NULL pointer dereference during mount

NULL pointer dereference during mount

[tao.peng]

Hi, Benny,

After rebasing to your pnfs-all-latest branch, I always got bellow NULL pointer dereference for both nfsv4 and pnfs. Do you know of any recent change that can cause this?

[   81.468606] BUG: unable to handle kernel NULL pointer dereference at 0000000000000030
[   81.468919] IP: [<ffffffff8120ca85>] selinux_set_mnt_opts+0x4f/0x3fb
[   81.469174] PGD 2ffc067 PUD 2e84067 PMD 0
[   81.469455] Oops: 0000 [#1] SMP
[   81.469455] CPU 0
[   81.469455] Modules linked in: blocklayoutdriver nfs lockd fscache auth_rpcgss nfs_acl ebtable_nat ebtables ipt_MASQUERADE ipt
able_nat nf_nat xt_CHECKSUM iptable_mangle bridge stp llc sunrpc ip6t_REJECT be2iscsi nf_conntrack_ipv6 nf_defrag_ipv6 iscsi_boot
_sysfs bnx2i ip6table_filter cnic uio ip6_tables cxgb3i libcxgbi cxgb3 mdio iscsi_tcp libiscsi_tcp libiscsi scsi_transport_iscsi
pcspkr e1000 i2c_piix4 ppdev shpchp i2c_core parport_pc parport vmw_balloon microcode ipv6 mptspi mptscsih mptbase scsi_transport
_spi [last unloaded: nfs]
[   81.469455]
[   81.469455] Pid: 1919, comm: mount.nfs Not tainted 3.1.0-rc3+ #3 VMware, Inc. VMware Virtual Platform/440BX Desktop Reference
Platform
[   81.469455] RIP: 0010:[<ffffffff8120ca85>]  [<ffffffff8120ca85>] selinux_set_mnt_opts+0x4f/0x3fb
[   81.469455] RSP: 0018:ffff88000d22fb28  EFLAGS: 00010296
[   81.469455] RAX: 0000000000000000 RBX: ffff880008460200 RCX: 0000000000000000
[   81.469455] RDX: 0000000000000000 RSI: ffff880008460390 RDI: ffff8800003dc800
[   81.469455] RBP: ffff88000d22fbb8 R08: 0000000000000001 R09: 0000000000000000
[   81.469455] R10: 0000ffff00066c0a R11: 0000000000000000 R12: ffff880000019a00
[   81.469455] R13: ffff8800003dc800 R14: ffff880008460390 R15: ffff8800003dc800
[   81.469455] FS:  00007ffebe637720(0000) GS:ffff880017400000(0000) knlGS:0000000000000000
[   81.469455] CS:  0010 DS: 0000 ES: 0000 CR0: 000000008005003b
[   81.469455] CR2: 0000000000000030 CR3: 000000000ce79000 CR4: 00000000000006f0
[   81.469455] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[   81.469455] DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400
[   81.469455] Process mount.nfs (pid: 1919, threadinfo ffff88000d22e000, task ffff88000cb08000)
[   81.469455] Stack:
[   81.469455]  ffffea0000031880 ffff880000000008 0000000000000000 ffff88000d22fb48
[   81.469455]  ffff88000d22fb78 ffffffffa02fe5ee 0000000180150011 ffff88000a308cc0
[   81.469455]  0000000000000001 0000000000000001 ffff88000d22fbc8 ffffffffa02caac5
[   81.469455] Call Trace:
[   81.469455]  [<ffffffffa02caac5>] ? nfs4_get_root+0x20f/0x242 [nfs]
[   81.469455]  [<ffffffff8120504e>] security_sb_set_mnt_opts+0x16/0x18
[   81.469455]  [<ffffffffa02ce6bd>] nfs4_remote_mount+0x17c/0x1ff [nfs]
[   81.469455]  [<ffffffff8112fde6>] mount_fs+0x69/0x157
[   81.469455]  [<ffffffff810fb000>] ? __alloc_percpu+0x10/0x12
[   81.469455]  [<ffffffff811444ca>] vfs_kern_mount+0x63/0xa0
[   81.469455]  [<ffffffffa02ce9b8>] nfs_do_root_mount+0x7b/0x9a [nfs]
[   81.469455]  [<ffffffffa02cec8f>] nfs4_try_mount+0x58/0xba [nfs]
[   81.469455]  [<ffffffffa02cfff1>] nfs_fs_mount+0x48b/0x6e9 [nfs]
[   81.469455]  [<ffffffff8112fde6>] mount_fs+0x69/0x157
[   81.469455]  [<ffffffff810fb000>] ? __alloc_percpu+0x10/0x12
[   81.469455]  [<ffffffff811444ca>] vfs_kern_mount+0x63/0xa0
[   81.469455]  [<ffffffff811451be>] do_kern_mount+0x4d/0xdf
[   81.469455]  [<ffffffff8114685d>] do_mount+0x63c/0x69f
[   81.469455]  [<ffffffff81146185>] ? copy_mount_options+0xcb/0x12e
[   81.469455]  [<ffffffff81146b6e>] sys_mount+0x88/0xc2
[   81.469455]  [<ffffffff8147e502>] system_call_fastpath+0x16/0x1b
[   81.469455] Code: 00 00 4c 8b a7 a8 01 00 00 49 89 fd 48 8b 56 08 48 89 45 a8 48 8b 47 28 48 89 55 80 48 8b 00 48 89 45 98 49 8b 04 24 48 8b 40 60
[   81.469455]  8b 40 30 48 8b 40 38 48 89 45 a0 48 8b 06 8b 76 10 48 89 45
[   81.469455] RIP  [<ffffffff8120ca85>] selinux_set_mnt_opts+0x4f/0x3fb
[   81.469455]  RSP <ffff88000d22fb28>
[   81.469455] CR2: 0000000000000030
[   81.513181] ---[ end trace dccee53f6c438566 ]---



Cheers,
-Bergwolf

RE: NULL pointer dereference during mount

[tao.peng]

> -----Original Message-----
> From: linux-nfs-owner@vger.kernel.org [mailto:linux-nfs-owner@vger.kernel.org]
> On Behalf Of tao.peng@emc.com
> Sent: Tuesday, September 13, 2011 7:55 PM
> To: bhalevy@tonian.com
> Cc: linux-nfs@vger.kernel.org
> Subject: NULL pointer dereference during mount
> 
> Hi, Benny,
> 
> After rebasing to your pnfs-all-latest branch, I always got bellow NULL pointer
> dereference for both nfsv4 and pnfs. Do you know of any recent change that can
> cause this?
Just verified that Trond's nfs-for-next branch is working. So should be something only in your tree.

> 
> [   81.468606] BUG: unable to handle kernel NULL pointer dereference at
> 0000000000000030
> [   81.468919] IP: [<ffffffff8120ca85>] selinux_set_mnt_opts+0x4f/0x3fb
> [   81.469174] PGD 2ffc067 PUD 2e84067 PMD 0
> [   81.469455] Oops: 0000 [#1] SMP
> [   81.469455] CPU 0
> [   81.469455] Modules linked in: blocklayoutdriver nfs lockd fscache auth_rpcgss
> nfs_acl ebtable_nat ebtables ipt_MASQUERADE ipt
> able_nat nf_nat xt_CHECKSUM iptable_mangle bridge stp llc sunrpc ip6t_REJECT
> be2iscsi nf_conntrack_ipv6 nf_defrag_ipv6 iscsi_boot
> _sysfs bnx2i ip6table_filter cnic uio ip6_tables cxgb3i libcxgbi cxgb3 mdio iscsi_tcp
> libiscsi_tcp libiscsi scsi_transport_iscsi
> pcspkr e1000 i2c_piix4 ppdev shpchp i2c_core parport_pc parport vmw_balloon
> microcode ipv6 mptspi mptscsih mptbase scsi_transport
> _spi [last unloaded: nfs]
> [   81.469455]
> [   81.469455] Pid: 1919, comm: mount.nfs Not tainted 3.1.0-rc3+ #3 VMware, Inc.
> VMware Virtual Platform/440BX Desktop Reference
> Platform
> [   81.469455] RIP: 0010:[<ffffffff8120ca85>]  [<ffffffff8120ca85>]
> selinux_set_mnt_opts+0x4f/0x3fb
> [   81.469455] RSP: 0018:ffff88000d22fb28  EFLAGS: 00010296
> [   81.469455] RAX: 0000000000000000 RBX: ffff880008460200 RCX:
> 0000000000000000
> [   81.469455] RDX: 0000000000000000 RSI: ffff880008460390 RDI:
> ffff8800003dc800
> [   81.469455] RBP: ffff88000d22fbb8 R08: 0000000000000001 R09:
> 0000000000000000
> [   81.469455] R10: 0000ffff00066c0a R11: 0000000000000000 R12:
> ffff880000019a00
> [   81.469455] R13: ffff8800003dc800 R14: ffff880008460390 R15:
> ffff8800003dc800
> [   81.469455] FS:  00007ffebe637720(0000) GS:ffff880017400000(0000)
> knlGS:0000000000000000
> [   81.469455] CS:  0010 DS: 0000 ES: 0000 CR0: 000000008005003b
> [   81.469455] CR2: 0000000000000030 CR3: 000000000ce79000 CR4:
> 00000000000006f0
> [   81.469455] DR0: 0000000000000000 DR1: 0000000000000000 DR2:
> 0000000000000000
> [   81.469455] DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7:
> 0000000000000400
> [   81.469455] Process mount.nfs (pid: 1919, threadinfo ffff88000d22e000, task
> ffff88000cb08000)
> [   81.469455] Stack:
> [   81.469455]  ffffea0000031880 ffff880000000008 0000000000000000
> ffff88000d22fb48
> [   81.469455]  ffff88000d22fb78 ffffffffa02fe5ee 0000000180150011
> ffff88000a308cc0
> [   81.469455]  0000000000000001 0000000000000001 ffff88000d22fbc8
> ffffffffa02caac5
> [   81.469455] Call Trace:
> [   81.469455]  [<ffffffffa02caac5>] ? nfs4_get_root+0x20f/0x242 [nfs]
> [   81.469455]  [<ffffffff8120504e>] security_sb_set_mnt_opts+0x16/0x18
> [   81.469455]  [<ffffffffa02ce6bd>] nfs4_remote_mount+0x17c/0x1ff [nfs]
> [   81.469455]  [<ffffffff8112fde6>] mount_fs+0x69/0x157
> [   81.469455]  [<ffffffff810fb000>] ? __alloc_percpu+0x10/0x12
> [   81.469455]  [<ffffffff811444ca>] vfs_kern_mount+0x63/0xa0
> [   81.469455]  [<ffffffffa02ce9b8>] nfs_do_root_mount+0x7b/0x9a [nfs]
> [   81.469455]  [<ffffffffa02cec8f>] nfs4_try_mount+0x58/0xba [nfs]
> [   81.469455]  [<ffffffffa02cfff1>] nfs_fs_mount+0x48b/0x6e9 [nfs]
> [   81.469455]  [<ffffffff8112fde6>] mount_fs+0x69/0x157
> [   81.469455]  [<ffffffff810fb000>] ? __alloc_percpu+0x10/0x12
> [   81.469455]  [<ffffffff811444ca>] vfs_kern_mount+0x63/0xa0
> [   81.469455]  [<ffffffff811451be>] do_kern_mount+0x4d/0xdf
> [   81.469455]  [<ffffffff8114685d>] do_mount+0x63c/0x69f
> [   81.469455]  [<ffffffff81146185>] ? copy_mount_options+0xcb/0x12e
> [   81.469455]  [<ffffffff81146b6e>] sys_mount+0x88/0xc2
> [   81.469455]  [<ffffffff8147e502>] system_call_fastpath+0x16/0x1b
> [   81.469455] Code: 00 00 4c 8b a7 a8 01 00 00 49 89 fd 48 8b 56 08 48 89 45 a8
> 48 8b 47 28 48 89 55 80 48 8b 00 48 89 45 98 49 8b 04 24 48 8b 40 60
> [   81.469455]  8b 40 30 48 8b 40 38 48 89 45 a0 48 8b 06 8b 76 10 48 89 45
> [   81.469455] RIP  [<ffffffff8120ca85>] selinux_set_mnt_opts+0x4f/0x3fb
> [   81.469455]  RSP <ffff88000d22fb28>
> [   81.469455] CR2: 0000000000000030
> [   81.513181] ---[ end trace dccee53f6c438566 ]---
> 
> 
> 
> Cheers,
> -Bergwolf
> 
> 
> --
> To unsubscribe from this list: send the line "unsubscribe linux-nfs" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: NULL pointer dereference during mount

[Benny Halevy]
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Re: NULL pointer dereference during mount

[Jim Rees]

tao.peng@emc.com wrote:

  After rebasing to your pnfs-all-latest branch, I always got bellow NULL
  pointer dereference for both nfsv4 and pnfs. Do you know of any recent
  change that can cause this?

"git diff" shows no change in the nfs client between my for-benny branch and
benny's pnfs-all-latest branch.  And the mount works for me.  But the first
thing I always do on a fresh Fedora install is disable selinux.  Maybe
something changed in selinux?

RE: NULL pointer dereference during mount

[tao.peng]

________________________________________
From: Jim Rees [rees@umich.edu]
Sent: Tuesday, September 13, 2011 8:42 PM
To: Peng, Tao
Cc: bhalevy@tonian.com; linux-nfs@vger.kernel.org
Subject: Re: NULL pointer dereference during mount

tao.peng@emc.com wrote:

  After rebasing to your pnfs-all-latest branch, I always got bellow NULL
  pointer dereference for both nfsv4 and pnfs. Do you know of any recent
  change that can cause this?

"git diff" shows no change in the nfs client between my for-benny branch and
benny's pnfs-all-latest branch.  And the mount works for me.  But the first
thing I always do on a fresh Fedora install is disable selinux.  Maybe
something changed in selinux?
[PT] Just tested 3.1-rc6 and it is working properly. So I think it should be caused by nfs changes.

Re: NULL pointer dereference during mount

[Benny Halevy]

I couldn't reproduce this with selinux neither disabled nor enforcing.
What's your selinux configuration?

Benny

On 2011-09-13 18:43, tao.peng@emc.com wrote:
> 
> ________________________________________
> From: Jim Rees [rees@umich.edu]
> Sent: Tuesday, September 13, 2011 8:42 PM
> To: Peng, Tao
> Cc: bhalevy@tonian.com; linux-nfs@vger.kernel.org
> Subject: Re: NULL pointer dereference during mount
> 
> tao.peng@emc.com wrote:
> 
>   After rebasing to your pnfs-all-latest branch, I always got bellow NULL
>   pointer dereference for both nfsv4 and pnfs. Do you know of any recent
>   change that can cause this?
> 
> "git diff" shows no change in the nfs client between my for-benny branch and
> benny's pnfs-all-latest branch.  And the mount works for me.  But the first
> thing I always do on a fresh Fedora install is disable selinux.  Maybe
> something changed in selinux?
> [PT] Just tested 3.1-rc6 and it is working properly. So I think it should be caused by nfs changes.--
> To unsubscribe from this list: send the line "unsubscribe linux-nfs" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html

RE: NULL pointer dereference during mount

[tao.peng]

> -----Original Message-----
> From: linux-nfs-owner@vger.kernel.org [mailto:linux-nfs-owner@vger.kernel.org]
> On Behalf Of Benny Halevy
> Sent: Wednesday, September 14, 2011 2:29 PM
> To: Peng, Tao
> Cc: rees@umich.edu; linux-nfs@vger.kernel.org
> Subject: Re: NULL pointer dereference during mount
> 
> I couldn't reproduce this with selinux neither disabled nor enforcing.
> What's your selinux configuration?
I didn't changed selinux config so it is the default on Fedora15
/etc/selinux/config:
SELINUX=enforcing
SELINUXTYPE=targeted

/etc/selinux/semanage.conf:
module-store = direct
expand-check=0
usepasswd=False

Cheers,
Tao
> 
> Benny
> 
> On 2011-09-13 18:43, tao.peng@emc.com wrote:
> >
> > ________________________________________
> > From: Jim Rees [rees@umich.edu]
> > Sent: Tuesday, September 13, 2011 8:42 PM
> > To: Peng, Tao
> > Cc: bhalevy@tonian.com; linux-nfs@vger.kernel.org
> > Subject: Re: NULL pointer dereference during mount
> >
> > tao.peng@emc.com wrote:
> >
> >   After rebasing to your pnfs-all-latest branch, I always got bellow NULL
> >   pointer dereference for both nfsv4 and pnfs. Do you know of any recent
> >   change that can cause this?
> >
> > "git diff" shows no change in the nfs client between my for-benny branch and
> > benny's pnfs-all-latest branch.  And the mount works for me.  But the first
> > thing I always do on a fresh Fedora install is disable selinux.  Maybe
> > something changed in selinux?
> > [PT] Just tested 3.1-rc6 and it is working properly. So I think it should be caused by
> nfs changes.--
> > To unsubscribe from this list: send the line "unsubscribe linux-nfs" in
> > the body of a message to majordomo@vger.kernel.org
> > More majordomo info at  http://vger.kernel.org/majordomo-info.html
> --
> To unsubscribe from this list: send the line "unsubscribe linux-nfs" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html

RE: NULL pointer dereference during mount

[tao.peng]

Hi, Benny,

OK, I found the problem. It is because I was building nfs module, instead of rebuilding the whole kernel. commit bea6cd74 in pnfs-all-latest branch changes struct super_block by adding s_pnfs_op pointer. Then sb->s_security of the new module is translated into accessing sb->s_xattr, which is NULL, and causes the oops.

After rebuilding the kernel, I don't get it any more. Sorry for the noise...

Cheers,
Tao
________________________________________
From: linux-nfs-owner@vger.kernel.org [linux-nfs-owner@vger.kernel.org] On Behalf Of tao.peng@emc.com [tao.peng@emc.com]
Sent: Wednesday, September 14, 2011 3:53 PM
To: bhalevy@tonian.com
Cc: rees@umich.edu; linux-nfs@vger.kernel.org
Subject: RE: NULL pointer dereference during mount

> -----Original Message-----
> From: linux-nfs-owner@vger.kernel.org [mailto:linux-nfs-owner@vger.kernel.org]
> On Behalf Of Benny Halevy
> Sent: Wednesday, September 14, 2011 2:29 PM
> To: Peng, Tao
> Cc: rees@umich.edu; linux-nfs@vger.kernel.org
> Subject: Re: NULL pointer dereference during mount
>
> I couldn't reproduce this with selinux neither disabled nor enforcing.
> What's your selinux configuration?
I didn't changed selinux config so it is the default on Fedora15
/etc/selinux/config:
SELINUX=enforcing
SELINUXTYPE=targeted

/etc/selinux/semanage.conf:
module-store = direct
expand-check=0
usepasswd=False

Cheers,
Tao
>
> Benny
>
> On 2011-09-13 18:43, tao.peng@emc.com wrote:
> >
> > ________________________________________
> > From: Jim Rees [rees@umich.edu]
> > Sent: Tuesday, September 13, 2011 8:42 PM
> > To: Peng, Tao
> > Cc: bhalevy@tonian.com; linux-nfs@vger.kernel.org
> > Subject: Re: NULL pointer dereference during mount
> >
> > tao.peng@emc.com wrote:
> >
> >   After rebasing to your pnfs-all-latest branch, I always got bellow NULL
> >   pointer dereference for both nfsv4 and pnfs. Do you know of any recent
> >   change that can cause this?
> >
> > "git diff" shows no change in the nfs client between my for-benny branch and
> > benny's pnfs-all-latest branch.  And the mount works for me.  But the first
> > thing I always do on a fresh Fedora install is disable selinux.  Maybe
> > something changed in selinux?
> > [PT] Just tested 3.1-rc6 and it is working properly. So I think it should be caused by
> nfs changes.--
> > To unsubscribe from this list: send the line "unsubscribe linux-nfs" in
> > the body of a message to majordomo@vger.kernel.org
> > More majordomo info at  http://vger.kernel.org/majordomo-info.html
> --
> To unsubscribe from this list: send the line "unsubscribe linux-nfs" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html

--
To unsubscribe from this list: send the line "unsubscribe linux-nfs" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: NULL pointer dereference during mount

[Benny Halevy]
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Re: NULL pointer dereference during mount

[Benny Halevy]

I couldn't reproduce this with selinux neither disabled nor enforcing.
What's your selinux configuration?

Benny

On 2011-09-13 18:43, tao.peng@emc.com wrote:
> 
> ________________________________________
> From: Jim Rees [rees@umich.edu]
> Sent: Tuesday, September 13, 2011 8:42 PM
> To: Peng, Tao
> Cc: bhalevy@tonian.com; linux-nfs@vger.kernel.org
> Subject: Re: NULL pointer dereference during mount
> 
> tao.peng@emc.com wrote:
> 
>   After rebasing to your pnfs-all-latest branch, I always got bellow NULL
>   pointer dereference for both nfsv4 and pnfs. Do you know of any recent
>   change that can cause this?
> 
> "git diff" shows no change in the nfs client between my for-benny branch and
> benny's pnfs-all-latest branch.  And the mount works for me.  But the first
> thing I always do on a fresh Fedora install is disable selinux.  Maybe
> something changed in selinux?
> [PT] Just tested 3.1-rc6 and it is working properly. So I think it should be caused by nfs changes.--
> To unsubscribe from this list: send the line "unsubscribe linux-nfs" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html