linux.git/arch/arm64/kernel/time.c, branch v4.11

arm64: ftrace: fix a stack tracer's output under function graph tracer

2015-12-21T17:26:02+00:00

Function graph tracer modifies a return address (LR) in a stack frame
to hook a function return. This will result in many useless entries
(return_to_handler) showing up in
 a) a stack tracer's output
 b) perf call graph (with perf record -g)
 c) dump_backtrace (at panic et al.)

For example, in case of a),
  $ echo function_graph > /sys/kernel/debug/tracing/current_tracer
  $ echo 1 > /proc/sys/kernel/stack_trace_enabled
  $ cat /sys/kernel/debug/tracing/stack_trace
        Depth    Size   Location    (54 entries)
        -----    ----   --------
  0)     4504      16   gic_raise_softirq+0x28/0x150
  1)     4488      80   smp_cross_call+0x38/0xb8
  2)     4408      48   return_to_handler+0x0/0x40
  3)     4360      32   return_to_handler+0x0/0x40
  ...

In case of b),
  $ echo function_graph > /sys/kernel/debug/tracing/current_tracer
  $ perf record -e mem:XXX:x -ag -- sleep 10
  $ perf report
                  ...
                  |          |          |--0.22%-- 0x550f8
                  |          |          |          0x10888
                  |          |          |          el0_svc_naked
                  |          |          |          sys_openat
                  |          |          |          return_to_handler
                  |          |          |          return_to_handler
                  ...

In case of c),
  $ echo function_graph > /sys/kernel/debug/tracing/current_tracer
  $ echo c > /proc/sysrq-trigger
  ...
  Call trace:
  [] sysrq_handle_crash+0x24/0x30
  [] return_to_handler+0x0/0x40
  [] return_to_handler+0x0/0x40
  ...

This patch replaces such entries with real addresses preserved in
current->ret_stack[] at unwind_frame(). This way, we can cover all
the cases.

Reviewed-by: Jungseok Lee 
Signed-off-by: AKASHI Takahiro 
[will: fixed minor context changes conflicting with irq stack bits]
Signed-off-by: Will Deacon

arm64: pass a task parameter to unwind_frame()

2015-12-21T17:26:01+00:00

Function graph tracer modifies a return address (LR) in a stack frame
to hook a function's return. This will result in many useless entries
(return_to_handler) showing up in a call stack list.
We will fix this problem in a later patch ("arm64: ftrace: fix a stack
tracer's output under function graph tracer"). But since real return
addresses are saved in ret_stack[] array in struct task_struct,
unwind functions need to be notified of, in addition to a stack pointer
address, which task is being traced in order to find out real return
addresses.

This patch extends unwind functions' interfaces by adding an extra
argument of a pointer to task_struct.

Signed-off-by: AKASHI Takahiro 
Signed-off-by: Will Deacon

clocksource: cosmetic: Drop OF 'dependency' from symbols

2015-10-01T00:18:39+00:00

Seeing the 'of' characters in a symbol that is being called from
ACPI seems to freak out people. So let's do a bit of pointless
renaming so that these folks do feel at home.

Signed-off-by: Marc Zyngier 
Acked-by: Catalin Marinas 
Reviewed-by: Hanjun Guo 
Acked-by: Thomas Gleixner 
Tested-by: Hanjun Guo 
Signed-off-by: Rafael J. Wysocki

clocksource / arm_arch_timer: Convert to ACPI probing

2015-10-01T00:18:39+00:00

It is now absolutely trivial to convert the arch timer driver to
use ACPI probing, just like its DT counterpart.

Let's enjoy another crapectomy.

Signed-off-by: Marc Zyngier 
Acked-by: Catalin Marinas 
Acked-by: Thomas Gleixner 
Tested-by: Hanjun Guo 
Signed-off-by: Rafael J. Wysocki

arm64: force CONFIG_SMP=y and remove redundant #ifdefs

2015-07-27T10:08:40+00:00

Nobody seems to be producing !SMP systems anymore, so this is just
becoming a source of kernel bugs, particularly if people want to use
coherent DMA with non-shared pages.

This patch forces CONFIG_SMP=y for arm64, removing a modest amount of
code in the process.

Signed-off-by: Will Deacon

clocksource / arch_timer: Parse GTDT to initialize arch timer

2015-03-26T15:13:07+00:00

Using the information presented by GTDT (Generic Timer Description Table)
to initialize the arch timer (not memory-mapped).

CC: Daniel Lezcano 
CC: Thomas Gleixner 
Originally-by: Amit Daniel Kachhap 
Tested-by: Suravee Suthikulpanit 
Tested-by: Yijing Wang 
Tested-by: Mark Langsdorf 
Tested-by: Jon Masters 
Tested-by: Timur Tabi 
Tested-by: Robert Richter 
Acked-by: Robert Richter 
Acked-by: Daniel Lezcano 
Reviewed-by: Grant Likely 
Signed-off-by: Hanjun Guo 
Signed-off-by: Will Deacon

arm64: kernel: initialize broadcast hrtimer based clock event device

2014-05-30T16:48:13+00:00

On platforms implementing CPU power management, the CPUidle subsystem
can allow CPUs to enter idle states where local timers logic is lost on power
down. To keep the software timers functional the kernel relies on an
always-on broadcast timer to be present in the platform to relay the
interrupt signalling the timer expiries.

For platforms implementing CPU core gating that do not implement an always-on
HW timer or implement it in a broken way, this patch adds code to initialize
the kernel hrtimer based clock event device upon boot (which can be chosen as
tick broadcast device by the kernel).
It relies on a dynamically chosen CPU to be always powered-up. This CPU then
relays the timer interrupt to CPUs in deep-idle states through its HW local
timer device.

Having a CPU always-on has implications on power management platform
capabilities and makes CPUidle suboptimal, since at least a CPU is kept
always in a shallow idle state by the kernel to relay timer interrupts,
but at least leaves the kernel with a functional system with some working
power management capabilities.

The hrtimer based clock event device is unconditionally registered, but
has the lowest possible rating such that any broadcast-capable HW clock
event device present will be chosen in preference as the tick broadcast
device.

Reviewed-by: Preeti U Murthy 
Acked-by: Will Deacon 
Acked-by: Mark Rutland 
Signed-off-by: Lorenzo Pieralisi 
Signed-off-by: Will Deacon

arm64: init: Move of_clk_init to time_init

2014-04-25T17:15:56+00:00

Clock providers should be initialized before clocksource_of_init.
If not, Clock source initialization can be fail to get the clock.

Acked-by: Will Deacon 
Signed-off-by: Chanho Min 
Signed-off-by: Will Deacon 
Signed-off-by: Catalin Marinas

arch_timer: Move to generic sched_clock framework

2013-10-09T23:54:10+00:00

Register with the generic sched_clock framework now that it
supports 64 bits. This fixes two problems with the current
sched_clock support for machines using the architected timers.
First off, we don't subtract the start value from subsequent
sched_clock calls so we can potentially start off with
sched_clock returning gigantic numbers. Second, there is no
support for suspend/resume handling so problems such as discussed
in 6a4dae5 (ARM: 7565/1: sched: stop sched_clock() during
suspend, 2012-10-23) can happen without this patch. Finally, it
allows us to move the sched_clock setup into drivers clocksource
out of the arch ports.

Cc: Christopher Covington 
Cc: Catalin Marinas 
Acked-by: Will Deacon 
Signed-off-by: Stephen Boyd 
Signed-off-by: John Stultz

arm64: kernel: compiling issue, need delete read_current_timer()

2013-06-10T16:58:20+00:00

Under arm64, we will calibrate the delay loop statically using a known
timer frequency, so delete read_current_timer(), or it will cause
compiling issue with allmodconfig.

The related error:
  ERROR: "read_current_timer" [lib/rbtree_test.ko] undefined!
  ERROR: "read_current_timer" [lib/interval_tree_test.ko] undefined!
  ERROR: "read_current_timer" [fs/ext4/ext4.ko] undefined!
  ERROR: "read_current_timer" [crypto/tcrypt.ko] undefined!

Signed-off-by: Chen Gang 
Acked-by: Marc Zyngier 
Signed-off-by: Catalin Marinas