TI中文支持网TI美女帅哥工程师你们好：
我们使用dsp的C6638处理器，芯片中有4核ARM，跑了Linux系统，在测试万兆网卡速率时报错如下
1.板子ping通PC机的万兆网卡.2。pc也能ping通板子的万兆网口
但是在使用iperf压力测试的时候
pc端执行如下命令iperf -c 192.168.11.14 -t 60
板子执行如下命令 iperf -s
错误如下：
[ 8571.558145] ————[ cut here ]————
[ 8571.562786] WARNING: at net/sched/sch_generic.c:255 dev_watchdog+0x274/0x294()
[ 8571.570019] NETDEV WATCHDOG: eth4 (keystone-netcp): transmit queue 0 timed out
[ 8571.577229] Modules linked in:
[ 8571.580300] CPU: 0 PID: 0 Comm: swapper/0 Not tainted 3.10.10 #2
[ 8571.586325] [<c00142c0>] (unwind_backtrace+0x0/0xf8) from [<c00117ac>] (show_stack+0x10/0x14)
[ 8571.594864] [<c00117ac>] (show_stack+0x10/0x14) from [<c0020074>] (warn_slowpath_common+0x4c/0x6c)
[ 8571.603842] [<c0020074>] (warn_slowpath_common+0x4c/0x6c) from [<c0020128>] (warn_slowpath_fmt+0x30/0x40)
[ 8571.613418] [<c0020128>] (warn_slowpath_fmt+0x30/0x40) from [<c03cc1d0>] (dev_watchdog+0x274/0x294)
[ 8571.622478] [<c03cc1d0>] (dev_watchdog+0x274/0x294) from [<c002d788>] (call_timer_fn+0x30/0xf8)
[ 8571.631186] [<c002d788>] (call_timer_fn+0x30/0xf8) from [<c002ed80>] (run_timer_softirq+0x1b0/0x260)
[ 8571.640331] [<c002ed80>] (run_timer_softirq+0x1b0/0x260) from [<c002730c>] (__do_softirq+0xe8/0x258)
[ 8571.649469] [<c002730c>] (__do_softirq+0xe8/0x258) from [<c0027754>] (irq_exit+0x58/0x90)
[ 8571.657640] [<c0027754>] (irq_exit+0x58/0x90) from [<c000e978>] (handle_IRQ+0x3c/0x94)
[ 8571.665563] [<c000e978>] (handle_IRQ+0x3c/0x94) from [<c0008520>] (gic_handle_irq+0x28/0x5c)
[ 8571.674006] [<c0008520>] (gic_handle_irq+0x28/0x5c) from [<c04d2980>] (__irq_svc+0x40/0x50)
[ 8571.682355] Exception stack(0xc0727f60 to 0xc0727fa8)
[ 8571.687403] 7f60: c0bb3918 00000000 003849ac 00000000 c0726000 c07641d0 c07639fb c072e454
[ 8571.695580] 7f80: c07639fb c04da350 c0726000 c0726000 00000000 c0727fa8 c000eae8 c000eaec
[ 8571.703753] 7fa0: 600f0013 ffffffff
[ 8571.707244] [<c04d2980>] (__irq_svc+0x40/0x50) from [<c000eaec>] (arch_cpu_idle+0x28/0x30)
[ 8571.715524] [<c000eaec>] (arch_cpu_idle+0x28/0x30) from [<c0053fe4>] (cpu_startup_entry+0x58/0x1dc)
[ 8571.724579] [<c0053fe4>] (cpu_startup_entry+0x58/0x1dc) from [<c06d4a70>] (start_kernel+0x300/0x358)
[ 8571.733708] —[ end trace f2a4f0f913be259f ]—
[ 8571.738332] net eth4: transmit timed out
[ 8581.558150] net eth4: transmit timed out
[ 8731.562093] dma dma2chan0: out of descriptors
[ 8731.566447] net eth4: failed to prep slave dma
[ 8731.571040] dma dma2chan0: out of descriptors
[ 8731.575393] net eth4: failed to prep slave dma
在帖子中发现有好多人提这个问题，但是一直没有找到正面的回答。好像是什么东西溢出导致万兆网口down的原因，不知道如何处理，请指教。
Shine：
请问单核跑的时候有这个问题吗？
user5315971：
回复 Shine:
就是单核跑的。好像是万兆网卡什么溢出了，之前我在试千兆网卡的时候，拷贝个1M以上的文件，千兆网卡就会报同样的错误，后来我将内核版本升级到了K2_LINUX_03.10.10_14.07.tar.gz，千兆网卡不会出现如上问题，但是测试发现万兆万卡有如下问题。
user5315971：
回复 Shine:
网口是通的，但是不明白为什么ping包时间长了会出现如下现象。
root@k2hk-evm:~# ping 192.168.10.11
PING 192.168.10.11 (192.168.10.11): 56 data bytes
64 bytes from 192.168.10.11: seq=0 ttl=128 time=0.341 ms
64 bytes from 192.168.10.11: seq=1 ttl=128 time=0.274 ms
64 bytes from 192.168.10.11: seq=2 ttl=128 time=0.291 ms
64 bytes from 192.168.10.11: seq=3 ttl=128 time=0.287 ms
64 bytes from 192.168.10.11: seq=4 ttl=128 time=0.278 ms
64 bytes from 192.168.10.11: seq=5 ttl=128 time=0.275 ms
64 bytes from 192.168.10.11: seq=6 ttl=128 time=0.287 ms
64 bytes from 192.168.10.11: seq=7 ttl=128 time=0.276 ms
64 bytes from 192.168.10.11: seq=8 ttl=128 time=0.277 ms
64 bytes from 192.168.10.11: seq=9 ttl=128 time=0.269 ms
64 bytes from 192.168.10.11: seq=10 ttl=128 time=0.277 ms
64 bytes from 192.168.10.11: seq=11 ttl=128 time=0.270 ms
64 bytes from 192.168.10.11: seq=12 ttl=128 time=0.277 ms
64 bytes from 192.168.10.11: seq=13 ttl=128 time=0.285 ms
64 bytes from 192.168.10.11: seq=14 ttl=128 time=0.297 ms
64 bytes from 192.168.10.11: seq=15 ttl=128 time=0.287 ms
64 bytes from 192.168.10.11: seq=16 ttl=128 time=0.278 ms
64 bytes from 192.168.10.11: seq=17 ttl=128 time=0.287 ms
64 bytes from 192.168.10.11: seq=18 ttl=128 time=0.289 ms
64 bytes from 192.168.10.11: seq=19 ttl=128 time=0.280 ms
64 bytes from 192.168.10.11: seq=20 ttl=128 time=0.287 ms
64 bytes from 192.168.10.11: seq=21 ttl=128 time=0.282 ms
64 bytes from 192.168.10.11: seq=22 ttl=128 time=0.279 ms
64 bytes from 192.168.10.11: seq=23 ttl=128 time=0.282 ms
64 bytes from 192.168.10.11: seq=24 ttl=128 time=0.268 ms
64 bytes from 192.168.10.11: seq=25 ttl=128 time=0.279 ms
64 bytes from 192.168.10.11: seq=26 ttl=128 time=0.276 ms
64 bytes from 192.168.10.11: seq=27 ttl=128 time=0.271 ms
64 bytes from 192.168.10.11: seq=28 ttl=128 time=0.269 ms
64 bytes from 192.168.10.11: seq=29 ttl=128 time=0.272 ms
64 bytes from 192.168.10.11: seq=30 ttl=128 time=0.293 ms
64 bytes from 192.168.10.11: seq=31 ttl=128 time=0.265 ms
64 bytes from 192.168.10.11: seq=32 ttl=128 time=0.274 ms
64 bytes from 192.168.10.11: seq=33 ttl=128 time=0.305 ms
64 bytes from 192.168.10.11: seq=34 ttl=128 time=0.274 ms
64 bytes from 192.168.10.11: seq=35 ttl=128 time=0.282 ms
64 bytes from 192.168.10.11: seq=36 ttl=128 time=0.276 ms
64 bytes from 192.168.10.11: seq=37 ttl=128 time=0.272 ms
64 bytes from 192.168.10.11: seq=38 ttl=128 time=0.274 ms
64 bytes from 192.168.10.11: seq=39 ttl=128 time=0.266 ms
64 bytes from 192.168.10.11: seq=40 ttl=128 time=0.294 ms
64 bytes from 192.168.10.11: seq=41 ttl=128 time=0.277 ms
64 bytes from 192.168.10.11: seq=42 ttl=128 time=0.275 ms
64 bytes from 192.168.10.11: seq=43 ttl=128 time=0.288 ms
64 bytes from 192.168.10.11: seq=44 ttl=128 time=0.271 ms
64 bytes from 192.168.10.11: seq=45 ttl=128 time=0.270 ms
64 bytes from 192.168.10.11: seq=46 ttl=128 time=0.274 ms
64 bytes from 192.168.10.11: seq=47 ttl=128 time=0.269 ms
64 bytes from 192.168.10.11: seq=48 ttl=128 time=0.275 ms
64 bytes from 192.168.10.11: seq=49 ttl=128 time=0.274 ms
64 bytes from 192.168.10.11: seq=50 ttl=128 time=0.273 ms
64 bytes from 192.168.10.11: seq=51 ttl=128 time=0.266 ms
64 bytes from 192.168.10.11: seq=52 ttl=128 time=0.271 ms
64 bytes from 192.168.10.11: seq=53 ttl=128 time=0.272 ms
64 bytes from 192.168.10.11: seq=54 ttl=128 time=0.284 ms
64 bytes from 192.168.10.11: seq=55 ttl=128 time=0.280 ms
64 bytes from 192.168.10.11: seq=56 ttl=128 time=0.280 ms
64 bytes from 192.168.10.11: seq=57 ttl=128 time=0.277 ms
64 bytes from 192.168.10.11: seq=58 ttl=128 time=0.266 ms
64 bytes from 192.168.10.11: seq=59 ttl=128 time=0.273 ms
64 bytes from 192.168.10.11: seq=60 ttl=128 time=0.274 ms
64 bytes from 192.168.10.11: seq=61 ttl=128 time=0.271 ms
64 bytes from 192.168.10.11: seq=62 ttl=128 time=0.283 ms
64 bytes from 192.168.10.11: seq=63 ttl=128 time=0.304 ms
64 bytes from 192.168.10.11: seq=64 ttl=128 time=0.265 ms
64 bytes from 192.168.10.11: seq=65 ttl=128 time=0.266 ms
64 bytes from 192.168.10.11: seq=66 ttl=128 time=0.278 ms
64 bytes from 192.168.10.11: seq=67 ttl=128 time=0.273 ms
64 bytes from 192.168.10.11: seq=68 ttl=128 time=0.265 ms
64 bytes from 192.168.10.11: seq=69 ttl=128 time=0.276 ms
64 bytes from 192.168.10.11: seq=70 ttl=128 time=0.284 ms
64 bytes from 192.168.10.11: seq=71 ttl=128 time=0.290 ms
64 bytes from 192.168.10.11: seq=72 ttl=128 time=0.285 ms
64 bytes from 192.168.10.11: seq=73 ttl=128 time=0.285 ms
64 bytes from 192.168.10.11: seq=74 ttl=128 time=0.290 ms
64 bytes from 192.168.10.11: seq=75 ttl=128 time=0.281 ms
64 bytes from 192.168.10.11: seq=76 ttl=128 time=0.285 ms
64 bytes from 192.168.10.11: seq=77 ttl=128 time=0.280 ms
64 bytes from 192.168.10.11: seq=78 ttl=128 time=1.791 ms
64 bytes from 192.168.10.11: seq=79 ttl=128 time=0.287 ms
64 bytes from 192.168.10.11: seq=80 ttl=128 time=0.294 ms
64 bytes from 192.168.10.11: seq=81 ttl=128 time=0.287 ms
64 bytes from 192.168.10.11: seq=82 ttl=128 time=0.292 ms
64 bytes from 192.168.10.11: seq=83 ttl=128 time=0.286 ms
[330.975186] ————[ cut here ]————
[330.979814] WARNING: at net/sched/sch_generic.c:255 dev_watchdog+0x274/0x294()
[330.987044] NETDEV WATCHDOG: eth4 (keystone-netcp): transmit queue 0 timed out
[330.994254] Modules linked in:
[330.997323] CPU: 0 PID: 0 Comm: swapper/0 Not tainted 3.10.10 #38
[331.003431] [<c00142c0>] (unwind_backtrace+0x0/0xf8) from [<c00117ac>] (show_stack+0x10/0x14)
[331.011970] [<c00117ac>] (show_stack+0x10/0x14) from [<c0020074>] (warn_slowpath_common+0x4c/0x6c)
[331.020937] [<c0020074>] (warn_slowpath_common+0x4c/0x6c) from [<c0020128>] (warn_slowpath_fmt+0x30/0x40)
[331.030509] [<c0020128>] (warn_slowpath_fmt+0x30/0x40) from [<c03af78c>] (dev_watchdog+0x274/0x294)
[331.039563] [<c03af78c>] (dev_watchdog+0x274/0x294) from [<c002d788>] (call_timer_fn+0x30/0xf8)
[331.048269] [<c002d788>] (call_timer_fn+0x30/0xf8) from [<c002ed80>] (run_timer_softirq+0x1b0/0x260)
[331.057411] [<c002ed80>] (run_timer_softirq+0x1b0/0x260) from [<c002730c>] (__do_softirq+0xe8/0x258)
[331.066548] [<c002730c>] (__do_softirq+0xe8/0x258) from [<c0027754>] (irq_exit+0x58/0x90)
[331.074720] [<c0027754>] (irq_exit+0x58/0x90) from [<c000e978>] (handle_IRQ+0x3c/0x94)
[331.082644] [<c000e978>] (handle_IRQ+0x3c/0x94) from [<c0008520>] (gic_handle_irq+0x28/0x5c)
[331.091088] [<c0008520>] (gic_handle_irq+0x28/0x5c) from [<c04b61c0>] (__irq_svc+0x40/0x50)
[331.099435] Exception stack(0xc06f5f60 to 0xc06f5fa8)
[331.104483] 5f60: c0b7f918 00000000 0001f008 00000000 c06f4000 c07312b0 c0730afb c06fc454
[331.112664] 5f80: c0730afb c04bd350 c06f4000 c06f4000 00000000 c06f5fa8 c000eae8 c000eaec
[331.120851] 5fa0: 600f0013 ffffffff
[331.124344] [<c04b61c0>] (__irq_svc+0x40/0x50) from [<c000eaec>] (arch_cpu_idle+0x28/0x30)
[331.132621] [<c000eaec>] (arch_cpu_idle+0x28/0x30) from [<c0053fe4>] (cpu_startup_entry+0x58/0x1dc)
[331.141679] [<c0053fe4>] (cpu_startup_entry+0x58/0x1dc) from [<c06a4a70>] (start_kernel+0x300/0x358)
[331.150809] —[ end trace fddb70a273ec2eaf ]—
[331.155432] net eth4: transmit timed out
[340.975190] net eth4: transmit timed out
64 bytes from 192.168.10.11: seq=84 ttl=128 time=15355.910 ms
[350.935189] net eth4: transmit timed out
[360.935189] net eth4: transmit timed out
[370.935189] net eth4: transmit timed out— 192.168.10.11 ping statistics —
135 packets transmitted, 85 packets received, 37% packet loss
round-trip min/avg/max = 0.265/180.951/15355.910 ms
root@k2hk-evm:~# [380.935189] net eth4: transmit timed out
[390.935189] net eth4: transmit timed out
user5315971：
回复 Shine:
美女工程师：
是否万兆网卡这里有什么补丁需要加上。请指教。
Shine：
回复 user5315971:
请关注下面e2e帖子。
e2e.ti.com/…/817372
Shine：
回复 Shine:
不知道您的linux kernel是什么版本？是TI的K2K EVM板还是自己的板子？从log看，iperf crash了，同时ping也creashe了。
在Linux kernel 3.10.72上我们没有碰到过这样的问题，我们在TI K2H EVM板上测试10GbE网口性能，ping和iperf运行了好几天都没有问题。我们也有客户用多块K2H跑10GbE网口也没有crash掉，客户的板子已经量产。
user1414547：
回复 Shine:
我的干脆就ping不通 ，为什么
