Will Deacon <will@xxxxxxxxxx> writes:
> On Fri, Aug 30, 2024 at 03:28:43PM -0700, Ankur Arora wrote:
>> Add architectural support for cpuidle-haltpoll driver by defining
>> arch_haltpoll_*().
>>
>> Also define ARCH_CPUIDLE_HALTPOLL to allow cpuidle-haltpoll to be
>> selected, and given that we have an optimized polling mechanism
>> in smp_cond_load*(), select ARCH_HAS_OPTIMIZED_POLL.
>>
>> smp_cond_load*() are implemented via LDXR, WFE, with LDXR loading
>> a memory region in exclusive state and the WFE waiting for any
>> stores to it.
>>
>> In the edge case -- no CPU stores to the waited region and there's no
>> interrupt -- the event-stream will provide the terminating condition
>> ensuring we don't wait forever, but because the event-stream runs at
>> a fixed frequency (configured at 10kHz) we might spend more time in
>> the polling stage than specified by cpuidle_poll_time().
>>
>> This would only happen in the last iteration, since overshooting the
>> poll_limit means the governor moves out of the polling stage.
>>
>> Tested-by: Haris Okanovic <harisokn@xxxxxxxxxx>
>> Tested-by: Misono Tomohiro <misono.tomohiro@xxxxxxxxxxx>
>> Signed-off-by: Ankur Arora <ankur.a.arora@xxxxxxxxxx>
>> ---
>> arch/arm64/Kconfig | 10 ++++++++++
>> arch/arm64/include/asm/cpuidle_haltpoll.h | 10 ++++++++++
>> arch/arm64/kernel/Makefile | 1 +
>> arch/arm64/kernel/cpuidle_haltpoll.c | 22 ++++++++++++++++++++++
>> 4 files changed, 43 insertions(+)
>> create mode 100644 arch/arm64/include/asm/cpuidle_haltpoll.h
>> create mode 100644 arch/arm64/kernel/cpuidle_haltpoll.c
>>
>> diff --git a/arch/arm64/Kconfig b/arch/arm64/Kconfig
>> index a2f8ff354ca6..9bd93ce2f9d9 100644
>> --- a/arch/arm64/Kconfig
>> +++ b/arch/arm64/Kconfig
>> @@ -36,6 +36,7 @@ config ARM64
>> select ARCH_HAS_MEMBARRIER_SYNC_CORE
>> select ARCH_HAS_NMI_SAFE_THIS_CPU_OPS
>> select ARCH_HAS_NON_OVERLAPPING_ADDRESS_SPACE
>> + select ARCH_HAS_OPTIMIZED_POLL
>> select ARCH_HAS_PTE_DEVMAP
>> select ARCH_HAS_PTE_SPECIAL
>> select ARCH_HAS_HW_PTE_YOUNG
>> @@ -2385,6 +2386,15 @@ config ARCH_HIBERNATION_HEADER
>> config ARCH_SUSPEND_POSSIBLE
>> def_bool y
>>
>> +config ARCH_CPUIDLE_HALTPOLL
>> + bool "Enable selection of the cpuidle-haltpoll driver"
>> + default n
>
> nit: this 'default n' line is redundant.
>
>> + help
>> + cpuidle-haltpoll allows for adaptive polling based on
>> + current load before entering the idle state.
>> +
>> + Some virtualized workloads benefit from using it.
>
> nit: This sentence is meaningless ^^.
Yeah. Yeah I think I added it to take care of a checkpatch warning.
But clearly it doesn't add anything useful. Will fix.
>> +
>> endmenu # "Power management options"
>>
>> menu "CPU Power Management"
>> diff --git a/arch/arm64/include/asm/cpuidle_haltpoll.h b/arch/arm64/include/asm/cpuidle_haltpoll.h
>> new file mode 100644
>> index 000000000000..ed615a99803b
>> --- /dev/null
>> +++ b/arch/arm64/include/asm/cpuidle_haltpoll.h
>> @@ -0,0 +1,10 @@
>> +/* SPDX-License-Identifier: GPL-2.0 */
>> +
>> +#ifndef _ARCH_HALTPOLL_H
>> +#define _ARCH_HALTPOLL_H
>> +
>> +static inline void arch_haltpoll_enable(unsigned int cpu) { }
>> +static inline void arch_haltpoll_disable(unsigned int cpu) { }
>> +
>> +bool arch_haltpoll_want(bool force);
>> +#endif
>> diff --git a/arch/arm64/kernel/Makefile b/arch/arm64/kernel/Makefile
>> index 2b112f3b7510..bbfb57eda2f1 100644
>> --- a/arch/arm64/kernel/Makefile
>> +++ b/arch/arm64/kernel/Makefile
>> @@ -70,6 +70,7 @@ obj-$(CONFIG_ARM64_PTR_AUTH) += pointer_auth.o
>> obj-$(CONFIG_ARM64_MTE) += mte.o
>> obj-y += vdso-wrap.o
>> obj-$(CONFIG_COMPAT_VDSO) += vdso32-wrap.o
>> +obj-$(CONFIG_ARCH_CPUIDLE_HALTPOLL) += cpuidle_haltpoll.o
>>
>> # Force dependency (vdso*-wrap.S includes vdso.so through incbin)
>> $(obj)/vdso-wrap.o: $(obj)/vdso/vdso.so
>> diff --git a/arch/arm64/kernel/cpuidle_haltpoll.c b/arch/arm64/kernel/cpuidle_haltpoll.c
>> new file mode 100644
>> index 000000000000..63fc5ebca79b
>> --- /dev/null
>> +++ b/arch/arm64/kernel/cpuidle_haltpoll.c
>> @@ -0,0 +1,22 @@
>> +// SPDX-License-Identifier: GPL-2.0
>> +
>> +#include <linux/kernel.h>
>> +#include <clocksource/arm_arch_timer.h>
>> +#include <asm/cpuidle_haltpoll.h>
>> +
>> +bool arch_haltpoll_want(bool force)
>> +{
>> + /*
>> + * Enabling haltpoll requires two things:
>> + *
>> + * - Event stream support to provide a terminating condition to the
>> + * WFE in the poll loop.
>> + *
>> + * - KVM support for arch_haltpoll_enable(), arch_haltpoll_disable().
>> + *
>> + * Given that the second is missing, allow haltpoll to only be force
>> + * loaded.
>> + */
>> + return (arch_timer_evtstrm_available() && false) || force;
>> +}
>> +EXPORT_SYMBOL_GPL(arch_haltpoll_want);
>
> This seems a bit over-the-top to justify a new C file. Just have a static
> inline in the header which returns 'force'. The '&& false' is misleading
> and unnecessary with the comment.
So, the only reason for doing it this way was that I wanted to encode the
arch_timer_evtstrm_available() dependency. But you are right that the
comment suffices since the check itself is not operative.
Will fix.
Thanks for reviewing.
--
ankur
