clocksource/drivers/imx-tpm: Fix next event not taking effect sometime

The value written into the TPM CnV can only be updated into the hardware
when the counter increases. Additional writes to the C

clocksource/drivers/imx-tpm: Fix next event not taking effect sometime

The value written into the TPM CnV can only be updated into the hardware
when the counter increases. Additional writes to the CnV write buffer are
ignored until the register has been updated. Therefore, we need to check
if the CnV has been updated before continuing. This may require waiting for
1 counter cycle in the worst case.

Cc: [email protected]
Fixes: 059ab7b82eec ("clocksource/drivers/imx-tpm: Add imx tpm timer support")
Signed-off-by: Jacky Bai <[email protected]>
Reviewed-by: Peng Fan <[email protected]>
Reviewed-by: Ye Li <[email protected]>
Reviewed-by: Jason Liu <[email protected]>
Signed-off-by: Frank Li <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Daniel Lezcano <[email protected]>

show more ...

clocksource/drivers/imx-tpm: Fix return -ETIME when delta exceeds INT_MAX

In tpm_set_next_event(delta), return -ETIME by wrong cast to int when delta
is larger than INT_MAX.

For example:

tpm_set_n

clocksource/drivers/imx-tpm: Fix return -ETIME when delta exceeds INT_MAX

In tpm_set_next_event(delta), return -ETIME by wrong cast to int when delta
is larger than INT_MAX.

For example:

tpm_set_next_event(delta = 0xffff_fffe)
{
...
next = tpm_read_counter(); // assume next is 0x10
next += delta; // next will 0xffff_fffe + 0x10 = 0x1_0000_000e
now = tpm_read_counter(); // now is 0x10
...

return (int)(next - now) <= 0 ? -ETIME : 0;
^^^^^^^^^^
0x1_0000_000e - 0x10 = 0xffff_fffe, which is -2 when
cast to int. So return -ETIME.
}

To fix this, introduce a 'prev' variable and check if 'now - prev' is
larger than delta.

Cc: [email protected]
Fixes: 059ab7b82eec ("clocksource/drivers/imx-tpm: Add imx tpm timer support")
Signed-off-by: Jacky Bai <[email protected]>
Reviewed-by: Peng Fan <[email protected]>
Reviewed-by: Ye Li <[email protected]>
Reviewed-by: Jason Liu <[email protected]>
Signed-off-by: Frank Li <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Daniel Lezcano <[email protected]>

show more ...

clocksource/drivers/imx-tpm: Move tpm_read_sched_clock() under CONFIG_ARM

When building this driver for an architecture other than ARCH=arm:

drivers/clocksource/timer-imx-tpm.c:78:20: error: unus

clocksource/drivers/imx-tpm: Move tpm_read_sched_clock() under CONFIG_ARM

When building this driver for an architecture other than ARCH=arm:

drivers/clocksource/timer-imx-tpm.c:78:20: error: unused function 'tpm_read_sched_clock' [-Werror,-Wunused-function]
static u64 notrace tpm_read_sched_clock(void)
^
1 error generated.

Move the function definition under the existing CONFIG_ARM section so
there is no more warning.

Fixes: 10720e120e2b ("clocksource/drivers/imx-tpm: Exclude sched clock for ARM64")
Signed-off-by: Nathan Chancellor <[email protected]>
Reviewed-by: Peng Fan <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Daniel Lezcano <[email protected]>

show more ...

clocksource/drivers/imx-tpm: Exclude sched clock for ARM64

For ARM64 platform such as i.MX8ULP which has ARMv8 generic timer as sched
clock, which is much faster compared with tpm sched clock. Readi

clocksource/drivers/imx-tpm: Exclude sched clock for ARM64

For ARM64 platform such as i.MX8ULP which has ARMv8 generic timer as sched
clock, which is much faster compared with tpm sched clock. Reading the
tpm count register in i.MX8ULP requires about 290ns, this is slow and
introduce scheduler latency. So exclude tpm sched clock for ARM64
platform.

Signed-off-by: Peng Fan <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Daniel Lezcano <[email protected]>

show more ...

clocksource/drivers/imx-tpm: Update name of clkevt

The tpm driver is not only for i.MX7ULP now, i.MX8ULP also use it. It
maybe also used by other i.MX variants, so update name to reflect it.

Signed

clocksource/drivers/imx-tpm: Update name of clkevt

The tpm driver is not only for i.MX7ULP now, i.MX8ULP also use it. It
maybe also used by other i.MX variants, so update name to reflect it.

Signed-off-by: Peng Fan <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Daniel Lezcano <[email protected]>

show more ...

clocksource/drivers/imx-tpm: Add CLOCK_EVT_FEAT_DYNIRQ

Add CLOCK_EVT_FEAT_DYNIRQ to allow the IRQ could be runtime set affinity
to the cores that needs wake up, otherwise saying core0 has to send
IP

clocksource/drivers/imx-tpm: Add CLOCK_EVT_FEAT_DYNIRQ

Add CLOCK_EVT_FEAT_DYNIRQ to allow the IRQ could be runtime set affinity
to the cores that needs wake up, otherwise saying core0 has to send
IPI to wakeup core1. With CLOCK_EVT_FEAT_DYNIRQ set, when broadcast
timer could wake up the cores, IPI is not needed.

Acked-by: Jacky Bai <[email protected]>
Signed-off-by: Peng Fan <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Daniel Lezcano <[email protected]>

show more ...

clocksource/drivers/imx-tpm: Mark two variable with __ro_after_init

counter_width and timer_base will not be updated after init, so mark
as __ro_after_init.

Signed-off-by: Peng Fan <[email protected]

clocksource/drivers/imx-tpm: Mark two variable with __ro_after_init

counter_width and timer_base will not be updated after init, so mark
as __ro_after_init.

Signed-off-by: Peng Fan <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Daniel Lezcano <[email protected]>

show more ...

clocksource/drivers/imx-tpm: Drop IRQF_IRQPOLL

Per the Documentation,
IRQF_IRQPOLL is used for polling (only the interrupt that is registered
first in a shared interrupt is considered for performanc

clocksource/drivers/imx-tpm: Drop IRQF_IRQPOLL

Per the Documentation,
IRQF_IRQPOLL is used for polling (only the interrupt that is registered
first in a shared interrupt is considered for performance reasons)
The TPM timer is not sharing interrupt with others, and pass irqpoll
not make sense for i.MX platform.

Signed-off-by: Peng Fan <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Daniel Lezcano <[email protected]>

show more ...

clocksource/drivers/imx-tpm: Add support for ARM64

Allows building and compile-testing the i.MX TPM driver also
for ARM64. The delay_timer is only supported on ARMv7.

Signed-off-by: Anson Huang <An

clocksource/drivers/imx-tpm: Add support for ARM64

Allows building and compile-testing the i.MX TPM driver also
for ARM64. The delay_timer is only supported on ARMv7.

Signed-off-by: Anson Huang <[email protected]>
Signed-off-by: Daniel Lezcano <[email protected]>
Link: https://lore.kernel.org/r/[email protected]

show more ...

clocksource/drivers/imx-tpm: Remove unused includes

There is nothing in use from of_address.h/of_irq.h, remove them.

Signed-off-by: Anson Huang <[email protected]>
Signed-off-by: Daniel Lezcano <

clocksource/drivers/imx-tpm: Remove unused includes

There is nothing in use from of_address.h/of_irq.h, remove them.

Signed-off-by: Anson Huang <[email protected]>
Signed-off-by: Daniel Lezcano <[email protected]>
Link: https://lore.kernel.org/r/[email protected]

show more ...

clocksource/drivers/timer-imx-tpm: Specify clock name for timer-of

i.MX TPM needs "ipg" clock for register access and "per" clock for
timer function, the driver gets "ipg" clock by searching the clo

clocksource/drivers/timer-imx-tpm: Specify clock name for timer-of

i.MX TPM needs "ipg" clock for register access and "per" clock for
timer function, the driver gets "ipg" clock by searching the clock
name, but timer-of initialization will get first clock in device
tree TPM node since no clock name specified in of_clk, that means
the "per" clock MUST be the first clock entry in device tree TPM
node, this patch specifies clock name for of_clk to avoid this
restriction, it makes TPM driver work properly with different sequence
of clock entries in device tree TPM node.

Signed-off-by: Anson Huang <[email protected]>
Signed-off-by: Daniel Lezcano <[email protected]>

show more ...

clocksource/drivers/timer-imx-tpm: Convert the driver to timer-of

Convert the driver to use the timer_of helpers. This allows to handle
timer base, clock and irq using common timer_of driver and it

clocksource/drivers/timer-imx-tpm: Convert the driver to timer-of

Convert the driver to use the timer_of helpers. This allows to handle
timer base, clock and irq using common timer_of driver and it
simplifies the code.

Signed-off-by: Anson Huang <[email protected]>
Signed-off-by: Daniel Lezcano <[email protected]>

show more ...

clocksource/drivers/timer-imx-tpm: Switch to SPDX identifier

Adopt the SPDX license identifier headers to ease license compliance
management.

Signed-off-by: Fabio Estevam <[email protected]>
Si

clocksource/drivers/timer-imx-tpm: Switch to SPDX identifier

Adopt the SPDX license identifier headers to ease license compliance
management.

Signed-off-by: Fabio Estevam <[email protected]>
Signed-off-by: Daniel Lezcano <[email protected]>

show more ...

clocksource/imx-tpm: Correct -ETIME return condition check

The additional brakects added to tpm_set_next_event's return value
computation causes (int) forced type conversion NOT taking effect, and t

clocksource/imx-tpm: Correct -ETIME return condition check

The additional brakects added to tpm_set_next_event's return value
computation causes (int) forced type conversion NOT taking effect, and the
incorrect value return will cause various system timer issue, like RCU
stall etc..

Remove the additional brackets to make sure tpm_set_next_event always
returns correct value.

Fixes: 059ab7b82eec ("clocksource/drivers/imx-tpm: Add imx tpm timer support")
Signed-off-by: Anson Huang <[email protected]>
Signed-off-by: Thomas Gleixner <[email protected]>
Acked-by: Dong Aisheng <[email protected]>
Cc: [email protected]
Cc: [email protected]
Cc: [email protected]
Link: https://lkml.kernel.org/r/[email protected]

show more ...

clocksource/drivers/imx-tpm: Add different counter width support

Different TPM modules have different width counters which is 16-bit or 32-bit,
the counter width can be read from TPM_PARAM register

clocksource/drivers/imx-tpm: Add different counter width support

Different TPM modules have different width counters which is 16-bit or 32-bit,
the counter width can be read from TPM_PARAM register bit[23:16], this patch
adds dynamic check for counter width to support both 16-bit and 32-bit TPM
modules.

Signed-off-by: Anson Huang <[email protected]>
Signed-off-by: Daniel Lezcano <[email protected]>

show more ...

clocksource/drivers/imx-tpm: Correct some registers operation flow

According to i.MX7ULP reference manual, TPM_SC_CPWMS can ONLY be written when
counter is disabled, TPM_SC_TOF is write-1-clear, TPM

clocksource/drivers/imx-tpm: Correct some registers operation flow

According to i.MX7ULP reference manual, TPM_SC_CPWMS can ONLY be written when
counter is disabled, TPM_SC_TOF is write-1-clear, TPM_C0SC_CHF is also
write-1-clear, correct these registers initialization flow;

Signed-off-by: Anson Huang <[email protected]>
Signed-off-by: Daniel Lezcano <[email protected]>

show more ...

clocksource/drivers/imx-tpm: Fix typo of clock name

The clock name should be ipg instead of igp.

Signed-off-by: Anson Huang <[email protected]>
Signed-off-by: Daniel Lezcano <daniel.lezcano@linar

clocksource/drivers/imx-tpm: Fix typo of clock name

The clock name should be ipg instead of igp.

Signed-off-by: Anson Huang <[email protected]>
Signed-off-by: Daniel Lezcano <[email protected]>

show more ...

clocksource/drivers/imx-tpm: Add imx tpm timer support

IMX Timer/PWM Module (TPM) supports both timer and pwm function while
this patch only adds the timer support. PWM would be added later.

The TP

clocksource/drivers/imx-tpm: Add imx tpm timer support

IMX Timer/PWM Module (TPM) supports both timer and pwm function while
this patch only adds the timer support. PWM would be added later.

The TPM counter, compare and capture registers are clocked by an
asynchronous clock that can remain enabled in low power modes.

NOTE: We observed in a very small probability, the bus fabric
contention between GPU and A7 may results a few cycles delay
of writing CNT registers which may cause the min_delta event got
missed, so we need add a ETIME check here in case it happened.

Cc: Daniel Lezcano <[email protected]>
Cc: Arnd Bergmann <[email protected]>
Cc: Thomas Gleixner <[email protected]>
Cc: Shawn Guo <[email protected]>
Cc: Anson Huang <[email protected]>
Cc: Bai Ping <[email protected]>
Signed-off-by: Dong Aisheng <[email protected]>
Signed-off-by: Daniel Lezcano <[email protected]>

show more ...