linux-kernel - Re: [PATCH v8 02/14] mm/gup: check every subpage of a compound page during isolation

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

Message-ID: <05a66361-214c-2afe-22e4-12862ea1e4e2@oracle.com>
Date:   Wed, 3 Feb 2021 13:22:18 +0000
From:   Joao Martins <joao.m.martins@...cle.com>
To:     Pavel Tatashin <pasha.tatashin@...een.com>, jgg@...pe.ca
Cc:     linux-kernel@...r.kernel.org, linux-mm@...ck.org,
        akpm@...ux-foundation.org, vbabka@...e.cz, mhocko@...e.com,
        david@...hat.com, osalvador@...e.de, dan.j.williams@...el.com,
        sashal@...nel.org, tyhicks@...ux.microsoft.com,
        iamjoonsoo.kim@....com, mike.kravetz@...cle.com,
        rostedt@...dmis.org, mingo@...hat.com, peterz@...radead.org,
        mgorman@...e.de, willy@...radead.org, rientjes@...gle.com,
        jhubbard@...dia.com, linux-doc@...r.kernel.org,
        ira.weiny@...el.com, linux-kselftest@...r.kernel.org,
        jmorris@...ei.org
Subject: Re: [PATCH v8 02/14] mm/gup: check every subpage of a compound page
 during isolation

On 1/25/21 7:47 PM, Pavel Tatashin wrote:
> When pages are isolated in check_and_migrate_movable_pages() we skip
> compound number of pages at a time. However, as Jason noted, it is
> not necessary correct that pages[i] corresponds to the pages that
> we skipped. This is because it is possible that the addresses in
> this range had split_huge_pmd()/split_huge_pud(), and these functions
> do not update the compound page metadata.
> 
> The problem can be reproduced if something like this occurs:
> 
> 1. User faulted huge pages.
> 2. split_huge_pmd() was called for some reason
> 3. User has unmapped some sub-pages in the range
> 4. User tries to longterm pin the addresses.
> 
> The resulting pages[i] might end-up having pages which are not compound
> size page aligned.
> 
> Fixes: aa712399c1e8 ("mm/gup: speed up check_and_migrate_cma_pages() on huge page")
> Reported-by: Jason Gunthorpe <jgg@...dia.com>
> Signed-off-by: Pavel Tatashin <pasha.tatashin@...een.com>
> Reviewed-by: Jason Gunthorpe <jgg@...dia.com>
> ---

[...]

>  		/*
>  		 * If we get a page from the CMA zone, since we are going to
>  		 * be pinning these entries, we might as well move them out
> @@ -1599,8 +1596,6 @@ static long check_and_migrate_cma_pages(struct mm_struct *mm,
>  				}
>  			}
>  		}
> -
> -		i += step;
>  	}
>  

With this, longterm gup will 'regress' for hugetlbfs e.g. from ~6k -> 32k usecs when
pinning a 16G hugetlb file.

Splitting can only occur on THP right? If so, perhaps we could retain the @step increment
for compound pages but when !is_transparent_hugepage(head) or just PageHuge(head) like:

+               if (!is_transparent_hugepage(head) && PageCompound(page))
+                       i += (compound_nr(head) - (pages[i] - head));

Or making specific to hugetlbfs:

+               if (PageHuge(head))
+                       i += (compound_nr(head) - (pages[i] - head));