Re: [PATCH v5 1/7] s390/percpu: Infrastructure for more efficient this_cpu operations

[Alexander Gordeev]

On Tue, May 26, 2026 at 07:56:56AM +0200, Heiko Carstens wrote:
> With the intended removal of PREEMPT_NONE this_cpu operations based on
> atomic instructions, guarded with preempt_disable()/preempt_enable() pairs
> become more expensive: the preempt_disable() / preempt_enable() pairs are
> not optimized away anymore during compile time.
> 
> In particular the conditional call to preempt_schedule_notrace() after
> preempt_enable() adds additional code and register pressure.
...
> 
> Signed-off-by: Heiko Carstens <hca@xxxxxxxxxxxxx>
> ---
>  arch/s390/include/asm/entry-percpu.h | 80 ++++++++++++++++++++++++++++
>  arch/s390/include/asm/lowcore.h      |  3 +-
>  arch/s390/include/asm/percpu.h       | 62 +++++++++++++++++++++
>  arch/s390/include/asm/ptrace.h       |  2 +
>  arch/s390/kernel/irq.c               | 24 ++++++---
>  arch/s390/kernel/nmi.c               |  5 ++
>  arch/s390/kernel/traps.c             |  5 ++
>  7 files changed, 173 insertions(+), 8 deletions(-)
>  create mode 100644 arch/s390/include/asm/entry-percpu.h

Reviewed-by: Alexander Gordeev <agordeev@xxxxxxxxxxxxx>