From: David Hildenbrand <david@redhat.com>
To: Vlastimil Babka <vbabka@suse.cz>, stable@vger.kernel.org
Cc: Peter Xu <peterx@redhat.com>,
linux-mm@kvack.org, Yuanzheng Song <songyuanzheng@huawei.com>
Subject: Re: [PATCH stable 5.15] mm/memory: add non-anonymous page check in the copy_present_page()
Date: Wed, 13 Nov 2024 17:45:43 +0100 [thread overview]
Message-ID: <99a77c9a-68ca-4445-bcbf-4681ca20a482@redhat.com> (raw)
In-Reply-To: <20241113163118.54834-2-vbabka@suse.cz>
On 13.11.24 17:31, Vlastimil Babka wrote:
> From: Yuanzheng Song <songyuanzheng@huawei.com>
>
> The vma->anon_vma of the child process may be NULL because
> the entire vma does not contain anonymous pages. In this
> case, a BUG will occur when the copy_present_page() passes
> a copy of a non-anonymous page of that vma to the
> page_add_new_anon_rmap() to set up new anonymous rmap.
>
> ------------[ cut here ]------------
> kernel BUG at mm/rmap.c:1052!
> Internal error: Oops - BUG: 0 [#1] SMP
> Modules linked in:
> CPU: 4 PID: 4652 Comm: test Not tainted 5.15.75 #1
> Hardware name: linux,dummy-virt (DT)
> pstate: 80000005 (Nzcv daif -PAN -UAO -TCO -DIT -SSBS BTYPE=--)
> pc : __page_set_anon_rmap+0xc0/0xe8
> lr : __page_set_anon_rmap+0xc0/0xe8
> sp : ffff80000e773860
> x29: ffff80000e773860 x28: fffffc13cf006ec0 x27: ffff04f3ccd68000
> x26: ffff04f3c5c33248 x25: 0000000010100073 x24: ffff04f3c53c0a80
> x23: 0000000020000000 x22: 0000000000000001 x21: 0000000020000000
> x20: fffffc13cf006ec0 x19: 0000000000000000 x18: 0000000000000000
> x17: 0000000000000000 x16: 0000000000000000 x15: 0000000000000000
> x14: 0000000000000000 x13: 0000000000000000 x12: 0000000000000000
> x11: 0000000000000000 x10: 0000000000000000 x9 : ffffdddc5581377c
> x8 : 0000000000000000 x7 : 0000000000000011 x6 : ffff2717a8433000
> x5 : ffff80000e773810 x4 : ffffdddc55400000 x3 : 0000000000000000
> x2 : ffffdddc56b20000 x1 : ffff04f3c9a48040 x0 : 0000000000000000
> Call trace:
> __page_set_anon_rmap+0xc0/0xe8
> page_add_new_anon_rmap+0x13c/0x200
> copy_pte_range+0x6b8/0x1018
> copy_page_range+0x3a8/0x5e0
> dup_mmap+0x3a0/0x6e8
> dup_mm+0x78/0x140
> copy_process+0x1528/0x1b08
> kernel_clone+0xac/0x610
> __do_sys_clone+0x78/0xb0
> __arm64_sys_clone+0x30/0x40
> invoke_syscall+0x68/0x170
> el0_svc_common.constprop.0+0x80/0x250
> do_el0_svc+0x48/0xb8
> el0_svc+0x48/0x1a8
> el0t_64_sync_handler+0xb0/0xb8
> el0t_64_sync+0x1a0/0x1a4
> Code: 97f899f4 f9400273 17ffffeb 97f899f1 (d4210000)
> ---[ end trace dc65e5edd0f362fa ]---
> Kernel panic - not syncing: Oops - BUG: Fatal exception
> SMP: stopping secondary CPUs
> Kernel Offset: 0x5ddc4d400000 from 0xffff800008000000
> PHYS_OFFSET: 0xfffffb0c80000000
> CPU features: 0x44000cf1,00000806
> Memory Limit: none
> ---[ end Kernel panic - not syncing: Oops - BUG: Fatal exception ]---
>
> This problem has been fixed by the commit <fb3d824d1a46>
> ("mm/rmap: split page_dup_rmap() into page_dup_file_rmap()
> and page_try_dup_anon_rmap()"), but still exists in the
> linux-5.15.y branch.
>
> This patch is not applicable to this version because
> of the large version differences. Therefore, fix it by
> adding non-anonymous page check in the copy_present_page().
>
> Cc: stable@vger.kernel.org
> Fixes: 70e806e4e645 ("mm: Do early cow for pinned pages during fork() for ptes")
> Signed-off-by: Yuanzheng Song <songyuanzheng@huawei.com>
> Signed-off-by: Vlastimil Babka <vbabka@suse.cz>
> ---
> Hi, this was posted in [1] but seems stable@ was not actually included
> in the recipients.
> The 5.10 version [2] was applied as 935a8b62021 but 5.15 is missing.
>
> [1] https://lore.kernel.org/all/20221028075244.3112566-1-songyuanzheng@huawei.com/T/#u
> [2] https://lore.kernel.org/all/20221028030705.2840539-1-songyuanzheng@huawei.com/
>
>
> mm/memory.c | 11 +++++++++++
> 1 file changed, 11 insertions(+)
>
> diff --git a/mm/memory.c b/mm/memory.c
> index 6d058973a97e..4785aecca9a8 100644
> --- a/mm/memory.c
> +++ b/mm/memory.c
> @@ -903,6 +903,17 @@ copy_present_page(struct vm_area_struct *dst_vma, struct vm_area_struct *src_vma
> if (likely(!page_needs_cow_for_dma(src_vma, page)))
> return 1;
>
> + /*
> + * The vma->anon_vma of the child process may be NULL
> + * because the entire vma does not contain anonymous pages.
> + * A BUG will occur when the copy_present_page() passes
> + * a copy of a non-anonymous page of that vma to the
> + * page_add_new_anon_rmap() to set up new anonymous rmap.
> + * Return 1 if the page is not an anonymous page.
> + */
> + if (!PageAnon(page))
> + return 1;
> +
> new_page = *prealloc;
> if (!new_page)
> return -EAGAIN;
Reviewed-by: David Hildenbrand <david@redhat.com>
--
Cheers,
David / dhildenb
next prev parent reply other threads:[~2024-11-13 16:45 UTC|newest]
Thread overview: 5+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-11-13 16:31 Vlastimil Babka
2024-11-13 16:45 ` David Hildenbrand [this message]
2024-11-15 5:17 ` Greg KH
2024-11-15 5:24 ` Patch "mm/memory: add non-anonymous page check in the copy_present_page()" has been added to the 5.15-stable tree gregkh
-- strict thread matches above, loose matches on Subject: below --
2022-10-28 7:52 [PATCH STABLE 5.15] mm/memory: add non-anonymous page check in the copy_present_page() Yuanzheng Song
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=99a77c9a-68ca-4445-bcbf-4681ca20a482@redhat.com \
--to=david@redhat.com \
--cc=linux-mm@kvack.org \
--cc=peterx@redhat.com \
--cc=songyuanzheng@huawei.com \
--cc=stable@vger.kernel.org \
--cc=vbabka@suse.cz \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox