Pull x86 TDX updates from Dave Hansen:
"The biggest change here is making TDX and kexec play nicely together.
Before this, the memory encryption hardware (which doesn't respect
cache coherency) could write back old cachelines on top of data in the
new kernel, so kexec and TDX were made mutually exclusive. This
removes the limitation.
There is also some work to tighten up a hardware bug workaround and
some MAINTAINERS updates.
- Make TDX and kexec work together
- Skip TDX bug workaround when the bug is not present
- Update maintainers entries"
* tag 'x86_tdx_for_6.18-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
x86/virt/tdx: Use precalculated TDVPR page physical address
KVM/TDX: Explicitly do WBINVD when no more TDX SEAMCALLs
x86/virt/tdx: Update the kexec section in the TDX documentation
x86/virt/tdx: Remove the !KEXEC_CORE dependency
x86/kexec: Disable kexec/kdump on platforms with TDX partial write erratum
x86/virt/tdx: Mark memory cache state incoherent when making SEAMCALL
x86/sme: Use percpu boolean to control WBINVD during kexec
x86/kexec: Consolidate relocate_kernel() function parameters
x86/tdx: Skip clearing reclaimed pages unless X86_BUG_TDX_PW_MCE is present
x86/tdx: Tidy reset_pamt functions
x86/tdx: Eliminate duplicate code in tdx_clear_page()
MAINTAINERS: Add KVM mail list to the TDX entry
MAINTAINERS: Add Rick Edgecombe as a TDX reviewer
MAINTAINERS: Update the file list in the TDX entry.
TL;DR:
Prepare to unify how TDX and SME do cache flushing during kexec by
making a percpu boolean control whether to do the WBINVD.
-- Background --
On SME platforms, dirty cacheline aliases with and without encryption
bit can coexist, and the CPU can flush them back to memory in random
order. During kexec, the caches must be flushed before jumping to the
new kernel otherwise the dirty cachelines could silently corrupt the
memory used by the new kernel due to different encryption property.
TDX also needs a cache flush during kexec for the same reason. It would
be good to have a generic way to flush the cache instead of scattering
checks for each feature all around.
When SME is enabled, the kernel basically encrypts all memory including
the kernel itself and a simple memory write from the kernel could dirty
cachelines. Currently, the kernel uses WBINVD to flush the cache for
SME during kexec in two places:
1) the one in stop_this_cpu() for all remote CPUs when the kexec-ing CPU
stops them;
2) the one in the relocate_kernel() where the kexec-ing CPU jumps to the
new kernel.
-- Solution --
Unlike SME, TDX can only dirty cachelines when it is used (i.e., when
SEAMCALLs are performed). Since there are no more SEAMCALLs after the
aforementioned WBINVDs, leverage this for TDX.
To unify the approach for SME and TDX, use a percpu boolean to indicate
the cache may be in an incoherent state and needs flushing during kexec,
and set the boolean for SME. TDX can then leverage it.
While SME could use a global flag (since it's enabled at early boot and
enabled on all CPUs), the percpu flag fits TDX better:
The percpu flag can be set when a CPU makes a SEAMCALL, and cleared when
another WBINVD on the CPU obviates the need for a kexec-time WBINVD.
Saving kexec-time WBINVD is valuable, because there is an existing
race[*] where kexec could proceed while another CPU is active. WBINVD
could make this race worse, so it's worth skipping it when possible.
-- Side effect to SME --
Today the first WBINVD in the stop_this_cpu() is performed when SME is
*supported* by the platform, and the second WBINVD is done in
relocate_kernel() when SME is *activated* by the kernel. Make things
simple by changing to do the second WBINVD when the platform supports
SME. This allows the kernel to simply turn on this percpu boolean when
bringing up a CPU by checking whether the platform supports SME.
No other functional change intended.
[*] The aforementioned race:
During kexec native_stop_other_cpus() is called to stop all remote CPUs
before jumping to the new kernel. native_stop_other_cpus() firstly
sends normal REBOOT vector IPIs to stop remote CPUs and waits them to
stop. If that times out, it sends NMI to stop the CPUs that are still
alive. The race happens when native_stop_other_cpus() has to send NMIs
and could potentially result in the system hang (for more information
please see [1]).
Signed-off-by: Kai Huang <kai.huang@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Dave Hansen <dave.hansen@linux.intel.com>
Reviewed-by: Tom Lendacky <thomas.lendacky@amd.com>
Reviewed-by: Borislav Petkov (AMD) <bp@alien8.de>
Tested-by: Tom Lendacky <thomas.lendacky@amd.com>
Link: https://lore.kernel.org/kvm/b963fcd60abe26c7ec5dc20b42f1a2ebbcc72397.1750934177.git.kai.huang@intel.com/ [1]
Link: https://lore.kernel.org/all/20250901160930.1785244-3-pbonzini%40redhat.com
With the introduction of clone3 in commit 7f192e3cd3 ("fork: add
clone3") the effective bit width of clone_flags on all architectures was
increased from 32-bit to 64-bit, with a new type of u64 for the flags.
However, for most consumers of clone_flags the interface was not
changed from the previous type of unsigned long.
While this works fine as long as none of the new 64-bit flag bits
(CLONE_CLEAR_SIGHAND and CLONE_INTO_CGROUP) are evaluated, this is still
undesirable in terms of the principle of least surprise.
Thus, this commit fixes all relevant interfaces of the copy_thread
function that is called from copy_process to consistently pass
clone_flags as u64, so that no truncation to 32-bit integers occurs on
32-bit architectures.
Signed-off-by: Simon Schuster <schuster.simon@siemens-energy.com>
Link: https://lore.kernel.org/20250901-nios2-implement-clone3-v2-3-53fcf5577d57@siemens-energy.com
Fixes: c5febea095 ("fork: Pass struct kernel_clone_args into copy_thread")
Acked-by: Guo Ren (Alibaba Damo Academy) <guoren@kernel.org>
Acked-by: Andreas Larsson <andreas@gaisler.com> # sparc
Acked-by: David Hildenbrand <david@redhat.com>
Acked-by: Geert Uytterhoeven <geert@linux-m68k.org> # m68k
Reviewed-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Christian Brauner <brauner@kernel.org>
Move the VERW clearing before the MONITOR so that VERW doesn't disarm it
and the machine never enters C1.
Original idea by Kim Phillips <kim.phillips@amd.com>.
Suggested-by: Andrew Cooper <andrew.cooper3@citrix.com>
Signed-off-by: Borislav Petkov (AMD) <bp@alien8.de>
io_bitmap_exit() is invoked from exit_thread() when a task exists or
when a fork fails. In the latter case the exit_thread() cleans up
resources which were allocated during fork().
io_bitmap_exit() invokes task_update_io_bitmap(), which in turn ends up
in tss_update_io_bitmap(). tss_update_io_bitmap() operates on the
current task. If current has TIF_IO_BITMAP set, but no bitmap installed,
tss_update_io_bitmap() crashes with a NULL pointer dereference.
There are two issues, which lead to that problem:
1) io_bitmap_exit() should not invoke task_update_io_bitmap() when
the task, which is cleaned up, is not the current task. That's a
clear indicator for a cleanup after a failed fork().
2) A task should not have TIF_IO_BITMAP set and neither a bitmap
installed nor IOPL emulation level 3 activated.
This happens when a kernel thread is created in the context of
a user space thread, which has TIF_IO_BITMAP set as the thread
flags are copied and the IO bitmap pointer is cleared.
Other than in the failed fork() case this has no impact because
kernel threads including IO workers never return to user space and
therefore never invoke tss_update_io_bitmap().
Cure this by adding the missing cleanups and checks:
1) Prevent io_bitmap_exit() to invoke task_update_io_bitmap() if
the to be cleaned up task is not the current task.
2) Clear TIF_IO_BITMAP in copy_thread() unconditionally. For user
space forks it is set later, when the IO bitmap is inherited in
io_bitmap_share().
For paranoia sake, add a warning into tss_update_io_bitmap() to catch
the case, when that code is invoked with inconsistent state.
Fixes: ea5f1cd7ab ("x86/ioperm: Remove bitmap if all permissions dropped")
Reported-by: syzbot+e2b1803445d236442e54@syzkaller.appspotmail.com
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Borislav Petkov (AMD) <bp@alien8.de>
Cc: stable@vger.kernel.org
Link: https://lore.kernel.org/87wmdceom2.ffs@tglx
The main CPUID header <asm/cpuid.h> was originally a storefront for the
headers:
<asm/cpuid/api.h>
<asm/cpuid/leaf_0x2_api.h>
Now that the latter CPUID(0x2) header has been merged into the former,
there is no practical difference between <asm/cpuid.h> and
<asm/cpuid/api.h>.
Migrate all users to the <asm/cpuid/api.h> header, in preparation of
the removal of <asm/cpuid.h>.
Don't remove <asm/cpuid.h> just yet, in case some new code in -next
started using it.
Suggested-by: Ingo Molnar <mingo@kernel.org>
Signed-off-by: Ahmed S. Darwish <darwi@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Cc: Andrew Cooper <andrew.cooper3@citrix.com>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: John Ogness <john.ogness@linutronix.de>
Cc: x86-cpuid@lists.linux.dev
Link: https://lore.kernel.org/r/20250508150240.172915-3-darwi@linutronix.de
Turn thread.fpu into a pointer. Since most FPU code internals work by passing
around the FPU pointer already, the code generation impact is small.
This allows us to remove the old kludge of task_struct being variable size:
struct task_struct {
...
/*
* New fields for task_struct should be added above here, so that
* they are included in the randomized portion of task_struct.
*/
randomized_struct_fields_end
/* CPU-specific state of this task: */
struct thread_struct thread;
/*
* WARNING: on x86, 'thread_struct' contains a variable-sized
* structure. It *MUST* be at the end of 'task_struct'.
*
* Do not put anything below here!
*/
};
... which creates a number of problems, such as requiring thread_struct to be
the last member of the struct - not allowing it to be struct-randomized, etc.
But the primary motivation is to allow the decoupling of task_struct from
hardware details (<asm/processor.h> in particular), and to eventually allow
the per-task infrastructure:
DECLARE_PER_TASK(type, name);
...
per_task(current, name) = val;
... which requires task_struct to be a constant size struct.
The fpu_thread_struct_whitelist() quirk to hardened usercopy can be removed,
now that the FPU structure is not embedded in the task struct anymore, which
reduces text footprint a bit.
Fixed-by: Oleg Nesterov <oleg@redhat.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Cc: Andy Lutomirski <luto@kernel.org>
Cc: Brian Gerst <brgerst@gmail.com>
Cc: Chang S. Bae <chang.seok.bae@intel.com>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Link: https://lore.kernel.org/r/20250409211127.3544993-4-mingo@kernel.org
The following commit, 12 years ago:
7e98b71920 ("x86, idle: Use static_cpu_has() for CLFLUSH workaround, add barriers")
added barriers around the CLFLUSH in mwait_idle_with_hints(), justified with:
... and add memory barriers around it since the documentation is explicit
that CLFLUSH is only ordered with respect to MFENCE.
This also triggered, 11 years ago, the same adjustment in:
f8e617f458 ("sched/idle/x86: Optimize unnecessary mwait_idle() resched IPIs")
during development, although it failed to get the static_cpu_has_bug() treatment.
X86_BUG_CLFLUSH_MONITOR (a.k.a the AAI65 errata) is specific to Intel CPUs,
and the SDM currently states:
Executions of the CLFLUSH instruction are ordered with respect to each
other and with respect to writes, locked read-modify-write instructions,
and fence instructions[1].
With footnote 1 reading:
Earlier versions of this manual specified that executions of the CLFLUSH
instruction were ordered only by the MFENCE instruction. All processors
implementing the CLFLUSH instruction also order it relative to the other
operations enumerated above.
i.e. The SDM was incorrect at the time, and barriers should not have been
inserted. Double checking the original AAI65 errata (not available from
intel.com any more) shows no mention of barriers either.
Note: If this were a general codepath, the MFENCEs would be needed, because
AMD CPUs of the same vintage do sport otherwise-unordered CLFLUSHs.
Remove the unnecessary barriers. Furthermore, use a plain alternative(),
rather than static_cpu_has_bug() and/or no optimisation. The workaround
is a single instruction.
Use an explicit %rax pointer rather than a general memory operand, because
MONITOR takes the pointer implicitly in the same way.
[ mingo: Cleaned up the commit a bit. ]
Fixes: 7e98b71920 ("x86, idle: Use static_cpu_has() for CLFLUSH workaround, add barriers")
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Dave Hansen <dave.hansen@intel.com>
Acked-by: Borislav Petkov (AMD) <bp@alien8.de>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Rik van Riel <riel@surriel.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andy Lutomirski <luto@kernel.org>
Cc: Brian Gerst <brgerst@gmail.com>
Cc: Juergen Gross <jgross@suse.com>
Cc: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Link: https://lore.kernel.org/r/20250402172458.1378112-1-andrew.cooper3@citrix.com
Direct HLT instruction execution causes #VEs for TDX VMs which is routed
to hypervisor via TDCALL. If HLT is executed in STI-shadow, resulting #VE
handler will enable interrupts before TDCALL is routed to hypervisor
leading to missed wakeup events, as current TDX spec doesn't expose
interruptibility state information to allow #VE handler to selectively
enable interrupts.
Commit bfe6ed0c67 ("x86/tdx: Add HLT support for TDX guests")
prevented the idle routines from executing HLT instruction in STI-shadow.
But it missed the paravirt routine which can be reached via this path
as an example:
kvm_wait() =>
safe_halt() =>
raw_safe_halt() =>
arch_safe_halt() =>
irq.safe_halt() =>
pv_native_safe_halt()
To reliably handle arch_safe_halt() for TDX VMs, introduce explicit
dependency on CONFIG_PARAVIRT and override paravirt halt()/safe_halt()
routines with TDX-safe versions that execute direct TDCALL and needed
interrupt flag updates. Executing direct TDCALL brings in additional
benefit of avoiding HLT related #VEs altogether.
As tested by Ryan Afranji:
"Tested with the specjbb2015 benchmark. It has heavy lock contention which leads
to many halt calls. TDX VMs suffered a poor score before this patchset.
Verified the major performance improvement with this patchset applied."
Fixes: bfe6ed0c67 ("x86/tdx: Add HLT support for TDX guests")
Signed-off-by: Vishal Annapurve <vannapurve@google.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Tested-by: Ryan Afranji <afranji@google.com>
Cc: Andy Lutomirski <luto@kernel.org>
Cc: Brian Gerst <brgerst@gmail.com>
Cc: Juergen Gross <jgross@suse.com>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Josh Poimboeuf <jpoimboe@redhat.com>
Cc: stable@vger.kernel.org
Link: https://lore.kernel.org/r/20250228014416.3925664-3-vannapurve@google.com
Pull x86/fpu updates from Ingo Molnar:
- Improve crypto performance by making kernel-mode FPU reliably usable
in softirqs ((Eric Biggers)
- Fully optimize out WARN_ON_FPU() (Eric Biggers)
- Initial steps to support Support Intel APX (Advanced Performance
Extensions) (Chang S. Bae)
- Fix KASAN for arch_dup_task_struct() (Benjamin Berg)
- Refine and simplify the FPU magic number check during signal return
(Chang S. Bae)
- Fix inconsistencies in guest FPU xfeatures (Chao Gao, Stanislav
Spassov)
- selftests/x86/xstate: Introduce common code for testing extended
states (Chang S. Bae)
- Misc fixes and cleanups (Borislav Petkov, Colin Ian King, Uros
Bizjak)
* tag 'x86-fpu-2025-03-22' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
x86/fpu/xstate: Fix inconsistencies in guest FPU xfeatures
x86/fpu: Clarify the "xa" symbolic name used in the XSTATE* macros
x86/fpu: Use XSAVE{,OPT,C,S} and XRSTOR{,S} mnemonics in xstate.h
x86/fpu: Improve crypto performance by making kernel-mode FPU reliably usable in softirqs
x86/fpu/xstate: Simplify print_xstate_features()
x86/fpu: Refine and simplify the magic number check during signal return
selftests/x86/xstate: Fix spelling mistake "hader" -> "header"
x86/fpu: Avoid copying dynamic FP state from init_task in arch_dup_task_struct()
vmlinux.lds.h: Remove entry to place init_task onto init_stack
selftests/x86/avx: Add AVX tests
selftests/x86/xstate: Clarify supported xstates
selftests/x86/xstate: Consolidate test invocations into a single entry
selftests/x86/xstate: Introduce signal ABI test
selftests/x86/xstate: Refactor ptrace ABI test
selftests/x86/xstate: Refactor context switching test
selftests/x86/xstate: Enumerate and name xstate components
selftests/x86/xstate: Refactor XSAVE helpers for general use
selftests/x86: Consolidate redundant signal helper functions
x86/fpu: Fix guest FPU state buffer allocation size
x86/fpu: Fully optimize out WARN_ON_FPU()
The init_task instance of struct task_struct is statically allocated and
may not contain the full FP state for userspace. As such, limit the copy
to the valid area of both init_task and 'dst' and ensure all memory is
initialized.
Note that the FP state is only needed for userspace, and as such it is
entirely reasonable for init_task to not contain parts of it.
Fixes: 5aaeb5c01c ("x86/fpu, sched: Introduce CONFIG_ARCH_WANTS_DYNAMIC_TASK_STRUCT and use it on x86")
Signed-off-by: Benjamin Berg <benjamin.berg@intel.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Cc: Andy Lutomirski <luto@kernel.org>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Link: https://lore.kernel.org/r/20250226133136.816901-1-benjamin@sipsolutions.net
----
v2:
- Fix code if arch_task_struct_size < sizeof(init_task) by using
memcpy_and_pad.
Linus Torvalds
Kai HuangandDave Hansen
Simon SchusterandChristian Brauner
Ingo Molnar
Borislav Petkov (AMD)
Thomas GleixnerandBorislav Petkov (AMD)
Ahmed S. DarwishandIngo Molnar
Oleg NesterovandIngo Molnar
Xin Li (Intel)andIngo Molnar
Andrew CooperandIngo Molnar
Vishal AnnapurveandIngo Molnar
Brian GerstandIngo Molnar
Benjamin BergandIngo Molnar