September 20, 2019, 04:49:52 pm

News:

Have you visited the Allwinner Chipset wiki? - http://linux-sunxi.org/


cubietruck NIC corrupting data?

Started by foef, May 25, 2014, 08:20:14 am

Previous topic - Next topic

Sloop

actually I have 960MHz configured in cpufrequtils. Is it safe/stable to use 1000 MHz, or will that value already lead the ct to crash?

-sent from my Raspberry Smartphone-


jiv

960mhz vs 1000mhz is almost same, i dont think you will notice difference.

synthomat

August 03, 2014, 10:01:29 am #47 Last Edit: August 03, 2014, 10:03:54 am by synthomat
Hey folks,

I'm experiencing the same problem running CTDebian2.4 with 3.4.98-sun7i+. It looks like downloads from cubie breaks earlier and more often than uploads to it.

I already tried different switches/routers, cables, PSUs, other kernels and distributions, underclocking etc. - but nothing make any difference :(

Sloop

did you read this thread??!

-sent from my Raspberry Smartphone-


synthomat

August 04, 2014, 07:09:21 am #49 Last Edit: August 04, 2014, 07:23:56 am by synthomat
Of course I did multiple times. Is the problem finally solved? Did I overlooked something?

Sloop

so, what actions did you try to fix that problem?

-sent from my Raspberry Smartphone-


synthomat

Quote from: Sloop on August 10, 2014, 04:11:09 amso, what actions did you try to fix that problem?


He?  ::)
Quote from: synthomat on August 03, 2014, 10:01:29 amI already tried different switches/routers, cables, PSUs, other kernels and distributions, underclocking etc. - but nothing make any difference :(


I also tried different .fex-files.

My current workaround solution is to use a 100mbit USB NIC. It works like charm but is not as fast as the Gbit NIC.

bug

September 13, 2014, 11:26:15 am #52 Last Edit: September 13, 2014, 11:32:52 am by bug
Quote from: synthomat on August 11, 2014, 09:23:13 am
My current workaround solution is to use a 100mbit USB NIC. It works like charm but is not as fast as the Gbit NIC.


I had the exact same problem. Tried different images, u-boot, kernels... One image did work though - lubuntu nand.

Observing it's script.bin, I did read that max_freq was set to 1008000000 in this distro, while it's usually set to 912000000.


< max_freq = 1008000000
---
> max_freq = 912000000


I did the change (bin2fex script.bin > script.fex / edit script.fex, change the value, save / fex2bin script.fex > script.bin) and right now the network is working like a charm.

Worth a try!

I spend several days trying all solutions until I found this one.

Hope it helps!

Sloop

I am wondering if these people ever have red this thread completely? everything is mentioned and detailled explained (including links). So what?

-sent from my Raspberry Smartphone-


bug

Quote from: Sloop on September 14, 2014, 04:47:34 pm
I am wondering if these people ever have red this thread completely? everything is mentioned and detailled explained (including links). So what?

-sent from my Raspberry Smartphone-


Well, maybe different people can have a different problem and the cpufreq trick didn't work for 'em?

Cheers

ssvb

Quote from: bug on September 19, 2014, 06:43:50 pm
Well, maybe different people can have a different problem and the cpufreq trick didn't work for 'em?

Sloop means that the other people have solved this problem ages ago and you are now re-discovering exactly the same solution :)

bug

September 20, 2014, 08:59:00 am #56 Last Edit: September 21, 2014, 08:50:28 am by bug
Quote from: ssvb on September 19, 2014, 06:54:52 pm
Sloop means that the other people have solved this problem ages ago and you are now re-discovering exactly the same solution :)


I did guess so too, but I didn't found this solution on this thread (cpufreq & voltage only).

Did just want to share in case it helped as this seems to be a recuring problem many people have. Good luck everyone!

(Edit: doesn't work for me in all cases though - even a 'perfect' (as defined in this thread) .bin script + correct cpufrequtils can let me have network issues with some distributions)