Re: CONFIG_PT_RECLAIM

From: David Hildenbrand
Date: Thu Feb 27 2025 - 04:54:32 EST

Next message: Igor Mammedov: "Re: [PATCH v4 00/14] Change ghes to use HEST-based offsets and add support for error inject"
Previous message: Bartosz Golaszewski: "Re: [PATCH v5 1/9] gpio: aggregator: protect driver attr handlers against module unload"
In reply to: Qi Zheng: "Re: CONFIG_PT_RECLAIM"
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

On 26.02.25 19:30, Johannes Weiner wrote:

Does PT_RECLAIM need to be configurable by the user?

Why not always try to free the page tables if the arch supports it?

I recall that the reasoning was that it might negatively affect some workloads so we'd be a bit careful first.

I'd be happy if we'd just unconditionally enable it.

--
Cheers,

David / dhildenb

Next message: Igor Mammedov: "Re: [PATCH v4 00/14] Change ghes to use HEST-based offsets and add support for error inject"
Previous message: Bartosz Golaszewski: "Re: [PATCH v5 1/9] gpio: aggregator: protect driver attr handlers against module unload"
In reply to: Qi Zheng: "Re: CONFIG_PT_RECLAIM"
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]