Messages in this thread | | | From | Raul Rangel <> | Date | Tue, 14 Jul 2020 10:16:07 -0600 | Subject | Re: UART/TTY console deadlock |
| |
Hey everyone, So it looks like this was a red herring. We were getting spurious interrupts which I suspect caused the console code to get called in some critical section. Once we fixed the spurious interrupts we no longer see the issue. Sorry for the noise.
Thanks, Raul
On Thu, Jul 9, 2020 at 7:22 AM Sergey Senozhatsky <sergey.senozhatsky@gmail.com> wrote: > > On (20/07/08 11:52), Petr Mladek wrote: > > > > I still believe that this is a false positive. > > I've no problems with that. The thing that I'm trying to improve here is the > fact that from lockdep's point of view there are no false positives. lockdep > reports the problem and then waves good bye. It disables itself. There might > be some real suspend locking problems on some particular setups, but it's > hard to tell, because console startup blows up the lockdep first. > > -ss
| |