Temperature dependent on DDR memory calibration ? Reason for kernel panic when the imx6q system get heat-up ?

[peteramond]

Hi All,

I have done DDR memory calibration for my imx6q custom hardware design in ambient temperature and got proper and reliable read,write,dqs gating values for DDR memory. And I updated u-boot configs according to those values. 

But when the system get heat up; as an example when the die temperature of the imx6 exceed 55C the system shows kernel panic. It is difficult to identify the reason using the errors pop-up in kernel panic using program counter. (May be this is not a reason in processor because imx6 can work up to 85C without de-clocking) 

1. Do we need to do DDR memory calibration in different temperatures before do the updates in u-boot ? Any dependency on memory calibration on temperature ?
2. How can we get temperature profile using few thermocouples which were placed on different areas of DDR memory chips, imx6 processor and etc by using imx6q libraries ?

Regards,

Peter.

[igorpadykov]

Hi Peter

1. if heating up the part causes more failures, then the drive strength is too low and the
signals may not rise/fall fast enough. There is no need to perform DDR memory calibration in
different temperatures, temperature influence is described in sect.7.2 ZQ Calibration Usage
AN4467 i.MX6 Series DDR Calibration
http://cache.nxp.com/docs/en/application-note/AN4467.pdf
2. please refer to AN4579 i.MX 6 Series Thermal Management Guidelines
http://cache.nxp.com/docs/en/application-note/AN4579.pdf

Best regards
igor
-----------------------------------------------------------------------------------------------------------------------
Note: If this post answers your question, please click the Correct Answer button. Thank you!
-----------------------------------------------------------------------------------------------------------------------

[peteramond]

Hi igor,

@igorpadykov

Thank you for your explanation. 

Today I encountered a new problem with my iMX6Q custom hardware design with Linux version 4.1.15-1.0.0_ga+yocto+g76c24a0

I have an issue with GPU during I launch cameras (System stuck).

Camera launch >> gst-launch-1.0 v4l2src device=/dev/video0 ! autovideosink

Error message >> [galcore]: GPU[0] hang, automatic recovery.

It shows the test patterns without any conflict. When I run

gst-launch-1.0 videotestsrc pattern=0 ! autovideosink

There are no any conflict.

What can be the cause of this issue ?

Regards,

Peter.