On Wed, 18 Oct 2023 08:23:45 +0300 Tony Lindgren <tony@xxxxxxxxxxx> wrote: > * Andreas Kemnade <andreas@xxxxxxxxxxxx> [231015 21:48]: > > On Sat, 7 Oct 2023 09:25:18 +0300 > > Tony Lindgren <tony@xxxxxxxxxxx> wrote: > > > If so then the following might be a fix, not familiar with runtime PM > > > done by ASoC though and not sure if some kind of locking would be > > > needed here. > > > > > just checked: that one fixes the regression. runtime suspends again. > > OK good to hear. So is there some fixes tag for this one or where did this > start happening? I'm not quite following how the dropping of platform data > could have affected this, maybe it just hid the problem because of > returning early? > yes I think so. From a perspective of OMAP[45] mith McBSP in master mode, applying "clk: ti: Fix missing omap4 mcbsp functional clock and aliases" on top of "ASoC: ti: omap-mcbsp: Ignore errors for getting fck_src" is a regression. For everyone else not. So "clk: ti: Fix missing omap4 mcbsp functional clock and aliases" did uncover a hidden problem, but of course it is the right step forward. Regards Andreas
