Received-SPF: pass (google.com: domain of linux-kernel+bounces-94105-linux.lists.archive=gmail.com@vger.kernel.org designates 147.75.80.249 as permitted sender) client-ip=147.75.80.249;
Precedence: bulk
MIME-Version: 1.0
References: <CABOYnLz8V-CMiZK0Gzz=eXf2G3E-psemp2pMZwZ_XJG53GawgA@mail.gmail.com>
 <CAKFNMomdU5RHVMt2CCXYMAb5oyjDwOVRitNM+XGGC65TQs1ECQ@mail.gmail.com>
 <CABOYnLxE86iTqTA3BOMLPHX5SeB--46S_4nec7H18H7B4oEi3w@mail.gmail.com> <CAKFNMomM0i1mOwkFsBta4rO+gDB1_LjSF_mENkB=PGF6a-tW-A@mail.gmail.com>
In-Reply-To: <CAKFNMomM0i1mOwkFsBta4rO+gDB1_LjSF_mENkB=PGF6a-tW-A@mail.gmail.com>
From: Ryusuke Konishi <konishi.ryusuke@gmail.com>
Date: Thu, 7 Mar 2024 00:03:19 +0900
Message-ID: <CAKFNMom7z+XG4J1sRC2aJ25aJZNzkUkrUtDnHa1LKDY-+0sbfQ@mail.gmail.com>
Subject: Re: [syzbot] [nilfs?] KMSAN: uninit-value in nilfs_add_checksums_on_logs
 (2)
To: xingwei lee <xrivendell7@gmail.com>
Cc: syzbot+47a017c46edb25eff048@syzkaller.appspotmail.com, 
	linux-fsdevel@vger.kernel.org, linux-kernel@vger.kernel.org, 
	linux-nilfs@vger.kernel.org, syzkaller-bugs@googlegroups.com
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

On Wed, Mar 6, 2024 at 4:20=E2=80=AFPM Ryusuke Konishi wrote:
>
> On Wed, Mar 6, 2024 at 4:07=E2=80=AFPM xingwei lee wrote:
> > On 3 Mar 2024, at 20:45, Ryusuke Konishi <konishi.ryusuke@gmail.com> wr=
ote:
> >
> > Hi, sorry for the delayed response.
> >
> > I test my reproducer in the linux 6.8-rc4 with KMSAN kernel config for =
one hours, it doesn=E2=80=99t trigger any crash or report as follows:
> >
> > [  315.607028][   T37] audit: type=3D1804 audit(1709708422.469:31293): =
pid=3D86478 uid=3D0 auid=3D0 ses=3D1 subj=3Dunconfined op=3Dinvalid_pcr cau=
se=3D0
> > [  315.608038][T86480] 884-0[86480]: segfault at 5c7ade ip 00000000005c=
7ade sp 00000000200001f8 error 14 likely on CPU 2 (core 2, socke)
> > [  315.611270][T86480] Code: Unable to access opcode bytes at 0x5c7ab4.
> > [  320.575680][   T37] kauditd_printk_skb: 1253 callbacks suppressed
> > [  320.575689][   T37] audit: type=3D1804 audit(1709708427.439:32130): =
pid=3D88573 uid=3D0 auid=3D0 ses=3D1 subj=3Dunconfined op=3Dinvalid_pcr cau=
se=3D0
> > [  320.576419][T88575] 884-0[88575]: segfault at 5c7ade ip 00000000005c=
7ade sp 00000000200001f8 error 14
> > [  320.576695][   T37] audit: type=3D1804 audit(1709708427.439:32131): =
pid=3D88574 uid=3D0 auid=3D0 ses=3D1 subj=3Dunconfined op=3Dinvalid_pcr cau=
se=3D0
> > [  320.579042][T88575]  likely on CPU 0 (core 0, socket 0)
> > [  320.584184][T88575] Code: Unable to access opcode bytes at 0x5c7ab4.
> > [  320.593832][   T37] audit: type=3D1804 audit(1709708427.459:32132): =
pid=3D88578 uid=3D0 auid=3D0 ses=3D1 subj=3Dunconfined op=3Dinvalid_pcr cau=
se=3D0
> > [  320.594549][T88580] 884-0[88580]: segfault at 5c7ade ip 00000000005c=
7ade sp 00000000200001f8 error 14 likely on CPU 1 (core 1, socke)
> > [  320.596256][   T37] audit: type=3D1804 audit(1709708427.459:32133): =
pid=3D88579 uid=3D0 auid=3D0 ses=3D1 subj=3Dunconfined op=3Dinvalid_pcr cau=
se=3D0
> > [  320.597901][T88580] Code: Unable to access opcode bytes at 0x5c7ab4.
> > [  320.610954][   T37] audit: type=3D1804 audit(1709708427.479:32134): =
pid=3D88583 uid=3D0 auid=3D0 ses=3D1 subj=3Dunconfined op=3Dinvalid_pcr cau=
se=3D0
> > [  320.611700][T88585] 884-0[88585]: segfault at 5c7ade ip 00000000005c=
7ade sp 00000000200001f8 error 14 likely on CPU 2 (core 2, socke)
> > [  320.613455][   T37] audit: type=3D1804 audit(1709708427.479:32135): =
pid=3D88584 uid=3D0 auid=3D0 ses=3D1 subj=3Dunconfined op=3Dinvalid_pcr cau=
se=3D0
> > [  320.615959][T88585] Code: Unable to access opcode bytes at 0x5c7ab4.
> > [  320.628571][   T37] audit: type=3D1804 audit(1709708427.489:32136): =
pid=3D88588 uid=3D0 auid=3D0 ses=3D1 subj=3Dunconfined op=3Dinvalid_pcr cau=
se=3D0
> > [  325.582663][   T37] kauditd_printk_skb: 1280 callbacks suppressed
> > [  325.582673][   T37] audit: type=3D1804 audit(1709708432.449:32990): =
pid=3D90727 uid=3D0 auid=3D0 ses=3D1 subj=3Dunconfined op=3Dinvalid_pcr cau=
se=3D0
> > [  325.583320][T90729] 884-0[90729]: segfault at 5c7ade ip 00000000005c=
7ade sp 00000000200001f8 error 14
> > [  325.583460][   T37] audit: type=3D1804 audit(1709708432.449:32991): =
pid=3D90728 uid=3D0 auid=3D0 ses=3D1 subj=3Dunconfined op=3Dinvalid_pcr cau=
se=3D0
> > [  325.585838][T90729]  likely on CPU 1 (core 1, socket 0)
> > [  325.590985][T90729] Code: Unable to access opcode bytes at 0x5c7ab4.
> > [  325.599620][   T37] audit: type=3D1804 audit(1709708432.459:32992): =
pid=3D90732 uid=3D0 auid=3D0 ses=3D1 subj=3Dunconfined op=3Dinvalid_pcr cau=
se=3D0
> > [  325.601818][T90734] 884-0[90734]: segfault at 5c7ade ip 00000000005c=
7ade sp 00000000200001f8 error 14
> > [  325.601827][   T37] audit: type=3D1804 audit(1709708432.459:32993): =
pid=3D90733 uid=3D0 auid=3D0 ses=3D1 subj=3Dunconfined op=3Dinvalid_pcr cau=
se=3D0
> > [  325.603945][T90734]  likely on CPU 2 (core 2, socket 0)
> > [  325.607037][T90734] Code: Unable to access opcode bytes at 0x5c7ab4.
> > [  325.617928][   T37] audit: type=3D1804 audit(1709708432.479:32994): =
pid=3D90737 uid=3D0 auid=3D0 ses=3D1 subj=3Dunconfined op=3Dinvalid_pcr cau=
se=3D0
> > [  325.618862][T90739] 884-0[90739]: segfault at 5c7ade ip 00000000005c=
7ade sp 00000000200001f8 error 14
> > [  325.620190][   T37] audit: type=3D1804 audit(1709708432.479:32995): =
pid=3D90738 uid=3D0 auid=3D0 ses=3D1 subj=3Dunconfined op=3Dinvalid_pcr cau=
se=3D0
> > [  325.623238][T90739]  likely on CPU 0 (core 0, socket 0)
> > [  325.623803][T90739] Code: Unable to access opcode bytes at 0x5c7ab4.
> > [  325.632693][   T37] audit: type=3D1804 audit(1709708432.499:32996): =
pid=3D90742 uid=3D0 auid=3D0 ses=3D1 subj=3Dunconfined op=3Dinvalid_pcr cau=
se=3D0
> >
> > It=E2=80=99s seems this issue have been fixed.
> >
> > I'd like to isolate that the issue is still not fixed with the latest
> > fixes, but I need to do some trial and error to reestablish a testable
> > (bootable) KMSAN-enabled kernel config.
> >
> > Thanks,
> > Ryusuke Konishi
> >
> >
> > I hope it helps.
> > Best regards
> > xingwei Lee
>
> Thank you!
> That helps a lot.
>
> Regards,
> Ryusuke Konishi

Ahh.  Looking at the February 28th syzbot crash, it appears that this
issue still exists in recent -rc releases.
So I'm going to investigate without closing it.

Regards,
Ryusuke Konishi