netdev - Re: BCM5709 hang and state dump...

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]
Date:	Fri, 22 Feb 2013 10:33:54 +0800
From:	Daniel J Blueman <daniel@...ascale-asia.com>
To:	Michael Chan <mchan@...adcom.com>
CC:	Eilon Greenstein <eilong@...adcom.com>,
	Steffen Persvold <sp@...ascale.com>, netdev@...r.kernel.org
Subject: Re: BCM5709 hang and state dump...

Hi Michael,

Thanks for your reply.

We'll probably be able to reproduce it next week and collect the output 
with your debug patches if useful.

Thanks again,
   Daniel

On 22/02/2013 05:59, Michael Chan wrote:
> On Thu, 2013-02-21 at 13:26 +0800, Daniel J Blueman wrote:
>> Hi Michael/Eilon,
>>
>> On a large system with 552 cores, 1.5TB memory and linux 3.7, under some
>> particular workloads, we've seem the Broadcom 5709 network controller
>> hang [1]. It's running boot code 6.2.0 and NCSI code 2.0.11.
>>
>> We suspect completion timeouts may be occurring due to possible starvation.
>>
>> Is there anything significant/indicative from the state dumped?
>
> The firmware state seems to be ok, although we see some MSIX interrupts
> being asserted internally which is a sign that they don't get serviced.
>
> Is this easily reproducible?  Can we send you some debug patches to dump
> more data?
>
> Thanks.
>
>>
>> Many thanks,
>>     Daniel
>>
>> --- [1]
>>
>> bnx2: Broadcom NetXtreme II Gigabit Ethernet Driver bnx2 v2.2.3 (June
>> 27, 2012)
>> bnx2 0000:01:00.0 eth0: Broadcom NetXtreme II BCM5709 1000Base-T (C0)
>> PCI Express found at mem fc000000, IRQ 44, node addr e4:1f:13:80:70:03
>> bnx2 0000:01:00.1: enabling device (0140 -> 0142)
>> bnx2 0000:01:00.0: irq 72 for MSI/MSI-X
>> bnx2 0000:01:00.0: irq 73 for MSI/MSI-X
>> bnx2 0000:01:00.0: irq 74 for MSI/MSI-X
>> bnx2 0000:01:00.0: irq 75 for MSI/MSI-X
>> bnx2 0000:01:00.0: irq 76 for MSI/MSI-X
>> bnx2 0000:01:00.0: irq 77 for MSI/MSI-X
>> bnx2 0000:01:00.0: irq 78 for MSI/MSI-X
>> bnx2 0000:01:00.0: irq 79 for MSI/MSI-X
>> bnx2 0000:01:00.0 eth0: using MSIX
>> bnx2 0000:01:00.0 eth0: NIC Copper Link is Up, 1000 Mbps full duplex
>>
>> <an hour later>
>>
>> bnx2 0000:01:00.0 eth0: <--- start FTQ dump --->
>> bnx2 0000:01:00.0 eth0: RV2P_PFTQ_CTL 00010000
>> bnx2 0000:01:00.0 eth0: RV2P_TFTQ_CTL 00020000
>> bnx2 0000:01:00.0 eth0: RV2P_MFTQ_CTL 00004000
>> bnx2 0000:01:00.0 eth0: TBDR_FTQ_CTL 00004000
>> bnx2 0000:01:00.0 eth0: TDMA_FTQ_CTL 00010000
>> bnx2 0000:01:00.0 eth0: TXP_FTQ_CTL 00010000
>> bnx2 0000:01:00.0 eth0: TXP_FTQ_CTL 00010000
>> bnx2 0000:01:00.0 eth0: TPAT_FTQ_CTL 00010000
>> bnx2 0000:01:00.0 eth0: RXP_CFTQ_CTL 00008000
>> bnx2 0000:01:00.0 eth0: RXP_FTQ_CTL 00100000
>> bnx2 0000:01:00.0 eth0: COM_COMXQ_FTQ_CTL 00010000
>> bnx2 0000:01:00.0 eth0: COM_COMTQ_FTQ_CTL 00020000
>> bnx2 0000:01:00.0 eth0: COM_COMQ_FTQ_CTL 00010000
>> bnx2 0000:01:00.0 eth0: CP_CPQ_FTQ_CTL 00004000
>> bnx2 0000:01:00.0 eth0: CPU states:
>> bnx2 0000:01:00.0 eth0: 045000 mode b84c state 80001000 evt_mask 500 pc
>> 8001284 pc 8001284 instr 8e260000
>> bnx2 0000:01:00.0 eth0: 085000 mode b84c state 80005000 evt_mask 500 pc
>> 8000a4c pc 8000a5c instr 38420001
>> bnx2 0000:01:00.0 eth0: 0c5000 mode b84c state 80001000 evt_mask 500 pc
>> 8004c20 pc 8004c10 instr 32050003
>> bnx2 0000:01:00.0 eth0: 105000 mode b8cc state 80008000 evt_mask 500 pc
>> 8000aa0 pc 8000aa0 instr 8c420020
>> bnx2 0000:01:00.0 eth0: 145000 mode b880 state 80000000 evt_mask 500 pc
>> 800d978 pc 8009c18 instr afbf001c
>> bnx2 0000:01:00.0 eth0: 185000 mode b8cc state 80000000 evt_mask 500 pc
>> 8000cb0 pc 8000c58 instr 8ce800e8
>> bnx2 0000:01:00.0 eth0: <--- end FTQ dump --->
>> bnx2 0000:01:00.0 eth0: <--- start TBDC dump --->
>> bnx2 0000:01:00.0 eth0: TBDC free cnt: 32
>> bnx2 0000:01:00.0 eth0: LINE   CID BIDX  CMD VALIDS
>> bnx2 0000:01:00.0 eth0: 00  001180 0f40  00  [0]
>> bnx2 0000:01:00.0 eth0: 01  001180 0f48  00  [0]
>> bnx2 0000:01:00.0 eth0: 02  1db680 af58  f6  [0]
>> bnx2 0000:01:00.0 eth0: 03  0ddd00 fb58  fd  [0]
>> bnx2 0000:01:00.0 eth0: 04  1fff80 ffc8  ef  [0]
>> bnx2 0000:01:00.0 eth0: 05  1e9f80 9fa8  cf  [0]
>> bnx2 0000:01:00.0 eth0: 06  1d7380 77e8  ff  [0]
>> bnx2 0000:01:00.0 eth0: 07  1ddf00 7bb0  fb  [0]
>> bnx2 0000:01:00.0 eth0: 08  1edb80 ff78  6f  [0]
>> bnx2 0000:01:00.0 eth0: 09  1e9e80 ee58  9e  [0]
>> bnx2 0000:01:00.0 eth0: 0a  17f780 fff8  74  [0]
>> bnx2 0000:01:00.0 eth0: 0b  1d7e00 6db8  fd  [0]
>> bnx2 0000:01:00.0 eth0: 0c  1f7780 bff0  cf  [0]
>> bnx2 0000:01:00.0 eth0: 0d  1bff80 bff8  ff  [0]
>> bnx2 0000:01:00.0 eth0: 0e  17ff80 3de0  fe  [0]
>> bnx2 0000:01:00.0 eth0: 0f  1ff780 98f0  ff  [0]
>> bnx2 0000:01:00.0 eth0: 10  1f7f80 ffd8  ee  [0]
>> bnx2 0000:01:00.0 eth0: 11  0e7780 eaa8  7f  [0]
>> bnx2 0000:01:00.0 eth0: 12  1f9980 fde8  f7  [0]
>> bnx2 0000:01:00.0 eth0: 13  07ef80 ffc8  77  [0]
>> bnx2 0000:01:00.0 eth0: 14  1fbf80 57e8  bf  [0]
>> bnx2 0000:01:00.0 eth0: 15  0fae80 df68  5b  [0]
>> bnx2 0000:01:00.0 eth0: 16  0fff80 7ff8  be  [0]
>> bnx2 0000:01:00.0 eth0: 17  1f7680 fed8  c6  [0]
>> bnx2 0000:01:00.0 eth0: 18  03e380 fe70  7b  [0]
>> bnx2 0000:01:00.0 eth0: 19  0bcd80 7db8  7f  [0]
>> bnx2 0000:01:00.0 eth0: 1a  0cb580 bbf0  ef  [0]
>> bnx2 0000:01:00.0 eth0: 1b  0dfd80 dbf8  fb  [0]
>> bnx2 0000:01:00.0 eth0: 1c  0bff80 7ff8  f3  [0]
>> bnx2 0000:01:00.0 eth0: 1d  0dfb80 f9f8  ec  [0]
>> bnx2 0000:01:00.0 eth0: 1e  1e6e80 9be8  f7  [0]
>> bnx2 0000:01:00.0 eth0: 1f  1faf80 db78  52  [0]
>> bnx2 0000:01:00.0 eth0: <--- end TBDC dump --->
>> bnx2 0000:01:00.0 eth0: DEBUG: intr_sem[0] PCI_CMD[00100546]
>> bnx2 0000:01:00.0 eth0: DEBUG: PCI_PM[19002008] PCI_MISC_CFG[92000088]
>> bnx2 0000:01:00.0 eth0: DEBUG: EMAC_TX_STATUS[00000008]
>> EMAC_RX_STATUS[00000000]
>> bnx2 0000:01:00.0 eth0: DEBUG: RPM_MGMT_PKT_CTRL[40000088]
>> bnx2 0000:01:00.0 eth0: DEBUG: HC_STATS_INTERRUPT_STATUS[010600f9]
>> bnx2 0000:01:00.0 eth0: DEBUG: PBA[00000000]
>> bnx2 0000:01:00.0 eth0: <--- start MCP states dump --->
>> bnx2 0000:01:00.0 eth0: DEBUG: MCP_STATE_P0[0003610e] MCP_STATE_P1[0003610e]
>> bnx2 0000:01:00.0 eth0: DEBUG: MCP mode[0000b880] state[80000000]
>> evt_mask[00000500]
>> bnx2 0000:01:00.0 eth0: DEBUG: pc[0800d31c] pc[0800b46c] instr[a023f35c]
>> bnx2 0000:01:00.0 eth0: DEBUG: shmem states:
>> bnx2 0000:01:00.0 eth0: DEBUG: drv_mb[01030003] fw_mb[00000003]
>> link_status[8000006f]
>> bnx2 0000:01:00.0 eth0: DEBUG: dev_info_signature[44564903]
>> reset_type[01005254]
>> bnx2 0000:01:00.0 eth0: DEBUG: 000001c0: 01005254 42530083 0003610e 00000000
>> bnx2 0000:01:00.0 eth0: DEBUG: 000003cc: 44444444 44444444 44444444 00000a14
>> bnx2 0000:01:00.0 eth0: DEBUG: 000003dc: 0004ffff 00000000 00000000 00000000
>> bnx2 0000:01:00.0 eth0: DEBUG: 000003ec: 00000000 00000000 00000000 00000000
>> bnx2 0000:01:00.0 eth0: DEBUG: 0x3fc[0000ffff]
>> bnx2 0000:01:00.0 eth0: <--- end MCP states dump --->
>> bnx2 0000:01:00.0 eth0: NIC Copper Link is Down
-- 
Daniel J Blueman
Principal Software Engineer, Numascale Asia
--
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