Hi all,
i am on a simple busybox-based rootfs, 4.14 imx-2.0.0 kernel, and observing a strange issue:
by a simple bash script, i am performing a simple read / write test from uart2 to uart6, and just after i perform a rpmsg m4 r/w test using the endpoint driver (ept created by echo 1 > /sys/class/rpmsg/channel_1/ept_new), after repeating 4 time the test, rpmsg channel read / write test hangs blocking execution, can exit by CTRL-.Z only and by ps i see rpmsg test process in R state.
If i remove the test on uart6, test uart2 to 5 + rpmsg works fine forever in a loop.
So seems uart6 has some special connection with rpmsg.
Also, in the imx6sx rm, after "Table 65-6. UART5 External Signals" i don't see any uart6 table, and this make me think uart6 is someway special.
I am going deeper, but if any help, welcome.
Regards,
angelo
Hi Angelo
uart6 is listed in Table 4-1. Pin Assignments i.MX 6SoloX Applications Processor Reference Manual
it is the same as other uarts. For rpmsg issue one can try to reproduce issue
on nxp sabre sd board with official linux from source.codeaurora.org/external/imx/linux-imx repository
Linux 4.14.98_2.2.0 Documentation
Best regards
igor
-----------------------------------------------------------------------------------------------------------------------
Note: If this post answers your question, please click the Correct Answer button. Thank you!
-----------------------------------------------------------------------------------------------------------------------
Hi Igor,
thanks ! I solved, since rpmsg dialog was stuck after touching uart6, i had a look into the m4 firmware, and it is using UART6 as debug uart. So, for my case, even if accessible from cortex-A side should not be touched.