Re: WARNING in is_bpf_text_address

From: Bart Van Assche
Date: Thu Aug 15 2019 - 21:42:33 EST


On 8/15/19 12:51 AM, Will Deacon wrote:
Hi Bart,

On Sat, Aug 10, 2019 at 05:24:06PM -0700, syzbot wrote:
syzbot has found a reproducer for the following crash on:

HEAD commit: 451577f3 Merge tag 'kbuild-fixes-v5.3-3' of git://git.kern..
git tree: upstream
console output: https://syzkaller.appspot.com/x/log.txt?x=120850a6600000
kernel config: https://syzkaller.appspot.com/x/.config?x=2031e7d221391b8a
dashboard link: https://syzkaller.appspot.com/bug?extid=bd3bba6ff3fcea7a6ec6
compiler: clang version 9.0.0 (/home/glider/llvm/clang
80fee25776c2fb61e74c1ecb1a523375c2500b69)
syz repro: https://syzkaller.appspot.com/x/repro.syz?x=130ffe4a600000
C reproducer: https://syzkaller.appspot.com/x/repro.c?x=17137d2c600000

The bug was bisected to:

commit a0b0fd53e1e67639b303b15939b9c653dbe7a8c4
Author: Bart Van Assche <bvanassche@xxxxxxx>
Date: Thu Feb 14 23:00:46 2019 +0000

locking/lockdep: Free lock classes that are no longer in use

bisection log: https://syzkaller.appspot.com/x/bisect.txt?x=152f6a9da00000
final crash: https://syzkaller.appspot.com/x/report.txt?x=172f6a9da00000
console output: https://syzkaller.appspot.com/x/log.txt?x=132f6a9da00000

I know you don't think much to these reports, but please could you have a
look (even if it's just to declare it a false positive)?

Hi Will,

Had you already noticed the following message?

https://lore.kernel.org/bpf/d76d7a63-7854-e92d-30cb-52546d333ffe@xxxxxxxxxxxxx/

From that message: "Hey Bart, don't think it's related in any way to your commit. I'll allocate some time on working on this issue today, thanks!"

Bart.