linux-kernel - Re: [PATCH RFC] mm: make try_to_unmap_one support batched unmap for anon large folios

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-Id: <20250523082517.3144-1-21cnbao@gmail.com>
Date: Fri, 23 May 2025 20:25:17 +1200
From: Barry Song <21cnbao@...il.com>
To: 21cnbao@...il.com,
	david@...hat.com
Cc: Liam.Howlett@...cle.com,
	akpm@...ux-foundation.org,
	baolin.wang@...ux.alibaba.com,
	bhe@...hat.com,
	chrisl@...nel.org,
	fan.ni@...sung.com,
	harry.yoo@...cle.com,
	kasong@...cent.com,
	linux-kernel@...r.kernel.org,
	linux-mm@...ck.org,
	lorenzo.stoakes@...cle.com,
	mhocko@...e.com,
	riel@...riel.com,
	rppt@...nel.org,
	ryan.roberts@....com,
	schatzberg.dan@...il.com,
	surenb@...gle.com,
	v-songbaohua@...o.com,
	vbabka@...e.cz,
	yukaixiong@...wei.com,
	zhengtangquan@...o.com
Subject: Re: [PATCH RFC] mm: make try_to_unmap_one support batched unmap for anon large folios

> @@ -2159,9 +2179,10 @@ static bool try_to_unmap_one(struct folio *folio, struct vm_area_struct *vma,
> 
> 			/* See folio_try_share_anon_rmap(): clear PTE first. */
>  			if (anon_exclusive &&
> -			    folio_try_share_anon_rmap_pte(folio, subpage)) {
> -				swap_free(entry);
> -				set_pte_at(mm, address, pvmw.pte, pteval);
> +			    __folio_try_share_anon_rmap(folio, subpage, nr_pages,
> +							RMAP_LEVEL_PTE)) {


I have a question for David regarding the __folio_try_share_anon_rmap()  
function. I noticed that it only clears the Exclusive flag for the first  
subpage, even when nr_pages > 1. Shouldn't we be clearing the Exclusive  
flag for all subpages in this case? Could this be a potential bug?

Thanks
Barry