[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CAFULd4bFbhOQ+mx2=pKqLbLznR-UFZamS4HhdUYZF2qFKNcDVg@mail.gmail.com>
Date: Mon, 25 Mar 2024 11:31:40 +0100
From: Uros Bizjak <ubizjak@...il.com>
To: kernel test robot <oliver.sang@...el.com>
Cc: oe-lkp@...ts.linux.dev, lkp@...el.com, linux-kernel@...r.kernel.org,
Ingo Molnar <mingo@...nel.org>, Nadav Amit <namit@...are.com>, Andy Lutomirski <luto@...nel.org>,
Brian Gerst <brgerst@...il.com>, Denys Vlasenko <dvlasenk@...hat.com>,
"H. Peter Anvin" <hpa@...or.com>, Linus Torvalds <torvalds@...ux-foundation.org>,
Peter Zijlstra <peterz@...radead.org>, Thomas Gleixner <tglx@...utronix.de>,
Josh Poimboeuf <jpoimboe@...hat.com>, linux-mm@...ck.org
Subject: Re: [linus:master] [x86/percpu] ca42563486: BUG:unable_to_handle_page_fault_for_address
On Mon, Mar 25, 2024 at 9:50 AM kernel test robot <oliver.sang@...el.com> wrote:
>
>
> hi, Uros Bizjak,
>
> we reported an early crash issue for this commit last Oct.
> https://lore.kernel.org/all/202310071301.a5113890-oliver.sang@intel.com/
>
> as you mentioned at that time
> "It is KASAN that is not compatible with named address spaces [1]."
>
> now we noticed the commit is merged into mainline, and we observed a different
> issue related with kcsan_setup_watchpoint. below detail report FYI.
>
> [1] https://lore.kernel.org/lkml/CAHk-=wi6U-O1wdPOESuCE6QO2OaPu0hEzaig0uDOU4L5CREhug@mail.gmail.com/
>
Yes, it is the same issue, but with KCSAN.
> Hello,
>
> kernel test robot noticed "BUG:unable_to_handle_page_fault_for_address" on:
>
> commit: ca4256348660cb2162668ec3d13d1f921d05374a ("x86/percpu: Use C for percpu read/write accessors")
> https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git master
>
> [test failed on linus/master b3603fcb79b1036acae10602bffc4855a4b9af80]
> [test failed on linux-next/master 226d3c72fcde130a99d760895ebdd20e78e02cb5]
>
> in testcase: boot
>
> compiler: gcc-12
> test machine: qemu-system-x86_64 -enable-kvm -cpu SandyBridge -smp 2 -m 16G
We already have the candidate fix for GCC [1], the fix will be in gcc-13.3+.
[1] https://gcc.gnu.org/bugzilla/show_bug.cgi?id=111736
Please find attached an emergency patch for mainline that disables
named address spaces when KCSAN is enabled. I'll post a formal patch
submission in a moment. The named AS will be re-enabled with KCSAN
with a patch via -tip tree.
Thanks,
Uros.
View attachment "p.diff.txt" of type "text/plain" (403 bytes)
Powered by blists - more mailing lists