On Wed, Jan 15, 2025 at 10:16 PM Kun Hu <
huk23@xxxxxxxxxxxxxx> wrote:
>
>
> > This is an already known issue and the relevant patch has been applied.
> > Please make sure that the following patch is applied to the kernel you tested.
> >
> > a5324b3a488d exfat: fix the infinite loop in __exfat_free_cluster()
> >
> > or try to reproduce it with linux-6.13-rc7.
>
> Hi Namjae,
Hi Kun,
>
> We still successfully reproduced it on the v6.13-rc7. Firstly, I apologize for taking up your time, I’m not sure if this is a significant issue since from the reproducer it kind of looks like it’s caused via fault injection.
>
>
> The syz_mount_image in the syscall reproducer mounts a randomly generated image and also has the potential to trigger an abnormal path to the file system. Specifically, the . /file0 file is crafted to contain invalid FAT table or bitmap information, it is possible to cause abnormal cyclic behavior in __exfat_free_cluster.
>
> Because p_chain->size is artificially constructed, if it has a large value, then exfat_clear_bitmap will be called frequently. As the call stack shows, the program eventually deadlocks in the loop in __exfat_free_cluster.
>
> This link is a link to our crash log in the rc7 kernel tree:
>
> Link:
https://github.com/pghk13/Kernel-Bug/blob/main/0103_6.13rc5_%E6%9C%AA%E6%8A%A5%E5%91%8A/%E6%9C%89%E7%9B%B8%E4%BC%BC%E6%A3%80%E7%B4%A2%E8%AE%B0%E5%BD%95/39-BUG_%20soft%20lockup%20in%20sys_unlink/crashlog0115_rc7.txt
>
> As I said earlier, I'm still consistently reporting the crash I found to you guys now because I'm not sure if this issue is useful to you. If it is not useful, please ignore it. I hope it doesn't take up too much of your time.
Can you check an attached patch ?
Thanks.
>
> ———
> Kun Hu
>
>
From 0bb26ac6aa65c9d9d41f19af305fb72c480fd1d6 Mon Sep 17 00:00:00 2001