This thread has been locked.
If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.
Hi Jim,
End of October is a reasonable time frame to expect Academy update for SDK10.0.
Meanwhile, do you mind to summarize the AM64x specific update that you are expecting? I know you have sent the requests several times in the past month or two, but it would be nice to have the items listed here so that we don't miss anything.
from AM6442: AM64X Linux v9.2 SDK - please clarify certain gadget files in release - Processors forum - Processors - TI E2E support forums Bin said "We will update the documentation to explain which Kconfig options are required for each devices, AM64x, AM62x, AM62Ax, AM62Px..."
from AM6442: AM64X/09_01_00_41 : IPC SafeIPC Echo - looking for SafeIPC transfer example between A53 cores in Linux and R5/M4 cores in FreeRTOS - Processors forum - Processors - TI E2E support forums there are disclaimers that Nick identified that should be melded in:
"What about using safeIPC on an AM64x running Linux?
1) safeIPC is ONLY supported for IPC between MCU+ cores. So we do NOT support safeIPC between Linux and an MCU+ core
2) So far, safeIPC has only been tested between MCU+ cores while Linux is NOT running on the A53 cores.
3) It is probably possible to have safeIPC between two MCU+ cores while Linux is running on the A53 cores (though this is technically not supported since it is not tested). The biggest thing here is that the resources used for safeIPC need to be firewalled, so Linux would need to avoid accessing any of those memory regions. I suspect that it is sufficient to reserve those firewalled memory regions in the Linux devicetree, but I have not verified myself."
from AM6442: Can the M4 core MCU_GPIO0 bus IO be assignable to the A53 coer? Is the capabilityOS and/or part number suffix dependent? - Processors forum - Processors - TI E2E support forums , Nick ahhas a good precaution to note:
"The resource config settings that are the default in the MCU+ SDK (i.e., SBL initialization/boot) may not match the resource config settings that are the default in the Linux SDK (i.e., SPL boot), so that's another potential "gotcha" to keep in mind."
from AM6442: Checked the AM64X TRM - can an A53 core controlled USB interface be only a USB 3.0 host, not a USB 3.0 device at the 5Gbps rate? - Processors forum - Processors - TI E2E support forums a key limitation statement that should be emphasized in Linux SDK use of USB:
"USB3.0 in device mode is not supported, only USB2.0 in device mode is supported."
from AM6442: Is there a Linux (A53) driver for the EQEP peripheral in AM64X - Processors forum - Processors - TI E2E support forums - i mentioned to Nick that we should followup on if it is ready. Some folks may want to use a Linux timeservice with an RTC IC connected to the A53 core to timestamp an encoder reading to 64 bit time/date representation (which will be great to get around the y2038 Linux situation)
wrt AM6442: MCU+SDK v9.1, CCS v12.6 - Multicore Development - remote core firmware that can be initialized by Linux - linker.cmd keeps getting deleted - Processors forum - Processors - TI E2E support forums - is the memory configurator through enough experience & testing that it can be used without having to resort to modifying and/or post-processing linker.cmd files?
Do we have a solution for the USB Linux hung at configuration situation as per AM6442: USB device configure cause Linux system hang - Processors forum - Processors - TI E2E support forums ?
I believe your advice from AM62x on reducing boot time is valuable to document as well - from PROCESSOR-SDK-AM62X: which drivers / tools i can disable in the kernel menuconfig in order to speed up the boot - Processors forum - Processors - TI E2E support forums
for SK-AM64B users - what you disclosed on "SK-AM64B J7 port is designed for USB host only, and this is what officially supported. However, it is technically trivial to use the J7 port for USB device mode (USB2.0) in Linux.." is valuable -- from SK-AM64B: AM64X academy - clarifications for USB 2.0 on SK-AM64B; request USB 3.0 in HOST mode on SK-AM64B - Processors forum - Processors - TI E2E support forums
I'll look for some more suggestions tomorrow
later
Jim
Bin:
please, should also check on
the resource management issue (i.e. deprecated host_cfg parameters ) situation documentwise wrt (1) PROCESSOR-SDK-AM64X: resource management question continue - Processors forum - Processors - TI E2E support forums
also Tushar had a Jira ticket dealing with interrupts and a faulty sysfw_boardcfg_rules.json setting: see (1) PROCESSOR-SDK-AM64X: boardconfig, rm and rules document dismatch. - Processors forum - Processors - TI E2E support forums
plus Tushar has an appropriate Jira ticket on this exchange on multicore & IPC pertinent issues that Nick was involved with (applies not just to AM62) see (2) PROCESSOR-SDK-AM62X: Multicore development, IPC Process, Multiple Endpoints, Reserved memory purpose - Processors forum - Processors - TI E2E support forums
also, I would like to know (check Nick) if the strategy of deciding whether to use spinlock, zerocopy, and/or rpc in multicore apps has changed recently with the A53 now using Linux 6.6 -- please see: (2) AM6422: How to use SPINLOCK between R5F0_0 FreeRTOS and A53 linux APP - Processors forum - Processors - TI E2E support forums
thanks,
Jim
Hi Jim,
Thanks for the addition list.
The SCI documentation related issues should be fixed in the SCI doc. They don't currently belong to the Academy.