| 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
| ||
|
Date: Mon, 8 May 2017 10:54:41 +0100
From: Joao Pinto <Joao.Pinto@...opsys.com>
To: Andy Shevchenko <andy.shevchenko@...il.com>,
Jan Kiszka <jan.kiszka@...mens.com>
CC: Joao Pinto <Joao.Pinto@...opsys.com>,
"David S. Miller" <davem@...emloft.net>,
Giuseppe CAVALLARO <peppe.cavallaro@...com>,
Alexandre TORGUE <alexandre.torgue@...com>,
netdev <netdev@...r.kernel.org>,
Linux Kernel Mailing List <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH v3 net-next 01/11] net: stmmac: prepare dma op mode config
for multiple queues
Hi Andy and Jan,
Às 10:36 AM de 5/8/2017, Andy Shevchenko escreveu:
> On Mon, May 8, 2017 at 9:56 AM, Jan Kiszka <jan.kiszka@...mens.com> wrote:
>> On 2017-03-15 12:04, Joao Pinto wrote:
>>> This patch prepares DMA Operation Mode configuration for multiple queues.
>>> The work consisted on breaking the DMA operation Mode configuration function
>>> into RX and TX scope and adapting its mechanism in stmmac_main.
>
>> Starting with this patch, the stmmac-based network adapters of the Intel
>> Quark SoC stop working. I'm getting an IP via DHCP, I can ping, but TCP
>> connections can no longer be established.
>>
>> Moving on a few patches (didn't bisect the exact one yet), the TX
>> watchdog starts to fire, and DHCP fails completely. And if I go to
>> current master in Linus tree (reverting an unrelated boot regression), I
>> even get a crash in stmmac_xmit.
>>
>> Here are some details about the hw from dma_cap POV, if this helps:
>>
>> ==============================
>> DMA HW features
>> ==============================
>> 10/100 Mbps: Y
>> 1000 Mbps: N
>> Half duplex: Y
>> Hash Filter: Y
>> Multiple MAC address registers: N
>> PCS (TBI/SGMII/RTBI PHY interfaces): N
>> SMA (MDIO) Interface: Y
>> PMT Remote wake up: N
>> PMT Magic Frame: N
>> RMON module: Y
>> IEEE 1588-2002 Time Stamp: N
>> IEEE 1588-2008 Advanced Time Stamp: Y
>> 802.3az - Energy-Efficient Ethernet (EEE): N
>> AV features: N
>> Checksum Offload in TX: Y
>> IP Checksum Offload (type1) in RX: N
>> IP Checksum Offload (type2) in RX: Y
>> RXFIFO > 2048bytes: Y
>> Number of Additional RX channel: 0
>> Number of Additional TX channel: 0
>> Enhanced descriptors: Y
>>
>> Given the number of different failure modes, my feeling is that there
>> are multiple regressions coming with these patches...
>>
>> I've tested on the IOT2000 board, but I suspect the Galileo Gen2 will be
>> affected equally. If you don't have access to any such device, let me
>> know what I can debug for you.
>
> JFYI: With today's linux-next when _kexec:ed_ kernel boots I tried and
> got the following:
>
>
> # ip a s
> 1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue qlen 1000
> link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
> inet 127.0.0.1/8 scope host lo
> valid_lft forever preferred_lft forever
> inet6 ::1/128 scope host
> valid_lft forever preferre[ 130.403995] random: fast init done
> d_lft forever
> 2: eth0: <BROADCAST,MULTICAST> mtu 1500 qdisc noop qlen 1000
> link/ether 00:00:00:00:00:00 brd ff:ff:ff:ff:ff:ff
> 3: eth1: <BROADCAST,MULTICAST> mtu 1500 qdisc noop qlen 1000
> link/ether 00:00:00:00:00:00 brd ff:ff:ff:ff:ff:ff
> 4: sit0@...E: <NOARP> mtu 1480 qdisc noop qlen 1000
> link/sit 0.0.0.0 brd 0.0.0.0
> # udhcpc -i eth0
> udhcpc: started, v1.26.2
> [ 140.825131] stmmaceth 0000:00:14.6 eth0: device MAC address 98:4f:ee:05:ac:47
> [ 140.834304] Generic PHY stmmac-a6:01: attached PHY driver [Generic
> PHY] (mii_bus:phy_addr=stmmac-a6:01, irq=-1)
> [ 140.930871] stmmaceth 0000:00:14.6 eth0: IEEE 1588-2008 Advanced
> Timestamp supported
> [ 140.941109] stmmaceth 0000:00:14.6 eth0: registered PTP clock
> [ 140.953626] IPv6: ADDRCONF(NETDEV_UP): eth0: link is not ready
> udhcpc: sending discover
> [ 142.979557] stmmaceth 0000:00:14.6 eth0: Link is Up - 100Mbps/Full
> - flow control off
> [ 142.988756] IPv6: ADDRCONF(NETDEV_CHANGE): eth0: link becomes ready
> [ 142.998810] BUG: unable to handle kernel NULL pointer dereference at (null)
> [ 143.006193] IP: stmmac_xmit+0xf1/0x1080
> [ 143.010168] *pde = 00000000
> [ 143.010177]
> [ 143.014762] Oops: 0002 [#1]
> [ 143.017672] Modules linked in: at24 nvmem_core pwm_pca9685
> [ 143.023338] CPU: 0 PID: 0 Comm: swapper Not tainted 4.11.0-next-20170508+ #2
> [ 143.030539] task: c8533580 task.stack: c852c000
> [ 143.035237] EIP: stmmac_xmit+0xf1/0x1080
> [ 143.039302] EFLAGS: 00010216 CPU: 0
> [ 143.042915] EAX: 00000000 EBX: 00000050 ECX: 00000000 EDX: ceb6a0c0
> [ 143.049326] ESI: 00000000 EDI: cdd16000 EBP: cdc25d70 ESP: cdc25d20
> [ 143.055735] DS: 007b ES: 007b FS: 0000 GS: 0000 SS: 0068
> [ 143.061271] CR0: 80050033 CR2: 00000000 CR3: 0eb5c000 CR4: 00100010
> [ 143.067671] Call Trace:
> [ 143.070238] <SOFTIRQ>
> [ 143.072763] dev_hard_start_xmit+0x7c/0x1a0
> [ 143.077120] sch_direct_xmit+0xf0/0x120
> [ 143.081130] __dev_queue_xmit+0x181/0x430
> [ 143.085311] ? eth_commit_mac_addr_change+0x20/0x20
> [ 143.090362] dev_queue_xmit+0xa/0x10
> [ 143.094100] neigh_resolve_output+0xdb/0x190
> [ 143.098561] ip6_finish_output2+0x184/0x500
> [ 143.102945] ip6_finish_output+0x91/0xe0
> [ 143.107057] ? ip6_finish_output+0x91/0xe0
> [ 143.111338] ip6_output+0x36/0x110
> [ 143.114924] ? ip6_fragment+0xb00/0xb00
> [ 143.118935] mld_sendpack+0x191/0x2b0
> [ 143.122769] ? mld_newpack+0xda/0x180
> [ 143.126598] ? ipv6_icmp_sysctl_init+0x30/0x30
> [ 143.131224] mld_ifc_timer_expire+0x158/0x240
> [ 143.135756] ? find_next_bit+0xa/0x10
> [ 143.139584] ? mld_dad_timer_expire+0x50/0x50
> [ 143.144112] call_timer_fn+0x2a/0xf0
> [ 143.147862] ? mld_dad_timer_expire+0x50/0x50
> [ 143.152395] run_timer_softirq+0x158/0x300
> [ 143.156668] ? file_free_rcu+0x1e/0x30
> [ 143.160589] __do_softirq+0xc4/0x200
> [ 143.164341] ? __hrtimer_tasklet_trampoline+0x30/0x30
> [ 143.169575] do_softirq_own_stack+0x1e/0x30
> [ 143.173902] </SOFTIRQ>
> [ 143.176502] irq_exit+0x95/0xa0
> [ 143.179812] smp_apic_timer_interrupt+0x31/0x40
> [ 143.184530] apic_timer_interrupt+0x32/0x40
> [ 143.188889] EIP: default_idle+0xc/0x70
> [ 143.192774] EFLAGS: 00000246 CPU: 0
> [ 143.196386] EAX: 00000000 EBX: 00000000 ECX: 00000001 EDX: 00000000
> [ 143.202795] ESI: 00000000 EDI: c8533580 EBP: c852df54 ESP: c852df4c
> [ 143.209205] DS: 007b ES: 007b FS: 0000 GS: 0000 SS: 0068
> [ 143.214780] arch_cpu_idle+0x9/0x10
> [ 143.218446] default_idle_call+0x17/0x30
> [ 143.222551] do_idle+0xed/0x130
> [ 143.225873] cpu_startup_entry+0x15/0x20
> [ 143.229965] rest_init+0x5c/0x60
> [ 143.233370] start_kernel+0x313/0x318
> [ 143.237221] i386_start_kernel+0x98/0x9c
> [ 143.241315] startup_32_smp+0x16b/0x16d
> [ 143.245289] Code: 84 45 06 00 00 c1 e2 05 03 94 c7 9c 09 00 00 89
> 55 b0 8b 45 c8 8b 75 bc 8b 55 d8 8d 1c 80 89
> 75 e4 c1 e3 03 8b 84 1f a4 09 00 00 <89> 14 b0 8b 87 40 0d 00 00 8b 40
> 24 85 c0 89 45 b8 0f 85 68 02
> [ 143.264746] EIP: stmmac_xmit+0xf1/0x1080 SS:ESP: 0068:cdc25d20
> [ 143.270727] CR2: 0000000000000000
> [ 143.274175] ---[ end trace 79da8ef70f8b98d7 ]---
> [ 143.278925] Kernel panic - not syncing: Fatal exception in interrupt
> [ 143.285433] Kernel Offset: 0x6a00000 from 0xc1000000 (relocation
> range: 0xc0000000-0xd05effff)
> [ 143.294268] ---[ end Kernel panic - not syncing: Fatal exception in interrupt
>
>
Are you using the same version of Ethernet IP, 10/100?
Could you please verify if the crash you are experiencing is this place?
https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c#n2956
I would say that for rather old IPs, the napi is not capable of giving a valid
queue number. Could you please print the queue index returned by this line?
https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c#n2948
Thank you.
Joao Pinto
Powered by blists - more mailing lists