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: <e6357e8e-5dd4-47a1-b7a4-603d32633692@redhat.com>
Date: Thu, 15 Aug 2024 18:39:19 +0200
From: David Hildenbrand <david@...hat.com>
To: Hao Ge <hao.ge@...ux.dev>, rppt@...nel.org, akpm@...ux-foundation.org,
 surenb@...gle.com, babka@...e.cz, kent.overstreet@...ux.dev, kees@...nel.org
Cc: linux-mm@...ck.org, linux-kernel@...r.kernel.org,
 Hao Ge <gehao@...inos.cn>
Subject: Re: [PATCH] codetag: debug: mark codetags for cma pages as empty

On 15.08.24 11:03, Hao Ge wrote:
> From: Hao Ge <gehao@...inos.cn>
> 
> Here are some warnning
> 
> [    1.342603] ------------[ cut here ]------------
> [    1.342608] alloc_tag was not set
> [    1.342623] WARNING: CPU: 0 PID: 1 at ./include/linux/alloc_tag.h:130 __free_pages+0x1a4/0x1b4
> [    1.342633] Modules linked in:
> [    1.342639] CPU: 0 UID: 0 PID: 1 Comm: swapper/0 Not tainted 6.11.0-rc3+ #12
> [    1.342645] pstate: 60400005 (nZCv daif +PAN -UAO -TCO -DIT -SSBS BTYPE=--)
> [    1.342650] pc : __free_pages+0x1a4/0x1b4
> [    1.342654] lr : __free_pages+0x1a4/0x1b4
> [    1.342658] sp : ffff8000833dbcb0
> [    1.342661] x29: ffff8000833dbcb0 x28: ffff000b34d9dac0 x27: ffff80008175f000
> [    1.342669] x26: 00000000000fc000 x25: ffff800082ac6028 x24: ffff80008252e000
> [    1.342675] x23: ffff0000c0660000 x22: 0000000000000000 x21: 0000000000000009
> [    1.342681] x20: 0000000000000000 x19: fffffd7fc3ac0000 x18: ffffffffffffffff
> [    1.342688] x17: ffff80008105bf68 x16: ffff800081841bb4 x15: 0720072007200720
> [    1.342694] x14: 0720072007200720 x13: 0720072007200720 x12: 0720072007200720
> [    1.342700] x11: 0720072007200720 x10: 0720072007200720 x9 : 0720072007200720
> [    1.342706] x8 : 0720072007200720 x7 : ffff80008237bad0 x6 : c0000000ffff7fff
> [    1.342712] x5 : ffff80008237ba78 x4 : ffff8000820bbad0 x3 : 0000000000000001
> [    1.342719] x2 : 13c63f521095dc00 x1 : 13c63f521095dc00 x0 : 0000000000000000
> [    1.342725] Call trace:
> [    1.342728]  __free_pages+0x1a4/0x1b4
> [    1.342732]  init_cma_reserved_pageblock+0x5c/0xac
> [    1.342739]  cma_init_reserved_areas+0x2b4/0x3f4
> [    1.342744]  do_one_initcall+0x54/0x368
> [    1.342749]  kernel_init_freeable+0x23c/0x450
> [    1.342755]  kernel_init+0x2c/0x144
> [    1.342760]  ret_from_fork+0x10/0x20
> [    1.342764] ---[ end trace 0000000000000000 ]---
> 
> To avoid debug warnings while freeing cma pages which were not
> allocated with usual allocators, mark their codetags as empty before
> freeing.
> 
> Reference commit d224eb0287fb ("codetag: debug: mark codetags
> for reserved pages as empty")
> 
> Fixes: 4b8736964640 ("mm/slab: add allocation accounting into slab allocation and free paths")
> Signed-off-by: Hao Ge <gehao@...inos.cn>
> ---
>   mm/mm_init.c | 10 ++++++++++
>   1 file changed, 10 insertions(+)
> 
> diff --git a/mm/mm_init.c b/mm/mm_init.c
> index 75c3bd42799b..d7d40a8d738f 100644
> --- a/mm/mm_init.c
> +++ b/mm/mm_init.c
> @@ -2243,6 +2243,16 @@ void __init init_cma_reserved_pageblock(struct page *page)
>   		set_page_count(p, 0);
>   	} while (++p, --i);
>   
> +	/* pages were reserved and not allocated */
> +	if (mem_alloc_profiling_enabled()) {
> +		union codetag_ref *ref = get_page_tag_ref(page);
> +
> +		if (ref) {
> +			set_codetag_empty(ref);
> +			put_page_tag_ref(ref);
> +		}
> +	}
> +
>   	set_pageblock_migratetype(page, MIGRATE_CMA);
>   	set_page_refcounted(page);
>   	__free_pages(page, pageblock_order);

See https://lkml.kernel.org/r/20240813150758.855881-2-surenb@google.com

-- 
Cheers,

David / dhildenb


Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ