[PATCH 09/16] tty: serial: 8250_dma: Add a TX trigger workaround for AM33xx
From: bigeasy@linutronix.de (Sebastian Andrzej Siewior)
Date: 2014-09-24 07:56:59
Also in:
linux-omap, linux-serial, lkml
Possibly related (same subject, not in this thread)
- 2014-09-29 · Re: [PATCH 09/16] tty: serial: 8250_dma: Add a TX trigger workaround for AM33xx · Sebastian Andrzej Siewior <bigeasy@linutronix.de>
- 2014-09-25 · Re: [PATCH 09/16] tty: serial: 8250_dma: Add a TX trigger workaround for AM33xx · Sebastian Andrzej Siewior <bigeasy@linutronix.de>
- 2014-09-22 · Re: [PATCH 09/16] tty: serial: 8250_dma: Add a TX trigger workaround for AM33xx · Heikki Krogerus <heikki.krogerus@linux.intel.com>
- 2014-09-19 · Re: [PATCH 09/16] tty: serial: 8250_dma: Add a TX trigger workaround for AM33xx · Peter Hurley <hidden>
- 2014-09-19 · Re: [PATCH 09/16] tty: serial: 8250_dma: Add a TX trigger workaround for AM33xx · Sebastian Andrzej Siewior <bigeasy@linutronix.de>
* Frans Klaver | 2014-09-22 11:28:54 [+0200]:
Wow, thanks for your work here. This does indeed sound hard to trap. I guess then we'd still have to answer the question why the yocto build calls set_termios() so often, but that's not on you then. Did you notice it even changing settings? We might want to fix this even if the kernel should be able to cope.
Yeah. I don't care if this is yocto's fault or not. If it is possible to stop the DMA transfer from userland with whatever method then it needs to be fixed in kernel so that this does happen.
Thanks again, Frans
Sebastian