linux-kernel - Re: [PATCH] mm: pgtable: make ptep

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: <CAG48ez3bKSbDMmz7gpBm9x86PDuvpKPOGaTg0C33qtmetM44Tg@mail.gmail.com>
Date: Tue, 19 Nov 2024 16:46:58 +0100
From: Jann Horn <jannh@...gle.com>
To: Qi Zheng <zhengqi.arch@...edance.com>
Cc: pasha.tatashin@...een.com, tongtiangen@...wei.com, 
	lorenzo.stoakes@...cle.com, david@...hat.com, ryan.roberts@....com, 
	peterx@...hat.com, jgg@...pe.ca, muchun.song@...ux.dev, 
	akpm@...ux-foundation.org, linux-kernel@...r.kernel.org, linux-mm@...ck.org
Subject: Re: [PATCH] mm: pgtable: make ptep_clear() non-atomic

On Tue, Nov 19, 2024 at 10:10 AM Qi Zheng <zhengqi.arch@...edance.com> wrote:
> In the generic ptep_get_and_clear() implementation, it is just a simple
> combination of ptep_get() and pte_clear(). But for some architectures
> (such as x86 and arm64, etc), the hardware will modify the A/D bits of the
> page table entry, so the ptep_get_and_clear() needs to be overwritten
> and implemented as an atomic operation to avoid contention, which has a
> performance cost.
>
> The commit d283d422c6c4 ("x86: mm: add x86_64 support for page table
> check") adds the ptep_clear() on the x86, and makes it call
> ptep_get_and_clear() when CONFIG_PAGE_TABLE_CHECK is enabled. The page
> table check feature does not actually care about the A/D bits, so only
> ptep_get() + pte_clear() should be called. But considering that the page
> table check is a debug option, this should not have much of an impact.
>
> But then the commit de8c8e52836d ("mm: page_table_check: add hooks to
> public helpers") changed ptep_clear() to unconditionally call
> ptep_get_and_clear(), so that the CONFIG_PAGE_TABLE_CHECK check can be
> put into the page table check stubs (in include/linux/page_table_check.h).
> This also cause performance loss to the kernel without
> CONFIG_PAGE_TABLE_CHECK enabled, which doesn't make sense.
>
> Currently ptep_clear() is only used in debug code and in khugepaged
> collapse paths, which are fairly expensive. So the cost of an extra atomic
> RMW operation does not matter. But this may be used for other paths in the
> future. After all, for the present pte entry, we need to call ptep_clear()
> instead of pte_clear() to ensure that PAGE_TABLE_CHECK works properly.
>
> So to be more precise, just calling ptep_get() and pte_clear() in the
> ptep_clear().
>
> Signed-off-by: Qi Zheng <zhengqi.arch@...edance.com>

Yeah, as you mentioned on the other thread, this seems like a
reasonable cleanup.

Reviewed-by: Jann Horn <jannh@...gle.com>