linux-kernel - Re: Kernel panic - not syncing: System is deadlocked on memory

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] [day] [month] [year] [list]

Message-ID: <01100196bf3f39f1-1e016101-be4b-4388-b7ee-b086346dec38-000000@eu-north-1.amazonses.com>
Date: Sun, 11 May 2025 12:08:34 +0000
From: Ozgur Kara <ozgur@...sey.org>
To: Yin Guoyu <y04609127@...il.com>
Cc: akpm@...ux-foundation.org, linux-mm@...ck.org, 
	linux-kernel@...r.kernel.org
Subject: Re: Kernel panic - not syncing: System is deadlocked on memory

Yin Guoyu <y04609127@...il.com>, 11 May 2025 Paz, 14:35 tarihinde şunu yazdı:
>
> Hi,
> When I test the Linux kernel (commit
> 38fec10eb60d687e30c8c6b5420d86e8149f7557), I encountered a kernel
> panic issue related to memory allocation, specifically in the
> alloc_pages_mpol function. The panic message indicates "System is
> deadlocked on memory," and the call trace points to
> __alloc_frozen_pages_noprof failing to allocate pages, leading to an
> OOM condition.
> The error originates in alloc_pages_mpol, which calls policy_nodemask
> to determine the nodemask and nid for memory allocation. The
> subsequent call to __alloc_frozen_pages_noprof fails when the system
> runs out of memory, with no swap available and no killable processes.
> This crash can be triggered by executing the provided C reproducer
> multiple times. The reproducer performs a series of system calls
> (e.g., file creation, perf_event_open, io_submit, and filesystem
> mounts), which likely stress the memory management system and expose
> the bug.

Hello,

did you open a bug for this?

https://bugzilla.kernel.org

And i'm not sure if this is a bug i gues you are using qemu and you
experienced OOM (out of memory) because there is no swap and  physical
ram is completely full.
it gets killed while kernel is initializing.

[  133.880979] Free swap  = 0kB
[  133.889391] Out of memory and no killable processes...

This was an expected kernel behavior when there was no swap.

Regards,

Ozgur

> This can be reproduced on:
> HEAD commit:
> 38fec10eb60d687e30c8c6b5420d86e8149f7557
> console output : https://pastebin.com/raw/T4dfWHSf
> kernel config : https://pastebin.com/raw/u0Efyj5P
> C reproducer : https://pastebin.com/raw/3hPqjF6M
>
>
>