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] [day] [month] [year] [list]
Message-Id: <5bfebbd4-d12e-4735-acab-549a7cf9604a@app.fastmail.com>
Date: Tue, 09 Jul 2024 21:33:01 +0200
From: "Arnd Bergmann" <arnd@...db.de>
To: "Jeff Johnson" <quic_jjohnson@...cinc.com>,
 "Kalle Valo" <kvalo@...nel.org>
Cc: linux-kernel@...r.kernel.org, ath12k@...ts.infradead.org,
 kasan-dev@...glegroups.com, "Andrey Ryabinin" <ryabinin.a.a@...il.com>,
 "Alexander Potapenko" <glider@...gle.com>,
 "Andrey Konovalov" <andreyknvl@...il.com>,
 "Dmitry Vyukov" <dvyukov@...gle.com>,
 "Vincenzo Frascino" <vincenzo.frascino@....com>
Subject: Re: crosstool: x86 kernel compiled with GCC 14.1 fails to boot

On Tue, Jul 9, 2024, at 17:29, Jeff Johnson wrote:
> On 7/8/2024 10:44 PM, Arnd Bergmann wrote:
>> On Tue, Jul 9, 2024, at 05:55, Jeff Johnson wrote:
>
> I picked my favorite to begin with, enabling KASAN (which in turn enabled a
> few others). The resulting kernel did not boot for me (just saw a black screen
> after the GRUB menu). Diff between working and non-working config is below.

Ok, good to know. I've added the KASAN developers to Cc now, maybe
they have already seen reports of x86 kernels failing with gcc-14?

> I then downloaded and built the config you supplied. With that I have the same
> behavior as my original config, the display is frozen with:
> Loading initial ramdisk ...

Interesting, so the same config that works for me fails on your
machine. I can see three possible reasons for this:

- qemu vs hardware -- Can you try running this kernel in
  qemu-system-x86_64 to see if that still boots

- kernel version -- it's possible that this is a known bug
  that was already fixed in the 6.10-rc7 kernel source I
  tried, or that your source tree has a new bug that I don't.
  Which version did you try?

- cross-compile vs native compile -- It's possible that my
  cross-built native x86_64 compiler has a bug that is not
  in natively built gcc binaries, or in the cross compiler
  I have on ARM. I've mostly ruled this one out by building
  the same kernel using the x86 compilers through qemu-user.

     Arnd

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ