Re: [PATCH v2 02,04,05/12] DHM: Dynamic Housekeeping Management

Re: [PATCH v2 02,04,05/12] DHM: Dynamic Housekeeping Management

[Qiliang Yuan]

Point taken regarding the notifier ordering and complexity. For v3, I will
replace the generic notifier chain in genirq and tick subsystems with
explicit, controlled function calls to ensure deterministic execution.

> Why is this exported? Are random modules allowed to invoke this?

The export of tick_nohz_full_mask in Patch 02 will be removed. I'll implement
a restricted helper instead to prevent unintended usage by modules.

On Tue, 14 Apr 2026 23:57:52 +0200, Thomas Gleixner wrote:
> I have no idea what DHEI is. Provide proper information and not magic
> acronyms.

Documentation will be updated to replace "DHEI" with standard kernel
nomenclature.

> Refactoring has to be done in a preparatory patch and not in the middle 
> of a feature.

I will reorganize the series by moving the necessary infrastructure changes
into preparatory patches.

> next_timer? Please pick variable names which are comprehensible and self
> explaining. Also why can't you re-use timer_cpu, which would be actually
> useful?

Variable naming will be adjusted for better clarity, and I will explore
re-using timer_cpu as suggested.

> That's not sufficient. Assume an interrupt was shut down before the
> change because there was no online CPU in the affinity mask, but now the
> affinity mask changes so there is an online CPU. What starts it up?
> Same the other way around.

The managed IRQ migration logic will be revised to correctly track and
re-evaluate the interrupt state across affinity changes.

Best regards,
Qiliang

---

To: Frederic Weisbecker <frederic@kernel.org>
Subject: Re: [PATCH v2 01,03,07,08,09/12] DHM Architecture and Cpuset Integration
Date: Mon, 20 Apr 2026 16:10:00 +0800

Hi Frederic,

On Wed, 15 Apr 2026 12:02:46 +0200, Frederic Weisbecker wrote: