Date: Thu, 11 Jan 2018 20:11:25 +0100
From: Peter Zijlstra <peterz@infradead.org>
To: Vince Weaver <vincent.weaver@maine.edu>
Cc: Ingo Molnar <mingo@kernel.org>, linux-kernel@vger.kernel.org,
        Ingo Molnar <mingo@redhat.com>,
        Arnaldo Carvalho de Melo <acme@kernel.org>,
        Thomas Gleixner <tglx@linutronix.de>
Subject: Re: perf: perf_fuzzer quickly locks up on 4.15-rc7
Message-ID: <20180111191125.GG2369@hirez.programming.kicks-ass.net>
References: <alpine.DEB.2.21.1801081528110.14364@macbook-air>
 <20180109102507.GG6176@hirez.programming.kicks-ass.net>
 <20180109132602.GA2369@hirez.programming.kicks-ass.net>
 <alpine.DEB.2.21.1801090841300.15884@macbook-air>
 <20180111091353.GW6176@hirez.programming.kicks-ass.net>
 <alpine.DEB.2.21.1801111022520.21701@macbook-air>
 <20180111153845.GD6176@hirez.programming.kicks-ass.net>
 <alpine.DEB.2.21.1801111139320.22460@macbook-air>
 <alpine.DEB.2.21.1801111144450.22460@macbook-air>
 <20180111170347.GG6176@hirez.programming.kicks-ass.net>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20180111170347.GG6176@hirez.programming.kicks-ass.net>
User-Agent: Mutt/1.9.2 (2017-12-15)
Sender: linux-kernel-owner@vger.kernel.org

On Thu, Jan 11, 2018 at 06:03:47PM +0100, Peter Zijlstra wrote:
> > On Thu, 11 Jan 2018, Vince Weaver wrote:
> > [  823.919729] BUG: unable to handle kernel paging request at ffff88011a7a1000
> > [  823.926928] IP: 0x7fbda0042b3c
> > 
> > I'm dumping vmlinux and can't find address 0x7fbda0042b3c at all.
> 
> That's a user address..

> > [  823.930123] PGD 22ea067 P4D 22ea067 PUD 22ed067 PMD 11a520063 PTE 800000011a7a1063
> > [  823.938012] Oops: 0002 [#1] SMP PTI
> > [  823.968898] CPU: 1 PID: 17688 Comm: perf_fuzzer Not tainted 4.15.0-rc7+ #211
> > [  823.976152] Hardware name: AOpen   DE7000/nMCP7ALPx-DE R1.06 Oct.19.2012, BIOS 080015  10/19/2012
> > [  823.985259] RIP: 0033:0x7fbda0042b3c
> > [  823.989002] RSP: 002b:00007ffed7fa46c8 EFLAGS: 00010246
> > [  823.994434] RAX: 000000000000462d RBX: 000000000000000c RCX: 00007fbda0042b3c
> > [  824.001786] RDX: 0000000000000000 RSI: 00007ffed7fa46dc RDI: 000000000000462d
> > [  824.009135] RBP: 00007ffed7fa46e0 R08: 00007fbda032e0e4 R09: 0000000000000000
> > [  824.016483] R10: 0000000000000000 R11: 0000000000000246 R12: 0000000000401950
> > [  824.023832] R13: 00007ffed7fa6ae0 R14: 0000000000000000 R15: 0000000000000000
> > [  824.031198] FS:  00007fbda0540700(0000) GS:ffff88011fc80000(0000) knlGS:0000000000000000
> > [  824.039530] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
> > [  824.045477] CR2: ffff88011a7a1000 CR3: 000000011a998000 CR4: 00000000000407e0
> > [  824.052827] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
> > [  824.060176] DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000600
> > [  824.072767] CR2: ffff88011a7a1000
> > [  824.076248] ---[ end trace 31d5336f74638197 ]--

So its faulting while looking up the userspace page tables, so
something buggered. The CR3 has bit 12 clear so its a kernel CR3, which
is also somewhat puzzling.

Oooh, right, its the CR3 from when we report the OOPS, which is less
than useful.

I can't get further than that the page-tables are screwed..