[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20230603043028.GE1128875@mit.edu>
Date: Sat, 3 Jun 2023 00:30:28 -0400
From: "Theodore Ts'o" <tytso@....edu>
To: Baokun Li <libaokun1@...wei.com>
Cc: syzbot <syzbot+list5ea887c46d22b2acf805@...kaller.appspotmail.com>,
adilger.kernel@...ger.ca, linux-ext4@...r.kernel.org,
linux-fsdevel@...r.kernel.org, linux-kernel@...r.kernel.org,
syzkaller-bugs@...glegroups.com, yangerkun <yangerkun@...wei.com>
Subject: Re: [syzbot] Monthly ext4 report (May 2023)
On Sat, Jun 03, 2023 at 10:05:44AM +0800, Baokun Li wrote:
>
> There are many sources of syzkaller issues, and the patches I send out
> to fix syzkaller issues add Reported-by to all but the ones that fix issues
> reported by our internal syzbot.
> However, there may be multiple syzbot reports for the same issue.
Yes, that happens a lot especially for Lockdep reports; depending on
how the reproducer triggers the locks in which order, there can often
be multiple different lockdep signatures, and Syzkaller can't tell
that they are all the same thing.
I tend to focus on syzbot reproducers on the upstream Linux instance,
rather than the Android-5.15 syzkaller instance. And that allows me
to use the ext4 subsystem dashboard available at:
https://syzkaller.appspot.com/upstream/s/ext4
It's against this list of reports that the Monthly ext4 report is
generated. So if people who are submitting fixes against syzkaller
reports, it would be nice if they were to check the ext4 dashboard
above to look for syzbot reports that might be also relevant to your
patch.
Thanks,
- Ted
Powered by blists - more mailing lists