e1000e 0000:00:19.0: DMA-API: device driver tries to free DMA memory it has not allocated [device address=0x0000000052ff084a] [size=90 bytes]

e1000e 0000:00:19.0: DMA-API: device driver tries to free DMA memory it has not allocated [device address=0x0000000052ff084a] [size=90 bytes]

[Ingo Molnar]

( e1000e Cc:s added. A new debug feature, CONFIG_DMA_API_DEBUG=y, 
  has triggered the warning below in e1000_put_txbuf(). )

* Joerg Roedel <joerg.roedel@amd.com> wrote:

> On Wed, Mar 18, 2009 at 10:38:47AM +0100, Ingo Molnar wrote:
> > e1000e 0000:00:19.0: irq 30 for MSI/MSI-X
> > e1000e 0000:00:19.0: irq 30 for MSI/MSI-X
> > ADDRCONF(NETDEV_UP): eth0: link is not ready
> > e1000e: eth0 NIC Link is Up 100 Mbps Full Duplex, Flow Control: RX/TX
> > 0000:00:19.0: eth0: 10/100 speed: disabling TSO
> > ADDRCONF(NETDEV_CHANGE): eth0: link becomes ready
> > ------------[ cut here ]------------
> > WARNING: at lib/dma-debug.c:461 check_unmap+0xd4/0x3dd() (Not tainted)
> > Hardware name: 2241B48
> > e1000e 0000:00:19.0: DMA-API: device driver tries to free DMA memory 
> > it has not allocated [device address=0x0000000052ff084a] [size=90 bytes]
> >  [<ffffffff811a6fb0>] check_unmap+0xd4/0x3dd
> >  [<ffffffff811a7406>] debug_dma_unmap_page+0x50/0x52
> >  [<ffffffffa00f4523>] pci_unmap_page+0x4e/0x57 [e1000e]
> >  [<ffffffffa00f455a>] e1000_put_txbuf+0x2e/0x4f [e1000e]
> >  [<ffffffffa00f4681>] e1000_clean_tx_irq+0xc8/0x2c2 [e1000e]
> >  [<ffffffffa00f809b>] ? e1000_clean+0x6b/0x246 [e1000e]
> >  [<ffffffffa00f80a7>] e1000_clean+0x77/0x246 [e1000e]
> >  [<ffffffff812f89dd>] net_rx_action+0xb6/0x1ee
> >  [<ffffffff812f8acc>] ? net_rx_action+0x1a5/0x1ee
> >  [<ffffffff81051353>] __do_softirq+0x94/0x179
> >  [<ffffffff810127ac>] call_softirq+0x1c/0x30
> >  [<ffffffff8101393e>] do_softirq+0x52/0xb9
> >  [<ffffffff81050f76>] irq_exit+0x53/0x90
> >  [<ffffffff81013c57>] do_IRQ+0x12c/0x151
> >  [<ffffffff81011e93>] ret_from_intr+0x0/0x2e
> >  <EOI>  [<ffffffff811fe088>] ? acpi_idle_enter_bm+0x287/0x2de
> >  [<ffffffff8106fbf9>] ? trace_hardirqs_on+0xd/0xf
> >  [<ffffffff811fe090>] ? acpi_idle_enter_bm+0x28f/0x2de
> >  [<ffffffff811fe088>] ? acpi_idle_enter_bm+0x287/0x2de
> >  [<ffffffff81399916>] ? __atomic_notifier_call_chain+0x0/0x86
> >  [<ffffffff812d62ad>] ? cpuidle_idle_call+0x8d/0xc4
> >  [<ffffffff810102c7>] ? cpu_idle+0x68/0xb3
> >  [<ffffffff81381817>] ? rest_init+0x6b/0x6d
> > ---[ end trace b8ae2341b2e9bbc2 ]---
> > 
> > i have some vague memories of this being discussed somewhere - it 
> > got fixed in the driver, right? If yes, do you have an URL to that 
> > fix? (if it's not upstream yet i will pick it up into 
> > tip:out-of-tree to not have already-fixed warnings)
> 
> We had this issue in the ixgbe driver and it was fixed there. As far as
> I know for e1000e this is a new issue.
> 
> Joerg
> 
> -- 
>            | Advanced Micro Devices GmbH
>  Operating | Karl-Hammerschmidt-Str. 34, 85609 Dornach bei München
>  System    | 
>  Research  | Geschäftsführer: Jochen Polster, Thomas M. McCoy, Giuliano Meroni
>  Center    | Sitz: Dornach, Gemeinde Aschheim, Landkreis München
>            | Registergericht München, HRB Nr. 43632
> 
> --
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at  http://www.tux.org/lkml/

Re: e1000e 0000:00:19.0: DMA-API: device driver tries to free DMA memory it has not allocated [device address=0x0000000052ff084a] [size=90 bytes]

[Lubomir Rintel]

On Wed, March 18, 2009 8:04 am, Ingo Molnar wrote:
>
> ( e1000e Cc:s added. A new debug feature, CONFIG_DMA_API_DEBUG=y,
>   has triggered the warning below in e1000_put_txbuf(). )

http://lkml.org/lkml/2009/3/2/318

-- 
Lubomir Rintel


------------------------------------------------------------------------------
Apps built with the Adobe(R) Flex(R) framework and Flex Builder(TM) are
powering Web 2.0 with engaging, cross-platform capabilities. Quickly and
easily build your RIAs with Flex Builder, the Eclipse(TM)based development
software that enables intelligent coding and step-through debugging.
Download the free 60 day trial. http://p.sf.net/sfu/www-adobe-com

Re: e1000e 0000:00:19.0: DMA-API: device driver tries to free DMA memory it has not allocated [device address=0x0000000052ff084a] [size=90 bytes]

[Ingo Molnar]

* Lubomir Rintel <lkundrak@v3.sk> wrote:

> On Wed, March 18, 2009 8:04 am, Ingo Molnar wrote:
> >
> > ( e1000e Cc:s added. A new debug feature, CONFIG_DMA_API_DEBUG=y,
> >   has triggered the warning below in e1000_put_txbuf(). )
> 
> http://lkml.org/lkml/2009/3/2/318

Thanks.

	Ingo

forcedeth 0000:00:0a.0: DMA-API: device driver frees DMA memory with wrong function [device address=0x0000000035992232] [size=42 bytes] [mapped as single] [unmapped as page]

[Ingo Molnar]

There's a similar-looking warning triggering for the forcedeth 
driver as well on another test-system (see the details below).

Does anyone know whether there's a fix for this too, or is it a
new warning?

	Ingo

----- Forwarded message from Ingo Molnar <mingo@elte.hu> -----

Date: Wed, 18 Mar 2009 12:20:26 +0100
From: Ingo Molnar <mingo@elte.hu>
To: Joerg Roedel <joerg.roedel@amd.com>
Subject: forcedeth 0000:00:0a.0: DMA-API: device driver frees DMA memory
	with wrong function [device address=0x0000000035992232] [size=42
	bytes] [mapped as single] [unmapped as page]
Cc: Ingo Molnar <mingo@redhat.com>, iommu@lists.linux-foundation.org,
	linux-kernel@vger.kernel.org


-tip testing found this DMA-debug assert:

[   19.648023] eth0: no link during initialization.
[   21.952553] eth0: link up.
[   22.684073] ------------[ cut here ]------------
[   22.688023] WARNING: at lib/dma-debug.c:562 check_unmap+0x30e/0x4b4()
[   22.688023] Hardware name: System Product Name
[   22.688023] 
forcedeth 0000:00:0a.0: DMA-API: device driver frees DMA memory with wrong function [device address=0x0000000035992232] [size=42 bytes] [mapped as single] [unmapped as page]
[   22.688023] Modules linked in:
[   22.688023] Pid: 1523, comm: arping Not tainted 2.6.29-rc8-tip #20999
[   22.688023] Call Trace:
[   22.688023]  [<c0140eae>] warn_slowpath+0x76/0xad
[   22.688023]  [<c0293a0b>] ? add_dma_entry+0x4b/0x51
[   22.688023]  [<c015eec5>] ? mark_lock+0x1c/0x16f
[   22.688023]  [<c015f1a4>] ? __lock_acquire+0x18c/0x296
[   22.688023]  [<c03473cc>] ? nv_start_xmit_optimized+0x3bd/0x3eb
[   22.688023]  [<c015eec5>] ? mark_lock+0x1c/0x16f
[   22.688023]  [<c015f1a4>] ? __lock_acquire+0x18c/0x296
[   22.688023]  [<c0293572>] check_unmap+0x30e/0x4b4
[   22.688023]  [<c055f7c3>] ? _spin_lock+0x27/0x2f
[   22.688023]  [<c0145ba1>] ? _local_bh_enable_ip+0xa1/0xa7
[   22.688023]  [<c049e0f8>] ? dev_queue_xmit+0x325/0x359
[   22.688023]  [<c0188922>] ? trace_hardirqs_on+0x21/0x23
[   22.688023]  [<c0145ba1>] ? _local_bh_enable_ip+0xa1/0xa7
[   22.688023]  [<c0145bc6>] ? local_bh_enable+0x10/0x12
[   22.688023]  [<c050225c>] ? packet_sendmsg+0x1b7/0x20a
[   22.688023]  [<c02939b8>] debug_dma_unmap_page+0x59/0x61
[   22.688023]  [<c034421f>] pci_unmap_page+0x5b/0x66
[   22.688023]  [<c03446fd>] nv_tx_done_optimized+0x46/0x1c5
[   22.688023]  [<c0345922>] nv_nic_irq_optimized+0xa7/0x233
[   22.688023]  [<c017706e>] handle_IRQ_event+0x85/0xdf
[   22.688023]  [<c01785df>] handle_fasteoi_irq+0x79/0xb9
[   22.688023]  [<c0119973>] handle_irq+0x1f/0x24
[   22.688023]  [<c0119332>] do_IRQ+0x45/0x88
[   22.688023]  [<c0118455>] common_interrupt+0x35/0x40
[   22.688023]  [<c0144515>] ? do_setitimer+0x160/0x2d9
[   22.688023]  [<c055f69b>] ? _spin_unlock_irq+0x29/0x2b
[   22.688023]  [<c0144515>] do_setitimer+0x160/0x2d9
[   22.688023]  [<c04926fa>] ? sys_socketcall+0xed/0x188
[   22.688023]  [<c014473b>] alarm_setitimer+0x39/0x58
[   22.688023]  [<c0149776>] sys_alarm+0x10/0x12
[   22.688023]  [<c0117d47>] sysenter_do_call+0x12/0x32
[   22.688023] ---[ end trace bf7fa25698012db6 ]---
[   22.688023] Mapped at:
[   22.688023]  [<c0293cc1>] debug_dma_map_page+0x6c/0x128
[   22.688023]  [<c0345bd0>] pci_map_single+0xb5/0xc1
[   22.688023]  [<c0347147>] nv_start_xmit_optimized+0x138/0x3eb
[   22.688023]  [<c049dc7c>] dev_hard_start_xmit+0x123/0x18f
[   22.688023]  [<c04acbb9>] __qdisc_run+0xd0/0x1b3
[   23.700081] arping used greatest stack depth: 5988 bytes left
[   24.545512] Adding 3911816k swap on /dev/sda2.  Priority:-1 extents:1 across:3911816k 

Config and full bootlog attached.

	Ingo

RE: e1000e 0000:00:19.0: DMA-API: device driver tries to free DMA memory it has not allocated [device address=0x0000000052ff084a] [size=90 bytes]

[Brandeburg, Jesse]

Ingo Molnar wrote:
> * Lubomir Rintel <lkundrak@v3.sk> wrote:
> 
>> On Wed, March 18, 2009 8:04 am, Ingo Molnar wrote:
>>> 
>>> ( e1000e Cc:s added. A new debug feature, CONFIG_DMA_API_DEBUG=y,
>>>   has triggered the warning below in e1000_put_txbuf(). )
>> 
>> http://lkml.org/lkml/2009/3/2/318

We are still looking for more testing from the general community of 
this patch, in particular if any tx hangs are reported.  Thanks in 
advance for any reports.

Re: e1000e 0000:00:19.0: DMA-API: device driver tries to free DMA memory it has not allocated [device address=0x0000000052ff084a] [size=90 bytes]

[Ingo Molnar]

* Brandeburg, Jesse <jesse.brandeburg@intel.com> wrote:

> Ingo Molnar wrote:
> > * Lubomir Rintel <lkundrak@v3.sk> wrote:
> > 
> >> On Wed, March 18, 2009 8:04 am, Ingo Molnar wrote:
> >>> 
> >>> ( e1000e Cc:s added. A new debug feature, CONFIG_DMA_API_DEBUG=y,
> >>>   has triggered the warning below in e1000_put_txbuf(). )
> >> 
> >> http://lkml.org/lkml/2009/3/2/318
> 
> We are still looking for more testing from the general community of 
> this patch, in particular if any tx hangs are reported.  Thanks in 
> advance for any reports.--

Yeah - already picked it up into tip:out-of-tree and it passed about 
a hundred bootup tests on various e1000e using systems already - 
with no problems whatsoever. Will let you know if anything breaks - 
e1000e has been rock solid for me in the .29 cycle.

	Ingo

------------------------------------------------------------------------------
Apps built with the Adobe(R) Flex(R) framework and Flex Builder(TM) are
powering Web 2.0 with engaging, cross-platform capabilities. Quickly and
easily build your RIAs with Flex Builder, the Eclipse(TM)based development
software that enables intelligent coding and step-through debugging.
Download the free 60 day trial. http://p.sf.net/sfu/www-adobe-com

Re: e1000e 0000:00:19.0: DMA-API: device driver tries to free DMA memory it has not allocated [device address=0x0000000052ff084a] [size=90 bytes]

[Brandeburg, Jesse]

On Wed, 18 Mar 2009, Ingo Molnar wrote:
> > >> http://lkml.org/lkml/2009/3/2/318
> > 
> > We are still looking for more testing from the general community of 
> > this patch, in particular if any tx hangs are reported.  Thanks in 
> > advance for any reports.--
> 
> Yeah - already picked it up into tip:out-of-tree and it passed about 
> a hundred bootup tests on various e1000e using systems already - 
> with no problems whatsoever. Will let you know if anything breaks - 
> e1000e has been rock solid for me in the .29 cycle.

thanks! did you pull in both e1000 and e1000e fixes?  We also have a 
similar fix in internal testing for igb.

Re: e1000e 0000:00:19.0: DMA-API: device driver tries to free DMA memory it has not allocated [device address=0x0000000052ff084a] [size=90 bytes]

[Ingo Molnar]

* Brandeburg, Jesse <jesse.brandeburg@intel.com> wrote:

> On Wed, 18 Mar 2009, Ingo Molnar wrote:
> > > >> http://lkml.org/lkml/2009/3/2/318
> > > 
> > > We are still looking for more testing from the general community of 
> > > this patch, in particular if any tx hangs are reported.  Thanks in 
> > > advance for any reports.--
> > 
> > Yeah - already picked it up into tip:out-of-tree and it passed about 
> > a hundred bootup tests on various e1000e using systems already - 
> > with no problems whatsoever. Will let you know if anything breaks - 
> > e1000e has been rock solid for me in the .29 cycle.
> 
> thanks! did you pull in both e1000 and e1000e fixes?  We also have a 
> similar fix in internal testing for igb.

yeah, i picked up both - albeit i've switched over all boxes to 
e1000e so e1000 will get little testing from me. Do you have an sha1 
for the igb fix by any chance?

	Ingo

------------------------------------------------------------------------------
Apps built with the Adobe(R) Flex(R) framework and Flex Builder(TM) are
powering Web 2.0 with engaging, cross-platform capabilities. Quickly and
easily build your RIAs with Flex Builder, the Eclipse(TM)based development
software that enables intelligent coding and step-through debugging.
Download the free 60 day trial. http://p.sf.net/sfu/www-adobe-com

3c59x 0000:00:0b.0: DMA-API: device driver maps memory from kernel text or rodata [addr=c0fffe54] [size=428]

[Ingo Molnar]

FYI, triggered a new type of DMA debug warning today, this time on 
an older box, in the boomerang/vortex 3c59x driver:

[  211.098328] eth0:  setting full-duplex.
[ 1745.974321] ------------[ cut here ]------------
[ 1745.980344] WARNING: at lib/dma-debug.c:627 check_for_illegal_area+0xa6/0xe4()
[ 1745.983613] 3c59x 0000:00:0b.0: DMA-API: device driver maps memory from kernel text or rodata [addr=c0fffe54] [size=428]
[ 1745.986521] Modules linked in:
[ 1745.986521] Pid: 19063, comm: distcc Not tainted 2.6.29-rc8-tip-02734-gfd4c260-dirty #729
[ 1745.993830] Call Trace:
[ 1745.996309]  [<c102e98b>] warn_slowpath+0x5e/0x74
[ 1746.001069]  [<c124de1e>] ? memcpy+0xe/0x31
[ 1746.005322]  [<c14eb801>] ? skb_put+0x62/0x7b
[ 1746.009732]  [<c15e0cf5>] ? _spin_unlock_irqrestore+0x19/0x25
[ 1746.011540]  [<c1006274>] ? enable_8259A_irq+0x45/0x48
[ 1746.016729]  [<c105ee38>] ? handle_level_irq+0xe3/0xef
[ 1746.021921]  [<c1004efb>] ? handle_irq+0x36/0x43
[ 1746.022582]  [<c1002c35>] ? restore_nocheck_notrace+0x0/0xe
[ 1746.028198]  [<c124e0e0>] ? trace_hardirqs_on_thunk+0xc/0x10
[ 1746.033900]  [<c1002c35>] ? restore_nocheck_notrace+0x0/0xe
[ 1746.035517]  [<c125724f>] check_for_illegal_area+0xa6/0xe4
[ 1746.041046]  [<c1258382>] debug_dma_map_page+0x12a/0x13b
[ 1746.043356]  [<c1316366>] boomerang_start_xmit+0x40b/0x54c
[ 1746.048900]  [<c109c226>] ? kmem_cache_free+0xf6/0xfe
[ 1746.054013]  [<c14f34ca>] dev_hard_start_xmit+0x101/0x171
[ 1746.055471]  [<c15081bc>] __qdisc_run+0xf4/0x21f
[ 1746.060143]  [<c14f5e02>] dev_queue_xmit+0x1e2/0x2dc
[ 1746.065162]  [<c1524fc8>] ip_finish_output+0x213/0x24d
[ 1746.066360]  [<c109a489>] ? check_object+0xfa/0x156
[ 1746.071291]  [<c15252f8>] ip_output+0x4e/0x51
[ 1746.075693]  [<c1523a51>] ip_local_out+0x39/0x42
[ 1746.080366]  [<c1525727>] ip_queue_xmit+0x2ce/0x314
[ 1746.085295]  [<c109d146>] ? __kmalloc_node_track_caller+0x165/0x172
[ 1746.087612]  [<c14ec135>] ? __alloc_skb+0x38/0x115
[ 1746.092458]  [<c15391de>] ? tcp_v4_md5_do_lookup+0xa/0x36
[ 1746.097916]  [<c15350ee>] ? __tcp_select_window+0xe/0x12f
[ 1746.099368]  [<c1535b2d>] ? tcp_options_write+0x138/0x1b7
[ 1746.104820]  [<c1536257>] tcp_transmit_skb+0x452/0x492
[ 1746.110009]  [<c1537824>] tcp_write_xmit+0x21d/0x2c4
[ 1746.111029]  [<c153791d>] __tcp_push_pending_frames+0x52/0x12e
[ 1746.116912]  [<c152b3a4>] do_tcp_sendpages+0x4a9/0x4d6
[ 1746.122112]  [<c152c036>] tcp_sendpage+0x54/0x69
[ 1746.122783]  [<c14e51f7>] sock_sendpage+0x34/0x3b
[ 1746.127547]  [<c10bc5bf>] pipe_to_sendpage+0x5b/0x66
[ 1746.132576]  [<c10bd2a2>] __splice_from_pipe+0x65/0x1db
[ 1746.137853]  [<c10bc564>] ? pipe_to_sendpage+0x0/0x66
[ 1746.138956]  [<c10bd864>] splice_from_pipe+0x56/0x6d
[ 1746.143975]  [<c10bd893>] generic_splice_sendpage+0x18/0x1a
[ 1746.149599]  [<c10bc564>] ? pipe_to_sendpage+0x0/0x66
[ 1746.150703]  [<c10bca34>] do_splice_from+0xc0/0xd1
[ 1746.155547]  [<c10bcbb1>] direct_splice_actor+0x1c/0x21
[ 1746.160835]  [<c10bcfd5>] splice_direct_to_actor+0x124/0x1f1
[ 1746.162841]  [<c10bcb95>] ? direct_splice_actor+0x0/0x21
[ 1746.168203]  [<c10bd0dc>] do_splice_direct+0x3a/0x55
[ 1746.173220]  [<c10a3110>] do_sendfile+0x1b0/0x23e
[ 1746.177980]  [<c10a38af>] sys_sendfile+0x5b/0xac
[ 1746.178649]  [<c1002b74>] sysenter_do_call+0x12/0x2b
[ 1746.183668] ---[ end trace e9b0fe468818fdae ]---

	Ingo

------------------------------------------------------------------------------
Apps built with the Adobe(R) Flex(R) framework and Flex Builder(TM) are
powering Web 2.0 with engaging, cross-platform capabilities. Quickly and
easily build your RIAs with Flex Builder, the Eclipse(TM)based development
software that enables intelligent coding and step-through debugging.
Download the free 60 day trial. http://p.sf.net/sfu/www-adobe-com