SIGSYS is primarily used in secure computing to notify tracer of syscall
events. This patch allows signal handler on compat task to get correct
information with SA_SIGINFO specified when this signal is delivered.
Reviewed-by: NKees Cook <keescook@chromium.org>
Signed-off-by: NAKASHI Takahiro <takahiro.akashi@linaro.org>
Signed-off-by: NWill Deacon <will.deacon@arm.com>

This patch allows compat task to issue seccomp() system call.
Reviewed-by: NKees Cook <keescook@chromium.org>
Signed-off-by: NAKASHI Takahiro <takahiro.akashi@linaro.org>
Signed-off-by: NWill Deacon <will.deacon@arm.com>

Those values (__NR_seccomp_*) are used solely in secure_computing()
to identify mode 1 system calls. If compat system calls have different
syscall numbers, asm/seccomp.h may override them.
Acked-by: NArnd Bergmann <arnd@arndb.de>
Reviewed-by: NKees Cook <keescook@chromium.org>
Signed-off-by: NAKASHI Takahiro <takahiro.akashi@linaro.org>
Signed-off-by: NWill Deacon <will.deacon@arm.com>

If tracer modifies a syscall number to -1, this traced system call should
be skipped with a return value specified in x0.
This patch implements this semantics.

Please note:
* syscall entry tracing and syscall exit tracing (ftrace tracepoint and
  audit) are always executed, if enabled, even when skipping a system call
  (that is, -1).
  In this way, we can avoid a potential bug where audit_syscall_entry()
  might be called without audit_syscall_exit() at the previous system call
  being called, that would cause OOPs in audit_syscall_entry().
Signed-off-by: NAKASHI Takahiro <takahiro.akashi@linaro.org>
[will: fixed up conflict with blr rework]
Signed-off-by: NWill Deacon <will.deacon@arm.com>

This regeset is intended to be used to get and set a system call number
while tracing.
There was some discussion about possible approaches to do so:

(1) modify x8 register with ptrace(PTRACE_SETREGSET) indirectly,
    and update regs->syscallno later on in syscall_trace_enter(), or
(2) define a dedicated regset for this purpose as on s390, or
(3) support ptrace(PTRACE_SET_SYSCALL) as on arch/arm

Thinking of the fact that user_pt_regs doesn't expose 'syscallno' to
tracer as well as that secure_computing() expects a changed syscall number,
especially case of -1, to be visible before this function returns in
syscall_trace_enter(), (1) doesn't work well.
We will take (2) since it looks much cleaner.
Signed-off-by: NAKASHI Takahiro <takahiro.akashi@linaro.org>
Signed-off-by: NWill Deacon <will.deacon@arm.com>

The head.text section is intended to be run at early bootup
before any of the regular kernel mappings have been setup.
Parts of head.text may be freed back into the buddy allocator
due to TEXT_OFFSET so for security requirements this memory
must not be executable. The suspend/resume/hotplug code path
requires some of these head.S functions to run however which
means they need to be executable. Support these conflicting
requirements by moving the few head.text functions that need
to be executable to the text section which has the appropriate
page table permissions.
Tested-by: NKees Cook <keescook@chromium.org>
Reviewed-by: NMark Rutland <mark.rutland@arm.com>
Tested-by: NMark Rutland <mark.rutland@arm.com>
Signed-off-by: NLaura Abbott <lauraa@codeaurora.org>
Signed-off-by: NWill Deacon <will.deacon@arm.com>

In the arm64 arch_static_branch implementation we place an A64 NOP into
the instruction stream and log relevant details to a jump_entry in a
__jump_table section. Later this may be replaced with an immediate
branch without link to the code for the unlikely case.

At init time, the core calls arch_jump_label_transform_static to
initialise the NOPs. On x86 this involves inserting the optimal NOP for
a given microarchitecture, but on arm64 we only use the architectural
NOP, and hence replace each NOP with the exact same NOP. This is
somewhat pointless.

Additionally, at module load time we don't call jump_label_apply_nops to
patch the optimal NOPs in, unlike other architectures, but get away with
this because we only use the architectural NOP anyway. A later notifier
will patch NOPs with branches as required.

Similarly to x86 commit 11570da1 (x86/jump-label: Do not bother
updating NOPs if they are correct), we can avoid patching NOPs with
identical NOPs. Given that we only use a single NOP encoding, this means
we can NOP-out the body of arch_jump_label_transform_static entirely. As
the default __weak arch_jump_label_transform_static implementation
performs a patch, we must use an empty function to achieve this.

Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: Jiang Liu <liuj97@gmail.com>
Cc: Laura Abbott <lauraa@codeaurora.org>
Acked-by: NWill Deacon <will.deacon@arm.com>
Signed-off-by: NMark Rutland <mark.rutland@arm.com>
Signed-off-by: NWill Deacon <will.deacon@arm.com>

In a similar manner to arm, it's useful to be able to dump the page
tables to verify permissions and memory types. Add a debugfs file
to check the page tables.
Acked-by: NSteve Capper <steve.capper@linaro.org>
Tested-by: NSteve Capper <steve.capper@linaro.org>
Reviewed-by: NMark Rutland <mark.rutland@arm.com>
Tested-by: NMark Rutland <mark.rutland@arm.com>
Signed-off-by: NLaura Abbott <lauraa@codeaurora.org>
[will: s/BUFFERABLE/NORMAL-NC/]
Signed-off-by: NWill Deacon <will.deacon@arm.com>

Every other architecture with permanent fixed addresses has
FIX_HOLE as the first entry. This seems to be designed as a
debugging aid but there are a couple of side effects of not
having FIX_HOLE:

- If the first fixed address is 0, fix_to_virt -> virt_to_fix
triggers a BUG_ON for the virtual address being equal to
FIXADDR_TOP
- fix_to_virt may return a value outside of FIXADDR_START
and FIXADDR_TOP which may look like a bug to a developer.

Match up with other architectures and make everything clearer
by adding FIX_HOLE.
Signed-off-by: NLaura Abbott <lauraa@codeaurora.org>
Signed-off-by: NWill Deacon <will.deacon@arm.com>

Consistently use the plural form for alternatives pr_fmt strings.
Signed-off-by: NWill Deacon <will.deacon@arm.com>

.exit.* sections may be subject to patching by the new alternatives
framework and so shouldn't be discarded at link-time. Without this patch,
such a section will result in the following linker error:

`.exit.text' referenced in section `.altinstructions' of
 drivers/built-in.o: defined in discarded section `.exit.text' of
drivers/built-in.o
Signed-off-by: NWill Deacon <will.deacon@arm.com>

The fixmap API was originally added for arm64 for
early_ioremap purposes. It can be used for other purposes too
so move the initialization from ioremap to somewhere more
generic. This makes it obvious where the fixmap is being set
up and allows for a cleaner implementation of __set_fixmap.
Reviewed-by: NKees Cook <keescook@chromium.org>
Acked-by: NMark Rutland <mark.rutland@arm.com>
Tested-by: NMark Rutland <mark.rutland@arm.com>
Tested-by: NKees Cook <keescook@chromium.org>
Signed-off-by: NLaura Abbott <lauraa@codeaurora.org>
Signed-off-by: NWill Deacon <will.deacon@arm.com>

The function cpu_resume currently lives in the .data section.
There's no reason for it to be there since we can use relative
instructions without a problem. Move a few cpu_resume data
structures out of the assembly file so the .data annotation
can be dropped completely and cpu_resume ends up in the read
only text section.
Reviewed-by: NKees Cook <keescook@chromium.org>
Reviewed-by: NMark Rutland <mark.rutland@arm.com>
Reviewed-by: NLorenzo Pieralisi <lorenzo.pieralisi@arm.com>
Tested-by: NMark Rutland <mark.rutland@arm.com>
Tested-by: NLorenzo Pieralisi <lorenzo.pieralisi@arm.com>
Tested-by: NKees Cook <keescook@chromium.org>
Acked-by: NArd Biesheuvel <ard.biesheuvel@linaro.org>
Signed-off-by: NLaura Abbott <lauraa@codeaurora.org>
Signed-off-by: NWill Deacon <will.deacon@arm.com>

The hyp stub vectors are currently loaded using adr. This
instruction has a +/- 1MB range for the loading address. If
the alignment for sections is changed the address may be more
than 1MB away, resulting in reclocation errors. Switch to using
adrp for getting the address to ensure we aren't affected by the
location of the __hyp_stub_vectors.
Acked-by: NArd Biesheuvel <ard.biesheuvel@linaro.org>
Acked-by: NMarc Zyngier <marc.zyngier@arm.com>
Tested-by: NMark Rutland <mark.rutland@arm.com>
Tested-by: NKees Cook <keescook@chromium.org>
Signed-off-by: NLaura Abbott <lauraa@codeaurora.org>
Signed-off-by: NWill Deacon <will.deacon@arm.com>

handle_arch_irq isn't actually text, it's just a function pointer.
It doesn't need to be stored in the text section and doing so
causes problesm if we ever want to make the kernel text read only.
Declare handle_arch_irq as a proper function pointer stored in
the data section.
Reviewed-by: NKees Cook <keescook@chromium.org>
Reviewed-by: NMark Rutland <mark.rutland@arm.com>
Acked-by: NArd Biesheuvel <ard.biesheuvel@linaro.org>
Tested-by: NMark Rutland <mark.rutland@arm.com>
Tested-by: NKees Cook <keescook@chromium.org>
Signed-off-by: NLaura Abbott <lauraa@codeaurora.org>
Signed-off-by: NWill Deacon <will.deacon@arm.com>

While we currently expect self-hosted debug support to be identical
across CPUs, we don't currently sanity check this.

This patch adds logging of the ID_AA64DFR{0,1}_EL1 values and associated
sanity checking code.

It's not clear to me whether we need to check PMUVer, TraceVer, and
DebugVer, as we don't currently rely on these fields at all.
Signed-off-by: NMark Rutland <mark.rutland@arm.com>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Acked-by: NWill Deacon <will.deacon@arm.com>
Signed-off-by: NWill Deacon <will.deacon@arm.com>

A missing newline in the WARN_TAINT_ONCE string results in ugly and
somewhat difficult to read output in the case of a sanity check failure,
as the next print does not appear on a new line:

  Unsupported CPU feature variation.Modules linked in:

This patch adds the missing newline, fixing the output formatting.
Signed-off-by: NMark Rutland <mark.rutland@arm.com>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Acked-by: NWill Deacon <will.deacon@arm.com>
Signed-off-by: NWill Deacon <will.deacon@arm.com>

It seems that Cortex-A53 r0p4 added support for AIFSR and ADFSR, and
ID_MMFR0.AuxReg has been updated accordingly to report this fact. As
Cortex-A53 could be paired with CPUs which do not implement these
registers (e.g. all current revisions of Cortex-A57), this may trigger a
sanity check failure at boot.

The AuxReg value describes the availability of the ACTLR, AIFSR, and
ADFSR registers, which are only of use to 32-bit guest OSs, and have
IMPLEMENTATION DEFINED contents. Given the nature of these registers it
is likely that KVM will need to trap accesses regardless of whether the
CPUs are heterogeneous.

This patch masks out the ID_MMFR0.AuxReg value from the sanity checks,
preventing spurious warnings at boot time.
Signed-off-by: NMark Rutland <mark.rutland@arm.com>
Reported-by: NAndre Przywara <andre.przywara@arm.com>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Acked-by: NWill Deacon <will.deacon@arm.com>
Cc: Marc Zyngier <marc.zyngier@arm.com>
Cc: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: NWill Deacon <will.deacon@arm.com>

The only requirement the scheduler has on cluster IDs is that they must
be unique.  When enumerating the topology based on MPIDR information the
kernel currently generates cluster IDs by using the first level of
affinity above the core ID (either level one or two depending on if the
core has multiple threads) however the ARMv8 architecture allows for up
to three levels of affinity.  This means that an ARMv8 system may
contain cores which have MPIDRs identical other than affinity level
three which with current code will cause us to report multiple cores
with the same identification to the scheduler in violation of its
uniqueness requirement.

Ensure that we do not violate the scheduler requirements on systems that
uses all the affinity levels by incorporating both affinity levels two
and three into the cluser ID when the cores are not threaded.

While no currently known hardware uses multi-level clusters it is better
to program defensively, this will help ease bringup of systems that have
them and will ensure that things like distribution install media do not
need to be respun to replace kernels in order to deploy such systems.
In the worst case the system will work but perform suboptimally until a
kernel modified to handle the new topology better is installed, in the
best case this will be an adequate description of such topologies for
the scheduler to perform well.
Signed-off-by: NMark Brown <broonie@linaro.org>
Signed-off-by: NWill Deacon <will.deacon@arm.com>

Not all of the errata we have workarounds for apply necessarily to all
SoCs, so people compiling a kernel for one very specific SoC may not
need to patch the kernel.
Introduce a new submenu in the "Platform selection" menu to allow
people to turn off certain bugs if they are not affected. By default
all of them are enabled.
Normal users or distribution kernels shouldn't bother to deselect any
bugs here, since the alternatives framework will take care of
patching them in only if needed.
Signed-off-by: NAndre Przywara <andre.przywara@arm.com>
[will: moved kconfig menu under `Kernel Features']
Signed-off-by: NWill Deacon <will.deacon@arm.com>

The ARM erratum 832075 applies to certain revisions of Cortex-A57,
one of the workarounds is to change device loads into using
load-aquire semantics.
This is achieved using the alternatives framework.
Signed-off-by: NAndre Przywara <andre.przywara@arm.com>
Signed-off-by: NWill Deacon <will.deacon@arm.com>

The ARM errata 819472, 826319, 827319 and 824069 define the same
workaround for these hardware issues in certain Cortex-A53 parts.
Use the new alternatives framework and the CPU MIDR detection to
patch "cache clean" into "cache clean and invalidate" instructions if
an affected CPU is detected at runtime.
Signed-off-by: NAndre Przywara <andre.przywara@arm.com>
[will: add __maybe_unused to squash gcc warning]
Signed-off-by: NWill Deacon <will.deacon@arm.com>

After each CPU has been started, we iterate through a list of
CPU features or bugs to detect CPUs which need (or could benefit
from) kernel code patches.
For each feature/bug there is a function which checks if that
particular CPU is affected. We will later provide some more generic
functions for common things like testing for certain MIDR ranges.
We do this for every CPU to cover big.LITTLE systems properly as
well.
If a certain feature/bug has been detected, the capability bit will
be set, so that later the call to apply_alternatives() will trigger
the actual code patching.
Signed-off-by: NAndre Przywara <andre.przywara@arm.com>
Signed-off-by: NWill Deacon <will.deacon@arm.com>

With a blatant copy of some x86 bits we introduce the alternative
runtime patching "framework" to arm64.
This is quite basic for now and we only provide the functions we need
at this time.
This is connected to the newly introduced feature bits.
Signed-off-by: NAndre Przywara <andre.przywara@arm.com>
Signed-off-by: NWill Deacon <will.deacon@arm.com>

For taking note if at least one CPU in the system needs a bug
workaround or would benefit from a code optimization, we create a new
bitmap to hold (artificial) feature bits.
Since elf_hwcap is part of the userland ABI, we keep it alone and
introduce a new data structure for that (along with some accessors).
Signed-off-by: NAndre Przywara <andre.przywara@arm.com>
Signed-off-by: NWill Deacon <will.deacon@arm.com>

update_insn_emulation_mode() returns 0 on success, so we should be
treating any non-zero values as failure, rather than the other way
around. Otherwise, writes to the sysctl file controlling the emulation
are ignored and immediately rolled back.
Reported-by: NGene Hackmann <ghackmann@google.com>
Signed-off-by: NWill Deacon <will.deacon@arm.com>

Translation faults that occur due to the input address being outside
of the address range mapped by the relevant base register are reported
as level 0 faults in ESR.DFSC.

If the faulting access cannot be resolved by the kernel (e.g. because
it is not mapped by a vma), then we report "input address range fault"
on the console. This was fine until we added support for 48-bit VAs,
which actually place PGDs at level 0 and can trigger faults for invalid
addresses that are within the range of the page tables.

This patch changes the string to report "level 0 translation fault",
which is far less confusing.
Signed-off-by: NWill Deacon <will.deacon@arm.com>

Having the instruction emulation submenu underneath "platform selection"
is a great way to hide options we don't want people to use, but somewhat
confusing when you stumble across it there.

Move the menuconfig option underneath "kernel features", where it makes
a bit more sense.
Signed-off-by: NWill Deacon <will.deacon@arm.com>

Introduce an event to trace the usage of emulated instructions. The
trace event is intended to help identify and encourage the migration
of legacy software using the emulation features.

Use this event to trace usage of swp and CP15 barrier emulation.
Acked-by: NSteven Rostedt <rostedt@goodmis.org>
Acked-by: NCatalin Marinas <catalin.marinas@arm.com>
Signed-off-by: NPunit Agrawal <punit.agrawal@arm.com>
Signed-off-by: NWill Deacon <will.deacon@arm.com>

The CP15 barrier instructions (CP15ISB, CP15DSB and CP15DMB) are
deprecated in the ARMv7 architecture, superseded by ISB, DSB and DMB
instructions respectively. Some implementations may provide the
ability to disable the CP15 barriers by disabling the CP15BEN bit in
SCTLR_EL1. If not enabled, the encodings for these instructions become
undefined.

To support legacy software using these instructions, this patch
register hooks to -
* emulate CP15 barriers and warn the user about their use
* toggle CP15BEN in SCTLR_EL1
Signed-off-by: NPunit Agrawal <punit.agrawal@arm.com>
Reviewed-by: NCatalin Marinas <catalin.marinas@arm.com>
Signed-off-by: NWill Deacon <will.deacon@arm.com>

The SWP instruction was deprecated in the ARMv6 architecture. The
ARMv7 multiprocessing extensions mandate that SWP/SWPB instructions
are treated as undefined from reset, with the ability to enable them
through the System Control Register SW bit. With ARMv8, the option to
enable these instructions through System Control Register was dropped
as well.

To support legacy applications using these instructions, port the
emulation of the SWP and SWPB instructions from the arm port to arm64.
Reviewed-by: NCatalin Marinas <catalin.marinas@arm.com>
Signed-off-by: NPunit Agrawal <punit.agrawal@arm.com>
Signed-off-by: NWill Deacon <will.deacon@arm.com>

Typically, providing support for legacy instructions requires
emulating the behaviour of instructions whose encodings have become
undefined. If the instructions haven't been removed from the
architecture, there maybe an option in the implementation to turn
on/off the support for these instructions.

Create common infrastructure to support legacy instruction
emulation. In addition to emulation, also provide an option to support
hardware execution when supported. The default execution mode (one of
undef, emulate, hw exeuction) is dependent on the state of the
instruction (deprecated or obsolete) in the architecture and
can specified at the time of registering the instruction handlers. The
runtime state of the emulation can be controlled by writing to
individual nodes in sysctl. The expected default behaviour is
documented as part of this patch.
Reviewed-by: NCatalin Marinas <catalin.marinas@arm.com>
Signed-off-by: NPunit Agrawal <punit.agrawal@arm.com>
Signed-off-by: NWill Deacon <will.deacon@arm.com>

Port support for AArch32 instruction condition code checking from arm
to arm64.
Signed-off-by: NPunit Agrawal <punit.agrawal@arm.com>
Reviewed-by: NCatalin Marinas <catalin.marinas@arm.com>
Signed-off-by: NWill Deacon <will.deacon@arm.com>

Add support to register hooks for undefined instructions. The handlers
will be called when the undefined instruction and the processor state
(as contained in pstate) match criteria used at registration.
Signed-off-by: NPunit Agrawal <punit.agrawal@arm.com>
Reviewed-by: NCatalin Marinas <catalin.marinas@arm.com>
Signed-off-by: NWill Deacon <will.deacon@arm.com>

The generic this_cpu operations disable interrupts to ensure that the
requested operation is protected from pre-emption. For arm64, this is
overkill and can hurt throughput and latency.

This patch provides arm64 specific implementations for the this_cpu
operations. Rather than disable interrupts, we use the exclusive
monitor or atomic operations as appropriate.

The following operations are implemented: add, add_return, and, or,
read, write, xchg. We also wire up a cmpxchg implementation from
cmpxchg.h.

Testing was performed using the percpu_test module and hackbench on a
Juno board running 3.18-rc4.
Signed-off-by: NSteve Capper <steve.capper@linaro.org>
Reviewed-by: NWill Deacon <will.deacon@arm.com>
Signed-off-by: NWill Deacon <will.deacon@arm.com>

We currently allocate different levels of page tables with a variety of
differing flags, and the PGALLOC_GFP flags, intended for use when
allocating any level of page table, are only used for ptes in
pte_alloc_one. On x86, PGALLOC_GFP is used for all page table
allocations.

Currently the major differences are:

* __GFP_NOTRACK -- Needed to ensure page tables are always accessible in
  the presence of kmemcheck to prevent recursive faults. Currently
  kmemcheck cannot be selected for arm64.

* __GFP_REPEAT -- Causes the allocator to try to reclaim pages and retry
  upon a failure to allocate.

* __GFP_ZERO -- Sometimes passed explicitly, sometimes zalloc variants
  are used.

While we've no encountered issues so far, it would be preferable to be
consistent. This patch ensures all levels of table are allocated in the
same manner, with PGALLOC_GFP.

Cc: Steve Capper <steve.capper@arm.com>
Acked-by: NCatalin Marinas <catalin.marinas@arm.com>
Signed-off-by: NMark Rutland <mark.rutland@arm.com>
Signed-off-by: NWill Deacon <will.deacon@arm.com>

Since commit 8a0a9bd4 ('random: make get_random_int() more
random'), get_random_int() returns a random value for each call,
so comment and hack introduced in mmap_rnd() as part of commit
1d18c47c ('arm64: MMU fault handling and page table management')
are incorrects.

Commit 1d18c47c seems to use the same hack introduced by
commit a5adc91a ('powerpc: Ensure random space between stack
and mmaps'), latter copied in commit 5a0efea0 ('sparc64: Sharpen
address space randomization calculations.').

But both architectures were cleaned up as part of commit
fa8cbaaf ('powerpc+sparc64/mm: Remove hack in mmap randomize
layout') as hack is no more needed since commit 8a0a9bd4.

So the present patch removes the comment and the hack around
get_random_int() on AArch64's mmap_rnd().

Cc: David S. Miller <davem@davemloft.net>
Cc: Anton Blanchard <anton@samba.org>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Acked-by: NWill Deacon <will.deacon@arm.com>
Acked-by: NDan McGee <dpmcgee@gmail.com>
Signed-off-by: NYann Droneaud <ydroneaud@opteya.com>
Signed-off-by: NWill Deacon <will.deacon@arm.com>

Commit a469abd0 (ARM: elf: add new hwcap for identifying atomic
ldrd/strd instructions) introduces HWCAP_ELF for 32-bit ARM
applications. As LPAE is always present on arm64, report the
corresponding compat HWCAP to user space.
Signed-off-by: NCatalin Marinas <catalin.marinas@arm.com>
Cc: <stable@vger.kernel.org> # 3.11+
Signed-off-by: NWill Deacon <will.deacon@arm.com>

On architectures with hardware broadcasting of TLB invalidation messages
, it makes sense to reduce the range of the mmu_gather structure when
unmapping page ranges based on the dirty address information passed to
tlb_remove_tlb_entry.

arm64 already does this by directly manipulating the start/end fields
of the gather structure, but this confuses the generic code which
does not expect these fields to change and can end up calculating
invalid, negative ranges when forcing a flush in zap_pte_range.

This patch moves the minimal range calculation out of the arm64 code
and into the generic implementation, simplifying zap_pte_range in the
process (which no longer needs to care about start/end, since they will
point to the appropriate ranges already). With the range being tracked
by core code, the need_flush flag is dropped in favour of checking that
the end of the range has actually been set.

Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Russell King - ARM Linux <linux@arm.linux.org.uk>
Cc: Michal Simek <monstr@monstr.eu>
Acked-by: NLinus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: NWill Deacon <will.deacon@arm.com>

The push/pop instructions can be suboptimal when saving/restoring large
amounts of data to/from the stack, for example on entry/exit from the
kernel. This is because:

  (1) They act on descending addresses (i.e. the newly decremented sp),
      which may defeat some hardware prefetchers

  (2) They introduce an implicit dependency between each instruction, as
      the sp has to be updated in order to resolve the address of the
      next access.

This patch removes the push/pop instructions from our kernel entry/exit
macros in favour of ldp/stp plus offset.
Signed-off-by: NWill Deacon <will.deacon@arm.com>