Aliasing attacks against CPU branch predictors can allow an attacker to
redirect speculative control flow on some CPUs and potentially divulge
information from one context to another.

This patch adds initial skeleton code behind a new Kconfig option to
enable implementation-specific mitigations against these attacks for
CPUs that are affected.
Co-developed-by: NMarc Zyngier <marc.zyngier@arm.com>
Signed-off-by: NWill Deacon <will.deacon@arm.com>
Signed-off-by: NCatalin Marinas <catalin.marinas@arm.com>

We will soon need to invoke a CPU-specific function pointer after changing
page tables, so move post_ttbr_update_workaround out into C code to make
this possible.
Signed-off-by: NMarc Zyngier <marc.zyngier@arm.com>
Signed-off-by: NWill Deacon <will.deacon@arm.com>
Signed-off-by: NCatalin Marinas <catalin.marinas@arm.com>

In order to invoke the CPU capability ->matches callback from the ->enable
callback for applying local-CPU workarounds, we need a handle on the
capability structure.

This patch passes a pointer to the capability structure to the ->enable
callback.
Reviewed-by: NSuzuki K Poulose <suzuki.poulose@arm.com>
Signed-off-by: NWill Deacon <will.deacon@arm.com>
Signed-off-by: NCatalin Marinas <catalin.marinas@arm.com>

For non-KASLR kernels where the KPTI behaviour has not been overridden
on the command line we can use ID_AA64PFR0_EL1.CSV3 to determine whether
or not we should unmap the kernel whilst running at EL0.
Reviewed-by: NSuzuki K Poulose <suzuki.poulose@arm.com>
Signed-off-by: NWill Deacon <will.deacon@arm.com>
Signed-off-by: NCatalin Marinas <catalin.marinas@arm.com>

Speculation attacks against the entry trampoline can potentially resteer
the speculative instruction stream through the indirect branch and into
arbitrary gadgets within the kernel.

This patch defends against these attacks by forcing a misprediction
through the return stack: a dummy BL instruction loads an entry into
the stack, so that the predicted program flow of the subsequent RET
instruction is to a branch-to-self instruction which is finally resolved
as a branch to the kernel vectors with speculation suppressed.
Signed-off-by: NWill Deacon <will.deacon@arm.com>
Signed-off-by: NCatalin Marinas <catalin.marinas@arm.com>

ARM v8.4 extensions add new neon instructions for performing a
multiplication of each FP16 element of one vector with the corresponding
FP16 element of a second vector, and to add or subtract this without an
intermediate rounding to the corresponding FP32 element in a third vector.

This patch detects this feature and let the userspace know about it via a
HWCAP bit and MRS emulation.

Cc: Dave Martin <Dave.Martin@arm.com>
Reviewed-by: NSuzuki K Poulose <suzuki.poulose@arm.com>
Signed-off-by: NDongjiu Geng <gengdongjiu@huawei.com>
Reviewed-by: NDave Martin <Dave.Martin@arm.com>
Signed-off-by: NCatalin Marinas <catalin.marinas@arm.com>

CPU_PM_CPU_IDLE_ENTER_RETENTION skips calling cpu_pm_enter() and
cpu_pm_exit(). By not calling cpu_pm functions in idle entry/exit
paths we can reduce the latency involved in entering and exiting
the low power idle state.

On ARM64 based Qualcomm server platform we measured below overhead
for calling cpu_pm_enter and cpu_pm_exit for retention states.

workload: stress --hdd #CPUs --hdd-bytes 32M  -t 30
	Average overhead of cpu_pm_enter - 1.2us
	Average overhead of cpu_pm_exit  - 3.1us
Acked-by: NSudeep Holla <sudeep.holla@arm.com>
Signed-off-by: NPrashanth Prakash <pprakash@codeaurora.org>
Signed-off-by: NCatalin Marinas <catalin.marinas@arm.com>

Currently, when using VA_BITS < 48, if the ID map text happens to be
placed in physical memory above VA_BITS, we increase the VA size (up to
48) and create a new table level, in order to map in the ID map text.
This is okay because the system always supports 48 bits of VA.

This patch extends the code such that if the system supports 52 bits of
VA, and the ID map text is placed that high up, then we increase the VA
size accordingly, up to 52.

One difference from the current implementation is that so far the
condition of VA_BITS < 48 has meant that the top level table is always
"full", with the maximum number of entries, and an extra table level is
always needed. Now, when VA_BITS = 48 (and using 64k pages), the top
level table is not full, and we simply need to increase the number of
entries in it, instead of creating a new table level.
Tested-by: NSuzuki K Poulose <suzuki.poulose@arm.com>
Reviewed-by: NSuzuki K Poulose <suzuki.poulose@arm.com>
Reviewed-by: NMarc Zyngier <marc.zyngier@arm.com>
Tested-by: NBob Picco <bob.picco@oracle.com>
Reviewed-by: NBob Picco <bob.picco@oracle.com>
Signed-off-by: NKristina Martsenko <kristina.martsenko@arm.com>
[catalin.marinas@arm.com: reduce arguments to __create_hyp_mappings()]
[catalin.marinas@arm.com: reworked/renamed __cpu_uses_extended_idmap_level()]
Signed-off-by: NCatalin Marinas <catalin.marinas@arm.com>

The top 4 bits of a 52-bit physical address are positioned at bits
12..15 of a page table entry. Introduce macros to convert between a
physical address and its placement in a table entry, and change all
macros/functions that access PTEs to use them.
Reviewed-by: NMarc Zyngier <marc.zyngier@arm.com>
Tested-by: NSuzuki K Poulose <suzuki.poulose@arm.com>
Reviewed-by: NSuzuki K Poulose <suzuki.poulose@arm.com>
Tested-by: NBob Picco <bob.picco@oracle.com>
Reviewed-by: NBob Picco <bob.picco@oracle.com>
Signed-off-by: NKristina Martsenko <kristina.martsenko@arm.com>
[catalin.marinas@arm.com: some long lines wrapped]
Signed-off-by: NCatalin Marinas <catalin.marinas@arm.com>

Instead of open coding the generation of page table entries, use the
macros/functions that exist for this - pfn_p*d and p*d_populate. Most
code in the kernel already uses these macros, this patch tries to fix
up the few places that don't. This is useful for the next patch in this
series, which needs to change the page table entry logic, and it's
better to have that logic in one place.

The KVM extended ID map is special, since we're creating a level above
CONFIG_PGTABLE_LEVELS and the required function isn't available. Leave
it as is and add a comment to explain it. (The normal kernel ID map code
doesn't need this change because its page tables are created in assembly
(__create_page_tables)).
Tested-by: NSuzuki K Poulose <suzuki.poulose@arm.com>
Reviewed-by: NSuzuki K Poulose <suzuki.poulose@arm.com>
Reviewed-by: NMarc Zyngier <marc.zyngier@arm.com>
Tested-by: NBob Picco <bob.picco@oracle.com>
Reviewed-by: NBob Picco <bob.picco@oracle.com>
Signed-off-by: NKristina Martsenko <kristina.martsenko@arm.com>
Signed-off-by: NCatalin Marinas <catalin.marinas@arm.com>

The top 4 bits of a 52-bit physical address are positioned at bits
12..15 in page table entries. Introduce a macro to move the bits there,
and change the early ID map and swapper table setup code to use it.
Tested-by: NSuzuki K Poulose <suzuki.poulose@arm.com>
Reviewed-by: NSuzuki K Poulose <suzuki.poulose@arm.com>
Reviewed-by: NMarc Zyngier <marc.zyngier@arm.com>
Tested-by: NBob Picco <bob.picco@oracle.com>
Reviewed-by: NBob Picco <bob.picco@oracle.com>
Signed-off-by: NKristina Martsenko <kristina.martsenko@arm.com>
[catalin.marinas@arm.com: additional comments for clarification]
Signed-off-by: NCatalin Marinas <catalin.marinas@arm.com>

The top 4 bits of a 52-bit physical address are positioned at bits 2..5
in the TTBR registers. Introduce a couple of macros to move the bits
there, and change all TTBR writers to use them.

Leave TTBR0 PAN code unchanged, to avoid complicating it. A system with
52-bit PA will have PAN anyway (because it's ARMv8.1 or later), and a
system without 52-bit PA can only use up to 48-bit PAs. A later patch in
this series will add a kconfig dependency to ensure PAN is configured.

In addition, when using 52-bit PA there is a special alignment
requirement on the top-level table. We don't currently have any VA_BITS
configuration that would violate the requirement, but one could be added
in the future, so add a compile-time BUG_ON to check for it.
Tested-by: NSuzuki K Poulose <suzuki.poulose@arm.com>
Reviewed-by: NSuzuki K Poulose <suzuki.poulose@arm.com>
Reviewed-by: NMarc Zyngier <marc.zyngier@arm.com>
Tested-by: NBob Picco <bob.picco@oracle.com>
Reviewed-by: NBob Picco <bob.picco@oracle.com>
Signed-off-by: NKristina Martsenko <kristina.martsenko@arm.com>
[catalin.marinas@arm.com: added TTBR_BADD_MASK_52 comment]
Signed-off-by: NCatalin Marinas <catalin.marinas@arm.com>

The literal pool entry for identifying the vectors base is the only piece
of information in the trampoline page that identifies the true location
of the kernel.

This patch moves it into a page-aligned region of the .rodata section
and maps this adjacent to the trampoline text via an additional fixmap
entry, which protects against any accidental leakage of the trampoline
contents.
Suggested-by: NArd Biesheuvel <ard.biesheuvel@linaro.org>
Tested-by: NLaura Abbott <labbott@redhat.com>
Tested-by: NShanker Donthineni <shankerd@codeaurora.org>
Signed-off-by: NWill Deacon <will.deacon@arm.com>

There are now a handful of open-coded masks to extract the ASID from a
TTBR value, so introduce a TTBR_ASID_MASK and use that instead.
Suggested-by: NMark Rutland <mark.rutland@arm.com>
Reviewed-by: NMark Rutland <mark.rutland@arm.com>
Tested-by: NLaura Abbott <labbott@redhat.com>
Tested-by: NShanker Donthineni <shankerd@codeaurora.org>
Signed-off-by: NWill Deacon <will.deacon@arm.com>

Allow explicit disabling of the entry trampoline on the kernel command
line (kpti=off) by adding a fake CPU feature (ARM64_UNMAP_KERNEL_AT_EL0)
that can be used to toggle the alternative sequences in our entry code and
avoid use of the trampoline altogether if desired. This also allows us to
make use of a static key in arm64_kernel_unmapped_at_el0().
Reviewed-by: NMark Rutland <mark.rutland@arm.com>
Tested-by: NLaura Abbott <labbott@redhat.com>
Tested-by: NShanker Donthineni <shankerd@codeaurora.org>
Signed-off-by: NWill Deacon <will.deacon@arm.com>

When unmapping the kernel at EL0, we use tpidrro_el0 as a scratch register
during exception entry from native tasks and subsequently zero it in
the kernel_ventry macro. We can therefore avoid zeroing tpidrro_el0
in the context-switch path for native tasks using the entry trampoline.
Reviewed-by: NMark Rutland <mark.rutland@arm.com>
Tested-by: NLaura Abbott <labbott@redhat.com>
Tested-by: NShanker Donthineni <shankerd@codeaurora.org>
Signed-off-by: NWill Deacon <will.deacon@arm.com>

We rely on an atomic swizzling of TTBR1 when transitioning from the entry
trampoline to the kernel proper on an exception. We can't rely on this
atomicity in the face of Falkor erratum #E1003, so on affected cores we
can issue a TLB invalidation to invalidate the walk cache prior to
jumping into the kernel. There is still the possibility of a TLB conflict
here due to conflicting walk cache entries prior to the invalidation, but
this doesn't appear to be the case on these CPUs in practice.
Reviewed-by: NMark Rutland <mark.rutland@arm.com>
Tested-by: NLaura Abbott <labbott@redhat.com>
Tested-by: NShanker Donthineni <shankerd@codeaurora.org>
Signed-off-by: NWill Deacon <will.deacon@arm.com>

Hook up the entry trampoline to our exception vectors so that all
exceptions from and returns to EL0 go via the trampoline, which swizzles
the vector base register accordingly. Transitioning to and from the
kernel clobbers x30, so we use tpidrro_el0 and far_el1 as scratch
registers for native tasks.
Reviewed-by: NMark Rutland <mark.rutland@arm.com>
Tested-by: NLaura Abbott <labbott@redhat.com>
Tested-by: NShanker Donthineni <shankerd@codeaurora.org>
Signed-off-by: NWill Deacon <will.deacon@arm.com>

We will need to treat exceptions from EL0 differently in kernel_ventry,
so rework the macro to take the exception level as an argument and
construct the branch target using that.
Reviewed-by: NMark Rutland <mark.rutland@arm.com>
Tested-by: NLaura Abbott <labbott@redhat.com>
Tested-by: NShanker Donthineni <shankerd@codeaurora.org>
Signed-off-by: NWill Deacon <will.deacon@arm.com>

The exception entry trampoline needs to be mapped at the same virtual
address in both the trampoline page table (which maps nothing else)
and also the kernel page table, so that we can swizzle TTBR1_EL1 on
exceptions from and return to EL0.

This patch maps the trampoline at a fixed virtual address in the fixmap
area of the kernel virtual address space, which allows the kernel proper
to be randomized with respect to the trampoline when KASLR is enabled.
Reviewed-by: NMark Rutland <mark.rutland@arm.com>
Tested-by: NLaura Abbott <labbott@redhat.com>
Tested-by: NShanker Donthineni <shankerd@codeaurora.org>
Signed-off-by: NWill Deacon <will.deacon@arm.com>

To allow unmapping of the kernel whilst running at EL0, we need to
point the exception vectors at an entry trampoline that can map/unmap
the kernel on entry/exit respectively.

This patch adds the trampoline page, although it is not yet plugged
into the vector table and is therefore unused.
Reviewed-by: NMark Rutland <mark.rutland@arm.com>
Tested-by: NLaura Abbott <labbott@redhat.com>
Tested-by: NShanker Donthineni <shankerd@codeaurora.org>
Signed-off-by: NWill Deacon <will.deacon@arm.com>

With the ASID now installed in TTBR1, we can re-enable ARM64_SW_TTBR0_PAN
by ensuring that we switch to a reserved ASID of zero when disabling
user access and restore the active user ASID on the uaccess enable path.
Reviewed-by: NMark Rutland <mark.rutland@arm.com>
Tested-by: NLaura Abbott <labbott@redhat.com>
Tested-by: NShanker Donthineni <shankerd@codeaurora.org>
Signed-off-by: NWill Deacon <will.deacon@arm.com>

The post_ttbr0_update_workaround hook applies to any change to TTBRx_EL1.
Since we're using TTBR1 for the ASID, rename the hook to make it clearer
as to what it's doing.
Reviewed-by: NMark Rutland <mark.rutland@arm.com>
Tested-by: NLaura Abbott <labbott@redhat.com>
Tested-by: NShanker Donthineni <shankerd@codeaurora.org>
Signed-off-by: NWill Deacon <will.deacon@arm.com>

When deciding whether to invalidate FPSIMD state cached in the cpu,
the backend function sve_flush_cpu_state() attempts to dereference
__this_cpu_read(fpsimd_last_state).  However, this is not safe:
there is no guarantee that this task_struct pointer is still valid,
because the task could have exited in the meantime.

This means that we need another means to get the appropriate value
of TIF_SVE for the associated task.

This patch solves this issue by adding a cached copy of the TIF_SVE
flag in fpsimd_last_state, which we can check without dereferencing
the task pointer.

In particular, although this patch is not a KVM fix per se, this
means that this check is now done safely in the KVM world switch
path (which is currently the only user of this code).
Signed-off-by: NDave Martin <Dave.Martin@arm.com>
Cc: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Cc: Christoffer Dall <christoffer.dall@linaro.org>
Cc: Marc Zyngier <marc.zyngier@arm.com>
Signed-off-by: NWill Deacon <will.deacon@arm.com>

There is currently some duplicate logic to associate current's
FPSIMD context with the cpu when loading FPSIMD state into the cpu
regs.

Subsequent patches will update that logic, so in order to ensure it
only needs to be done in one place, this patch factors the relevant
code out into a new function fpsimd_bind_to_cpu().
Signed-off-by: NDave Martin <Dave.Martin@arm.com>
Reviewed-by: NArd Biesheuvel <ard.biesheuvel@linaro.org>
Signed-off-by: NWill Deacon <will.deacon@arm.com>

Currently, loading of a task's fpsimd state into the CPU registers
is skipped if that task's state is already present in the registers
of that CPU.

However, the code relies on the struct fpsimd_state * (and by
extension struct task_struct *) to unambiguously identify a task.

There is a particular case in which this doesn't work reliably:
when a task exits, its task_struct may be recycled to describe a
new task.

Consider the following scenario:

 1) Task P loads its fpsimd state onto cpu C.
        per_cpu(fpsimd_last_state, C) := P;
        P->thread.fpsimd_state.cpu := C;

 2) Task X is scheduled onto C and loads its fpsimd state on C.
        per_cpu(fpsimd_last_state, C) := X;
        X->thread.fpsimd_state.cpu := C;

 3) X exits, causing X's task_struct to be freed.

 4) P forks a new child T, which obtains X's recycled task_struct.
	T == X.
	T->thread.fpsimd_state.cpu == C (inherited from P).

 5) T is scheduled on C.
	T's fpsimd state is not loaded, because
	per_cpu(fpsimd_last_state, C) == T (== X) &&
	T->thread.fpsimd_state.cpu == C.

        (This is the check performed by fpsimd_thread_switch().)

So, T gets X's registers because the last registers loaded onto C
were those of X, in (2).

This patch fixes the problem by ensuring that the sched-in check
fails in (5): fpsimd_flush_task_state(T) is called when T is
forked, so that T->thread.fpsimd_state.cpu == C cannot be true.
This relies on the fact that T is not schedulable until after
copy_thread() completes.

Once T's fpsimd state has been loaded on some CPU C there may still
be other cpus D for which per_cpu(fpsimd_last_state, D) ==
&X->thread.fpsimd_state.  But D is necessarily != C in this case,
and the check in (5) must fail.

An alternative fix would be to do refcounting on task_struct.  This
would result in each CPU holding a reference to the last task whose
fpsimd state was loaded there.  It's not clear whether this is
preferable, and it involves higher overhead than the fix proposed
in this patch.  It would also move all the task_struct freeing
work into the context switch critical section, or otherwise some
deferred cleanup mechanism would need to be introduced, neither of
which seems obviously justified.

Cc: <stable@vger.kernel.org>
Fixes: 005f78cd ("arm64: defer reloading a task's FPSIMD state to userland resume")
Signed-off-by: NDave Martin <Dave.Martin@arm.com>
Reviewed-by: NArd Biesheuvel <ard.biesheuvel@linaro.org>
[will: word-smithed the comment so it makes more sense]
Signed-off-by: NWill Deacon <will.deacon@arm.com>

Building the kernel with an LTO-enabled GCC spits out the following "const"
warning for the cpu_ops code:

  mm/percpu.c:2168:20: error: pcpu_fc_names causes a section type conflict
  with dt_supported_cpu_ops
  const char * const pcpu_fc_names[PCPU_FC_NR] __initconst = {
          ^
  arch/arm64/kernel/cpu_ops.c:34:37: note: ‘dt_supported_cpu_ops’ was declared here
  static const struct cpu_operations *dt_supported_cpu_ops[] __initconst = {

Fix it by adding missed const qualifiers.
Signed-off-by: NYury Norov <ynorov@caviumnetworks.com>
Reviewed-by: NNick Desaulniers <ndesaulniers@google.com>
Signed-off-by: NWill Deacon <will.deacon@arm.com>

bus access read/write events are not supported in A73, based on the
Cortex-A73 TRM r0p2, section 11.9 Events (pages 11-457 to 11-460).

Fixes: 5561b6c5 "arm64: perf: add support for Cortex-A73"
Acked-by: NJulien Thierry <julien.thierry@arm.com>
Signed-off-by: NXu YiPing <xuyiping@hisilicon.com>
Signed-off-by: NWill Deacon <will.deacon@arm.com>

The fpsimd_update_current_state() function is responsible for
loading the FPSIMD state from the user signal frame into the
current task during sigreturn.  When implementing support for SVE,
conditional code was added to this function in order to handle the
case where SVE state need to be loaded for the task and merged with
the FPSIMD data from the signal frame; however, the FPSIMD-only
case was unintentionally dropped.

As a result of this, sigreturn does not currently restore the
FPSIMD state of the task, except in the case where the system
supports SVE and the signal frame contains SVE state in addition to
FPSIMD state.

This patch fixes this bug by making the copy-in of the FPSIMD data
from the signal frame to thread_struct unconditional.

This remains a performance regression from v4.14, since the FPSIMD
state is now copied into thread_struct and then loaded back,
instead of _only_ being loaded into the CPU FPSIMD registers.
However, it is essential to call task_fpsimd_load() here anyway in
order to ensure that the SVE enable bit in CPACR_EL1 is set
correctly before returning to userspace.  This could use some
refactoring, but since sigreturn is not a fast path I have kept
this patch as a pure fix and left the refactoring for later.

Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Fixes: 8cd969d2 ("arm64/sve: Signal handling support")
Reported-by: NAlex Bennée <alex.bennee@linaro.org>
Tested-by: NAlex Bennée <alex.bennee@linaro.org>
Reviewed-by: NAlex Bennée <alex.bennee@linaro.org>
Signed-off-by: NDave Martin <Dave.Martin@arm.com>
Signed-off-by: NWill Deacon <will.deacon@arm.com>

When building the arm64 kernel with both CONFIG_ARM64_MODULE_PLTS and
CONFIG_DYNAMIC_FTRACE enabled, the ftrace-mod.o object file is built
with the kernel and contains a trampoline that is linked into each
module, so that modules can be loaded far away from the kernel and
still reach the ftrace entry point in the core kernel with an ordinary
relative branch, as is emitted by the compiler instrumentation code
dynamic ftrace relies on.

In order to be able to build out of tree modules, this object file
needs to be included into the linux-headers or linux-devel packages,
which is undesirable, as it makes arm64 a special case (although a
precedent does exist for 32-bit PPC).

Given that the trampoline essentially consists of a PLT entry, let's
not bother with a source or object file for it, and simply patch it
in whenever the trampoline is being populated, using the existing
PLT support routines.

Cc: <stable@vger.kernel.org>
Signed-off-by: NArd Biesheuvel <ard.biesheuvel@linaro.org>
Signed-off-by: NWill Deacon <will.deacon@arm.com>

To allow the ftrace trampoline code to reuse the PLT entry routines,
factor it out and move it into asm/module.h.

Cc: <stable@vger.kernel.org>
Signed-off-by: NArd Biesheuvel <ard.biesheuvel@linaro.org>
Signed-off-by: NWill Deacon <will.deacon@arm.com>

This patch enables detection of hardware SVE support via the
cpufeatures framework, and reports its presence to the kernel and
userspace via the new ARM64_SVE cpucap and HWCAP_SVE hwcap
respectively.

Userspace can also detect SVE using ID_AA64PFR0_EL1, using the
cpufeatures MRS emulation.

When running on hardware that supports SVE, this enables runtime
kernel support for SVE, and allows user tasks to execute SVE
instructions and make of the of the SVE-specific user/kernel
interface extensions implemented by this series.
Signed-off-by: NDave Martin <Dave.Martin@arm.com>
Reviewed-by: NSuzuki K Poulose <suzuki.poulose@arm.com>
Reviewed-by: NCatalin Marinas <catalin.marinas@arm.com>
Signed-off-by: NWill Deacon <will.deacon@arm.com>

Until KVM has full SVE support, guests must not be allowed to
execute SVE instructions.

This patch enables the necessary traps, and also ensures that the
traps are disabled again on exit from the guest so that the host
can still use SVE if it wants to.

On guest exit, high bits of the SVE Zn registers may have been
clobbered as a side-effect the execution of FPSIMD instructions in
the guest.  The existing KVM host FPSIMD restore code is not
sufficient to restore these bits, so this patch explicitly marks
the CPU as not containing cached vector state for any task, thus
forcing a reload on the next return to userspace.  This is an
interim measure, in advance of adding full SVE awareness to KVM.

This marking of cached vector state in the CPU as invalid is done
using __this_cpu_write(fpsimd_last_state, NULL) in fpsimd.c.  Due
to the repeated use of this rather obscure operation, it makes
sense to factor it out as a separate helper with a clearer name.
This patch factors it out as fpsimd_flush_cpu_state(), and ports
all callers to use it.

As a side effect of this refactoring, a this_cpu_write() in
fpsimd_cpu_pm_notifier() is changed to __this_cpu_write().  This
should be fine, since cpu_pm_enter() is supposed to be called only
with interrupts disabled.
Signed-off-by: NDave Martin <Dave.Martin@arm.com>
Reviewed-by: NAlex Bennée <alex.bennee@linaro.org>
Reviewed-by: NChristoffer Dall <christoffer.dall@linaro.org>
Acked-by: NCatalin Marinas <catalin.marinas@arm.com>
Acked-by: NMarc Zyngier <marc.zyngier@arm.com>
Cc: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Signed-off-by: NWill Deacon <will.deacon@arm.com>

Because of the effect of SVE on the size of the signal frame, the
default vector length used for new processes involves a tradeoff
between performance of SVE-enabled software on the one hand, and
reliability of non-SVE-aware software on the other hand.

For this reason, the best choice depends on the repertoire of
userspace software in use and is thus best left up to distro
maintainers, sysadmins and developers.

If CONFIG_SYSCTL and CONFIG_PROC_SYSCTL are enabled, this patch
exposes the default vector length in
/proc/sys/abi/sve_default_vector_length, where boot scripts or the
adventurous can poke it.

In common with other arm64 ABI sysctls, this control is currently
global: setting it requires CAP_SYS_ADMIN in the root user
namespace, but the value set is effective for subsequent execs in
all namespaces.  The control only affects _new_ processes, however:
changing it does not affect the vector length of any existing
process.

The intended usage model is that if userspace is known to be fully
SVE-tolerant (or a developer is curious to find out) then this
parameter can be cranked up during system startup.
Signed-off-by: NDave Martin <Dave.Martin@arm.com>
Reviewed-by: NAlex Bennée <alex.bennee@linaro.org>
Reviewed-by: NCatalin Marinas <catalin.marinas@arm.com>
Signed-off-by: NWill Deacon <will.deacon@arm.com>

This patch adds two arm64-specific prctls, to permit userspace to
control its vector length:

 * PR_SVE_SET_VL: set the thread's SVE vector length and vector
   length inheritance mode.

 * PR_SVE_GET_VL: get the same information.

Although these prctls resemble instruction set features in the SVE
architecture, they provide additional control: the vector length
inheritance mode is Linux-specific and nothing to do with the
architecture, and the architecture does not permit EL0 to set its
own vector length directly.  Both can be used in portable tools
without requiring the use of SVE instructions.
Signed-off-by: NDave Martin <Dave.Martin@arm.com>
Reviewed-by: NCatalin Marinas <catalin.marinas@arm.com>
Cc: Alex Bennée <alex.bennee@linaro.org>
[will: Fixed up prctl constants to avoid clash with PDEATHSIG]
Signed-off-by: NWill Deacon <will.deacon@arm.com>

This patch defines and implements a new regset NT_ARM_SVE, which
describes a thread's SVE register state.  This allows a debugger to
manipulate the SVE state, as well as being included in ELF
coredumps for post-mortem debugging.

Because the regset size and layout are dependent on the thread's
current vector length, it is not possible to define a C struct to
describe the regset contents as is done for existing regsets.
Instead, and for the same reasons, NT_ARM_SVE is based on the
freeform variable-layout approach used for the SVE signal frame.

Additionally, to reduce debug overhead when debugging threads that
might or might not have live SVE register state, NT_ARM_SVE may be
presented in one of two different formats: the old struct
user_fpsimd_state format is embedded for describing the state of a
thread with no live SVE state, whereas a new variable-layout
structure is embedded for describing live SVE state.  This avoids a
debugger needing to poll NT_PRFPREG in addition to NT_ARM_SVE, and
allows existing userspace code to handle the non-SVE case without
too much modification.

For this to work, NT_ARM_SVE is defined with a fixed-format header
of type struct user_sve_header, which the recipient can use to
figure out the content, size and layout of the reset of the regset.
Accessor macros are defined to allow the vector-length-dependent
parts of the regset to be manipulated.
Signed-off-by: NAlan Hayward <alan.hayward@arm.com>
Signed-off-by: NDave Martin <Dave.Martin@arm.com>
Reviewed-by: NCatalin Marinas <catalin.marinas@arm.com>
Cc: Alex Bennée <alex.bennee@linaro.org>
Cc: Okamoto Takayuki <tokamoto@jp.fujitsu.com>
Signed-off-by: NWill Deacon <will.deacon@arm.com>

The EFI runtime services ABI allows EFI to make free use of the
FPSIMD registers during EFI runtime service calls, subject to the
callee-save requirements of the AArch64 procedure call standard.

However, the SVE architecture allows upper bits of the SVE vector
registers to be zeroed as a side-effect of FPSIMD V-register
writes.  This means that the SVE vector registers must be saved in
their entirety in order to avoid data loss: non-SVE-aware EFI
implementations cannot restore them correctly.

The non-IRQ case is already handled gracefully by
kernel_neon_begin().  For the IRQ case, this patch allocates a
suitable per-CPU stash buffer for the full SVE register state and
uses it to preserve the affected registers around EFI calls.  It is
currently unclear how the EFI runtime services ABI will be
clarified with respect to SVE, so it safest to assume that the
predicate registers and FFR must be saved and restored too.

No attempt is made to restore the restore the vector length after
a call, for now.  It is deemed rather insane for EFI to change it,
and contemporary EFI implementations certainly won't.
Signed-off-by: NDave Martin <Dave.Martin@arm.com>
Reviewed-by: NAlex Bennée <alex.bennee@linaro.org>
Reviewed-by: NCatalin Marinas <catalin.marinas@arm.com>
Cc: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Signed-off-by: NWill Deacon <will.deacon@arm.com>

Kernel-mode NEON will corrupt the SVE vector registers, due to the
way they alias the FPSIMD vector registers in the hardware.

This patch ensures that any live SVE register content for the task
is saved by kernel_neon_begin().  The data will be restored in the
usual way on return to userspace.
Signed-off-by: NDave Martin <Dave.Martin@arm.com>
Reviewed-by: NArd Biesheuvel <ard.biesheuvel@linaro.org>
Reviewed-by: NAlex Bennée <alex.bennee@linaro.org>
Reviewed-by: NCatalin Marinas <catalin.marinas@arm.com>
Signed-off-by: NWill Deacon <will.deacon@arm.com>

This patch uses the cpufeatures framework to determine common SVE
capabilities and vector lengths, and configures the runtime SVE
support code appropriately.

ZCR_ELx is not really a feature register, but it is convenient to
use it as a template for recording the maximum vector length
supported by a CPU, using the LEN field.  This field is similar to
a feature field in that it is a contiguous bitfield for which we
want to determine the minimum system-wide value.  This patch adds
ZCR as a pseudo-register in cpuinfo/cpufeatures, with appropriate
custom code to populate it.  Finding the minimum supported value of
the LEN field is left to the cpufeatures framework in the usual
way.

The meaning of ID_AA64ZFR0_EL1 is not architecturally defined yet,
so for now we just require it to be zero.

Note that much of this code is dormant and SVE still won't be used
yet, since system_supports_sve() remains hardwired to false.
Signed-off-by: NDave Martin <Dave.Martin@arm.com>
Reviewed-by: NSuzuki K Poulose <suzuki.poulose@arm.com>
Reviewed-by: NCatalin Marinas <catalin.marinas@arm.com>
Cc: Alex Bennée <alex.bennee@linaro.org>
Signed-off-by: NWill Deacon <will.deacon@arm.com>

update_cpu_features() currently cannot tell whether it is being
called during early or late secondary boot.  This doesn't
desperately matter for anything it currently does.

However, SVE will need to know here whether the set of available
vector lengths is known or still to be determined when booting a
CPU, so that it can be updated appropriately.

This patch simply moves the sys_caps_initialised stuff to the top
of the file so that it can be used more widely.  There doesn't seem
to be a more obvious place to put it.
Signed-off-by: NDave Martin <Dave.Martin@arm.com>
Reviewed-by: NAlex Bennée <alex.bennee@linaro.org>
Reviewed-by: NSuzuki K Poulose <suzuki.poulose@arm.com>
Acked-by: NCatalin Marinas <catalin.marinas@arm.com>
Signed-off-by: NWill Deacon <will.deacon@arm.com>