linux-kernel - Re: [PATCH v7 2/4] mm/rmap: add helper to restart pgtable walk on changes

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: <b0d551a9-ee70-46a7-a0d6-c422f0baf91e@redhat.com>
Date: Thu, 13 Jun 2024 10:30:15 +0200
From: David Hildenbrand <david@...hat.com>
To: Lance Yang <ioworker0@...il.com>, akpm@...ux-foundation.org
Cc: willy@...radead.org, sj@...nel.org, baolin.wang@...ux.alibaba.com,
 maskray@...gle.com, ziy@...dia.com, ryan.roberts@....com, 21cnbao@...il.com,
 mhocko@...e.com, fengwei.yin@...el.com, zokeefe@...gle.com,
 shy828301@...il.com, xiehuan09@...il.com, libang.li@...group.com,
 wangkefeng.wang@...wei.com, songmuchun@...edance.com, peterx@...hat.com,
 minchan@...nel.org, linux-mm@...ck.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH v7 2/4] mm/rmap: add helper to restart pgtable walk on
 changes

On 10.06.24 14:02, Lance Yang wrote:
> Introduce the page_vma_mapped_walk_restart() helper to handle scenarios
> where the page table walk needs to be restarted due to changes in the page
> table, such as when a PMD is split. It releases the PTL held during the
> previous walk and resets the state, allowing a new walk to start at the
> current address stored in pvmw->address.
> 
> Suggested-by: David Hildenbrand <david@...hat.com>
> Signed-off-by: Lance Yang <ioworker0@...il.com>
> ---
>   include/linux/rmap.h | 22 ++++++++++++++++++++++
>   1 file changed, 22 insertions(+)
> 
> diff --git a/include/linux/rmap.h b/include/linux/rmap.h
> index 7229b9baf20d..5f18509610cc 100644
> --- a/include/linux/rmap.h
> +++ b/include/linux/rmap.h
> @@ -710,6 +710,28 @@ static inline void page_vma_mapped_walk_done(struct page_vma_mapped_walk *pvmw)
>   		spin_unlock(pvmw->ptl);
>   }
>   
> +/**
> + * page_vma_mapped_walk_restart - Restart the page table walk.
> + * @pvmw: Pointer to struct page_vma_mapped_walk.
> + *
> + * It restarts the page table walk when changes occur in the page
> + * table, such as splitting a PMD. Ensures that the PTL held during
> + * the previous walk is released and resets the state to allow for
> + * a new walk starting at the current address stored in pvmw->address.
> + */
> +static inline void
> +page_vma_mapped_walk_restart(struct page_vma_mapped_walk *pvmw)
> +{
> +	WARN_ON_ONCE(!pvmw->pmd);

Can we have this more general, like

WARN_ON_ONCE(!pvmw->pmd && !pvmw->pte);

And then setting both to NULL below?


> +	WARN_ON_ONCE(!pvmw->ptl);

This is confusing: you check for ptl below. What would be clearer is

if (likely(pvmw->ptl))
	spin_unlock(pvmw->ptl);
else
	WARN_ON_ONCE(1);


> +
> +	if (pvmw->ptl)
> +		spin_unlock(pvmw->ptl);
> +
> +	pvmw->ptl = NULL;
> +	pvmw->pmd = NULL;
> +}
> +
>   bool page_vma_mapped_walk(struct page_vma_mapped_walk *pvmw);
>   
>   /*

I'd suggest squashing that into the next patch.

-- 
Cheers,

David / dhildenb