocation path.
This new logioc cleanly handles the case of normal, successful
CLONE_VFORK's skipping cleaning up their shadow stack's on exit as well.
So remove the existing, vfork shadow stack freeing logic. This is in
deactivate_mm() where vfork_done is used to tell if it is a vfork child
that can skip cleaning up the thread shadow stack.
Fixes: b2926a36b97a ("x86/shstk: Handle thread shadow stack")
Reported-by: H.J. Lu <hjl.tools@gmail.com>
Signed-off-by: Rick Edgecombe <rick.p.edgecombe@intel.com>
Signed-off-by: Dave Hansen <dave.hansen@linux.intel.com>
Tested-by: H.J. Lu <hjl.tools@gmail.com>
Link: https://lore.kernel.org/all/20230908203655.543765-2-rick.p.edgecombe%40intel.com
|
|
The build with CONFIG_UNWINDER_FRAME_POINTER=y is broken for
current exceptions feature as it assumes ORC unwinder specific fields in
the unwind_state. Disable exceptions when frame_pointer unwinder is
enabled for now.
Fixes: fd5d27b70188 ("arch/x86: Implement arch_bpf_stack_walk")
Reported-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Kumar Kartikeya Dwivedi <memxor@gmail.com>
Link: https://lore.kernel.org/r/20230918155233.297024-4-memxor@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
|
|
If the user has requested no SRSO mitigation, other mitigations can use
the lighter-weight SBPB instead of IBPB.
Fixes: fb3bd914b3ec ("x86/srso: Add a Speculative RAS Overflow mitigation")
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/b20820c3cfd1003171135ec8d762a0b957348497.1693889988.git.jpoimboe@kernel.org
|
|
To support live migration, the hypervisor sets the "lowest common
denominator" of features. Probing the microcode isn't allowed because
any detected features might go away after a migration.
As Andy Cooper states:
"Linux must not probe microcode when virtualised. What it may see
instantaneously on boot (owing to MSR_PRED_CMD being fully passed
through) is not accurate for the lifetime of the VM."
Rely on the hypervisor to set the needed IBPB_BRTYPE and SBPB bits.
Fixes: 1b5277c0ea0b ("x86/srso: Add SRSO_NO support")
Suggested-by: Andrew Cooper <andrew.cooper3@citrix.com>
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>
Reviewed-by: Andrew Cooper <andrew.cooper3@citrix.com>
Acked-by: Borislav Petkov (AMD) <bp@alien8.de>
Link: https://lore.kernel.org/r/3938a7209606c045a3f50305d201d840e8c834c7.1693889988.git.jpoimboe@kernel.org
|
|
Booting with mitigations=off incorrectly prevents the
X86_FEATURE_{IBPB_BRTYPE,SBPB} CPUID bits from getting set.
Also, future CPUs without X86_BUG_SRSO might still have IBPB with branch
type prediction flushing, in which case SBPB should be used instead of
IBPB. The current code doesn't allow for that.
Also, cpu_has_ibpb_brtype_microcode() has some surprising side effects
and the setting of these feature bits really doesn't belong in the
mitigation code anyway. Move it to earlier.
Fixes: fb3bd914b3ec ("x86/srso: Add a Speculative RAS Overflow mitigation")
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>
Reviewed-by: Nikolay Borisov <nik.borisov@suse.com>
Reviewed-by: Borislav Petkov (AMD) <bp@alien8.de>
Acked-by: Borislav Petkov (AMD) <bp@alien8.de>
Link: https://lore.kernel.org/r/869a1709abfe13b673bdd10c2f4332ca253a40bc.1693889988.git.jpoimboe@kernel.org
|
|
Reading the 'spec_rstack_overflow' sysfs file can trigger an unnecessary
MSR write, and possibly even a (handled) exception if the microcode
hasn't been updated.
Avoid all that by just checking X86_FEATURE_IBPB_BRTYPE instead, which
gets set by srso_select_mitigation() if the updated microcode exists.
Fixes: fb3bd914b3ec ("x86/srso: Add a Speculative RAS Overflow mitigation")
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>
Reviewed-by: Nikolay Borisov <nik.borisov@suse.com>
Acked-by: Borislav Petkov (AMD) <bp@alien8.de>
Link: https://lore.kernel.org/r/27d128899cb8aee9eb2b57ddc996742b0c1d776b.1693889988.git.jpoimboe@kernel.org
|
|
`strncpy` is deprecated for use on NUL-terminated destination strings [1].
`efi_loader_signature` has space for 4 bytes. We are copying "Xen" (3 bytes)
plus a NUL-byte which makes 4 total bytes. With that being said, there is
currently not a bug with the current `strncpy()` implementation in terms of
buffer overreads but we should favor a more robust string interface
either way.
A suitable replacement is `strscpy` [2] due to the fact that it guarantees
NUL-termination on the destination buffer while being functionally the
same in this case.
Link: www.kernel.org/doc/html/latest/process/deprecated.html#strncpy-on-nul-terminated-strings[1]
Link: https://manpages.debian.org/testing/linux-manual-4.8/strscpy.9.en.html [2]
Link: https://github.com/KSPP/linux/issues/90
Cc: linux-hardening@vger.kernel.org
Cc: Kees Cook <keescook@chromium.org>
Signed-off-by: Justin Stitt <justinstitt@google.com>
Reviewed-by: Kees Cook <keescook@chromium.org>
Link: https://lore.kernel.org/r/20230911-strncpy-arch-x86-xen-efi-c-v1-1-96ab2bba2feb@google.com
Signed-off-by: Juergen Gross <jgross@suse.com>
|
|
When running as a paravirtualized guest under Xen, Linux is using
"lazy mode" for issuing hypercalls which don't need to take immediate
effect in order to improve performance (examples are e.g. multiple
PTE changes).
There are two different lazy modes defined: MMU and CPU lazy mode.
Today it is not possible to nest multiple lazy mode sections, even if
they are of the same kind. A recent change in memory management added
nesting of MMU lazy mode sections, resulting in a regression when
running as Xen PV guest.
Technically there is no reason why nesting of multiple sections of the
same kind of lazy mode shouldn't be allowed. So add support for that
for fixing the regression.
Fixes: bcc6cc832573 ("mm: add default definition of set_ptes()")
Signed-off-by: Juergen Gross <jgross@suse.com>
Reviewed-by: Boris Ostrovsky <boris.ostrovsky@oracle.com>
Link: https://lore.kernel.org/r/20230913113828.18421-4-jgross@suse.com
Signed-off-by: Juergen Gross <jgross@suse.com>
|
|
Only Xen is using the paravirt lazy mode code, so it can be moved to
Xen specific sources.
This allows to make some of the functions static or to merge them into
their only call sites.
While at it do a rename from "paravirt" to "xen" for all moved
specifiers.
No functional change.
Signed-off-by: Juergen Gross <jgross@suse.com>
Reviewed-by: Boris Ostrovsky <boris.ostrovsky@oracle.com>
Link: https://lore.kernel.org/r/20230913113828.18421-3-jgross@suse.com
Signed-off-by: Juergen Gross <jgross@suse.com>
|
|
There are several functions involved for performing the functionality
of evtchn_do_upcall():
- __xen_evtchn_do_upcall() doing the real work
- xen_hvm_evtchn_do_upcall() just being a wrapper for
__xen_evtchn_do_upcall(), exposed for external callers
- xen_evtchn_do_upcall() calling __xen_evtchn_do_upcall(), too, but
without any user
Simplify this maze by:
- removing the unused xen_evtchn_do_upcall()
- removing xen_hvm_evtchn_do_upcall() as the only left caller of
__xen_evtchn_do_upcall(), while renaming __xen_evtchn_do_upcall() to
xen_evtchn_do_upcall()
Signed-off-by: Juergen Gross <jgross@suse.com>
Reviewed-by: Boris Ostrovsky <boris.ostrovsky@oracle.com>
Reviewed-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Juergen Gross <jgross@suse.com>
|
|
The following commit:
bc08b449ee14 ("lockref: implement lockless reference count updates using cmpxchg()")
enabled lockless reference count updates using cmpxchg() only for x86_64,
and left x86_32 behind due to inability to detect support for
cmpxchg8b instruction.
Nowadays, we can use CONFIG_X86_CMPXCHG64 for this purpose. Also,
by using try_cmpxchg64() instead of cmpxchg64() in the CMPXCHG_LOOP macro,
the compiler actually produces sane code, improving the
lockref_get_not_zero() main loop from:
eb: 8d 48 01 lea 0x1(%eax),%ecx
ee: 85 c0 test %eax,%eax
f0: 7e 2f jle 121 <lockref_get_not_zero+0x71>
f2: 8b 44 24 10 mov 0x10(%esp),%eax
f6: 8b 54 24 14 mov 0x14(%esp),%edx
fa: 8b 74 24 08 mov 0x8(%esp),%esi
fe: f0 0f c7 0e lock cmpxchg8b (%esi)
102: 8b 7c 24 14 mov 0x14(%esp),%edi
106: 89 c1 mov %eax,%ecx
108: 89 c3 mov %eax,%ebx
10a: 8b 74 24 10 mov 0x10(%esp),%esi
10e: 89 d0 mov %edx,%eax
110: 31 fa xor %edi,%edx
112: 31 ce xor %ecx,%esi
114: 09 f2 or %esi,%edx
116: 75 58 jne 170 <lockref_get_not_zero+0xc0>
to:
350: 8d 4f 01 lea 0x1(%edi),%ecx
353: 85 ff test %edi,%edi
355: 7e 79 jle 3d0 <lockref_get_not_zero+0xb0>
357: f0 0f c7 0e lock cmpxchg8b (%esi)
35b: 75 53 jne 3b0 <lockref_get_not_zero+0x90>
Signed-off-by: Uros Bizjak <ubizjak@gmail.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Linus Torvalds <torvalds@linux-foundation.org>
Link: https://lore.kernel.org/r/20230918184050.9180-1-ubizjak@gmail.com
|
|
Building UML with KASAN fails since commit 69d4c0d32186 ("entry, kasan,
x86: Disallow overriding mem*() functions") with the following errors:
$ tools/testing/kunit/kunit.py run --kconfig_add CONFIG_KASAN=y
...
ld: mm/kasan/shadow.o: in function `memset':
shadow.c:(.text+0x40): multiple definition of `memset';
arch/x86/lib/memset_64.o:(.noinstr.text+0x0): first defined here
ld: mm/kasan/shadow.o: in function `memmove':
shadow.c:(.text+0x90): multiple definition of `memmove';
arch/x86/lib/memmove_64.o:(.noinstr.text+0x0): first defined here
ld: mm/kasan/shadow.o: in function `memcpy':
shadow.c:(.text+0x110): multiple definition of `memcpy';
arch/x86/lib/memcpy_64.o:(.noinstr.text+0x0): first defined here
UML does not use GENERIC_ENTRY and is still supposed to be allowed to
override the mem*() functions, so use weak aliases in that case.
Fixes: 69d4c0d32186 ("entry, kasan, x86: Disallow overriding mem*() functions")
Signed-off-by: Vincent Whitchurch <vincent.whitchurch@axis.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Link: https://lore.kernel.org/r/20230918-uml-kasan-v3-1-7ad6db477df6@axis.com
|
|
The code calling ima_free_kexec_buffer() runs long after the memblock
allocator has already been torn down, potentially resulting in a use
after free in memblock_isolate_range().
With KASAN or KFENCE, this use after free will result in a BUG
from the idle task, and a subsequent kernel panic.
Switch ima_free_kexec_buffer() over to memblock_free_late() to avoid
that bug.
Fixes: fee3ff99bc67 ("powerpc: Move arch independent ima kexec functions to drivers/of/kexec.c")
Suggested-by: Mike Rappoport <rppt@kernel.org>
Signed-off-by: Rik van Riel <riel@surriel.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Link: https://lore.kernel.org/r/20230817135558.67274c83@imladris.surriel.com
|
|
LKP reported below build warning:
vmlinux.o: warning: objtool: __tdx_hypercall+0x128: __tdx_hypercall_failed() is missing a __noreturn annotation
The __tdx_hypercall_failed() function definition already has __noreturn
annotation, but it turns out the __noreturn must be annotated to the
function declaration.
PeterZ explains:
"FWIW, the reason being that...
The point of noreturn is that the caller should know to stop generating
code. For that the declaration needs the attribute, because call sites
typically do not have access to the function definition in C."
Add __noreturn annotation to the declaration of __tdx_hypercall_failed()
to fix. It's not a bad idea to document the __noreturn nature at the
definition site either, so keep the annotation at the definition.
Note <asm/shared/tdx.h> is also included by TDX related assembly files.
Include <linux/compiler_attributes.h> only in case of !__ASSEMBLY__
otherwise compiling assembly file would trigger build error.
Also, following the objtool documentation, add __tdx_hypercall_failed()
to "tools/objtool/noreturns.h".
Fixes: c641cfb5c157 ("x86/tdx: Make TDX_HYPERCALL asm similar to TDX_MODULE_CALL")
Reported-by: kernel test robot <lkp@intel.com>
Signed-off-by: Kai Huang <kai.huang@intel.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Link: https://lore.kernel.org/r/20230918041858.331234-1-kai.huang@intel.com
Closes: https://lore.kernel.org/oe-kbuild-all/202309140828.9RdmlH2Z-lkp@intel.com/
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull x86 fixes from Ingo Molnar:
"Misc fixes:
- Fix an UV boot crash
- Skip spurious ENDBR generation on _THIS_IP_
- Fix ENDBR use in putuser() asm methods
- Fix corner case boot crashes on 5-level paging
- and fix a false positive WARNING on LTO kernels"
* tag 'x86-urgent-2023-09-17' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
x86/purgatory: Remove LTO flags
x86/boot/compressed: Reserve more memory for page tables
x86/ibt: Avoid duplicate ENDBR in __put_user_nocheck*()
x86/ibt: Suppress spurious ENDBR
x86/platform/uv: Use alternate source for socket to node data
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull scheduler fixes from Ingo Molnar:
"Fix a performance regression on large SMT systems, an Intel SMT4
balancing bug, and a topology setup bug on (Intel) hybrid processors"
* tag 'sched-urgent-2023-09-17' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
x86/sched: Restore the SD_ASYM_PACKING flag in the DIE domain
sched/fair: Fix SMT4 group_smt_balance handling
sched/fair: Optimize should_we_balance() for large SMT systems
|
|
Align x86 with other EFI architectures, and increase the section
alignment to the EFI page size (4k), so that firmware is able to honour
the section permission attributes and map code read-only and data
non-executable.
There are a number of requirements that have to be taken into account:
- the sign tools get cranky when there are gaps between sections in the
file view of the image
- the virtual offset of each section must be aligned to the image's
section alignment
- the file offset *and size* of each section must be aligned to the
image's file alignment
- the image size must be aligned to the section alignment
- each section's virtual offset must be greater than or equal to the
size of the headers.
In order to meet all these requirements, while avoiding the need for
lots of padding to accommodate the .compat section, the latter is placed
at an arbitrary offset towards the end of the image, but aligned to the
minimum file alignment (512 bytes). The space before the .text section
is therefore distributed between the PE header, the .setup section and
the .compat section, leaving no gaps in the file coverage, making the
signing tools happy.
Signed-off-by: Ard Biesheuvel <ardb@kernel.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Link: https://lore.kernel.org/r/20230915171623.655440-18-ardb@google.com
|
|
Describe the code and data of the decompressor binary using separate
.text and .data PE/COFF sections, so that we will be able to map them
using restricted permissions once we increase the section and file
alignment sufficiently. This avoids the need for memory mappings that
are writable and executable at the same time, which is something that
is best avoided for security reasons.
Signed-off-by: Ard Biesheuvel <ardb@kernel.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Link: https://lore.kernel.org/r/20230915171623.655440-17-ardb@google.com
|
|
Ancient buggy EFI loaders may have required a .reloc section to be
present at some point in time, but this has not been true for a long
time so the .reloc section can just be dropped.
Signed-off-by: Ard Biesheuvel <ardb@kernel.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Link: https://lore.kernel.org/r/20230915171623.655440-16-ardb@google.com
|
|
Now that the size of the setup block is visible to the assembler, it is
possible to populate the PE/COFF header fields from the asm code
directly, instead of poking the values into the binary using the build
tool. This will make it easier to reorganize the section layout without
having to tweak the build tool in lockstep.
This change has no impact on the resulting bzImage binary.
Signed-off-by: Ard Biesheuvel <ardb@kernel.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Link: https://lore.kernel.org/r/20230915171623.655440-15-ardb@google.com
|
|
Tweak the linker script so that the value of _edata represents the
decompressor binary's file size rounded up to the appropriate alignment.
This removes the need to calculate it in the build tool, and will make
it easier to refer to the file size from the header directly in
subsequent changes to the PE header layout.
While adding _edata to the sed regex that parses the compressed
vmlinux's symbol list, tweak the regex a bit for conciseness.
This change has no impact on the resulting bzImage binary when
configured with CONFIG_EFI_STUB=y.
Signed-off-by: Ard Biesheuvel <ardb@kernel.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Link: https://lore.kernel.org/r/20230915171623.655440-14-ardb@google.com
|
|
The setup block contains the real mode startup code that is used when
booting from a legacy BIOS, along with the boot_params/setup_data that
is used by legacy x86 bootloaders to pass the command line and initial
ramdisk parameters, among other things.
The setup block also contains the PE/COFF header of the entire combined
image, which includes the compressed kernel image, the decompressor and
the EFI stub.
This PE header describes the layout of the executable image in memory,
and currently, the fact that the setup block precedes it makes it rather
fiddly to get the right values into the right place in the final image.
Let's make things a bit easier by defining the setup_size in the linker
script so it can be referenced from the asm code directly, rather than
having to rely on the build tool to calculate it. For the time being,
add 64 bytes of fixed padding for the .reloc and .compat sections - this
will be removed in a subsequent patch after the PE/COFF header has been
reorganized.
This change has no impact on the resulting bzImage binary when
configured with CONFIG_EFI_MIXED=y.
Signed-off-by: Ard Biesheuvel <ardb@kernel.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Link: https://lore.kernel.org/r/20230915171623.655440-13-ardb@google.com
|
