Using the physical counter allows KVM to retain the offset between the
virtual and physical counter as long as it is actively running a VCPU.

As soon as a VCPU is released, another thread is scheduled or we start
running userspace applications, we reset the offset to 0, so that
userspace accessing the virtual timer can still read the virtual counter
and get the same view of time as the kernel.

This opens up potential improvements for KVM performance, but we have to
make a few adjustments to preserve system consistency.

Currently get_cycles() is hardwired to arch_counter_get_cntvct() on
arm64, but as we move to using the physical timer for the in-kernel
time-keeping on systems that boot in EL2, we should use the same counter
for get_cycles() as for other in-kernel timekeeping operations.

Similarly, implementations of arch_timer_set_next_event_phys() is
modified to use the counter specific to the timer being programmed.

VHE kernels or kernels continuing to use the virtual timer are
unaffected.

Cc: Will Deacon <will.deacon@arm.com>
Cc: Mark Rutland <mark.rutland@arm.com>
Acked-by: NCatalin Marinas <catalin.marinas@arm.com>
Acked-by: NMarc Zyngier <marc.zyngier@arm.com>
Signed-off-by: NChristoffer Dall <cdall@linaro.org>

As we are about to use the physical counter on arm64 systems that have
KVM support, implement arch_counter_get_cntpct() and the associated
errata workaround functionality for stable timer reads.

Cc: Will Deacon <will.deacon@arm.com>
Cc: Mark Rutland <mark.rutland@arm.com>
Acked-by: NCatalin Marinas <catalin.marinas@arm.com>
Acked-by: NMarc Zyngier <marc.zyngier@arm.com>
Signed-off-by: NChristoffer Dall <christoffer.dall@linaro.org>

The loop to find the best memory frame in arch_timer_mem_acpi_init()
initializes the loop counter with itself ('i = i'), which is suspicious
in the first place and pointed out by clang. The loop condition is
'i < timer_count' and a prior for loop exits when 'i' reaches
'timer_count', therefore the second loop is never executed.

Initialize the loop counter with 0 to iterate over all timers, which
supposedly was the intention before the typo monster attacked.

Fixes: c2743a36 ("clocksource: arm_arch_timer: add GTDT support for memory-mapped timer")
Signed-off-by: NMatthias Kaehlcke <mka@chromium.org>
Reported-by: NArd Biesheuvel <ard.biesheuvel@linaro.org>
Acked-by: NMark Rutland <mark.rutland@arm.com>
Signed-off-by: NDaniel Lezcano <daniel.lezcano@linaro.org>

Use the new static_branch_enable_cpuslocked() function to switch
the workaround static key on the CPU hotplug path.
Signed-off-by: NMarc Zyngier <marc.zyngier@arm.com>
Signed-off-by: NPeter Zijlstra (Intel) <peterz@infradead.org>
Cc: Leo Yan <leo.yan@linaro.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: linux-arm-kernel@lists.infradead.org
Link: http://lkml.kernel.org/r/20170801080257.5056-5-marc.zyngier@arm.comSigned-off-by: NIngo Molnar <mingo@kernel.org>

The macro name is now renamed to 'TIMER_ACPI_DECLARE' for consistency
with the CLOCKSOURCE_OF_DECLARE => TIMER_OF_DECLARE change.
Signed-off-by: NDaniel Lezcano <daniel.lezcano@linaro.org>
Reviewed-by: NLinus Walleij <linus.walleij@linaro.org>

The CLOCKSOURCE_OF_DECLARE macro is used widely for the timers to declare the
clocksource at early stage. However, this macro is also used to initialize
the clockevent if any, or the clockevent only.

It was originally suggested to declare another macro to initialize a
clockevent, so in order to separate the two entities even they belong to the
same IP. This was not accepted because of the impact on the DT where splitting
a clocksource/clockevent definition does not make sense as it is a Linux
concept not a hardware description.

On the other side, the clocksource has not interrupt declared while the
clockevent has, so it is easy from the driver to know if the description is
for a clockevent or a clocksource, IOW it could be implemented at the driver
level.

So instead of dealing with a named clocksource macro, let's use a more generic
one: TIMER_OF_DECLARE.

The patch has not functional changes.
Signed-off-by: NDaniel Lezcano <daniel.lezcano@linaro.org>
Acked-by: NHeiko Stuebner <heiko@sntech.de>
Acked-by: NNeil Armstrong <narmstrong@baylibre.com>
Acked-by: NArnd Bergmann <arnd@arndb.de>
Acked-by: NMatthias Brugger <matthias.bgg@gmail.com>
Reviewed-by: NLinus Walleij <linus.walleij@linaro.org>

Fix boot warning 'Trying to vfree() nonexistent vm area'
from arch_timer_mem_of_init().

Refactored code attempts to read and iounmap using address frame
instead of address ioremap(frame->cntbase).

Fixes: c389d701 ("clocksource: arm_arch_timer: split MMIO timer probing.")
Signed-off-by: NFrank Rowand <frank.rowand@sony.com>
Reviewed-by: NFu Wei <fu.wei@linaro.org>
Acked-by: NMarc Zyngier <marc.zyngier@arm.com>
Signed-off-by: NDaniel Lezcano <daniel.lezcano@linaro.org>

arch_timer_mem_find_best_frame() looks through ARCH_TIMER_MEM_MAX_FRAMES
frames even after finding matches to ensure the best frame is chosen,
which means the variable frame will point to the last valid frame but
not necessarily the best frame.

On Juno, we get the following error as the wrong frame is returned as the
best frame from arch_timer_mem_find_best_frame():

  arch_timer: Unable to map frame @ 0x0000000000000000
  arch_timer: Frame missing phys irq.
  Failed to initialize '/timer@2a810000': -22

Fix the issue by correctly returning the best frame from
arch_timer_mem_find_best_frame().

Fixes: c389d701 ("clocksource: arm_arch_timer: split MMIO timer probing.")
Signed-off-by: NSudeep Holla <sudeep.holla@arm.com>
Acked-by: NMark Rutland <mark.rutland@arm.com>
Cc: Marc Zyngier <marc.zyngier@arm.com>
Cc: Daniel Lezcano <daniel.lezcano@linaro.org>
Cc: linux-arm-kernel@lists.infradead.org
Link: http://lkml.kernel.org/r/1494246747-17267-1-git-send-email-sudeep.holla@arm.comSigned-off-by: NThomas Gleixner <tglx@linutronix.de>

In some rare randconfig builds, we end up with two functions being entirely
unused:

drivers/clocksource/arm_arch_timer.c:342:12: error: 'erratum_set_next_event_tval_phys' defined but not used [-Werror=unused-function]
 static int erratum_set_next_event_tval_phys(unsigned long evt,
drivers/clocksource/arm_arch_timer.c:335:12: error: 'erratum_set_next_event_tval_virt' defined but not used [-Werror=unused-function]
 static int erratum_set_next_event_tval_virt(unsigned long evt,

We could add an #ifdef around them, but we would already have to check for
several symbols there and there is a chance this would get more complicated
over time, so marking them as __maybe_unused is the simplest way to avoid the
harmless warnings.

Fixes: 01d3e3ff ("arm64: arch_timer: Rework the set_next_event workarounds")
Signed-off-by: NArnd Bergmann <arnd@arndb.de>
Acked-by: NMarc Zyngier <marc.zyngier@arm.com>
Acked-by: NDaniel Lezcano <daniel.lezcano@linaro.org>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Will Deacon <will.deacon@arm.com>
Cc: linux-arm-kernel@lists.infradead.org
Cc: Christoffer Dall <christoffer.dall@linaro.org>
Link: http://lkml.kernel.org/r/20170419173737.3846098-1-arnd@arndb.deSigned-off-by: NThomas Gleixner <tglx@linutronix.de>

The patch add memory-mapped timer register support by using the
information provided by the new GTDT driver of ACPI.
Signed-off-by: NFu Wei <fu.wei@linaro.org>
Reviewed-by: NHanjun Guo <hanjun.guo@linaro.org>
[Mark: verify CNTFRQ, only register the first frame]
Signed-off-by: NMark Rutland <mark.rutland@arm.com>

The patch update arm_arch_timer driver to use the function
provided by the new GTDT driver of ACPI.
By this way, arm_arch_timer.c can be simplified, and separate
all the ACPI GTDT knowledge from this timer driver.
Signed-off-by: NFu Wei <fu.wei@linaro.org>
Signed-off-by: NHanjun Guo <hanjun.guo@linaro.org>
Tested-by: NXiongfeng Wang <wangxiongfeng2@huawei.com>
Reviewed-by: NHanjun Guo <hanjun.guo@linaro.org>
Tested-by: NHanjun Guo <hanjun.guo@linaro.org>
Signed-off-by: NMark Rutland <mark.rutland@arm.com>

Currently the code to probe MMIO architected timers mixes DT parsing with
actual poking of hardware. This makes the code harder than necessary to
understand, and makes it difficult to add support for probing via ACPI.

This patch splits the DT parsing from HW probing. The DT parsing now
lives in arch_timer_mem_of_init(), which fills in an arch_timer_mem
structure that it hands to probing functions that can be reused for ACPI
support.

Since the rate detection logic will be slight different when using ACPI,
the probing is performed as a number of steps. This results in more code
for the moment, and some arguably redundant work, but simplifies matters
considerably when ACPI support is added.
Signed-off-by: NFu Wei <fu.wei@linaro.org>
[Mark: refactor the probing split]
Signed-off-by: NMark Rutland <mark.rutland@arm.com>

To cleanly split code paths specific to ACPI or DT at a higher level,
this patch removes arch_timer_init(), folding the relevant
parts of its logic into existing callers.

This pathes the way for further rework, and saves a few lines.
Signed-off-by: NFu Wei <fu.wei@linaro.org>
Reviewed-by: NHanjun Guo <hanjun.guo@linaro.org>
[Mark: reword commit message]
Signed-off-by: NMark Rutland <mark.rutland@arm.com>

When booting with DT, it's possible for timer nodes to be probed in any
order. Some common initialisation needs to occur after all nodes have
been probed, and arch_timer_common_init() has code to detect when this
has happened.

This logic is DT-specific, and it would be best to factor it out of the
common code that will be shared with ACPI.

This patch folds this into the existing arch_timer_needs_probing(),
which is renamed to arch_timer_needs_of_probing(), and no longer takes
any arguments. This is only called when using DT, and not when using
ACPI, which will have a deterministic probe order.
Signed-off-by: NFu Wei <fu.wei@linaro.org>
Reviewed-by: NHanjun Guo <hanjun.guo@linaro.org>
[Mark: reword commit message]
Signed-off-by: NMark Rutland <mark.rutland@arm.com>

For historical reasons, rate detection when probing via DT is somewhat
convoluted. We tried to package this up in arch_timer_detect_rate(), but
with the addition of ACPI worse, and gets in the way of stringent rate
checking when ACPI is used.

This patch makes arch_timer_detect_rate() specific to DT, ripping out
ACPI logic. In preparation for rework of the MMIO timer probing, the
reading of the relevant CNTFRQ register is factored out to callers. The
function is then renamed to arch_timer_of_configure_rate(), which better
represents its new place in the world.

Comments are added in the DT and ACPI probe paths to explain this.
Signed-off-by: NFu Wei <fu.wei@linaro.org>
[Mark: reword commit message]
Signed-off-by: NMark Rutland <mark.rutland@arm.com>

Currently, the arch timer driver uses ARCH_TIMER_PHYS_SECURE_PPI to mean
the driver will use the secure PPI *and* potentially also use the
non-secure PPI. This is somewhat confusing.

For arm64 it never makes sense to use the secure PPI, but we do anyway,
inheriting this behaviour from 32-bit arm. For ACPI, we may not even
have a valid secure PPI, so we need to be able to only request the
non-secure PPI.

To that end, this patch reworks the timer driver so that we can request
the non-secure PPI alone. The PPI selection is split out into a new
function, arch_timer_select_ppi(), and verification of the selected PPI
is shifted out to callers (as DT may select the PPI by other means and
must handle this anyway).

We now consistently use arch_timer_has_nonsecure_ppi() to determine
whether we must manage a non-secure PPI *in addition* to a secure PPI.
When we only have a non-secure PPI, this returns false.
Signed-off-by: NFu Wei <fu.wei@linaro.org>
Tested-by: NXiongfeng Wang <wangxiongfeng2@huawei.com>
[Mark: reword commit message]
Signed-off-by: NMark Rutland <mark.rutland@arm.com>

This patch add a new enum "arch_timer_spi_nr" and use it in the driver.
Just for code's readability, no functional change.
Signed-off-by: NFu Wei <fu.wei@linaro.org>
Acked-by: NMark Rutland <mark.rutland@arm.com>
Signed-off-by: NMark Rutland <mark.rutland@arm.com>

To support the arm_arch_timer via ACPI we need to share defines and enums
between the driver and the ACPI parser code.
So we split out the relevant defines and enums into arm_arch_timer.h.

No functional change.
Signed-off-by: NFu Wei <fu.wei@linaro.org>
Acked-by: NMark Rutland <mark.rutland@arm.com>
Tested-by: NXiongfeng Wang <wangxiongfeng2@huawei.com>
Signed-off-by: NMark Rutland <mark.rutland@arm.com>

In preparation for moving the PPI enum out into a header, rename the
enum and its constituent values these so they are namespaced w.r.t. the
arch timer. This will aid consistency and avoid potential name clashes
when this move occurs.

No functional change.
Signed-off-by: NFu Wei <fu.wei@linaro.org>
Tested-by: NXiongfeng Wang <wangxiongfeng2@huawei.com>
[Mark: reword commit message]
Signed-off-by: NMark Rutland <mark.rutland@arm.com>

In preparation for moving the type macros out into a header, rename
these so they are namespaced w.r.t. the arch timer. We'll apply the same
prefix to other definitions in subsequent patches. This will aid
consistency and avoid potential name clahses when this move occurs.

No functional change.
Signed-off-by: NFu Wei <fu.wei@linaro.org>
Tested-by: NXiongfeng Wang <wangxiongfeng2@huawei.com>
[Mark: reword commit message]
Signed-off-by: NMark Rutland <mark.rutland@arm.com>

Almost all string in the arm_arch_timer driver duplicate an common
prefix (though a few do not). For consistency, it would be better to use
pr_fmt(), and always use this prefix. At the same time, we may as well
clean up some whitespace issues in arch_timer_banner and
arch_timer_init.

No functional change.
Signed-off-by: NFu Wei <fu.wei@linaro.org>
Acked-by: NMark Rutland <mark.rutland@arm.com>
Tested-by: NXiongfeng Wang <wangxiongfeng2@huawei.com>
[Mark: reword commit message]
Signed-off-by: NMark Rutland <mark.rutland@arm.com>

Printing with pr_* functions requires adding line break manually.
Signed-off-by: NRafał Miłecki <rafal@milecki.pl>
Acked-by: NThierry Reding <treding@nvidia.com>
Signed-off-by: NDaniel Lezcano <daniel.lezcano@linaro.org>

In order to deal with ACPI enabled platforms suffering from the
HISILICON_ERRATUM_161010101, let's add the required OEM data that
allow the workaround to be enabled.
Acked-by: NThomas Gleixner <tglx@linutronix.de>
Tested-by: Ndann frazier <dann.frazier@canonical.com>
Tested-by: NHanjun Guo <hanjun.guo@linaro.org>
Reviewed-by: NHanjun Guo <hanjun.guo@linaro.org>
Signed-off-by: NMarc Zyngier <marc.zyngier@arm.com>

Just as we're able to identify a broken platform using some DT
information, let's enable a way to spot the offenders with ACPI.

The difference is that we can only match on some OEM info instead
of implementation-specific properties. So in order to avoid the
insane multiplication of errata structures, we allow an array
of OEM descriptions to be attached to an erratum structure.
Acked-by: NThomas Gleixner <tglx@linutronix.de>
Tested-by: Ndann frazier <dann.frazier@canonical.com>
Tested-by: NHanjun Guo <hanjun.guo@linaro.org>
Reviewed-by: NHanjun Guo <hanjun.guo@linaro.org>
Signed-off-by: NMarc Zyngier <marc.zyngier@arm.com>

Cortex-A73 (all versions) counter read can return a wrong value
when the counter crosses a 32bit boundary.

The workaround involves performing the read twice, and to return
one or the other depending on whether a transition has taken place.
Acked-by: NThomas Gleixner <tglx@linutronix.de>
Signed-off-by: NMarc Zyngier <marc.zyngier@arm.com>

Userspace being allowed to use read CNTVCT_EL0 anytime (and not
only in the VDSO), we need to enable trapping whenever a cntvct
workaround is enabled on a given CPU.
Acked-by: NThomas Gleixner <tglx@linutronix.de>
Signed-off-by: NMarc Zyngier <marc.zyngier@arm.com>

As we're about to allow per CPU cntkctl_el1 configuration, we cannot
rely on the register value to be common when performing power
management.

Let's turn saved_cntkctl into a per-cpu variable.
Acked-by: NThomas Gleixner <tglx@linutronix.de>
Signed-off-by: NMarc Zyngier <marc.zyngier@arm.com>

In order to access clocksource_counter from the errata handling code,
move it (together with the related structures and functions) towards
the top of the file.
Acked-by: NThomas Gleixner <tglx@linutronix.de>
Signed-off-by: NMarc Zyngier <marc.zyngier@arm.com>

Instead of applying a CPU-specific workaround to all CPUs in the system,
allow it to only affect a subset of them (typical big-little case).

This is done by turning the erratum pointer into a per-CPU variable.
Acked-by: NThomas Gleixner <tglx@linutronix.de>
Signed-off-by: NMarc Zyngier <marc.zyngier@arm.com>

The way we work around errata affecting set_next_event is not very
nice, at it imposes this workaround on errata that do not need it.

Add new workaround hooks and let the existing workarounds use them.
Acked-by: NThomas Gleixner <tglx@linutronix.de>
Signed-off-by: NMarc Zyngier <marc.zyngier@arm.com>

Let's move the handling of workarounds affecting set_next_event
to the affected function, instead of overriding the pointers
as an afterthough. Yes, this is an extra indirection on the
erratum handling path, but the HW is busted anyway.

This will allow for some more flexibility later.
Acked-by: NThomas Gleixner <tglx@linutronix.de>
Signed-off-by: NMarc Zyngier <marc.zyngier@arm.com>

As we're about to move things around, let's start with the low
level read/write functions. This allows us to use these functions
in the errata handling code without having to use forward declaration
of static functions.
Acked-by: NThomas Gleixner <tglx@linutronix.de>
Acked-by: NMark Rutland <mark.rutland@arm.com>
Acked-by: NDaniel Lezcano <daniel.lezcano@linaro.org>
Signed-off-by: NMarc Zyngier <marc.zyngier@arm.com>

Should we ever have a workaround for an erratum that is detected using
a capability and affecting a particular CPU, it'd be nice to have
a way to probe them directly.
Acked-by: NThomas Gleixner <tglx@linutronix.de>
Signed-off-by: NMarc Zyngier <marc.zyngier@arm.com>

We're currently stuck with DT when it comes to handling errata, which
is pretty restrictive. In order to make things more flexible, let's
introduce an infrastructure that could support alternative discovery
methods. No change in functionality.
Acked-by: NThomas Gleixner <tglx@linutronix.de>
Reviewed-by: NHanjun Guo <hanjun.guo@linaro.org>
Signed-off-by: NMarc Zyngier <marc.zyngier@arm.com>

We are going to split <linux/sched/clock.h> out of <linux/sched.h>, which
will have to be picked up from other headers and .c files.

Create a trivial placeholder <linux/sched/clock.h> file that just
maps to <linux/sched.h> to make this patch obviously correct and
bisectable.

Include the new header in the files that are going to need it.
Acked-by: NLinus Torvalds <torvalds@linux-foundation.org>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: linux-kernel@vger.kernel.org
Signed-off-by: NIngo Molnar <mingo@kernel.org>

The object cyclecounter of type cyclecounter is only modified during
initialization in arch_counter_register. So it can be marked
__ro_after_init.
Signed-off-by: NBhumika Goyal <bhumirks@gmail.com>
Cc: mark.rutland@arm.com
Cc: keescook@chromium.org
Cc: marc.zyngier@arm.com
Cc: daniel.lezcano@linaro.org
Cc: julia.lawall@lip6.fr
Cc: linux-arm-kernel@lists.infradead.org
Link: http://lkml.kernel.org/r/1486840818-22214-1-git-send-email-bhumirks@gmail.comSigned-off-by: NThomas Gleixner <tglx@linutronix.de>

Erratum Hisilicon-161010101 says that the ARM generic timer counter "has
the potential to contain an erroneous value when the timer value
changes". Accesses to TVAL (both read and write) are also affected due
to the implicit counter read. Accesses to CVAL are not affected.

The workaround is to reread the system count registers until the value
of the second read is larger than the first one by less than 32, the
system counter can be guaranteed not to return wrong value twice by
back-to-back read and the error value is always larger than the correct
one by 32. Writes to TVAL are replaced with an equivalent write to CVAL.
Signed-off-by: NDing Tianhong <dingtianhong@huawei.com>
[Mark: split patch, fix Kconfig, reword commit message]
Signed-off-by: NMark Rutland <mark.rutland@arm.com>
Acked-by: NDaniel Lezcano <daniel.lezcano@linaro.org>
Signed-off-by: NDaniel Lezcano <daniel.lezcano@linaro.org>

Currently we have code inline in the arch timer probe path to cater for
Freescale erratum A-008585, complete with ifdeffery. This is a little
ugly, and will get worse as we try to add more errata handling.

This patch refactors the handling of Freescale erratum A-008585. Now the
erratum is described in a generic arch_timer_erratum_workaround
structure, and the probe path can iterate over these to detect errata
and enable workarounds.

This will simplify the addition and maintenance of code handling
Hisilicon erratum 161010101.
Signed-off-by: NDing Tianhong <dingtianhong@huawei.com>
[Mark: split patch, correct Kconfig, reword commit message]
Signed-off-by: NMark Rutland <mark.rutland@arm.com>
Acked-by: NDaniel Lezcano <daniel.lezcano@linaro.org>
Signed-off-by: NDaniel Lezcano <daniel.lezcano@linaro.org>

Having a command line option to flip the errata handling for a
particular erratum is a little bit unusual, and it's vastly superior to
pass this in the DT. By common consensus, it's best to kill off the
command line parameter.
Signed-off-by: NDing Tianhong <dingtianhong@huawei.com>
[Mark: split patch, reword commit message]
Signed-off-by: NMark Rutland <mark.rutland@arm.com>
Acked-by: NDaniel Lezcano <daniel.lezcano@linaro.org>
Signed-off-by: NDaniel Lezcano <daniel.lezcano@linaro.org>

There is no point in having an extra type for extra confusion. u64 is
unambiguous.

Conversion was done with the following coccinelle script:

@rem@
@@
-typedef u64 cycle_t;

@fix@
typedef cycle_t;
@@
-cycle_t
+u64
Signed-off-by: NThomas Gleixner <tglx@linutronix.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: John Stultz <john.stultz@linaro.org>