When the kernel is compiled with pointer auth instructions, the boot CPU
needs to start using address auth very early, so change the cpucap to
account for this.

Pointer auth must be enabled before we call C functions, because it is
not possible to enter a function with pointer auth disabled and exit it
with pointer auth enabled. Note, mismatches between architected and
IMPDEF algorithms will still be caught by the cpufeature framework (the
separate *_ARCH and *_IMP_DEF cpucaps).

Note the change in behavior: if the boot CPU has address auth and a
late CPU does not, then the late CPU is parked by the cpufeature
framework. This is possible as kernel will only have NOP space intructions
for PAC so such mismatched late cpu will silently ignore those
instructions in C functions. Also, if the boot CPU does not have address
auth and the late CPU has then the late cpu will still boot but with
ptrauth feature disabled.

Leave generic authentication as a "system scope" cpucap for now, since
initially the kernel will only use address authentication.
Reviewed-by: NKees Cook <keescook@chromium.org>
Reviewed-by: NSuzuki K Poulose <suzuki.poulose@arm.com>
Reviewed-by: NVincenzo Frascino <Vincenzo.Frascino@arm.com>
Signed-off-by: NKristina Martsenko <kristina.martsenko@arm.com>
[Amit: Re-worked ptrauth setup logic, comments]
Signed-off-by: NAmit Daniel Kachhap <amit.kachhap@arm.com>
Signed-off-by: NCatalin Marinas <catalin.marinas@arm.com>

Each system capability can be of either boot, local, or system scope,
depending on when the state of the capability is finalized. When we
detect a conflict on a late CPU, we either offline the CPU or panic the
system. We currently always panic if the conflict is caused by a boot
scope capability, and offline the CPU if the conflict is caused by a
local or system scope capability.

We're going to want to add a new capability (for pointer authentication)
which needs to be boot scope but doesn't need to panic the system when a
conflict is detected. So add a new flag to specify whether the
capability requires the system to panic or not. Current boot scope
capabilities are updated to set the flag, so there should be no
functional change as a result of this patch.
Signed-off-by: NAmit Daniel Kachhap <amit.kachhap@arm.com>
Signed-off-by: NKristina Martsenko <kristina.martsenko@arm.com>
Reviewed-by: NVincenzo Frascino <vincenzo.frascino@arm.com>
Reviewed-by: NSuzuki K Poulose <suzuki.poulose@arm.com>
Reviewed-by: NKees Cook <keescook@chromium.org>
Signed-off-by: NCatalin Marinas <catalin.marinas@arm.com>

These helpers are used only by functions inside cpufeature.c and
hence makes sense to be moved from cpufeature.h to cpufeature.c as
they are not expected to be used globally.

This change helps in reducing the header file size as well as to add
future cpu capability types without confusion. Only a cpu capability
type macro is sufficient to expose those capabilities globally.
Signed-off-by: NAmit Daniel Kachhap <amit.kachhap@arm.com>
Reviewed-by: NVincenzo Frascino <Vincenzo.Frascino@arm.com>
Signed-off-by: NCatalin Marinas <catalin.marinas@arm.com>

This patch allows __cpu_setup to be invoked with one of these flags,
ARM64_CPU_BOOT_PRIMARY, ARM64_CPU_BOOT_SECONDARY or ARM64_CPU_RUNTIME.
This is required as some cpufeatures need different handling during
different scenarios.

The input parameter in x0 is preserved till the end to be used inside
this function.

There should be no functional change with this patch and is useful
for the subsequent ptrauth patch which utilizes it. Some upcoming
arm cpufeatures can also utilize these flags.
Suggested-by: NJames Morse <james.morse@arm.com>
Signed-off-by: NAmit Daniel Kachhap <amit.kachhap@arm.com>
Reviewed-by: NVincenzo Frascino <Vincenzo.Frascino@arm.com>
Reviewed-by: NJames Morse <james.morse@arm.com>
Reviewed-by: NSuzuki K Poulose <suzuki.poulose@arm.com>
Signed-off-by: NCatalin Marinas <catalin.marinas@arm.com>

As we're going to enable pointer auth within the kernel and use a
different APIAKey for the kernel itself, so move the user APIAKey
switch to EL0 exception return.

The other 4 keys could remain switched during task switch, but are also
moved to keep things consistent.
Reviewed-by: NKees Cook <keescook@chromium.org>
Reviewed-by: NJames Morse <james.morse@arm.com>
Reviewed-by: NVincenzo Frascino <Vincenzo.Frascino@arm.com>
Signed-off-by: NKristina Martsenko <kristina.martsenko@arm.com>
[Amit: commit msg, re-positioned the patch, comments]
Signed-off-by: NAmit Daniel Kachhap <amit.kachhap@arm.com>
Signed-off-by: NCatalin Marinas <catalin.marinas@arm.com>

We currently enable ptrauth for userspace, but do not use it within the
kernel. We're going to enable it for the kernel, and will need to manage
a separate set of ptrauth keys for the kernel.

We currently keep all 5 keys in struct ptrauth_keys. However, as the
kernel will only need to use 1 key, it is a bit wasteful to allocate a
whole ptrauth_keys struct for every thread.

Therefore, a subsequent patch will define a separate struct, with only 1
key, for the kernel. In preparation for that, rename the existing struct
(and associated macros and functions) to reflect that they are specific
to userspace.
Acked-by: NCatalin Marinas <catalin.marinas@arm.com>
Reviewed-by: NVincenzo Frascino <Vincenzo.Frascino@arm.com>
Signed-off-by: NKristina Martsenko <kristina.martsenko@arm.com>
[Amit: Re-positioned the patch to reduce the diff]
Signed-off-by: NAmit Daniel Kachhap <amit.kachhap@arm.com>
Signed-off-by: NCatalin Marinas <catalin.marinas@arm.com>

To enable pointer auth for the kernel, we're going to need to check for
the presence of address auth and generic auth using alternative_if. We
currently have two cpucaps for each, but alternative_if needs to check a
single cpucap. So define meta-capabilities that are present when either
of the current two capabilities is present.

Leave the existing four cpucaps in place, as they are still needed to
check for mismatched systems where one CPU has the architected algorithm
but another has the IMP DEF algorithm.

Note, the meta-capabilities were present before but were removed in
commit a56005d3 ("arm64: cpufeature: Reduce number of pointer auth
CPU caps from 6 to 4") and commit 1e013d06 ("arm64: cpufeature: Rework
ptr auth hwcaps using multi_entry_cap_matches"), as they were not needed
then. Note, unlike before, the current patch checks the cpucap values
directly, instead of reading the CPU ID register value.
Reviewed-by: NSuzuki K Poulose <suzuki.poulose@arm.com>
Reviewed-by: NKees Cook <keescook@chromium.org>
Reviewed-by: NVincenzo Frascino <Vincenzo.Frascino@arm.com>
Signed-off-by: NKristina Martsenko <kristina.martsenko@arm.com>
[Amit: commit message and macro rebase, use __system_matches_cap]
Signed-off-by: NAmit Daniel Kachhap <amit.kachhap@arm.com>
Signed-off-by: NCatalin Marinas <catalin.marinas@arm.com>

Add brackets around the evaluation of the 'addr' parameter to the
untagged_addr() macro so that the cast to 'u64' applies to the result
of the expression.

Cc: <stable@vger.kernel.org>
Fixes: 597399d0 ("arm64: tags: Preserve tags for addresses translated via TTBR1")
Reported-by: NLinus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: NWill Deacon <will@kernel.org>

Commit e0d5896b ("arm64: lse: fix LSE atomics with LLVM's integrated
assembler") broke the build when clang is used in connjunction with the
binutils assembler ("-no-integrated-as"). This happens because
__LSE_PREAMBLE is defined as ".arch armv8-a+lse", which overrides the
version of the CPU architecture passed via the "-march" paramter to gas:

$ aarch64-none-linux-gnu-as -EL -I ./arch/arm64/include
                                -I ./arch/arm64/include/generated
                                -I ./include -I ./include
                                -I ./arch/arm64/include/uapi
                                -I ./arch/arm64/include/generated/uapi
                                -I ./include/uapi -I ./include/generated/uapi
                                -I ./init -I ./init
                                -march=armv8.3-a -o init/do_mounts.o
                                /tmp/do_mounts-d7992a.s
/tmp/do_mounts-d7992a.s: Assembler messages:
/tmp/do_mounts-d7992a.s:1959: Error: selected processor does not support `autiasp'
/tmp/do_mounts-d7992a.s:2021: Error: selected processor does not support `paciasp'
/tmp/do_mounts-d7992a.s:2157: Error: selected processor does not support `autiasp'
/tmp/do_mounts-d7992a.s:2175: Error: selected processor does not support `paciasp'
/tmp/do_mounts-d7992a.s:2494: Error: selected processor does not support `autiasp'

Fix the issue by replacing ".arch armv8-a+lse" with ".arch_extension lse".
Sami confirms that the clang integrated assembler does now support the
'.arch_extension' directive, so this change will be fine even for LTO
builds in future.

Fixes: e0d5896b ("arm64: lse: fix LSE atomics with LLVM's integrated assembler")
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: Will Deacon <will@kernel.org>
Reported-by: NAmit Kachhap <Amit.Kachhap@arm.com>
Tested-by: NSami Tolvanen <samitolvanen@google.com>
Signed-off-by: NVincenzo Frascino <vincenzo.frascino@arm.com>
Signed-off-by: NWill Deacon <will@kernel.org>

The commit f5bfdc8e ("locking/osq: Use optimized spinning loop for
arm64") introduced a warning from Clang because vcpu_is_preempted() is
compiled away,

kernel/locking/osq_lock.c:25:19: warning: unused function 'node_cpu'
[-Wunused-function]
static inline int node_cpu(struct optimistic_spin_node *node)
                  ^
1 warning generated.

Fix it by converting vcpu_is_preempted() to a static inline function.

Fixes: f5bfdc8e ("locking/osq: Use optimized spinning loop for arm64")
Acked-by: NWaiman Long <longman@redhat.com>
Signed-off-by: NQian Cai <cai@lca.pw>
Signed-off-by: NWill Deacon <will@kernel.org>

There is a redundant do_sp_pc_abort() declaration in exceptions.h which can
be removed. Also do_el0_ia_bp_hardening() as been already been dropped with
the commit bfe29874 ("arm64: entry-common: don't touch daif before
bp-hardening") and hence does not need a declaration any more. This should
not introduce any functional change.

Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: Will Deacon <will@kernel.org>
Cc: James Morse <james.morse@arm.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: linux-kernel@vger.kernel.org
Acked-by: NMark Rutland <mark.rutland@arm.com>
Signed-off-by: NAnshuman Khandual <anshuman.khandual@arm.com>
Signed-off-by: NWill Deacon <will@kernel.org>

V{PEND,PROP}BASER registers are actually located in VLPI_base frame
of the *redistributor*. Rename their accessors to reflect this fact.

No functional changes.
Signed-off-by: NZenghui Yu <yuzenghui@huawei.com>
Signed-off-by: NMarc Zyngier <maz@kernel.org>
Link: https://lore.kernel.org/r/20200206075711.1275-7-yuzenghui@huawei.com

dma-continuguous.h is generic for all architectures except arm32 which has
its own version.

Similar change was done for msi.h by commit a1b39bae
("asm-generic: Make msi.h a mandatory include/asm header")
Suggested-by: NChristoph Hellwig <hch@infradead.org>
Link: https://lore.kernel.org/linux-arm-kernel/20200117080446.GA8980@lst.de/T/#m92bb56b04161057635d4142e1b3b9b6b0a70122eSigned-off-by: NMichal Simek <michal.simek@xilinx.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Acked-by: NThomas Gleixner <tglx@linutronix.de>
Acked-by: NArnd Bergmann <arnd@arndb.de>
Acked-by: Paul Walmsley <paul.walmsley@sifive.com> # for arch/riscv

Now walk_page_range() can walk kernel page tables, we can switch the arm64
ptdump code over to using it, simplifying the code.

Link: http://lkml.kernel.org/r/20191218162402.45610-22-steven.price@arm.comSigned-off-by: NSteven Price <steven.price@arm.com>
Reviewed-by: NCatalin Marinas <catalin.marinas@arm.com>
Cc: Albert Ou <aou@eecs.berkeley.edu>
Cc: Alexandre Ghiti <alex@ghiti.fr>
Cc: Andy Lutomirski <luto@kernel.org>
Cc: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Cc: Arnd Bergmann <arnd@arndb.de>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Christian Borntraeger <borntraeger@de.ibm.com>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: David S. Miller <davem@davemloft.net>
Cc: Heiko Carstens <heiko.carstens@de.ibm.com>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: James Hogan <jhogan@kernel.org>
Cc: James Morse <james.morse@arm.com>
Cc: Jerome Glisse <jglisse@redhat.com>
Cc: "Liang, Kan" <kan.liang@linux.intel.com>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: Paul Burton <paul.burton@mips.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Paul Walmsley <paul.walmsley@sifive.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Ralf Baechle <ralf@linux-mips.org>
Cc: Russell King <linux@armlinux.org.uk>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Vasily Gorbik <gor@linux.ibm.com>
Cc: Vineet Gupta <vgupta@synopsys.com>
Cc: Will Deacon <will@kernel.org>
Cc: Zong Li <zong.li@sifive.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

walk_page_range() is going to be allowed to walk page tables other than
those of user space.  For this it needs to know when it has reached a
'leaf' entry in the page tables.  This information will be provided by the
p?d_leaf() functions/macros.

For arm64, we already have p?d_sect() macros which we can reuse for
p?d_leaf().

pud_sect() is defined as a dummy function when CONFIG_PGTABLE_LEVELS < 3
or CONFIG_ARM64_64K_PAGES is defined.  However when the kernel is
configured this way then architecturally it isn't allowed to have a large
page at this level, and any code using these page walking macros is
implicitly relying on the page size/number of levels being the same as the
kernel.  So it is safe to reuse this for p?d_leaf() as it is an
architectural restriction.

Link: http://lkml.kernel.org/r/20191218162402.45610-5-steven.price@arm.comSigned-off-by: NSteven Price <steven.price@arm.com>
Acked-by: NCatalin Marinas <catalin.marinas@arm.com>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: Will Deacon <will@kernel.org>
Cc: Albert Ou <aou@eecs.berkeley.edu>
Cc: Alexandre Ghiti <alex@ghiti.fr>
Cc: Andy Lutomirski <luto@kernel.org>
Cc: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Cc: Arnd Bergmann <arnd@arndb.de>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Christian Borntraeger <borntraeger@de.ibm.com>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: David S. Miller <davem@davemloft.net>
Cc: Heiko Carstens <heiko.carstens@de.ibm.com>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: James Hogan <jhogan@kernel.org>
Cc: James Morse <james.morse@arm.com>
Cc: Jerome Glisse <jglisse@redhat.com>
Cc: "Liang, Kan" <kan.liang@linux.intel.com>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: Paul Burton <paul.burton@mips.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Paul Walmsley <paul.walmsley@sifive.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Ralf Baechle <ralf@linux-mips.org>
Cc: Russell King <linux@armlinux.org.uk>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Vasily Gorbik <gor@linux.ibm.com>
Cc: Vineet Gupta <vgupta@synopsys.com>
Cc: Zong Li <zong.li@sifive.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

For ring-based dirty log tracking, it will be more efficient to account
writes during schedule-out or schedule-in to the currently running VCPU.
We would like to do it even if the write doesn't use the current VCPU's
address space, as is the case for cached writes (see commit 4e335d9e,
"Revert "KVM: Support vCPU-based gfn->hva cache"", 2017-05-02).

Therefore, add a mechanism to track the currently-loaded kvm_vcpu struct.
There is already something similar in KVM/ARM; one important difference
is that kvm_arch_vcpu_{load,put} have two callers in virt/kvm/kvm_main.c:
we have to update both the architecture-independent vcpu_{load,put} and
the preempt notifiers.

Another change made in the process is to allow using kvm_get_running_vcpu()
in preemptible code.  This is allowed because preempt notifiers ensure
that the value does not change even after the VCPU thread is migrated.
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>
Reviewed-by: NPaolo Bonzini <pbonzini@redhat.com>
Signed-off-by: NPeter Xu <peterx@redhat.com>
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

Remove kvm_arch_vcpu_init() and kvm_arch_vcpu_uninit() now that all
arch specific implementations are nops.
Acked-by: NChristoffer Dall <christoffer.dall@arm.com>
Signed-off-by: NSean Christopherson <sean.j.christopherson@intel.com>
Reviewed-by: NCornelia Huck <cohuck@redhat.com>
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

Add an arm specific hook to free the arm64-only sve_state.  Doing so
eliminates the last functional code from kvm_arch_vcpu_uninit() across
all architectures and paves the way for removing kvm_arch_vcpu_init()
and kvm_arch_vcpu_uninit() entirely.
Signed-off-by: NSean Christopherson <sean.j.christopherson@intel.com>
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

Our MMIO handling is a bit odd, in the sense that it uses an
intermediate per-vcpu structure to store the various decoded
information that describe the access.

But the same information is readily available in the HSR/ESR_EL2
field, and we actually use this field to populate the structure.

Let's simplify the whole thing by getting rid of the superfluous
structure and save a (tiny) bit of space in the vcpu structure.

[32bit fix courtesy of Olof Johansson <olof@lixom.net>]
Signed-off-by: NMarc Zyngier <maz@kernel.org>

Since commit:

  d44f1b8d ("arm64: KVM/mm: Move SEA handling behind a single 'claim' interface")

... the top-level APEI SEA handler has the shape:

1. current_flags = arch_local_save_flags()
2. local_daif_restore(DAIF_ERRCTX)
3. <GHES handler>
4. local_daif_restore(current_flags)

However, since commit:

  4a503217 ("arm64: irqflags: Use ICC_PMR_EL1 for interrupt masking")

... when pseudo-NMIs (pNMIs) are in use, arch_local_save_flags() will save
the PMR value rather than the DAIF flags.

The combination of these two commits means that the APEI SEA handler will
erroneously attempt to restore the PMR value into DAIF. Fix this by
factoring local_daif_save_flags() out of local_daif_save(), so that we
can consistently save DAIF in step #1, regardless of whether pNMIs are in
use.

Both commits were introduced concurrently in v5.0.

Cc: <stable@vger.kernel.org>
Fixes: 4a503217 ("arm64: irqflags: Use ICC_PMR_EL1 for interrupt masking")
Fixes: d44f1b8d ("arm64: KVM/mm: Move SEA handling behind a single 'claim' interface")
Signed-off-by: NMark Rutland <mark.rutland@arm.com>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: James Morse <james.morse@arm.com>
Cc: Julien Thierry <julien.thierry.kdev@gmail.com>
Cc: Will Deacon <will@kernel.org>
Signed-off-by: NWill Deacon <will@kernel.org>

GICv4.1 defines a new VPE table that is potentially shared between
both the ITSs and the redistributors, following complicated affinity
rules.

To make things more confusing, the programming of this table at
the redistributor level is reusing the GICv4.0 GICR_VPROPBASER register
for something completely different.

The code flow is somewhat complexified by the need to respect the
affinities required by the HW, meaning that tables can either be
inherited from a previously discovered ITS or redistributor.
Signed-off-by: NMarc Zyngier <maz@kernel.org>
Reviewed-by: NZenghui Yu <yuzenghui@huawei.com>
Link: https://lore.kernel.org/r/20191224111055.11836-6-maz@kernel.org

When seeding KALSR on a system where we have architecture level random
number generation make use of that entropy, mixing it in with the seed
passed by the bootloader. Since this is run very early in init before
feature detection is complete we open code rather than use archrandom.h.
Signed-off-by: NMark Brown <broonie@kernel.org>
Reviewed-by: NMark Rutland <mark.rutland@arm.com>
Reviewed-by: NArd Biesheuvel <ardb@kernel.org>
Signed-off-by: NWill Deacon <will@kernel.org>

Expose the ID_AA64ISAR0.RNDR field to userspace, as the RNG system
registers are always available at EL0.

Implement arch_get_random_seed_long using RNDR.  Given that the
TRNG is likely to be a shared resource between cores, and VMs,
do not explicitly force re-seeding with RNDRRS.  In order to avoid
code complexity and potential issues with hetrogenous systems only
provide values after cpufeature has finalized the system capabilities.
Signed-off-by: NRichard Henderson <richard.henderson@linaro.org>
[Modified to only function after cpufeature has finalized the system
capabilities and move all the code into the header -- broonie]
Signed-off-by: NMark Brown <broonie@kernel.org>
Reviewed-by: NMark Rutland <mark.rutland@arm.com>
Reviewed-by: NArd Biesheuvel <ardb@kernel.org>
[will: Advertise HWCAP via /proc/cpuinfo]
Signed-off-by: NWill Deacon <will@kernel.org>

Confusingly, there are three SPSR layouts that a kernel may need to deal
with:

(1) An AArch64 SPSR_ELx view of an AArch64 pstate
(2) An AArch64 SPSR_ELx view of an AArch32 pstate
(3) An AArch32 SPSR_* view of an AArch32 pstate

When the KVM AArch32 support code deals with SPSR_{EL2,HYP}, it's either
dealing with #2 or #3 consistently. On arm64 the PSR_AA32_* definitions
match the AArch64 SPSR_ELx view, and on arm the PSR_AA32_* definitions
match the AArch32 SPSR_* view.

However, when we inject an exception into an AArch32 guest, we have to
synthesize the AArch32 SPSR_* that the guest will see. Thus, an AArch64
host needs to synthesize layout #3 from layout #2.

This patch adds a new host_spsr_to_spsr32() helper for this, and makes
use of it in the KVM AArch32 support code. For arm64 we need to shuffle
the DIT bit around, and remove the SS bit, while for arm we can use the
value as-is.

I've open-coded the bit manipulation for now to avoid having to rework
the existing PSR_* definitions into PSR64_AA32_* and PSR32_AA32_*
definitions. I hope to perform a more thorough refactoring in future so
that we can handle pstate view manipulation more consistently across the
kernel tree.
Signed-off-by: NMark Rutland <mark.rutland@arm.com>
Signed-off-by: NMarc Zyngier <maz@kernel.org>
Reviewed-by: NAlexandru Elisei <alexandru.elisei@arm.com>
Cc: stable@vger.kernel.org
Link: https://lore.kernel.org/r/20200108134324.46500-4-mark.rutland@arm.com

When KVM injects an exception into a guest, it generates the CPSR value
from scratch, configuring CPSR.{M,A,I,T,E}, and setting all other
bits to zero.

This isn't correct, as the architecture specifies that some CPSR bits
are (conditionally) cleared or set upon an exception, and others are
unchanged from the original context.

This patch adds logic to match the architectural behaviour. To make this
simple to follow/audit/extend, documentation references are provided,
and bits are configured in order of their layout in SPSR_EL2. This
layout can be seen in the diagram on ARM DDI 0487E.a page C5-426.

Note that this code is used by both arm and arm64, and is intended to
fuction with the SPSR_EL2 and SPSR_HYP layouts.
Signed-off-by: NMark Rutland <mark.rutland@arm.com>
Signed-off-by: NMarc Zyngier <maz@kernel.org>
Reviewed-by: NAlexandru Elisei <alexandru.elisei@arm.com>
Cc: stable@vger.kernel.org
Link: https://lore.kernel.org/r/20200108134324.46500-3-mark.rutland@arm.com

On AArch64 you can do a sign-extended load to either a 32-bit or 64-bit
register, and we should only sign extend the register up to the width of
the register as specified in the operation (by using the 32-bit Wn or
64-bit Xn register specifier).

As it turns out, the architecture provides this decoding information in
the SF ("Sixty-Four" -- how cute...) bit.

Let's take advantage of this with the usual 32-bit/64-bit header file
dance and do the right thing on AArch64 hosts.
Signed-off-by: NChristoffer Dall <christoffer.dall@arm.com>
Signed-off-by: NMarc Zyngier <maz@kernel.org>
Cc: stable@vger.kernel.org
Link: https://lore.kernel.org/r/20191212195055.5541-1-christoffer.dall@arm.com

/* Background. */
For a very long time, extending openat(2) with new features has been
incredibly frustrating. This stems from the fact that openat(2) is
possibly the most famous counter-example to the mantra "don't silently
accept garbage from userspace" -- it doesn't check whether unknown flags
are present[1].

This means that (generally) the addition of new flags to openat(2) has
been fraught with backwards-compatibility issues (O_TMPFILE has to be
defined as __O_TMPFILE|O_DIRECTORY|[O_RDWR or O_WRONLY] to ensure old
kernels gave errors, since it's insecure to silently ignore the
flag[2]). All new security-related flags therefore have a tough road to
being added to openat(2).

Userspace also has a hard time figuring out whether a particular flag is
supported on a particular kernel. While it is now possible with
contemporary kernels (thanks to [3]), older kernels will expose unknown
flag bits through fcntl(F_GETFL). Giving a clear -EINVAL during
openat(2) time matches modern syscall designs and is far more
fool-proof.

In addition, the newly-added path resolution restriction LOOKUP flags
(which we would like to expose to user-space) don't feel related to the
pre-existing O_* flag set -- they affect all components of path lookup.
We'd therefore like to add a new flag argument.

Adding a new syscall allows us to finally fix the flag-ignoring problem,
and we can make it extensible enough so that we will hopefully never
need an openat3(2).

/* Syscall Prototype. */
  /*
   * open_how is an extensible structure (similar in interface to
   * clone3(2) or sched_setattr(2)). The size parameter must be set to
   * sizeof(struct open_how), to allow for future extensions. All future
   * extensions will be appended to open_how, with their zero value
   * acting as a no-op default.
   */
  struct open_how { /* ... */ };

  int openat2(int dfd, const char *pathname,
              struct open_how *how, size_t size);

/* Description. */
The initial version of 'struct open_how' contains the following fields:

  flags
    Used to specify openat(2)-style flags. However, any unknown flag
    bits or otherwise incorrect flag combinations (like O_PATH|O_RDWR)
    will result in -EINVAL. In addition, this field is 64-bits wide to
    allow for more O_ flags than currently permitted with openat(2).

  mode
    The file mode for O_CREAT or O_TMPFILE.

    Must be set to zero if flags does not contain O_CREAT or O_TMPFILE.

  resolve
    Restrict path resolution (in contrast to O_* flags they affect all
    path components). The current set of flags are as follows (at the
    moment, all of the RESOLVE_ flags are implemented as just passing
    the corresponding LOOKUP_ flag).

    RESOLVE_NO_XDEV       => LOOKUP_NO_XDEV
    RESOLVE_NO_SYMLINKS   => LOOKUP_NO_SYMLINKS
    RESOLVE_NO_MAGICLINKS => LOOKUP_NO_MAGICLINKS
    RESOLVE_BENEATH       => LOOKUP_BENEATH
    RESOLVE_IN_ROOT       => LOOKUP_IN_ROOT

open_how does not contain an embedded size field, because it is of
little benefit (userspace can figure out the kernel open_how size at
runtime fairly easily without it). It also only contains u64s (even
though ->mode arguably should be a u16) to avoid having padding fields
which are never used in the future.

Note that as a result of the new how->flags handling, O_PATH|O_TMPFILE
is no longer permitted for openat(2). As far as I can tell, this has
always been a bug and appears to not be used by userspace (and I've not
seen any problems on my machines by disallowing it). If it turns out
this breaks something, we can special-case it and only permit it for
openat(2) but not openat2(2).

After input from Florian Weimer, the new open_how and flag definitions
are inside a separate header from uapi/linux/fcntl.h, to avoid problems
that glibc has with importing that header.

/* Testing. */
In a follow-up patch there are over 200 selftests which ensure that this
syscall has the correct semantics and will correctly handle several
attack scenarios.

In addition, I've written a userspace library[4] which provides
convenient wrappers around openat2(RESOLVE_IN_ROOT) (this is necessary
because no other syscalls support RESOLVE_IN_ROOT, and thus lots of care
must be taken when using RESOLVE_IN_ROOT'd file descriptors with other
syscalls). During the development of this patch, I've run numerous
verification tests using libpathrs (showing that the API is reasonably
usable by userspace).

/* Future Work. */
Additional RESOLVE_ flags have been suggested during the review period.
These can be easily implemented separately (such as blocking auto-mount
during resolution).

Furthermore, there are some other proposed changes to the openat(2)
interface (the most obvious example is magic-link hardening[5]) which
would be a good opportunity to add a way for userspace to restrict how
O_PATH file descriptors can be re-opened.

Another possible avenue of future work would be some kind of
CHECK_FIELDS[6] flag which causes the kernel to indicate to userspace
which openat2(2) flags and fields are supported by the current kernel
(to avoid userspace having to go through several guesses to figure it
out).

[1]: https://lwn.net/Articles/588444/
[2]: https://lore.kernel.org/lkml/CA+55aFyyxJL1LyXZeBsf2ypriraj5ut1XkNDsunRBqgVjZU_6Q@mail.gmail.com
[3]: commit 629e014b ("fs: completely ignore unknown open flags")
[4]: https://sourceware.org/bugzilla/show_bug.cgi?id=17523
[5]: https://lore.kernel.org/lkml/20190930183316.10190-2-cyphar@cyphar.com/
[6]: https://youtu.be/ggD-eb3yPVsSuggested-by: NChristian Brauner <christian.brauner@ubuntu.com>
Signed-off-by: NAleksa Sarai <cyphar@cyphar.com>
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

For most of the exception entry code, <foo>_handler() is the first C
function called from the entry assembly in entry-common.c, and external
functions handling the bulk of the logic are called do_<foo>().

For consistency, apply this scheme to el0_svc_handler and
el0_svc_compat_handler, renaming them to do_el0_svc and
do_el0_svc_compat respectively.

There should be no functional change as a result of this patch.
Signed-off-by: NMark Rutland <mark.rutland@arm.com>
Reviewed-by: NAnshuman Khandual <anshuman.khandual@arm.com>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: James Morse <james.morse@arm.com>
Cc: Will Deacon <will@kernel.org>
Signed-off-by: NWill Deacon <will@kernel.org>

These days arm64 kernels are always SMP, and thus smp_dmb is an
overly-long way of writing dmb. Naturally, no-one uses it.

Remove the unused macro.
Signed-off-by: NMark Rutland <mark.rutland@arm.com>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: Will Deacon <will@kernel.org>
Signed-off-by: NWill Deacon <will@kernel.org>

We haven't needed the inherit_daif macro since commit:

  ed3768db ("arm64: entry: convert el1_sync to C")

... which converted all callers to C and the local_daif_inherit
function.

Remove the unused macro.
Signed-off-by: NMark 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>
Signed-off-by: NWill Deacon <will@kernel.org>

Currently, the arm64 __cpu_setup has hard-coded constants for the memory
attributes that go into the MAIR_EL1 register. Define proper macros in
asm/sysreg.h and make use of them in proc.S.
Signed-off-by: NCatalin Marinas <catalin.marinas@arm.com>
Signed-off-by: NWill Deacon <will@kernel.org>

The "silver" KRYO3XX and KRYO4XX CPU cores are not affected by Spectre
variant 2. Add them to spectre_v2 safe list to correct the spurious
ARM_SMCCC_ARCH_WORKAROUND_1 warning and vulnerability status reported
under sysfs.
Reviewed-by: NStephen Boyd <swboyd@chromium.org>
Tested-by: NStephen Boyd <swboyd@chromium.org>
Signed-off-by: NSai Prakash Ranjan <saiprakash.ranjan@codeaurora.org>
[will: tweaked commit message to remove stale mention of "gold" cores]
Signed-off-by: NWill Deacon <will@kernel.org>

Arm64 has a more optimized spinning loop (atomic_cond_read_acquire)
using wfe for spinlock that can boost performance of sibling threads
by putting the current cpu to a wait state that is broken only when
the monitored variable changes or an external event happens.

OSQ has a more complicated spinning loop. Besides the lock value, it
also checks for need_resched() and vcpu_is_preempted(). The check for
need_resched() is not a problem as it is only set by the tick interrupt
handler. That will be detected by the spinning cpu right after iret.

The vcpu_is_preempted() check, however, is a problem as changes to the
preempt state of of previous node will not affect the wait state. For
ARM64, vcpu_is_preempted is not currently defined and so is a no-op.
Will has indicated that he is planning to para-virtualize wfe instead
of defining vcpu_is_preempted for PV support. So just add a comment in
arch/arm64/include/asm/spinlock.h to indicate that vcpu_is_preempted()
should not be defined as suggested.

On a 2-socket 56-core 224-thread ARM64 system, a kernel mutex locking
microbenchmark was run for 10s with and without the patch. The
performance numbers before patch were:

Running locktest with mutex [runtime = 10s, load = 1]
Threads = 224, Min/Mean/Max = 316/123,143/2,121,269
Threads = 224, Total Rate = 2,757 kop/s; Percpu Rate = 12 kop/s

After patch, the numbers were:

Running locktest with mutex [runtime = 10s, load = 1]
Threads = 224, Min/Mean/Max = 334/147,836/1,304,787
Threads = 224, Total Rate = 3,311 kop/s; Percpu Rate = 15 kop/s

So there was about 20% performance improvement.
Signed-off-by: NWaiman Long <longman@redhat.com>
Signed-off-by: NPeter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: NWill Deacon <will@kernel.org>
Link: https://lkml.kernel.org/r/20200113150735.21956-1-longman@redhat.com

LLVM's integrated assembler fails with the following error when
building KVM:

  <inline asm>:12:6: error: expected absolute expression
   .if kvm_update_va_mask == 0
       ^
  <inline asm>:21:6: error: expected absolute expression
   .if kvm_update_va_mask == 0
       ^
  <inline asm>:24:2: error: unrecognized instruction mnemonic
          NOT_AN_INSTRUCTION
          ^
  LLVM ERROR: Error parsing inline asm

These errors come from ALTERNATIVE_CB and __ALTERNATIVE_CFG,
which test for the existence of the callback parameter in inline
assembly using the following expression:

  " .if " __stringify(cb) " == 0\n"

This works with GNU as, but isn't supported by LLVM. This change
splits __ALTERNATIVE_CFG and ALTINSTR_ENTRY into separate macros
to fix the LLVM build.

Link: https://github.com/ClangBuiltLinux/linux/issues/472Signed-off-by: NSami Tolvanen <samitolvanen@google.com>
Tested-by: NNick Desaulniers <ndesaulniers@google.com>
Reviewed-by: NKees Cook <keescook@chromium.org>
Signed-off-by: NWill Deacon <will@kernel.org>

Unlike gcc, clang considers each inline assembly block to be independent
and therefore, when using the integrated assembler for inline assembly,
any preambles that enable features must be repeated in each block.

This change defines __LSE_PREAMBLE and adds it to each inline assembly
block that has LSE instructions, which allows them to be compiled also
with clang's assembler.

Link: https://github.com/ClangBuiltLinux/linux/issues/671Signed-off-by: NSami Tolvanen <samitolvanen@google.com>
Tested-by: NAndrew Murray <andrew.murray@arm.com>
Tested-by: NKees Cook <keescook@chromium.org>
Reviewed-by: NAndrew Murray <andrew.murray@arm.com>
Reviewed-by: NKees Cook <keescook@chromium.org>
Reviewed-by: NNick Desaulniers <ndesaulniers@google.com>
Signed-off-by: NWill Deacon <will@kernel.org>

Apparently there exist certain workloads which rely heavily on software
checksumming, for which the generic do_csum() implementation becomes a
significant bottleneck. Therefore let's give arm64 its own optimised
version - for ease of maintenance this foregoes assembly or intrisics,
and is thus not actually arm64-specific, but does rely heavily on C
idioms that translate well to the A64 ISA and the typical load/store
capabilities of most ARMv8 CPU cores.

The resulting increase in checksum throughput scales nicely with buffer
size, tending towards 4x for a small in-order core (Cortex-A53), and up
to 6x or more for an aggressive big core (Ampere eMAG).
Reported-by: NLingyan Huang <huanglingyan2@huawei.com>
Tested-by: NLingyan Huang <huanglingyan2@huawei.com>
Signed-off-by: NRobin Murphy <robin.murphy@arm.com>
Signed-off-by: NWill Deacon <will@kernel.org>

Cortex-A55 erratum 1530923 allows TLB entries to be allocated as a
result of a speculative AT instruction. This may happen in the middle of
a guest world switch while the relevant VMSA configuration is in an
inconsistent state, leading to erroneous content being allocated into
TLBs.

The same workaround as is used for Cortex-A76 erratum 1165522
(WORKAROUND_SPECULATIVE_AT_VHE) can be used here. Note that this
mandates the use of VHE on affected parts.
Acked-by: NMarc Zyngier <maz@kernel.org>
Signed-off-by: NSteven Price <steven.price@arm.com>
Signed-off-by: NWill Deacon <will@kernel.org>

To match SPECULATIVE_AT_VHE let's also have a generic name for the NVHE
variant.
Acked-by: NMarc Zyngier <maz@kernel.org>
Reviewed-by: NSuzuki K Poulose <suzuki.poulose@arm.com>
Signed-off-by: NSteven Price <steven.price@arm.com>
Signed-off-by: NWill Deacon <will@kernel.org>

Cortex-A55 is affected by a similar erratum, so rename the existing
workaround for errarum 1165522 so it can be used for both errata.
Acked-by: NMarc Zyngier <maz@kernel.org>
Reviewed-by: NSuzuki K Poulose <suzuki.poulose@arm.com>
Signed-off-by: NSteven Price <steven.price@arm.com>
Signed-off-by: NWill Deacon <will@kernel.org>

Refactor the code which checks to see if we need to use non-global
mappings to use a variable instead of checking with the CPU capabilities
each time, doing the initial check for KPTI early in boot before we
start allocating memory so we still avoid transitioning to non-global
mappings in common cases.

Since this variable always matches our decision about non-global
mappings this means we can also combine arm64_kernel_use_ng_mappings()
and arm64_unmap_kernel_at_el0() into a single function, the variable
simply stores the result and the decision code is elsewhere. We could
just have the users check the variable directly but having a function
makes it clear that these uses are read-only.

The result is that we simplify the code a bit and reduces the amount of
code executed at runtime.
Signed-off-by: NMark Brown <broonie@kernel.org>
Reviewed-by: NSuzuki K Poulose <suzuki.poulose@arm.com>
Signed-off-by: NWill Deacon <will@kernel.org>