On Mon, Mar 10, 2025 at 5:43 PM Mathieu Poirier <mathieu.poirier@xxxxxxxxxx> wrote: > > Good day, > > On Wed, Mar 05, 2025 at 12:00:36PM +0200, Daniel Baluta wrote: > > DSP on i.MX8MP doesn't have a direct reset line so according to hardware > > design team in order to handle assert/deassert/reset functionality we > > need to use a combination of control bits from two modules. Audio block > > control module for Run/Stall control of the DSP and DAP module in order > > to do software reset. > > > > Initial design accessed RunStall bits from audio block control using the > > syscon interface but this got NACKed by community because of abusing the > > syscon interface [1]. > > > > So instead we use the reset controller API. > > I am good with this patchset. The only thing I'd like so see modified is the > above changelog. The second paragraph doesn't belong there, i.e it is not > useful to remark on a design that was not accepted. Instead please concentrate > on _why_ we are moving from the current design to using the reset controller > API. > > This can go in the 6.15 merge window if you send me a V5 fast enough. Thanks for review. Reworded the commit message and send V5. Thanks, Daniel.
