On 17.03.21 11:05, Oscar Salvador wrote:
On Wed, Mar 17, 2021 at 10:48:31AM +0100, David Hildenbrand wrote:
I was preparing v5, and I wanted to be sure I understood you here.
Right you are that the in-use page check can be dropped, as those pages
can
be migrated away, and the Hugetlb page check can also be dropped since
isolate_migratepages_range is now capable of dealing with those kind of
pages.
b) Similarly, check for gigantic pages and/or movability/migratability.
I lost you here.
isolate_or_dissolve_huge_page() already bails out on hugetlb-gigantic
pages.
Or do you mean to place an upfront check here? (hstate_is_gigantic())?
Yes. But I prefer a) and keeping it simple here -- just doing basic sanity
checks (online, zone, PageReserved()) that are absolutely necessary.
Ok, I am probably dense as I understood as if you were lean towards having
a) + b).
Sorry, I meant either a) or b) :)
That is what I have as the last patch of the patchset:
From e97175b7d4970cbdcbafcf8c398f72a571e817b0 Mon Sep 17 00:00:00 2001
From: Oscar Salvador <osalvador@xxxxxxx>
Date: Thu, 18 Mar 2021 05:03:18 +0100
Subject: [PATCH] mm,page_alloc: Drop unnecesary checks from
pfn_range_valid_contig
pfn_range_valid_contig() bails out when it finds an in-use page or a
hugetlb page, among other things.
We can drop the in-use page check since __alloc_contig_pages can migrate
away those pages, and the hugetlb page check can go too since
isolate_migratepages_range is now capable of dealing with hugetlb pages.
Might want to mention that the existing checks were racy either way :)
Signed-off-by: Oscar Salvador <osalvador@xxxxxxx>
---
mm/page_alloc.c | 6 ------
1 file changed, 6 deletions(-)
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 4cb455355f6d..50d73e68b79e 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -8685,12 +8685,6 @@ static bool pfn_range_valid_contig(struct zone *z, unsigned long start_pfn,
if (PageReserved(page))
return false;
-
- if (page_count(page) > 0)
- return false;
-
- if (PageHuge(page))
- return false;
}
return true;
}
--
Thanks,
David / dhildenb
