thesofproject / sof-docs

Documentation for SOF
Other
16 stars 72 forks source link

[FEATURE] rimage IPC4 manifest documentation #455

Open cujomalainey opened 1 year ago

cujomalainey commented 1 year ago

detailed documents explaining how all these fields correlate and how they are used, what consequences they have on the runtime environments

lgirdwood commented 1 year ago

@cujomalainey I've added @mmaka1 and @mwasko the real experts on IPC4 and @jsarha expert on topology2· Can you add more context on any other parts where we are missing data and we can then priorize and update.

cujomalainey commented 1 year ago

@lgirdwood basically everything in https://github.com/thesofproject/sof-docs/pull/451 was unknown and/or confusing to me

mwasko commented 1 year ago

@lgirdwood basically everything in #451 was unknown and/or confusing to me

I did some digging and we already have more meaningful description of extended manifest fields in converged-modules Readme @RanderWang can you update #451 using data from that Readme?

I am aware that there might still be some gaps and questions but lets first fill in the data we have and as a next step I will work on missing information.

RanderWang commented 1 year ago

@mwasko thanks, update the Readme to PR 451

lgirdwood commented 1 year ago

@jsarha @ranj063 are we done now for topology docs ? If so, can we check the topology box above. @mwasko @mmaka1 are we complete on manifest data ? If so lets check teh box above. Thanks all.

jsarha commented 1 year ago

@jsarha @ranj063 are we done now for topology docs ? If so, can we check the topology box above.

@lgirdwood , The situation with the topology2 docs has not changed. The topology2 language itself is very well documented here: https://thesofproject.github.io/latest/developer_guides/topology2/topology2.html

However, the structure of the topology2 sources and how they produce the topology binaries is only partly documented by sporadic comments in the sources. My plan was to implement a doxygen filter for topology2 sources and edit the comments to a structured document, that can be extended as needed and can evolve with the sources. However I dropped that work at the end of may, when it was obvious that I can not get it to any conclusion before the end of Q2, and there was still some work to do for topology2 cleanup. I am respinning that work now as soon as can.