Date:	Sat, 12 Jan 2008 11:15:53 +0300
From:	slavon@...telecom.ru
To:	"Kok, Auke" <auke-jan.h.kok@...el.com>
Cc:	netdev@...r.kernel.org
Subject: Re: e1000_clean_tx_irq: Detected Tx Unit Hang - it's bug?

>> Hello all.
>> Some time in dmesg i see this:
>>
>> [16121.400422] e1000: eth0: e1000_clean_tx_irq: Detected Tx Unit Hang
>> [16121.400426] Tx Queue <0>
>> [16121.400427] TDH <28>
>> [16121.400429] TDT <28>
>> [16121.400430] next_to_use <28>
>> [16121.400431] next_to_clean <7d>
>> [16121.400433] buffer_info[next_to_clean]
>> [16121.400434] time_stamp <17b949>
>> [16121.400435] next_to_watch <7d>
>> [16121.400437] jiffies <17ba57>
>> [16121.400438] next_to_watch.status <1>
>
> might be a bug. What kernel version are you using?

Hello.

Now i try 2.6.24-rc7-git2 - its have NAPI patches that work GREAT!

Have many messages like up in dmeseg.

Also more info for You
fw ~ # ethtool -S eth0

NIC statistics:
rx_packets: 1048831452
tx_packets: 28418
rx_bytes: 644208597062
tx_bytes: 3458632
rx_broadcast: 1840
tx_broadcast: 5
rx_multicast: 0
tx_multicast: 0
rx_errors: 1663
tx_errors: 0
tx_dropped: 0
multicast: 0
collisions: 0
rx_length_errors: 0
rx_over_errors: 0
rx_crc_errors: 5172
rx_frame_errors: 0
rx_no_buffer_count: 153305
rx_missed_errors: 883176
tx_aborted_errors: 0
tx_carrier_errors: 0
tx_fifo_errors: 0
tx_heartbeat_errors: 0
tx_window_errors: 0
tx_abort_late_coll: 0
tx_deferred_ok: 0
tx_single_coll_ok: 0
tx_multi_coll_ok: 0
tx_timeout_count: 0
tx_restart_queue: 0
rx_long_length_errors: 0
rx_short_length_errors: 0
rx_align_errors: 0
tx_tcp_seg_good: 10
tx_tcp_seg_failed: 0
rx_flow_control_xon: 0
rx_flow_control_xoff: 0
tx_flow_control_xon: 0
tx_flow_control_xoff: 0
rx_long_byte_count: 644208597062
rx_csum_offload_good: 1030747822
rx_csum_offload_errors: 7086
rx_header_split: 0
alloc_rx_buff_failed: 0
tx_smbus: 0
rx_smbus: 0
dropped_smbus: 0


That my dmesg:
[ 5280.282257] e1000: eth0: e1000_clean_tx_irq: Detected Tx Unit Hang
[ 5280.282264] Tx Queue <0>
[ 5280.282265] TDH <67>
[ 5280.282265] TDT <a9>
[ 5280.282266] next_to_use <a9>
[ 5280.282267] next_to_clean <fe>
[ 5280.282268] buffer_info[next_to_clean]
[ 5280.282269] time_stamp <76c86>
[ 5280.282270] next_to_watch <fe>
[ 5280.282271] jiffies <76d33>
[ 5280.282272] next_to_watch.status <1>
[ 272.396662] opreport[5752]: segfault at b7be1010 eip 080b396e esp  
bfba9480 error 4
[ 2264.085353] htb: too many events !
[ 3376.658037] htb: too many events !
[ 5724.766531] e1000: eth0: e1000_clean_tx_irq: Detected Tx Unit Hang
[ 5724.766533] Tx Queue <0>
[ 5724.766534] TDH <23>
[ 5724.766535] TDT <23>
[ 5724.766536] next_to_use <23>
[ 5724.766537] next_to_clean <fd>
[ 5724.766537] buffer_info[next_to_clean]
[ 5724.766538] time_stamp <50db81>
[ 5724.766539] next_to_watch <fd>
[ 5724.766540] jiffies <50e3ab>
[ 5724.766541] next_to_watch.status <1>
[ 5724.767771] htb: too many events !
[10527.197125] htb: too many events !
[11744.405451] htb: too many events !
[12925.662880] htb: too many events !
[14123.186443] htb: too many events !
[15333.652189] htb: too many events !
[16522.524045] htb: too many events !
[17722.572386] htb: too many events !
[18925.121034] htb: too many events !
[20117.689133] htb: too many events !
[20117.690436] e1000: eth0: e1000_clean_tx_irq: Detected Tx Unit Hang
[20117.690438] Tx Queue <0>
[20117.690439] TDH <8>
[20117.690440] TDT <4b>
[20117.690441] next_to_use <4b>
[20117.690442] next_to_clean <a0>
[20117.690443] buffer_info[next_to_clean]
[20117.690443] time_stamp <128b414>
[20117.690444] next_to_watch <a0>
[20117.690445] jiffies <128bc3f>
[20117.690446] next_to_watch.status <1>
[21314.498859] htb: too many events !
[22192.855455] htb: too many events !
[22520.640345] htb: too many events !
[23712.083687] htb: too many events !
[24912.407418] htb: too many events !
[26128.954774] htb: too many events !
[27314.726708] htb: too many events !
[28517.307859] htb: too many events !
[28517.309179] e1000: eth0: e1000_clean_tx_irq: Detected Tx Unit Hang
[28517.309181] Tx Queue <0>
[28517.309182] TDH <1a>
[28517.309183] TDT <4f>
[28517.309184] next_to_use <4f>
[28517.309185] next_to_clean <a4>
[28517.309186] buffer_info[next_to_clean]
[28517.309187] time_stamp <1a5c7c3>
[28517.309187] next_to_watch <a4>
[28517.309188] jiffies <1a5cf20>
[28517.309189] next_to_watch.status <1>
[30922.659736] htb: too many events !
[30922.661064] e1000: eth0: e1000_clean_tx_irq: Detected Tx Unit Hang
[30922.661066] Tx Queue <0>
[30922.661067] TDH <51>
[30922.661068] TDT <76>
[30922.661069] next_to_use <76>
[30922.661070] next_to_clean <cb>
[30922.661071] buffer_info[next_to_clean]
[30922.661071] time_stamp <1c9ad65>
[30922.661072] next_to_watch <cb>
[30922.661073] jiffies <1c9b459>
[30922.661074] next_to_watch.status <1>
[32124.687849] htb: too many events !
[33345.636940] htb: too many events !
[34528.356531] htb: too many events !
[34528.358024] e1000: eth0: e1000_clean_tx_irq: Detected Tx Unit Hang
[34528.358027] Tx Queue <0>
[34528.358028] TDH <fa>
[34528.358029] TDT <3>
[34528.358029] next_to_use <3>
[34528.358030] next_to_clean <58>
[34528.358031] buffer_info[next_to_clean]
[34528.358032] time_stamp <1ff85ab>
[34528.358033] next_to_watch <58>
[34528.358034] jiffies <1ff8d27>
[34528.358035] next_to_watch.status <1>
[35734.669698] e1000: eth0: e1000_clean_tx_irq: Detected Tx Unit Hang
[35734.669700] Tx Queue <0>
[35734.669701] TDH <5f>
[35734.669702] TDT <5f>
[35734.669703] next_to_use <5f>
[35734.669704] next_to_clean <bf>
[35734.669705] buffer_info[next_to_clean]
[35734.669705] time_stamp <211994b>
[35734.669706] next_to_watch <bf>
[35734.669707] jiffies <211a017>
[35734.669708] next_to_watch.status <1>
[35734.672161] htb: too many events !
[36955.160096] htb: too many events !
[38118.346906] htb: too many events !
[39330.213684] htb: too many events !
[39330.214762] e1000: eth0: e1000_clean_tx_irq: Detected Tx Unit Hang
[39330.214764] Tx Queue <0>
[39330.214765] TDH <eb>
[39330.214766] TDT <45>
[39330.214767] next_to_use <45>
[39330.214767] next_to_clean <9a>
[39330.214768] buffer_info[next_to_clean]
[39330.214769] time_stamp <2477328>
[39330.214770] next_to_watch <9a>
[39330.214771] jiffies <2477b16>
[39330.214772] next_to_watch.status <1>
[40551.898491] htb: too many events !
[40551.899264] e1000: eth0: e1000_clean_tx_irq: Detected Tx Unit Hang
[40551.899266] Tx Queue <0>
[40551.899267] TDH <bf>
[40551.899268] TDT <ca>
[40551.899269] next_to_use <ca>
[40551.899269] next_to_clean <1f>
[40551.899270] buffer_info[next_to_clean]
[40551.899271] time_stamp <259c212>
[40551.899272] next_to_watch <1f>
[40551.899273] jiffies <259c91e>
[40551.899274] next_to_watch.status <1>
[41724.774002] htb: too many events !
[42920.657519] htb: too many events !
[44143.420904] htb: too many events !
[45303.157047] htb: too many events !
[46492.258329] htb: too many events !
[47694.779740] htb: too many events !
[58471.571956] e1000: eth0: e1000_clean_tx_irq: Detected Tx Unit Hang
[58471.571959] Tx Queue <0>
[58471.571960] TDH <7b>
[58471.571960] TDT <7b>
[58471.571961] next_to_use <7b>
[58471.571962] next_to_clean <d0>
[58471.571963] buffer_info[next_to_clean]
[58471.571964] time_stamp <36516dd>
[58471.571965] next_to_watch <d0>
[58471.571966] jiffies <36531cc>
[58471.571967] next_to_watch.status <1>
[59669.526000] e1000: eth0: e1000_clean_tx_irq: Detected Tx Unit Hang
[59669.526002] Tx Queue <0>
[59669.526003] TDH <6>
[59669.526004] TDT <1e>
[59669.526005] next_to_use <1e>
[59669.526006] next_to_clean <73>
[59669.526007] buffer_info[next_to_clean]
[59669.526008] time_stamp <37718da>
[59669.526009] next_to_watch <73>
[59669.526009] jiffies <3773434>
[59669.526010] next_to_watch.status <1>
[60865.254499] NETDEV WATCHDOG: eth0: transmit timed out
[60865.254584] e1000: eth0: e1000_clean_tx_irq: Detected Tx Unit Hang
[60865.254585] Tx Queue <0>
[60865.254586] TDH <ed>
[60865.254587] TDT <ed>
[60865.254588] next_to_use <ed>
[60865.254589] next_to_clean <4c>
[60865.254590] buffer_info[next_to_clean]
[60865.254590] time_stamp <38919bb>
[60865.254591] next_to_watch <4c>
[60865.254592] jiffies <38934a2>
[60865.254593] next_to_watch.status <1>
[62064.876121] e1000: eth0: e1000_clean_tx_irq: Detected Tx Unit Hang
[62064.876123] Tx Queue <0>
[62064.876124] TDH <6a>
[62064.876125] TDT <71>
[62064.876126] next_to_use <71>
[62064.876126] next_to_clean <c6>
[62064.876127] buffer_info[next_to_clean]
[62064.876128] time_stamp <39b221d>
[62064.876129] next_to_watch <c6>
[62064.876130] jiffies <39b3d07>
[62064.876131] next_to_watch.status <1>
[63261.288625] e1000: eth0: e1000_clean_tx_irq: Detected Tx Unit Hang
[63261.288628] Tx Queue <0>
[63261.288629] TDH <b9>
[63261.288630] TDT <b9>
[63261.288631] next_to_use <b9>
[63261.288631] next_to_clean <e>
[63261.288632] buffer_info[next_to_clean]
[63261.288633] time_stamp <3ad1b46>
[63261.288634] next_to_watch <e>
[63261.288635] jiffies <3ad3b43>
[63261.288636] next_to_watch.status <1>
[64459.486793] e1000: eth0: e1000_clean_tx_irq: Detected Tx Unit Hang
[64459.486796] Tx Queue <0>
[64459.486797] TDH <19>
[64459.486798] TDT <19>
[64459.486798] next_to_use <19>
[64459.486799] next_to_clean <ae>
[64459.486800] buffer_info[next_to_clean]
[64459.486801] time_stamp <3bf25e5>
[64459.486802] next_to_watch <ae>
[64459.486803] jiffies <3bf422f>
[64459.486804] next_to_watch.status <1>
[65663.246067] e1000: eth0: e1000_clean_tx_irq: Detected Tx Unit Hang
[65663.246069] Tx Queue <0>
[65663.246070] TDH <c3>
[65663.246071] TDT <17>
[65663.246072] next_to_use <17>
[65663.246073] next_to_clean <6c>
[65663.246074] buffer_info[next_to_clean]
[65663.246075] time_stamp <3d141e5>
[65663.246076] next_to_watch <6c>
[65663.246077] jiffies <3d15cca>
[65663.246077] next_to_watch.status <1>
[66855.416474] e1000: eth0: e1000_clean_tx_irq: Detected Tx Unit Hang
[66855.416477] Tx Queue <0>
[66855.416478] TDH <e5>
[66855.416479] TDT <e5>
[66855.416480] next_to_use <e5>
[66855.416480] next_to_clean <96>
[66855.416481] buffer_info[next_to_clean]
[66855.416482] time_stamp <3e333f4>
[66855.416483] next_to_watch <96>
[66855.416484] jiffies <3e35090>
[66855.416485] next_to_watch.status <1>
[68054.040533] e1000: eth0: e1000_clean_tx_irq: Detected Tx Unit Hang
[68054.040536] Tx Queue <0>
[68054.040537] TDH <b2>
[68054.040538] TDT <2>
[68054.040538] next_to_use <2>
[68054.040539] next_to_clean <57>
[68054.040540] buffer_info[next_to_clean]
[68054.040541] time_stamp <3f546cd>
[68054.040542] next_to_watch <57>
[68054.040542] jiffies <3f560f6>
[68054.040543] next_to_watch.status <1>
[69260.670653] e1000: eth0: e1000_clean_tx_irq: Detected Tx Unit Hang
[69260.670656] Tx Queue <0>
[69260.670657] TDH <83>
[69260.670657] TDT <e9>
[69260.670658] next_to_use <e9>
[69260.670659] next_to_clean <3e>
[69260.670660] buffer_info[next_to_clean]
[69260.670661] time_stamp <4077a70>
[69260.670662] next_to_watch <3e>
[69260.670663] jiffies <40795d1>
[69260.670663] next_to_watch.status <1>
[70450.618127] e1000: eth0: e1000_clean_tx_irq: Detected Tx Unit Hang
[70450.618130] Tx Queue <0>
[70450.618131] TDH <e5>
[70450.618131] TDT <35>
[70450.618132] next_to_use <35>
[70450.618133] next_to_clean <8a>
[70450.618134] buffer_info[next_to_clean]
[70450.618135] time_stamp <4196ed3>
[70450.618136] next_to_watch <8a>
[70450.618137] jiffies <41987ab>
[70450.618137] next_to_watch.status <1>
[71648.909256] e1000: eth0: e1000_clean_tx_irq: Detected Tx Unit Hang
[71648.909258] Tx Queue <0>
[71648.909259] TDH <41>
[71648.909260] TDT <42>
[71648.909261] next_to_use <42>
[71648.909262] next_to_clean <97>
[71648.909262] buffer_info[next_to_clean]
[71648.909263] time_stamp <42b8542>
[71648.909264] next_to_watch <97>
[71648.909265] jiffies <42b9fa1>
[71648.909266] next_to_watch.status <1>
[72851.697294] e1000: eth0: e1000_clean_tx_irq: Detected Tx Unit Hang
[72851.697297] Tx Queue <0>
[72851.697298] TDH <30>
[72851.697298] TDT <54>
[72851.697299] next_to_use <54>
[72851.697300] next_to_clean <a9>
[72851.697301] buffer_info[next_to_clean]
[72851.697302] time_stamp <43dadd4>
[72851.697303] next_to_watch <a9>
[72851.697304] jiffies <43dc870>
[72851.697305] next_to_watch.status <1>
[75245.677954] e1000: eth0: e1000_clean_tx_irq: Detected Tx Unit Hang
[75245.677957] Tx Queue <0>
[75245.677958] TDH <85>
[75245.677958] TDT <b5>
[75245.677959] next_to_use <b5>
[75245.677960] next_to_clean <a>
[75245.677961] buffer_info[next_to_clean]
[75245.677962] time_stamp <461dd97>
[75245.677963] next_to_watch <a>
[75245.677964] jiffies <461f952>
[75245.677965] next_to_watch.status <1>
[76448.839007] e1000: eth0: e1000_clean_tx_irq: Detected Tx Unit Hang
[76448.839009] Tx Queue <0>
[76448.839011] TDH <18>
[76448.839012] TDT <68>
[76448.839012] next_to_use <68>
[76448.839013] next_to_clean <bd>
[76448.839014] buffer_info[next_to_clean]
[76448.839015] time_stamp <4740b7f>
[76448.839016] next_to_watch <bd>
[76448.839017] jiffies <4742788>
[76448.839018] next_to_watch.status <1>
[77641.472192] e1000: eth0: e1000_clean_tx_irq: Detected Tx Unit Hang
[77641.472195] Tx Queue <0>
[77641.472197] TDH <f>
[77641.472198] TDT <6a>
[77641.472198] next_to_use <6a>
[77641.472199] next_to_clean <d7>
[77641.472200] buffer_info[next_to_clean]
[77641.472201] time_stamp <48612f3>
[77641.472202] next_to_watch <d7>
[77641.472203] jiffies <4862c1f>
[77641.472204] next_to_watch.status <1>
[78841.540399] e1000: eth0: e1000_clean_tx_irq: Detected Tx Unit Hang
[78841.540401] Tx Queue <0>
[78841.540402] TDH <49>
[78841.540403] TDT <49>
[78841.540404] next_to_use <49>
[78841.540405] next_to_clean <40>
[78841.540406] buffer_info[next_to_clean]
[78841.540406] time_stamp <49832d4>
[78841.540407] next_to_watch <40>
[78841.540408] jiffies <4984da3>
[78841.540409] next_to_watch.status <1>
[80047.286824] e1000: eth0: e1000_clean_tx_irq: Detected Tx Unit Hang
[80047.286827] Tx Queue <0>
[80047.286828] TDH <fc>
[80047.286829] TDT <37>
[80047.286830] next_to_use <37>
[80047.286830] next_to_clean <8c>
[80047.286831] buffer_info[next_to_clean]
[80047.286832] time_stamp <4aa66e4>
[80047.286833] next_to_watch <8c>
[80047.286834] jiffies <4aa812b>
[80047.286835] next_to_watch.status <1>
[81240.261877] e1000: eth0: e1000_clean_tx_irq: Detected Tx Unit Hang
[81240.261879] Tx Queue <0>
[81240.261880] TDH <8f>
[81240.261881] TDT <be>
[81240.261882] next_to_use <be>
[81240.261883] next_to_clean <13>
[81240.261884] buffer_info[next_to_clean]
[81240.261884] time_stamp <4bc5b32>
[81240.261885] next_to_watch <13>
[81240.261886] jiffies <4bc73d9>
[81240.261887] next_to_watch.status <1>
[82440.249836] e1000: eth0: e1000_clean_tx_irq: Detected Tx Unit Hang
[82440.249838] Tx Queue <0>
[82440.249839] TDH <9d>
[82440.249840] TDT <9d>
[82440.249841] next_to_use <9d>
[82440.249842] next_to_clean <f2>
[82440.249842] buffer_info[next_to_clean]
[82440.249843] time_stamp <4ce6735>
[82440.249844] next_to_watch <f2>
[82440.249845] jiffies <4ce8149>
[82440.249846] next_to_watch.status <1>



> it appears the tx
> handler was
> just sitting idle and this message might be bogus, which is one of
> the things that
> we fixed recently.
>
> Auke
>
>


----------------------------------------------------------------
This message was sent using IMP, the Internet Messaging Program.

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

Powered by blists - more mailing lists