Message-ID: <CAHp75Vfe0pXTEvdt7Z3-DgMTFdC5B3106ubj_D+_ng0vNCUaRA@mail.gmail.com>
Date:   Mon, 8 May 2017 12:36:10 +0300
From:   Andy Shevchenko <andy.shevchenko@...il.com>
To:     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

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


-- 
With Best Regards,
Andy Shevchenko

Powered by blists - more mailing lists