On Fri, 31 Oct 2014 07:03:19 -0700 (PDT)
Dennis Kerrisk <[email protected]> wrote:
> 2 rounds of testing and all is goof. Any other ideas?
Do you mean 'good'? Just two rounds of lima-memtester is not long
enough. But on the other hand, A10-Lime was never reported to have
DRAM reliability problems so far.
The results of the voltage/frequency test are actually more interesting:
http://linux-sunxi.org/Hardware_Reliability_Tests#Reliability_of_cpufreq_voltage.2Ffrequency_settings
This one was really failing at least on two A10-Lime boards earlier,
and you can find more details in the mailing list archives:
https://www.mail-archive.com/[email protected]/msg04343.html
The kernel needs to have 'userspace' cpufreq governor enabled, which
is a minor inconvenience because the sunxi-3.4 kernel defconfigs don't
seem to have it. But as a quick test, you can alternatively run the
following commands in the console:
cd /tmp && wget http://linux-sunxi.org/images/8/83/A10-LIME.jpg
wget http://people.freedesktop.org/~siamashka/files/20140504/djpeg-static
chmod +x djpeg-static
while true ; do ./djpeg-static A10-LIME.jpg | md5sum ; done
If identical md5 hashes are printed on each line, then everything
is fine. If the md5 hashes are different, then some data corruption
is happening. Let the test run until it prints 100-1000 lines of
output.
Please also report the hardware revision of your A10-Lime (it should
be printed on the board near the A10 chip) and a bit more information
about your u-boot/kernel.
Thanks.
--
Best regards,
Siarhei Siamashka
--
You received this message because you are subscribed to the Google Groups
"linux-sunxi" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
For more options, visit https://groups.google.com/d/optout.
