JFFS2 locking issue on 3.14 + patches

[Joakim Tjernlund]

"linux-mtd" <linux-mtd-bounces at lists.infradead.org> wrote on 2014/04/28 
17:57:12:

> From: Michal Simek <monstr at monstr.eu>
> To: linux-mtd at lists.infradead.org, Brian Norris 
<computersforpeace at gmail.com>, Li Zefan <lizefan at huawei.com>, Wang Guoli 
<andy.wangguoli at huawei.com>, Kamlakant Patel 
<kamlakant.patel at broadcom.com>, Ajesh Kunhipurayil Vijayan 
<ajesh at broadcom.com>, 
> Date: 2014/04/28 18:01
> Subject: JFFS2 locking issue on 3.14 + patches
> Sent by: "linux-mtd" <linux-mtd-bounces at lists.infradead.org>
> 
> Hi,
> 
> we are having this problem with JFFS2 rootfs. It is v3.14 kernel
> with these 5 patches which cherry-picked from 3.15.
> 
> 41bf1a2 jffs2: Fix crash due to truncation of csize
> 3367da5 jffs2: Fix segmentation fault found in stress test
> 01887a3 jffs2: unlock f->sem on error in jffs2_new_inode()
> 13b546d jffs2: avoid soft-lockup in jffs2_reserve_space_gc()
> 3ead957 jffs2: remove from wait queue after schedule()
> 
> There were some discussions on mtd mailing list about these locking
> issues which are probably not resolved yet.
> 
> Here is the full branch I do use if someone is interested.
> 
http://git.monstr.eu/?p=linux-2.6-microblaze.git;a=shortlog;h=refs/heads/xilinx/master-next

> I have also attached .config file for reference.
> 
> I have tested jffs2 rootfs on previous kernel versions and it was 
working
> fine that's why I think there is no problem with jffs2 rootfs itself.
> 
> on unicpu system is less amount of CRC faults.

Wild guess, JFFS2 lacks readpages(). Perhaps you need to add it?

    Jocke