linux-kernel - Re: 5.4-rc1 boot regression with kmemleak enabled

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <20191105182212.GF22987@arrakis.emea.arm.com>
Date:   Tue, 5 Nov 2019 18:22:13 +0000
From:   Catalin Marinas <catalin.marinas@....com>
To:     Amir Goldstein <amir73il@...il.com>
Cc:     Theodore Tso <tytso@....edu>, fstests <fstests@...r.kernel.org>,
        Matthew Wilcox <willy@...radead.org>,
        Michal Hocko <mhocko@...nel.org>, Qian Cai <cai@....pw>,
        Andrew Morton <akpm@...ux-foundation.org>,
        Linus Torvalds <torvalds@...ux-foundation.org>,
        Linux MM <linux-mm@...ck.org>,
        linux-kernel <linux-kernel@...r.kernel.org>,
        Thomas Gleixner <tglx@...utronix.de>
Subject: Re: 5.4-rc1 boot regression with kmemleak enabled

On Tue, Nov 05, 2019 at 08:17:11PM +0200, Amir Goldstein wrote:
> On Tue, Nov 5, 2019 at 5:31 PM Catalin Marinas <catalin.marinas@....com> wrote:
> >
> > On Tue, Nov 05, 2019 at 02:33:48PM +0200, Amir Goldstein wrote:
> > > On Tue, Nov 5, 2019 at 1:54 PM Catalin Marinas <catalin.marinas@....com> wrote:
> > > > (sorry if you got this message twice; our SMTP server went bust)
> > > >
> > > > On Tue, Nov 05, 2019 at 09:14:06AM +0200, Amir Goldstein wrote:
> > > > > My kvm-xfstests [1] VM doesn't boot with kmemleak enabled since commit
> > > > > c5665868183f ("mm: kmemleak: use the memory pool for early allocations").
> > > > >
> > > > > There is no console output when running:
> > > > >
> > > > > $ kvm -boot order=c -net none -machine type=pc,accel=kvm:tcg -cpu host \
> > > > >     -drive file=$ROOTFS,if=virtio,snapshot=on -vga none -nographic \
> > > > >     -smp 2 -m 2048 -serial mon:stdio --kernel $KERNEL \
> > > > >     --append 'root=/dev/vda console=ttyS0,115200'
> > > >
> > > > This was fixed in 5.4-rc4, see commit 2abd839aa7e6 ("kmemleak: Do not
> > > > corrupt the object_list during clean-up").
> > >
> > > Did not fix my issue.
> > > Still not booting with 5.4-rc6.
> > > Any other suggestions?
> >
> > Can you pass an earlyprintk=ttyS0,115200 (if that's the correct x86
> > syntax) on the kernel command line? It may print some early messages
> > that would help with debugging.
[...]
> [    0.022796] BUG: unable to handle page fault for address: 0000000000001ff0
> [    0.023682] #PF: supervisor read access in kernel mode
> [    0.024341] #PF: error_code(0x0000) - not-present page
> [    0.025000] PGD 0 P4D 0 
> [    0.025326] Oops: 0000 [#1] SMP PTI
> [    0.025775] CPU: 0 PID: 0 Comm: swapper Not tainted 5.4.0-rc6-xfstests #4302
> [    0.026683] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.10.2-1ubuntu1 04/01/2014
> [    0.027836] RIP: 0010:get_stack_info+0xa7/0x146

Ah, it looks very similar to this report:

http://lkml.kernel.org/r/20191019114421.GK9698@uranus.lan

Thomas had a patch here:

https://lore.kernel.org/linux-mm/alpine.DEB.2.21.1910231950590.1852@nanos.tec.linutronix.de/

but not sure whether it has hit mainline yet.

-- 
Catalin