On Wed, Sep 06, 2023 at 04:46:55PM +0300, Andy Shevchenko wrote: > On Tue, Sep 05, 2023 at 05:27:23PM -0500, Stephen Boyd wrote: > > Quoting Mika Westerberg (2023-08-31 22:50:11) > > > On Wed, Aug 30, 2023 at 06:14:01PM -0700, Stephen Boyd wrote: > > > > It's possible for the polling loop in busy_loop() to get scheduled away > > > > for a long time. > > > > > > > > status = ipc_read_status(scu); > > > > <long time scheduled away> > > > > if (!(status & IPC_STATUS_BUSY)) > > > > > > How can the status bit change here as we are the only user and the SCU > > > access is serialized by ipclock? > > > > I don't know how the SCU works. I thought that IPC_STATUS_BUSY bit was > > cleared by the SCU when it was done processing. With that assumption, I > > tried to show that the status is read and then the process schedules > > away for a long time and has an outdated view of the busy bit. > > We probably have different versions of firmwares for the different SoC > generations. But I _think_ that you are right, the SCU firmware should > clear the bit when it's done. Yes, IIRC it does. Okay I see the (potential, although quite unlikely) problem now. Thanks!
