linux-kernel - Re: [PATCH] do not try to sync identity map for non-mapped pages

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: <20130407133346.GD31299@pd.tnic>
Date:	Sun, 7 Apr 2013 15:33:46 +0200
From:	Borislav Petkov <bp@...en8.de>
To:	Dave Hansen <dave@...1.net>
Cc:	linux-kernel@...r.kernel.org, hpa@...ux.intel.com,
	penguin-kernel@...ove.sakura.ne.jp, x86@...nel.org,
	akpm@...ux-foundation.org, suresh.b.siddha@...el.com,
	khlebnikov@...nvz.org, bhelgaas@...gle.com
Subject: Re: [PATCH] do not try to sync identity map for non-mapped pages

Hey Dave,

On Thu, Mar 07, 2013 at 08:31:51AM -0800, Dave Hansen wrote:
> 
> The original bug reporter says this fixes it for him, so I'm
> broadening the cc list a bit.  I assume this should just get
> sucked in to the x86 tree.

looks like we haven't whacked all the moles - I keep seeing this when
testing 32-bit builds in qemu on latest Linus + tip. I'd guess this is
still that /dev/mem accessing thing called wdm.

I'm still wondering though whether we should BUG_ON on a /dev/mem
access?

I've added debug output to show why we're triggering:

[  471.102902] slow_virt_to_phys((void *)x): 0x0, phys_addr: 0x37bfe000
[  471.119500] ------------[ cut here ]------------
[  471.119500] kernel BUG at arch/x86/mm/physaddr.c:85!
[  471.119500] invalid opcode: 0000 [#1] PREEMPT SMP 
[  471.119500] Modules linked in:
[  471.119500] Pid: 1571, comm: wdm Not tainted 3.9.0-rc5+ #42 Bochs Bochs
[  471.119500] EIP: 0060:[<c1032f56>] EFLAGS: 00000206 CPU: 0
[  471.119500] EIP is at __phys_addr+0x86/0xb0
[  471.119500] EAX: 37bfe000 EBX: 37bfe000 ECX: 00000001 EDX: 37bfe000
[  471.119500] ESI: f7bfe000 EDI: 00002000 EBP: f67f1f3c ESP: f67f1f28
[  471.119500]  DS: 007b ES: 007b FS: 00d8 GS: 0033 SS: 0068
[  471.119500] CR0: 8005003b CR2: bfeb12d4 CR3: 35edd000 CR4: 000006f0
[  471.119500] DR0: 00000000 DR1: 00000000 DR2: 00000000 DR3: 00000000
[  471.119500] DR6: 00000000 DR7: 00000000
[  471.119500] Process wdm (pid: 1571, ti=f67f0000 task=f5c90000 task.ti=f67f0000)
[  471.119500] Stack:
[  471.119500]  c16b1f2c 00000000 37bfe000 00000000 00002000 f67f1f64 c131d074 00002000
[  471.119500]  00000246 bfeb12ec 00000000 00000000 f67a6c40 c131d040 00002000 f67f1f8c
[  471.119500]  c1129c85 f67f1f98 f67f0000 f5ebe864 c131d040 00000020 f67a6c40 00000000
[  471.119500] Call Trace:
[  471.119500]  [<c131d074>] read_mem+0x34/0x100
[  471.119500]  [<c131d040>] ? write_mem+0x110/0x110
[  471.119500]  [<c1129c85>] vfs_read+0x85/0x130
[  471.119500]  [<c131d040>] ? write_mem+0x110/0x110
[  471.119500]  [<c1129e87>] sys_read+0x47/0xa0
[  471.119500]  [<c1546e5e>] sysenter_do_call+0x12/0x36
[  471.119500] Code: 0b a1 88 ae 0c c2 05 00 00 80 00 39 c6 72 bb a1 ac 1a 76 c1 2d 00 a0 3e 00 25 00 00 e0 ff 2d 00 20 00 00 39 c6 73 a3 0f 0b 0f 0b <0f> 0b 89 f0 e8 41 ca ff ff 89 5c 24 08 89 44 24 04 c7 04 24 2c
[  471.119500] EIP: [<c1032f56>] __phys_addr+0x86/0xb0 SS:ESP 0068:f67f1f28
[  471.508967] ---[ end trace 5fc00ac35d61284a ]---

Hmmm.

-- 
Regards/Gruss,
    Boris.

Sent from a fat crate under my desk. Formatting is fine.
--
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/