Pull scheduler updates from Thomas Gleixner:
- migrate_disable/enable() support which originates from the RT tree
and is now a prerequisite for the new preemptible kmap_local() API
which aims to replace kmap_atomic().
- A fair amount of topology and NUMA related improvements
- Improvements for the frequency invariant calculations
- Enhanced robustness for the global CPU priority tracking and decision
making
- The usual small fixes and enhancements all over the place
* tag 'sched-core-2020-12-14' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (61 commits)
sched/fair: Trivial correction of the newidle_balance() comment
sched/fair: Clear SMT siblings after determining the core is not idle
sched: Fix kernel-doc markup
x86: Print ratio freq_max/freq_base used in frequency invariance calculations
x86, sched: Use midpoint of max_boost and max_P for frequency invariance on AMD EPYC
x86, sched: Calculate frequency invariance for AMD systems
irq_work: Optimize irq_work_single()
smp: Cleanup smp_call_function*()
irq_work: Cleanup
sched: Limit the amount of NUMA imbalance that can exist at fork time
sched/numa: Allow a floating imbalance between NUMA nodes
sched: Avoid unnecessary calculation of load imbalance at clone time
sched/numa: Rename nr_running and break out the magic number
sched: Make migrate_disable/enable() independent of RT
sched/topology: Condition EAS enablement on FIE support
arm64: Rebuild sched domains on invariance status changes
sched/topology,schedutil: Wrap sched domains rebuild
sched/uclamp: Allow to reset a task uclamp constraint value
sched/core: Fix typos in comments
Documentation: scheduler: fix information on arch SD flags, sched_domain and sched_debug
...
Pull arm64 updates from Catalin Marinas:
- Expose tag address bits in siginfo. The original arm64 ABI did not
expose any of the bits 63:56 of a tagged address in siginfo. In the
presence of user ASAN or MTE, this information may be useful. The
implementation is generic to other architectures supporting tags
(like SPARC ADI, subject to wiring up the arch code). The user will
have to opt in via sigaction(SA_EXPOSE_TAGBITS) so that the extra
bits, if available, become visible in si_addr.
- Default to 32-bit wide ZONE_DMA. Previously, ZONE_DMA was set to the
lowest 1GB to cope with the Raspberry Pi 4 limitations, to the
detriment of other platforms. With these changes, the kernel scans
the Device Tree dma-ranges and the ACPI IORT information before
deciding on a smaller ZONE_DMA.
- Strengthen READ_ONCE() to acquire when CONFIG_LTO=y. When building
with LTO, there is an increased risk of the compiler converting an
address dependency headed by a READ_ONCE() invocation into a control
dependency and consequently allowing for harmful reordering by the
CPU.
- Add CPPC FFH support using arm64 AMU counters.
- set_fs() removal on arm64. This renders the User Access Override
(UAO) ARMv8 feature unnecessary.
- Perf updates: PMU driver for the ARM DMC-620 memory controller, sysfs
identifier file for SMMUv3, stop event counters support for i.MX8MP,
enable the perf events-based hard lockup detector.
- Reorganise the kernel VA space slightly so that 52-bit VA
configurations can use more virtual address space.
- Improve the robustness of the arm64 memory offline event notifier.
- Pad the Image header to 64K following the EFI header definition
updated recently to increase the section alignment to 64K.
- Support CONFIG_CMDLINE_EXTEND on arm64.
- Do not use tagged PC in the kernel (TCR_EL1.TBID1==1), freeing up 8
bits for PtrAuth.
- Switch to vmapped shadow call stacks.
- Miscellaneous clean-ups.
* tag 'arm64-upstream' of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux: (78 commits)
perf/imx_ddr: Add system PMU identifier for userspace
bindings: perf: imx-ddr: add compatible string
arm64: Fix build failure when HARDLOCKUP_DETECTOR_PERF is enabled
arm64: mte: fix prctl(PR_GET_TAGGED_ADDR_CTRL) if TCF0=NONE
arm64: mark __system_matches_cap as __maybe_unused
arm64: uaccess: remove vestigal UAO support
arm64: uaccess: remove redundant PAN toggling
arm64: uaccess: remove addr_limit_user_check()
arm64: uaccess: remove set_fs()
arm64: uaccess cleanup macro naming
arm64: uaccess: split user/kernel routines
arm64: uaccess: refactor __{get,put}_user
arm64: uaccess: simplify __copy_user_flushcache()
arm64: uaccess: rename privileged uaccess routines
arm64: sdei: explicitly simulate PAN/UAO entry
arm64: sdei: move uaccess logic to arch/arm64/
arm64: head.S: always initialize PSTATE
arm64: head.S: cleanup SCTLR_ELx initialization
arm64: head.S: rename el2_setup -> init_kernel_el
arm64: add C wrappers for SET_PSTATE_*()
...
* for-next/kvm-build-fix:
: Fix KVM build issues with 64K pages
KVM: arm64: Fix build error in user_mem_abort()
* for-next/va-refactor:
: VA layout changes
arm64: mm: don't assume struct page is always 64 bytes
Documentation/arm64: fix RST layout of memory.rst
arm64: mm: tidy up top of kernel VA space
arm64: mm: make vmemmap region a projection of the linear region
arm64: mm: extend linear region for 52-bit VA configurations
* for-next/lto:
: Upgrade READ_ONCE() to RCpc acquire on arm64 with LTO
arm64: lto: Strengthen READ_ONCE() to acquire when CONFIG_LTO=y
arm64: alternatives: Remove READ_ONCE() usage during patch operation
arm64: cpufeatures: Add capability for LDAPR instruction
arm64: alternatives: Split up alternative.h
arm64: uaccess: move uao_* alternatives to asm-uaccess.h
* for-next/mem-hotplug:
: Memory hotplug improvements
arm64/mm/hotplug: Ensure early memory sections are all online
arm64/mm/hotplug: Enable MEM_OFFLINE event handling
arm64/mm/hotplug: Register boot memory hot remove notifier earlier
arm64: mm: account for hotplug memory when randomizing the linear region
* for-next/cppc-ffh:
: Add CPPC FFH support using arm64 AMU counters
arm64: abort counter_read_on_cpu() when irqs_disabled()
arm64: implement CPPC FFH support using AMUs
arm64: split counter validation function
arm64: wrap and generalise counter read functions
* for-next/pad-image-header:
: Pad Image header to 64KB and unmap it
arm64: head: tidy up the Image header definition
arm64/head: avoid symbol names pointing into first 64 KB of kernel image
arm64: omit [_text, _stext) from permanent kernel mapping
* for-next/zone-dma-default-32-bit:
: Default to 32-bit wide ZONE_DMA (previously reduced to 1GB for RPi4)
of: unittest: Fix build on architectures without CONFIG_OF_ADDRESS
mm: Remove examples from enum zone_type comment
arm64: mm: Set ZONE_DMA size based on early IORT scan
arm64: mm: Set ZONE_DMA size based on devicetree's dma-ranges
of: unittest: Add test for of_dma_get_max_cpu_address()
of/address: Introduce of_dma_get_max_cpu_address()
arm64: mm: Move zone_dma_bits initialization into zone_sizes_init()
arm64: mm: Move reserve_crashkernel() into mem_init()
arm64: Force NO_BLOCK_MAPPINGS if crashkernel reservation is required
arm64: Ignore any DMA offsets in the max_zone_phys() calculation
* for-next/signal-tag-bits:
: Expose the FAR_EL1 tag bits in siginfo
arm64: expose FAR_EL1 tag bits in siginfo
signal: define the SA_EXPOSE_TAGBITS bit in sa_flags
signal: define the SA_UNSUPPORTED bit in sa_flags
arch: provide better documentation for the arch-specific SA_* flags
signal: clear non-uapi flag bits when passing/returning sa_flags
arch: move SA_* definitions to generic headers
parisc: start using signal-defs.h
parisc: Drop parisc special case for __sighandler_t
* for-next/cmdline-extended:
: Add support for CONFIG_CMDLINE_EXTENDED
arm64: Extend the kernel command line from the bootloader
arm64: kaslr: Refactor early init command line parsing
Conflict resolution gone astray results in the kernel not booting
on VHE-capable HW when VHE support is disabled. Thankfully spotted
by David.
Reported-by: David Brazdil <dbrazdil@google.com>
Signed-off-by: Marc Zyngier <maz@kernel.org>
When compiling with __KVM_NVHE_HYPERVISOR__, redefine per_cpu_offset()
to __hyp_per_cpu_offset() which looks up the base of the nVHE per-CPU
region of the given cpu and computes its offset from the
.hyp.data..percpu section.
This enables use of per_cpu_ptr() helpers in nVHE hyp code. Until now
only this_cpu_ptr() was supported by setting TPIDR_EL2.
Signed-off-by: David Brazdil <dbrazdil@google.com>
Signed-off-by: Marc Zyngier <maz@kernel.org>
Link: https://lore.kernel.org/r/20201202184122.26046-14-dbrazdil@google.com
Add rules for renaming the .data..ro_after_init ELF section in KVM nVHE
object files to .hyp.data..ro_after_init, linking it into the kernel
and mapping it in hyp at runtime.
The section is RW to the host, then mapped RO in hyp. The expectation is
that the host populates the variables in the section and they are never
changed by hyp afterwards.
Signed-off-by: David Brazdil <dbrazdil@google.com>
Signed-off-by: Marc Zyngier <maz@kernel.org>
Link: https://lore.kernel.org/r/20201202184122.26046-13-dbrazdil@google.com
MAIR_EL2 and TCR_EL2 are currently initialized from their _EL1 values.
This will not work once KVM starts intercepting PSCI ON/SUSPEND SMCs
and initializing EL2 state before EL1 state.
Obtain the EL1 values during KVM init and store them in the init params
struct. The struct will stay in memory and can be used when booting new
cores.
Take the opportunity to move copying the T0SZ value from idmap_t0sz in
KVM init rather than in .hyp.idmap.text. This avoids the need for the
idmap_t0sz symbol alias.
Signed-off-by: David Brazdil <dbrazdil@google.com>
Signed-off-by: Marc Zyngier <maz@kernel.org>
Link: https://lore.kernel.org/r/20201202184122.26046-12-dbrazdil@google.com
Once we start initializing KVM on newly booted cores before the rest of
the kernel, parameters to __do_hyp_init will need to be provided by EL2
rather than EL1. At that point it will not be possible to pass its three
arguments directly because PSCI_CPU_ON only supports one context
argument.
Refactor __do_hyp_init to accept its parameters in a struct. This
prepares the code for KVM booting cores as well as removes any limits on
the number of __do_hyp_init arguments.
Signed-off-by: David Brazdil <dbrazdil@google.com>
Signed-off-by: Marc Zyngier <maz@kernel.org>
Link: https://lore.kernel.org/r/20201202184122.26046-11-dbrazdil@google.com
When a CPU is booted in EL2, the kernel checks for VHE support and
initializes the CPU core accordingly. For nVHE it also installs the stub
vectors and drops down to EL1.
Once KVM gains the ability to boot cores without going through the
kernel entry point, it will need to initialize the CPU the same way.
Extract the relevant bits of el2_setup into an init_el2_state macro
with an argument specifying whether to initialize for VHE or nVHE.
The following ifdefs are removed:
* CONFIG_ARM_GIC_V3 - always selected on arm64
* CONFIG_COMPAT - hstr_el2 can be set even without 32-bit support
No functional change intended. Size of el2_setup increased by
148 bytes due to duplication.
Signed-off-by: David Brazdil <dbrazdil@google.com>
[maz: reworked to fit the new PSTATE initial setup code]
Signed-off-by: Marc Zyngier <maz@kernel.org>
Link: https://lore.kernel.org/r/20201202184122.26046-9-dbrazdil@google.com
CPU index should never be negative. Change the signature of
(set_)cpu_logical_map to take an unsigned int.
This still works even if the users treat the CPU index as an int,
and will allow the hypervisor's implementation to check that the index
is valid with a single upper-bound check.
Signed-off-by: David Brazdil <dbrazdil@google.com>
Signed-off-by: Marc Zyngier <maz@kernel.org>
Link: https://lore.kernel.org/r/20201202184122.26046-8-dbrazdil@google.com
Now that the PAN toggling has been removed, the only user of
__system_matches_cap() is has_generic_auth(), which is only built when
CONFIG_ARM64_PTR_AUTH is selected, and Qian reports that this results in
a build-time warning when CONFIG_ARM64_PTR_AUTH is not selected:
| arch/arm64/kernel/cpufeature.c:2649:13: warning: '__system_matches_cap' defined but not used [-Wunused-function]
| static bool __system_matches_cap(unsigned int n)
| ^~~~~~~~~~~~~~~~~~~~
It's tricky to restructure things to prevent this, so let's mark
__system_matches_cap() as __maybe_unused, as we used to do for the other
user of __system_matches_cap() which we just removed.
Reported-by: Qian Cai <qcai@redhat.com>
Suggested-by: Qian Cai <qcai@redhat.com>
Signed-off-by: Mark Rutland <mark.rutland@arm.com>
Link: https://lore.kernel.org/r/20201203152403.26100-1-mark.rutland@arm.com
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
Pull arm64 fixes from Will Deacon:
"I'm sad to say that we've got an unusually large arm64 fixes pull for
rc7 which addresses numerous significant instrumentation issues with
our entry code.
Without these patches, lockdep is hopelessly unreliable in some
configurations [1,2] and syzkaller is therefore not a lot of use
because it's so noisy.
Although much of this has always been broken, it appears to have been
exposed more readily by other changes such as 044d0d6de9 ("lockdep:
Only trace IRQ edges") and general lockdep improvements around IRQ
tracing and NMIs.
Fixing this properly required moving much of the instrumentation hooks
from our entry assembly into C, which Mark has been working on for the
last few weeks. We're not quite ready to move to the recently added
generic functions yet, but the code here has been deliberately written
to mimic that closely so we can look at cleaning things up once we
have a bit more breathing room.
Having said all that, the second version of these patches was posted
last week and I pushed it into our CI (kernelci and cki) along with a
commit which forced on PROVE_LOCKING, NOHZ_FULL and
CONTEXT_TRACKING_FORCE. The result? We found a real bug in the
md/raid10 code [3].
Oh, and there's also a really silly typo patch that's unrelated.
Summary:
- Fix numerous issues with instrumentation and exception entry
- Fix hideous typo in unused register field definition"
[1] https://lore.kernel.org/r/CACT4Y+aAzoJ48Mh1wNYD17pJqyEcDnrxGfApir=-j171TnQXhw@mail.gmail.com
[2] https://lore.kernel.org/r/20201119193819.GA2601289@elver.google.com
[3] https://lore.kernel.org/r/94c76d5e-466a-bc5f-e6c2-a11b65c39f83@redhat.com
* tag 'arm64-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux:
arm64: mte: Fix typo in macro definition
arm64: entry: fix EL1 debug transitions
arm64: entry: fix NMI {user, kernel}->kernel transitions
arm64: entry: fix non-NMI kernel<->kernel transitions
arm64: ptrace: prepare for EL1 irq/rcu tracking
arm64: entry: fix non-NMI user<->kernel transitions
arm64: entry: move el1 irq/nmi logic to C
arm64: entry: prepare ret_to_user for function call
arm64: entry: move enter_from_user_mode to entry-common.c
arm64: entry: mark entry code as noinstr
arm64: mark idle code as noinstr
arm64: syscall: exit userspace before unmasking exceptions
Now that set_fs() is gone, addr_limit_user_check() is redundant. Remove
the checks and associated thread flag.
To ensure that _TIF_WORK_MASK can be used as an immediate value in an
AND instruction (as it is in `ret_to_user`), TIF_MTE_ASYNC_FAULT is
renumbered to keep the constituent bits of _TIF_WORK_MASK contiguous.
There should be no functional change as a result of this patch.
Signed-off-by: Mark Rutland <mark.rutland@arm.com>
Cc: Christoph Hellwig <hch@lst.de>
Cc: James Morse <james.morse@arm.com>
Cc: Will Deacon <will@kernel.org>
Link: https://lore.kernel.org/r/20201202131558.39270-11-mark.rutland@arm.com
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
Now that the uaccess primitives dont take addr_limit into account, we
have no need to manipulate this via set_fs() and get_fs(). Remove
support for these, along with some infrastructure this renders
redundant.
We no longer need to flip UAO to access kernel memory under KERNEL_DS,
and head.S unconditionally clears UAO for all kernel configurations via
an ERET in init_kernel_el. Thus, we don't need to dynamically flip UAO,
nor do we need to context-switch it. However, we still need to adjust
PAN during SDEI entry.
Masking of __user pointers no longer needs to use the dynamic value of
addr_limit, and can use a constant derived from the maximum possible
userspace task size. A new TASK_SIZE_MAX constant is introduced for
this, which is also used by core code. In configurations supporting
52-bit VAs, this may include a region of unusable VA space above a
48-bit TTBR0 limit, but never includes any portion of TTBR1.
Note that TASK_SIZE_MAX is an exclusive limit, while USER_DS and
KERNEL_DS were inclusive limits, and is converted to a mask by
subtracting one.
As the SDEI entry code repurposes the otherwise unnecessary
pt_regs::orig_addr_limit field to store the TTBR1 of the interrupted
context, for now we rename that to pt_regs::sdei_ttbr1. In future we can
consider factoring that out.
Signed-off-by: Mark Rutland <mark.rutland@arm.com>
Acked-by: James Morse <james.morse@arm.com>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Will Deacon <will@kernel.org>
Link: https://lore.kernel.org/r/20201202131558.39270-10-mark.rutland@arm.com
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
We currently have many uaccess_*{enable,disable}*() variants, which
subsequent patches will cut down as part of removing set_fs() and
friends. Once this simplification is made, most uaccess routines will
only need to ensure that the user page tables are mapped in TTBR0, as is
currently dealt with by uaccess_ttbr0_{enable,disable}().
The existing uaccess_{enable,disable}() routines ensure that user page
tables are mapped in TTBR0, and also disable PAN protections, which is
necessary to be able to use atomics on user memory, but also permit
unrelated privileged accesses to access user memory.
As preparatory step, let's rename uaccess_{enable,disable}() to
uaccess_{enable,disable}_privileged(), highlighting this caveat and
discouraging wider misuse. Subsequent patches can reuse the
uaccess_{enable,disable}() naming for the common case of ensuring the
user page tables are mapped in TTBR0.
There should be no functional change as a result of this patch.
Signed-off-by: Mark Rutland <mark.rutland@arm.com>
Cc: Christoph Hellwig <hch@lst.de>
Cc: James Morse <james.morse@arm.com>
Cc: Will Deacon <will@kernel.org>
Link: https://lore.kernel.org/r/20201202131558.39270-5-mark.rutland@arm.com
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
In preparation for removing addr_limit and set_fs() we must decouple the
SDEI PAN/UAO manipulation from the uaccess code, and explicitly
reinitialize these as required.
SDEI enters the kernel with a non-architectural exception, and prior to
the most recent revision of the specification (ARM DEN 0054B), PSTATE
bits (e.g. PAN, UAO) are not manipulated in the same way as for
architectural exceptions. Notably, older versions of the spec can be
read ambiguously as to whether PSTATE bits are inherited unchanged from
the interrupted context or whether they are generated from scratch, with
TF-A doing the latter.
We have three cases to consider:
1) The existing TF-A implementation of SDEI will clear PAN and clear UAO
(along with other bits in PSTATE) when delivering an SDEI exception.
2) In theory, implementations of SDEI prior to revision B could inherit
PAN and UAO (along with other bits in PSTATE) unchanged from the
interrupted context. However, in practice such implementations do not
exist.
3) Going forward, new implementations of SDEI must clear UAO, and
depending on SCTLR_ELx.SPAN must either inherit or set PAN.
As we can ignore (2) we can assume that upon SDEI entry, UAO is always
clear, though PAN may be clear, inherited, or set per SCTLR_ELx.SPAN.
Therefore, we must explicitly initialize PAN, but do not need to do
anything for UAO.
Considering what we need to do:
* When set_fs() is removed, force_uaccess_begin() will have no HW
side-effects. As this only clears UAO, which we can assume has already
been cleared upon entry, this is not a problem. We do not need to add
code to manipulate UAO explicitly.
* PAN may be cleared upon entry (in case 1 above), so where a kernel is
built to use PAN and this is supported by all CPUs, the kernel must
set PAN upon entry to ensure expected behaviour.
* PAN may be inherited from the interrupted context (in case 3 above),
and so where a kernel is not built to use PAN or where PAN support is
not uniform across CPUs, the kernel must clear PAN to ensure expected
behaviour.
This patch reworks the SDEI code accordingly, explicitly setting PAN to
the expected state in all cases. To cater for the cases where the kernel
does not use PAN or this is not uniformly supported by hardware we add a
new cpu_has_pan() helper which can be used regardless of whether the
kernel is built to use PAN.
The existing system_uses_ttbr0_pan() is redefined in terms of
system_uses_hw_pan() both for clarity and as a minor optimization when
HW PAN is not selected.
Signed-off-by: Mark Rutland <mark.rutland@arm.com>
Reviewed-by: James Morse <james.morse@arm.com>
Cc: James Morse <james.morse@arm.com>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Will Deacon <will@kernel.org>
Link: https://lore.kernel.org/r/20201202131558.39270-3-mark.rutland@arm.com
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
The SDEI support code is split across arch/arm64/ and drivers/firmware/,
largley this is split so that the arch-specific portions are under
arch/arm64, and the management logic is under drivers/firmware/.
However, exception entry fixups are currently under drivers/firmware.
Let's move the exception entry fixups under arch/arm64/. This
de-clutters the management logic, and puts all the arch-specific
portions in one place. Doing this also allows the fixups to be applied
earlier, so things like PAN and UAO will be in a known good state before
we run other logic. This will also make subsequent refactoring easier.
Signed-off-by: Mark Rutland <mark.rutland@arm.com>
Reviewed-by: James Morse <james.morse@arm.com>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Will Deacon <will@kernel.org>
Link: https://lore.kernel.org/r/20201202131558.39270-2-mark.rutland@arm.com
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
As with SCTLR_ELx and other control registers, some PSTATE bits are
UNKNOWN out-of-reset, and we may not be able to rely on hardware or
firmware to initialize them to our liking prior to entry to the kernel,
e.g. in the primary/secondary boot paths and return from idle/suspend.
It would be more robust (and easier to reason about) if we consistently
initialized PSTATE to a default value, as we do with control registers.
This will ensure that the kernel is not adversely affected by bits it is
not aware of, e.g. when support for a feature such as PAN/UAO is
disabled.
This patch ensures that PSTATE is consistently initialized at boot time
via an ERET. This is not intended to relax the existing requirements
(e.g. DAIF bits must still be set prior to entering the kernel). For
features detected dynamically (which may require system-wide support),
it is still necessary to subsequently modify PSTATE.
As ERET is not always a Context Synchronization Event, an ISB is placed
before each exception return to ensure updates to control registers have
taken effect. This handles the kernel being entered with SCTLR_ELx.EOS
clear (or any future control bits being in an UNKNOWN state).
Signed-off-by: Mark Rutland <mark.rutland@arm.com>
Cc: Christoph Hellwig <hch@lst.de>
Cc: James Morse <james.morse@arm.com>
Cc: Will Deacon <will@kernel.org>
Link: https://lore.kernel.org/r/20201113124937.20574-6-mark.rutland@arm.com
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
For a while now el2_setup has performed some basic initialization of EL1
even when the kernel is booted at EL1, so the name is a little
misleading. Further, some comments are stale as with VHE it doesn't drop
the CPU to EL1.
To clarify things, rename el2_setup to init_kernel_el, and update
comments to be clearer as to the function's purpose.
There should be no functional change as a result of this patch.
Signed-off-by: Mark Rutland <mark.rutland@arm.com>
Cc: Christoph Hellwig <hch@lst.de>
Cc: James Morse <james.morse@arm.com>
Cc: Will Deacon <will@kernel.org>
Link: https://lore.kernel.org/r/20201113124937.20574-4-mark.rutland@arm.com
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
For consistency, all tasks have a pt_regs reserved at the highest
portion of their task stack. Among other things, this ensures that a
task's SP is always pointing within its stack rather than pointing
immediately past the end.
While it is never legitimate to ERET from a kthread, we take pains to
initialize pt_regs for kthreads as if this were legitimate. As this is
never legitimate, the effects of an erroneous return are rarely tested.
Let's simplify things by initializing a kthread's pt_regs such that an
ERET is caught as an illegal exception return, and removing the explicit
initialization of other exception context. Note that as
spectre_v4_enable_task_mitigation() only manipulates the PSTATE within
the unused regs this is safe to remove.
As user tasks will have their exception context initialized via
start_thread() or start_compat_thread(), this should only impact cases
where something has gone very wrong and we'd like that to be clearly
indicated.
Signed-off-by: Mark Rutland <mark.rutland@arm.com>
Cc: Christoph Hellwig <hch@lst.de>
Cc: James Morse <james.morse@arm.com>
Cc: Will Deacon <will@kernel.org>
Link: https://lore.kernel.org/r/20201113124937.20574-2-mark.rutland@arm.com
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
Handling all combinations of the VMAP_STACK and SHADOW_CALL_STACK options
in sdei_arch_get_entry_point() makes the code difficult to read,
particularly when considering the error and cleanup paths.
Move the checking of these options into the callee functions, so that
they return early if the relevant option is not enabled.
Signed-off-by: Will Deacon <will@kernel.org>
In debug_exception_enter() and debug_exception_exit() we trace hardirqs
on/off while RCU isn't guaranteed to be watching, and we don't save and
restore the hardirq state, and so may return with this having changed.
Handle this appropriately with new entry/exit helpers which do the bare
minimum to ensure this is appropriately maintained, without marking
debug exceptions as NMIs. These are placed in entry-common.c with the
other entry/exit helpers.
In future we'll want to reconsider whether some debug exceptions should
be NMIs, but this will require a significant refactoring, and for now
this should prevent issues with lockdep and RCU.
Signed-off-by: Mark Rutland <mark.rutland@arm.com>
Cc: Catalin Marins <catalin.marinas@arm.com>
Cc: James Morse <james.morse@arm.com>
Cc: Will Deacon <will@kernel.org>
Link: https://lore.kernel.org/r/20201130115950.22492-12-mark.rutland@arm.com
Signed-off-by: Will Deacon <will@kernel.org>
Exceptions which can be taken at (almost) any time are consdiered to be
NMIs. On arm64 that includes:
* SDEI events
* GICv3 Pseudo-NMIs
* Kernel stack overflows
* Unexpected/unhandled exceptions
... but currently debug exceptions (BRKs, breakpoints, watchpoints,
single-step) are not considered NMIs.
As these can be taken at any time, kernel features (lockdep, RCU,
ftrace) may not be in a consistent kernel state. For example, we may
take an NMI from the idle code or partway through an entry/exit path.
While nmi_enter() and nmi_exit() handle most of this state, notably they
don't save/restore the lockdep state across an NMI being taken and
handled. When interrupts are enabled and an NMI is taken, lockdep may
see interrupts become disabled within the NMI code, but not see
interrupts become enabled when returning from the NMI, leaving lockdep
believing interrupts are disabled when they are actually disabled.
The x86 code handles this in idtentry_{enter,exit}_nmi(), which will
shortly be moved to the generic entry code. As we can't use either yet,
we copy the x86 approach in arm64-specific helpers. All the NMI
entrypoints are marked as noinstr to prevent any instrumentation
handling code being invoked before the state has been corrected.
Signed-off-by: Mark Rutland <mark.rutland@arm.com>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: James Morse <james.morse@arm.com>
Cc: Will Deacon <will@kernel.org>
Link: https://lore.kernel.org/r/20201130115950.22492-11-mark.rutland@arm.com
Signed-off-by: Will Deacon <will@kernel.org>
There are periods in kernel mode when RCU is not watching and/or the
scheduler tick is disabled, but we can still take exceptions such as
interrupts. The arm64 exception handlers do not account for this, and
it's possible that RCU is not watching while an exception handler runs.
The x86/generic entry code handles this by ensuring that all (non-NMI)
kernel exception handlers call irqentry_enter() and irqentry_exit(),
which handle RCU, lockdep, and IRQ flag tracing. We can't yet move to
the generic entry code, and already hadnle the user<->kernel transitions
elsewhere, so we add new kernel<->kernel transition helpers alog the
lines of the generic entry code.
Since we now track interrupts becoming masked when an exception is
taken, local_daif_inherit() is modified to track interrupts becoming
re-enabled when the original context is inherited. To balance the
entry/exit paths, each handler masks all DAIF exceptions before
exit_to_kernel_mode().
Signed-off-by: Mark Rutland <mark.rutland@arm.com>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: James Morse <james.morse@arm.com>
Cc: Will Deacon <will@kernel.org>
Link: https://lore.kernel.org/r/20201130115950.22492-10-mark.rutland@arm.com
Signed-off-by: Will Deacon <will@kernel.org>
When built with PROVE_LOCKING, NO_HZ_FULL, and CONTEXT_TRACKING_FORCE
will WARN() at boot time that interrupts are enabled when we call
context_tracking_user_enter(), despite the DAIF flags indicating that
IRQs are masked.
The problem is that we're not tracking IRQ flag changes accurately, and
so lockdep believes interrupts are enabled when they are not (and
vice-versa). We can shuffle things so to make this more accurate. For
kernel->user transitions there are a number of constraints we need to
consider:
1) When we call __context_tracking_user_enter() HW IRQs must be disabled
and lockdep must be up-to-date with this.
2) Userspace should be treated as having IRQs enabled from the PoV of
both lockdep and tracing.
3) As context_tracking_user_enter() stops RCU from watching, we cannot
use RCU after calling it.
4) IRQ flag tracing and lockdep have state that must be manipulated
before RCU is disabled.
... with similar constraints applying for user->kernel transitions, with
the ordering reversed.
The generic entry code has enter_from_user_mode() and
exit_to_user_mode() helpers to handle this. We can't use those directly,
so we add arm64 copies for now (without the instrumentation markers
which aren't used on arm64). These replace the existing user_exit() and
user_exit_irqoff() calls spread throughout handlers, and the exception
unmasking is left as-is.
Note that:
* The accounting for debug exceptions from userspace now happens in
el0_dbg() and ret_to_user(), so this is removed from
debug_exception_enter() and debug_exception_exit(). As
user_exit_irqoff() wakes RCU, the userspace-specific check is removed.
* The accounting for syscalls now happens in el0_svc(),
el0_svc_compat(), and ret_to_user(), so this is removed from
el0_svc_common(). This does not adversely affect the workaround for
erratum 1463225, as this does not depend on any of the state tracking.
* In ret_to_user() we mask interrupts with local_daif_mask(), and so we
need to inform lockdep and tracing. Here a trace_hardirqs_off() is
sufficient and safe as we have not yet exited kernel context and RCU
is usable.
* As PROVE_LOCKING selects TRACE_IRQFLAGS, the ifdeferry in entry.S only
needs to check for the latter.
* EL0 SError handling will be dealt with in a subsequent patch, as this
needs to be treated as an NMI.
Prior to this patch, booting an appropriately-configured kernel would
result in spats as below:
| DEBUG_LOCKS_WARN_ON(lockdep_hardirqs_enabled())
| WARNING: CPU: 2 PID: 1 at kernel/locking/lockdep.c:5280 check_flags.part.54+0x1dc/0x1f0
| Modules linked in:
| CPU: 2 PID: 1 Comm: init Not tainted 5.10.0-rc3 #3
| Hardware name: linux,dummy-virt (DT)
| pstate: 804003c5 (Nzcv DAIF +PAN -UAO -TCO BTYPE=--)
| pc : check_flags.part.54+0x1dc/0x1f0
| lr : check_flags.part.54+0x1dc/0x1f0
| sp : ffff80001003bd80
| x29: ffff80001003bd80 x28: ffff66ce801e0000
| x27: 00000000ffffffff x26: 00000000000003c0
| x25: 0000000000000000 x24: ffffc31842527258
| x23: ffffc31842491368 x22: ffffc3184282d000
| x21: 0000000000000000 x20: 0000000000000001
| x19: ffffc318432ce000 x18: 0080000000000000
| x17: 0000000000000000 x16: ffffc31840f18a78
| x15: 0000000000000001 x14: ffffc3184285c810
| x13: 0000000000000001 x12: 0000000000000000
| x11: ffffc318415857a0 x10: ffffc318406614c0
| x9 : ffffc318415857a0 x8 : ffffc31841f1d000
| x7 : 647261685f706564 x6 : ffffc3183ff7c66c
| x5 : ffff66ce801e0000 x4 : 0000000000000000
| x3 : ffffc3183fe00000 x2 : ffffc31841500000
| x1 : e956dc24146b3500 x0 : 0000000000000000
| Call trace:
| check_flags.part.54+0x1dc/0x1f0
| lock_is_held_type+0x10c/0x188
| rcu_read_lock_sched_held+0x70/0x98
| __context_tracking_enter+0x310/0x350
| context_tracking_enter.part.3+0x5c/0xc8
| context_tracking_user_enter+0x6c/0x80
| finish_ret_to_user+0x2c/0x13cr
Signed-off-by: Mark Rutland <mark.rutland@arm.com>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: James Morse <james.morse@arm.com>
Cc: Will Deacon <will@kernel.org>
Link: https://lore.kernel.org/r/20201130115950.22492-8-mark.rutland@arm.com
Signed-off-by: Will Deacon <will@kernel.org>
In preparation for reworking the EL1 irq/nmi entry code, move the
existing logic to C. We no longer need the asm_nmi_enter() and
asm_nmi_exit() wrappers, so these are removed. The new C functions are
marked noinstr, which prevents compiler instrumentation and runtime
probing.
In subsequent patches we'll want the new C helpers to be called in all
cases, so we don't bother wrapping the calls with ifdeferry. Even when
the new C functions are stubs the trivial calls are unlikely to have a
measurable impact on the IRQ or NMI paths anyway.
Prototypes are added to <asm/exception.h> as otherwise (in some
configurations) GCC will complain about the lack of a forward
declaration. We already do this for existing function, e.g.
enter_from_user_mode().
The new helpers are marked as noinstr (which prevents all
instrumentation, tracing, and kprobes). Otherwise, there should be no
functional change as a result of this patch.
Signed-off-by: Mark Rutland <mark.rutland@arm.com>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: James Morse <james.morse@arm.com>
Cc: Will Deacon <will@kernel.org>
Link: https://lore.kernel.org/r/20201130115950.22492-7-mark.rutland@arm.com
Signed-off-by: Will Deacon <will@kernel.org>
In a subsequent patch ret_to_user will need to make a C function call
(in some configurations) which may clobber x0-x18 at the start of the
finish_ret_to_user block, before enable_step_tsk consumes the flags
loaded into x1.
In preparation for this, let's load the flags into x19, which is
preserved across C function calls. This avoids a redundant reload of the
flags and ensures we operate on a consistent shapshot regardless.
There should be no functional change as a result of this patch. At this
point of the entry/exit paths we only need to preserve x28 (tsk) and the
sp, and x19 is free for this use.
Signed-off-by: Mark Rutland <mark.rutland@arm.com>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: James Morse <james.morse@arm.com>
Cc: Will Deacon <will@kernel.org>
Link: https://lore.kernel.org/r/20201130115950.22492-6-mark.rutland@arm.com
Signed-off-by: Will Deacon <will@kernel.org>
In later patches we'll want to extend enter_from_user_mode() and add a
corresponding exit_to_user_mode(). As these will be common for all
entries/exits from userspace, it'd be better for these to live in
entry-common.c with the rest of the entry logic.
This patch moves enter_from_user_mode() into entry-common.c. As with
other functions in entry-common.c it is marked as noinstr (which
prevents all instrumentation, tracing, and kprobes) but there are no
other functional changes.
Signed-off-by: Mark Rutland <mark.rutland@arm.com>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: James Morse <james.morse@arm.com>
Cc: Will Deacon <will@kernel.org>
Link: https://lore.kernel.org/r/20201130115950.22492-5-mark.rutland@arm.com
Signed-off-by: Will Deacon <will@kernel.org>
Functions in entry-common.c are marked as notrace and NOKPROBE_SYMBOL(),
but they're still subject to other instrumentation which may rely on
lockdep/rcu/context-tracking being up-to-date, and may cause nested
exceptions (e.g. for WARN/BUG or KASAN's use of BRK) which will corrupt
exceptions registers which have not yet been read.
Prevent this by marking all functions in entry-common.c as noinstr to
prevent compiler instrumentation. This also blacklists the functions for
tracing and kprobes, so we don't need to handle that separately.
Functions elsewhere will be dealt with in subsequent patches.
Signed-off-by: Mark Rutland <mark.rutland@arm.com>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: James Morse <james.morse@arm.com>
Cc: Will Deacon <will@kernel.org>
Link: https://lore.kernel.org/r/20201130115950.22492-4-mark.rutland@arm.com
Signed-off-by: Will Deacon <will@kernel.org>
Core code disables RCU when calling arch_cpu_idle(), so it's not safe
for arch_cpu_idle() or its calees to be instrumented, as the
instrumentation callbacks may attempt to use RCU or other features which
are unsafe to use in this context.
Mark them noinstr to prevent issues.
The use of local_irq_enable() in arch_cpu_idle() is similarly
problematic, and the "sched/idle: Fix arch_cpu_idle() vs tracing" patch
queued in the tip tree addresses that case.
Reported-by: Marco Elver <elver@google.com>
Signed-off-by: Mark Rutland <mark.rutland@arm.com>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: James Morse <james.morse@arm.com>
Cc: Will Deacon <will@kernel.org>
Link: https://lore.kernel.org/r/20201130115950.22492-3-mark.rutland@arm.com
Signed-off-by: Will Deacon <will@kernel.org>
In el0_svc_common() we unmask exceptions before we call user_exit(), and
so there's a window where an IRQ or debug exception can be taken while
RCU is not watching. In do_debug_exception() we account for this in via
debug_exception_{enter,exit}(), but in the el1_irq asm we do not and we
call trace functions which rely on RCU before we have a guarantee that
RCU is watching.
Let's avoid this by having el0_svc_common() exit userspace before
unmasking exceptions, matching what we do for all other EL0 entry paths.
We can use user_exit_irqoff() to avoid the pointless save/restore of IRQ
flags while we're sure exceptions are masked in DAIF.
The workaround for Cortex-A76 erratum 1463225 may trigger a debug
exception before this point, but the debug code invoked in this case is
safe even when RCU is not watching.
Signed-off-by: Mark Rutland <mark.rutland@arm.com>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: James Morse <james.morse@arm.com>
Cc: Will Deacon <will@kernel.org>
Link: https://lore.kernel.org/r/20201130115950.22492-2-mark.rutland@arm.com
Signed-off-by: Will Deacon <will@kernel.org>
Pull locking fixes from Thomas Gleixner:
"Two more places which invoke tracing from RCU disabled regions in the
idle path.
Similar to the entry path the low level idle functions have to be
non-instrumentable"
* tag 'locking-urgent-2020-11-29' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
intel_idle: Fix intel_idle() vs tracing
sched/idle: Fix arch_cpu_idle() vs tracing
Our Meltdown mitigation state isn't exposed outside of cpufeature.c,
contrary to the rest of the Spectre mitigation state. As we are going
to use it in KVM, expose a arm64_get_meltdown_state() helper which
returns the same possible values as arm64_get_spectre_v?_state().
Signed-off-by: Marc Zyngier <maz@kernel.org>
