NetBSD Problem Report #52602

From www@NetBSD.org  Sun Oct  8 16:15:39 2017
Return-Path: <www@NetBSD.org>
Received: from mail.netbsd.org (mail.netbsd.org [199.233.217.200])
	(using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits))
	(Client CN "mail.NetBSD.org", Issuer "mail.NetBSD.org CA" (not verified))
	by mollari.NetBSD.org (Postfix) with ESMTPS id 273F67A20B
	for <gnats-bugs@gnats.NetBSD.org>; Sun,  8 Oct 2017 16:15:39 +0000 (UTC)
Message-Id: <20171008161538.2D6647A2B0@mollari.NetBSD.org>
Date: Sun,  8 Oct 2017 16:15:38 +0000 (UTC)
From: venture37@geeklan.co.uk
Reply-To: venture37@geeklan.co.uk
To: gnats-bugs@NetBSD.org
Subject: Panic when trying to list probes after dtrace_sdt.kmod has been loaded
X-Send-Pr-Version: www-1.0

>Number:         52602
>Category:       kern
>Synopsis:       Panic when trying to list probes after dtrace_sdt.kmod has been loaded
>Confidential:   no
>Severity:       serious
>Priority:       medium
>Responsible:    kern-bug-people
>State:          closed
>Class:          sw-bug
>Submitter-Id:   net
>Arrival-Date:   Sun Oct 08 16:20:00 +0000 2017
>Closed-Date:    Mon Oct 09 01:13:24 +0000 2017
>Last-Modified:  Mon Oct 09 01:13:24 +0000 2017
>Originator:     Sevan Janiyan
>Release:        HEAD/201710080310Z
>Organization:
>Environment:
GENERIC i386 kernel
>Description:
system panics with 
uvm_fault(0xc12f7b40, 0xdc41f000, 1) -> 0xe
uvm_fault(0xc12f7b40, 0xdc41f000, 1) -> 0xe
fatal page fault in supervisor mode
trap type 6 code 0 eip 0xc011d265 cs 0x8 eflag 0x10246 cr2 0xdc41f007 ilevel 0 esp 0xc0f777c8
curlwp 0xc4c902a0 pid 34 lid 1 lowest kstack 0xdcc032c8
kernel: supervisor trap mode page fault, code=0
Stopped in pid 34.1 (dtrace) at netbsd:trap+0x9e6: movzbl 0(%eax),%ecx
>How-To-Repeat:
modload dtrace_sdt.kmod
run dtrace -l
>Fix:

>Release-Note:

>Audit-Trail:
From: Paul Goyette <paul@whooppee.com>
To: gnats-bugs@NetBSD.org
Cc: kern-bug-people@netbsd.org, gnats-admin@netbsd.org, netbsd-bugs@netbsd.org
Subject: Re: kern/52602: Panic when trying to list probes after dtrace_sdt.kmod
 has been loaded
Date: Mon, 9 Oct 2017 06:13:39 +0800 (+08)

 >> Description:
 > system panics with
 > uvm_fault(0xc12f7b40, 0xdc41f000, 1) -> 0xe
 > uvm_fault(0xc12f7b40, 0xdc41f000, 1) -> 0xe
 > fatal page fault in supervisor mode
 > trap type 6 code 0 eip 0xc011d265 cs 0x8 eflag 0x10246 cr2 0xdc41f007 ilevel 0 esp 0xc0f777c8
 > curlwp 0xc4c902a0 pid 34 lid 1 lowest kstack 0xdcc032c8
 > kernel: supervisor trap mode page fault, code=0
 > Stopped in pid 34.1 (dtrace) at netbsd:trap+0x9e6: movzbl 0(%eax),%ecx


 Hmmm, runs fine in qemu with a 8.99.3 kernel from Oct 10 sources.

 >> How-To-Repeat:
 > modload dtrace_sdt.kmod
 > run dtrace -l

 Presumably, the actual commands you execute to reproduce this bug are:

  	modload dtrace_sdt
  	dtrace -l

 If I try to run "modload dtrace_sdt.kmod" I get a "no such file or 
 directory" error (since the ".kmod" suffix and module-path prefix are 
 automatically added by the modload code).




 +------------------+--------------------------+----------------------------+
 | Paul Goyette     | PGP Key fingerprint:     | E-mail addresses:          |
 | (Retired)        | FA29 0E3B 35AF E8AE 6651 | paul at whooppee dot com   |
 | Kernel Developer | 0786 F758 55DE 53BA 7731 | pgoyette at netbsd dot org |
 +------------------+--------------------------+----------------------------+

From: Paul Goyette <paul@whooppee.com>
To: gnats-bugs@NetBSD.org
Cc: kern-bug-people@netbsd.org, gnats-admin@netbsd.org, netbsd-bugs@netbsd.org
Subject: Re: kern/52602: Panic when trying to list probes after dtrace_sdt.kmod
 has been loaded
Date: Mon, 9 Oct 2017 06:32:24 +0800 (+08)

 > Hmmm, runs fine in qemu with a 8.99.3 kernel from Oct 10 sources.

 Just noting that the original PR/failure was filed for i386 kernel, 
 while my success was on amd64.  This is probably significant!


 +------------------+--------------------------+----------------------------+
 | Paul Goyette     | PGP Key fingerprint:     | E-mail addresses:          |
 | (Retired)        | FA29 0E3B 35AF E8AE 6651 | paul at whooppee dot com   |
 | Kernel Developer | 0786 F758 55DE 53BA 7731 | pgoyette at netbsd dot org |
 +------------------+--------------------------+----------------------------+

From: Sevan Janiyan <venture37@geeklan.co.uk>
To: gnats-bugs@NetBSD.org
Cc: 
Subject: Re: kern/52602: Panic when trying to list probes after
 dtrace_sdt.kmod has been loaded
Date: Sun, 8 Oct 2017 23:47:56 +0100

 On 08/10/2017 23:15, Paul Goyette wrote:
 >  Presumably, the actual commands you execute to reproduce this bug are:
 >  
 >   	modload dtrace_sdt
 >   	dtrace -l



 modload /stand/i386/8.99.3/modules/dtrace_sdt/dtrace_sdt.kmod
 dtrace -l

From: Sevan Janiyan <venture37@geeklan.co.uk>
To: gnats-bugs@NetBSD.org
Cc: 
Subject: Re: kern/52602: Panic when trying to list probes after
 dtrace_sdt.kmod has been loaded
Date: Sun, 8 Oct 2017 23:50:53 +0100

 On 08/10/2017 23:35, Paul Goyette wrote:
 >  Just noting that the original PR/failure was filed for i386 kernel, 
 >  while my success was on amd64.  This is probably significant!

 Indeed, I'm unable to trigger the same issue on amd64, albeit with a
 slightly older build of HEAD, built on the 29th of September.
 On i386, I'm using the very latest build available this morning.


 Sevan

From: Paul Goyette <paul@whooppee.com>
To: gnats-bugs@NetBSD.org
Cc: kern-bug-people@netbsd.org, gnats-admin@netbsd.org, netbsd-bugs@netbsd.org
Subject: Re: kern/52602: Panic when trying to list probes after dtrace_sdt.kmod
 has been loaded
Date: Mon, 9 Oct 2017 08:00:04 +0800 (+08)

 On Mon, 9 Oct 2017, Paul Goyette wrote:

 >> Hmmm, runs fine in qemu with a 8.99.3 kernel from Oct 10 sources.
 >
 > Just noting that the original PR/failure was filed for i386 kernel, while
 > my success was on amd64.  This is probably significant!

 I just built a clean i386 release and installed it in a qemu VM.  I was 
 able to successfully load the dtrace_sdt module and execute the 'dtrace 
 -l' command:

 # uname -a
 NetBSD  8.99.3 NetBSD 8.99.3 (GENERIC) #0: Sun Oct  8 23:16:27 UTC 2017  paul@speedy.whooppee.com:/build/netbsd-local/obj/i386/sys/arch/i386/compile/GENERIC i386
 # modload dtrace_sdt
 dtrace: XXX dtrace_dof_property not implemented (name=dof-data-0)
 # dtrace -l
     ID   PROVIDER            MODULE                       FUNCTION NAME
      1     dtrace                                                  BEGIN
      2     dtrace                                                  END
      3     dtrace                                                  ERROR
 ....



 +------------------+--------------------------+----------------------------+
 | Paul Goyette     | PGP Key fingerprint:     | E-mail addresses:          |
 | (Retired)        | FA29 0E3B 35AF E8AE 6651 | paul at whooppee dot com   |
 | Kernel Developer | 0786 F758 55DE 53BA 7731 | pgoyette at netbsd dot org |
 +------------------+--------------------------+----------------------------+

From: Paul Goyette <paul@whooppee.com>
To: gnats-bugs@NetBSD.org
Cc: kern-bug-people@netbsd.org, gnats-admin@netbsd.org, netbsd-bugs@netbsd.org
Subject: Re: kern/52602: Panic when trying to list probes after dtrace_sdt.kmod
 has been loaded
Date: Mon, 9 Oct 2017 09:03:11 +0800 (+08)

 From a photo taken by Sevan of the original crash backtrace, manually 
 gtranscribed by me...

 ?(eb070970,de272cf8,5,c0945af0,c368d408,c4dbdc60,0,de272d08,c09a309a,0) at dc2a400b
 dtrace_open(cf00,0,3,2008,c4dbda80,0,c427ce70,c8c1436c,cf,c08bac23) at dtrace:dtrace_open+0x42
 spec_open(de272db0,6,de272dc4,c0f096b4,c0c1436c,3,caef2900,de272ca0,de272etc,c09a90f2) at netbsd:spec_open+0x28c
 VOP_OPEN(c8c1436c,3,caef2900,3,c8c1436c,caef2900,0,60,de272e14,c08f12de) at netbsd:VOP_OPEN+0x32
 vn_open(de272ea0,3,86c,de272cb4,c46d83d4,c6531680,3,0,c8d1ffe8,c3fa2400) at netbsd:vn_open+0x2e1
 do_open(c4dbda80,0,c8d1ffe8,2,bfb5ea7c,de272f34,0,c8d1ffe0,c4dbda80,c125ea84) at netbsd:do_open+0xab
 do_sys_openat(2,bfb5ea7c,de272f34,de272fa8,de272f9c,c015560d,c4dbda80,de272f68,de272f60,c46d83d4) at netbsd:do_sys_openat+0x77
 sys_open(c4dbda80,de272f68,de272f60,c46d83d4,de270000,de272f60,de272f68,5,0,0) at sys_open+0x2c
 syscall() at netbsd:syscall+0x1ed
 --- syscall (number 5) ---
 b4671e37:


 +------------------+--------------------------+----------------------------+
 | Paul Goyette     | PGP Key fingerprint:     | E-mail addresses:          |
 | (Retired)        | FA29 0E3B 35AF E8AE 6651 | paul at whooppee dot com   |
 | Kernel Developer | 0786 F758 55DE 53BA 7731 | pgoyette at netbsd dot org |
 +------------------+--------------------------+----------------------------+

State-Changed-From-To: open->closed
State-Changed-By: sevan@NetBSD.org
State-Changed-When: Mon, 09 Oct 2017 01:13:24 +0000
State-Changed-Why:
With an update to coreboot, it's no longer possible to trigger this crash.


>Unformatted:

Home	PR Database Search