[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <4ff13fd3-3bfe-4c29-81dc-5e6c7aaa7af5@suse.cz>
Date: Fri, 30 May 2025 19:06:43 +0200
From: Vlastimil Babka <vbabka@...e.cz>
To: Zi Yan <ziy@...dia.com>, David Hildenbrand <david@...hat.com>,
Johannes Weiner <hannes@...xchg.org>, linux-mm@...ck.org
Cc: Andrew Morton <akpm@...ux-foundation.org>,
Oscar Salvador <osalvador@...e.de>,
Baolin Wang <baolin.wang@...ux.alibaba.com>,
"Kirill A . Shutemov" <kirill.shutemov@...ux.intel.com>,
Mel Gorman <mgorman@...hsingularity.net>,
Suren Baghdasaryan <surenb@...gle.com>, Michal Hocko <mhocko@...e.com>,
Brendan Jackman <jackmanb@...gle.com>, Richard Chang
<richardycc@...gle.com>, linux-kernel@...r.kernel.org
Subject: Re: [PATCH v6 3/6] mm/page_alloc: add support for initializing
pageblock as isolated.
On 5/30/25 18:22, Zi Yan wrote:
> MIGRATE_ISOLATE is a standalone bit, so a pageblock cannot be initialized
> to just MIGRATE_ISOLATE. Add init_pageblock_migratetype() to enable
> initialize a pageblock with a migratetype and isolated.
>
> Signed-off-by: Zi Yan <ziy@...dia.com>
> Reviewed-by: Vlastimil Babka <vbabka@...e.cz>
> +void __meminit init_pageblock_migratetype(struct page *page,
> + enum migratetype migratetype,
> + bool isolate)
> +{
> + unsigned long mask = MIGRATETYPE_MASK;
> + unsigned long flags = migratetype;
> +
> + if (unlikely(page_group_by_mobility_disabled &&
> + migratetype < MIGRATE_PCPTYPES))
> + migratetype = MIGRATE_UNMOVABLE;
> +
> +#ifdef CONFIG_MEMORY_ISOLATION
> + if (migratetype == MIGRATE_ISOLATE) {
> + VM_WARN_ONCE(
> + 1,
> + "Set isolate=true to isolate pageblock with a migratetype");
> + return;
> + }
> + if (isolate) {
> + mask = MIGRATETYPE_AND_ISO_MASK;
> + flags |= BIT(PB_migrate_isolate);
> + }
> +#endif
> + __set_pfnblock_flags_mask(page, page_to_pfn(page), flags, mask);
Nit: I think this could have also used MIGRATETYPE_AND_ISO_MASK unconditionally?
> +}
> +
> #ifdef CONFIG_DEBUG_VM
> static int page_outside_zone_boundaries(struct zone *zone, struct page *page)
> {
Powered by blists - more mailing lists