All of lore.kernel.org
 help / color / mirror / Atom feed
* Regression since 4a96895f74c9633b51427fd080ab70fa62b65bc4
@ 2019-09-28  5:59 Nuno Gonçalves
  2019-09-29  6:46 ` Nuno Gonçalves
  0 siblings, 1 reply; 10+ messages in thread
From: Nuno Gonçalves @ 2019-09-28  5:59 UTC (permalink / raw)
  To: linux-serial, yegorslists
  Cc: sr, andriy.shevchenko, mika.westerberg, giulio.benetti, gregkh

On 32 bit arm platforms (am335x and sunxi) I get unexpected bytes on
serial ports after this commit.

I use CTS (so: for input flow control).

I've observed two different issues, one where I get just a few missed
or added bytes about once a minute, and another where I get totally
unexpected chars until I reopen the port.

I will have a hard time to analyse and give a better description of
how exactly the port is failing, so maybe you have a idea of what is
the issue with this patch.

I've reverted this on top of 5.3.1 and consistently got the issue fixed.

Thanks,
Nuno

commit 4a96895f74c9633b51427fd080ab70fa62b65bc4
Author: Yegor Yefremov <yegorslists@googlemail.com>
Date:   Thu Jun 20 08:24:20 2019 +0200

    tty/serial/8250: use mctrl_gpio helpers

    This patch permits the usage for GPIOs to control
    the CTS/RTS/DTR/DSR/DCD/RI signals.

    Changed by Stefan:
    Only call mctrl_gpio_init(), if the device has no ACPI companion device
    to not break existing ACPI based systems. Also only use the mctrl_gpio_
    functions when "gpios" is available.

    Use MSR / MCR <-> TIOCM wrapper functions.

    Signed-off-by: Yegor Yefremov <yegorslists@googlemail.com>
    Signed-off-by: Stefan Roese <sr@denx.de>
    Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
    Reviewed-by: Mika Westerberg <mika.westerberg@linux.intel.com>
    Tested-by: Yegor Yefremov <yegorslists@googlemail.com>
    Cc: Mika Westerberg <mika.westerberg@linux.intel.com>
    Cc: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
    Cc: Giulio Benetti <giulio.benetti@micronovasrl.com>
    Cc: Yegor Yefremov <yegorslists@googlemail.com>
    Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
    Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

^ permalink raw reply	[flat|nested] 10+ messages in thread
* Re: Regression since 4a96895f74c9633b51427fd080ab70fa62b65bc4
  2019-09-28  5:59 Regression since 4a96895f74c9633b51427fd080ab70fa62b65bc4 Nuno Gonçalves
@ 2019-09-29  6:46 ` Nuno Gonçalves
  2019-10-01 20:02   ` Yegor Yefremov
  0 siblings, 1 reply; 10+ messages in thread
From: Nuno Gonçalves @ 2019-09-29  6:46 UTC (permalink / raw)
  To: linux-serial, yegorslists
  Cc: sr, andriy.shevchenko, mika.westerberg, giulio.benetti, gregkh

I have since found that I am affected by two apparently different
regressions since 5.3.

4a96895f74c9633b51427fd080ab70fa62b65bc4 is only confirmed to affect
me on a omap8250 where the issue is that there are a few char errors
every few seconds. I can't confirm if chars are added, removed or
changed, for now I only can confirm my state machine looses tracking.

I will make a separate report from the problem identified on sunxi.

On Sat, Sep 28, 2019 at 7:59 AM Nuno Gonçalves <nunojpg@gmail.com> wrote:
>
> On 32 bit arm platforms (am335x and sunxi) I get unexpected bytes on
> serial ports after this commit.
>
> I use CTS (so: for input flow control).
>
> I've observed two different issues, one where I get just a few missed
> or added bytes about once a minute, and another where I get totally
> unexpected chars until I reopen the port.
>
> I will have a hard time to analyse and give a better description of
> how exactly the port is failing, so maybe you have a idea of what is
> the issue with this patch.
>
> I've reverted this on top of 5.3.1 and consistently got the issue fixed.
>
> Thanks,
> Nuno
>
> commit 4a96895f74c9633b51427fd080ab70fa62b65bc4
> Author: Yegor Yefremov <yegorslists@googlemail.com>
> Date:   Thu Jun 20 08:24:20 2019 +0200
>
>     tty/serial/8250: use mctrl_gpio helpers
>
>     This patch permits the usage for GPIOs to control
>     the CTS/RTS/DTR/DSR/DCD/RI signals.
>
>     Changed by Stefan:
>     Only call mctrl_gpio_init(), if the device has no ACPI companion device
>     to not break existing ACPI based systems. Also only use the mctrl_gpio_
>     functions when "gpios" is available.
>
>     Use MSR / MCR <-> TIOCM wrapper functions.
>
>     Signed-off-by: Yegor Yefremov <yegorslists@googlemail.com>
>     Signed-off-by: Stefan Roese <sr@denx.de>
>     Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
>     Reviewed-by: Mika Westerberg <mika.westerberg@linux.intel.com>
>     Tested-by: Yegor Yefremov <yegorslists@googlemail.com>
>     Cc: Mika Westerberg <mika.westerberg@linux.intel.com>
>     Cc: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
>     Cc: Giulio Benetti <giulio.benetti@micronovasrl.com>
>     Cc: Yegor Yefremov <yegorslists@googlemail.com>
>     Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
>     Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

^ permalink raw reply	[flat|nested] 10+ messages in thread
* Re: Regression since 4a96895f74c9633b51427fd080ab70fa62b65bc4
  2019-09-29  6:46 ` Nuno Gonçalves
@ 2019-10-01 20:02   ` Yegor Yefremov
  2019-10-01 21:03     ` Nuno Gonçalves
  0 siblings, 1 reply; 10+ messages in thread
From: Yegor Yefremov @ 2019-10-01 20:02 UTC (permalink / raw)
  To: Nuno Gonçalves
  Cc: linux-serial, Stefan Roese, Andy Shevchenko, Mika Westerberg,
	Greg KH, giulio.benetti

Hi Nuno,

On Sun, Sep 29, 2019 at 8:46 AM Nuno Gonçalves <nunojpg@gmail.com> wrote:
>
> I have since found that I am affected by two apparently different
> regressions since 5.3.
>
> 4a96895f74c9633b51427fd080ab70fa62b65bc4 is only confirmed to affect
> me on a omap8250 where the issue is that there are a few char errors
> every few seconds. I can't confirm if chars are added, removed or
> changed, for now I only can confirm my state machine looses tracking.

Can you see kernel warnings about an overrun on OMAP UART?

> I will make a separate report from the problem identified on sunxi.
>
> On Sat, Sep 28, 2019 at 7:59 AM Nuno Gonçalves <nunojpg@gmail.com> wrote:
> >
> > On 32 bit arm platforms (am335x and sunxi) I get unexpected bytes on
> > serial ports after this commit.
> >
> > I use CTS (so: for input flow control).
> >
> > I've observed two different issues, one where I get just a few missed
> > or added bytes about once a minute, and another where I get totally
> > unexpected chars until I reopen the port.
> >
> > I will have a hard time to analyse and give a better description of
> > how exactly the port is failing, so maybe you have a idea of what is
> > the issue with this patch.
> >
> > I've reverted this on top of 5.3.1 and consistently got the issue fixed.
> >
> > Thanks,
> > Nuno
> >
> > commit 4a96895f74c9633b51427fd080ab70fa62b65bc4
> > Author: Yegor Yefremov <yegorslists@googlemail.com>
> > Date:   Thu Jun 20 08:24:20 2019 +0200
> >
> >     tty/serial/8250: use mctrl_gpio helpers
> >
> >     This patch permits the usage for GPIOs to control
> >     the CTS/RTS/DTR/DSR/DCD/RI signals.
> >
> >     Changed by Stefan:
> >     Only call mctrl_gpio_init(), if the device has no ACPI companion device
> >     to not break existing ACPI based systems. Also only use the mctrl_gpio_
> >     functions when "gpios" is available.
> >
> >     Use MSR / MCR <-> TIOCM wrapper functions.
> >
> >     Signed-off-by: Yegor Yefremov <yegorslists@googlemail.com>
> >     Signed-off-by: Stefan Roese <sr@denx.de>
> >     Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
> >     Reviewed-by: Mika Westerberg <mika.westerberg@linux.intel.com>
> >     Tested-by: Yegor Yefremov <yegorslists@googlemail.com>
> >     Cc: Mika Westerberg <mika.westerberg@linux.intel.com>
> >     Cc: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
> >     Cc: Giulio Benetti <giulio.benetti@micronovasrl.com>
> >     Cc: Yegor Yefremov <yegorslists@googlemail.com>
> >     Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
> >     Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

^ permalink raw reply	[flat|nested] 10+ messages in thread
* Re: Regression since 4a96895f74c9633b51427fd080ab70fa62b65bc4
  2019-10-01 20:02   ` Yegor Yefremov
@ 2019-10-01 21:03     ` Nuno Gonçalves
  2019-10-17  9:19       ` Nuno Gonçalves
  0 siblings, 1 reply; 10+ messages in thread
From: Nuno Gonçalves @ 2019-10-01 21:03 UTC (permalink / raw)
  To: Yegor Yefremov
  Cc: linux-serial, Stefan Roese, Andy Shevchenko, Mika Westerberg,
	Greg KH, giulio.benetti

Yes, I remember seeing those warnings.

On Tue, Oct 1, 2019 at 10:02 PM Yegor Yefremov
<yegorslists@googlemail.com> wrote:
>
> Hi Nuno,
>
> On Sun, Sep 29, 2019 at 8:46 AM Nuno Gonçalves <nunojpg@gmail.com> wrote:
> >
> > I have since found that I am affected by two apparently different
> > regressions since 5.3.
> >
> > 4a96895f74c9633b51427fd080ab70fa62b65bc4 is only confirmed to affect
> > me on a omap8250 where the issue is that there are a few char errors
> > every few seconds. I can't confirm if chars are added, removed or
> > changed, for now I only can confirm my state machine looses tracking.
>
> Can you see kernel warnings about an overrun on OMAP UART?
>
> > I will make a separate report from the problem identified on sunxi.
> >
> > On Sat, Sep 28, 2019 at 7:59 AM Nuno Gonçalves <nunojpg@gmail.com> wrote:
> > >
> > > On 32 bit arm platforms (am335x and sunxi) I get unexpected bytes on
> > > serial ports after this commit.
> > >
> > > I use CTS (so: for input flow control).
> > >
> > > I've observed two different issues, one where I get just a few missed
> > > or added bytes about once a minute, and another where I get totally
> > > unexpected chars until I reopen the port.
> > >
> > > I will have a hard time to analyse and give a better description of
> > > how exactly the port is failing, so maybe you have a idea of what is
> > > the issue with this patch.
> > >
> > > I've reverted this on top of 5.3.1 and consistently got the issue fixed.
> > >
> > > Thanks,
> > > Nuno
> > >
> > > commit 4a96895f74c9633b51427fd080ab70fa62b65bc4
> > > Author: Yegor Yefremov <yegorslists@googlemail.com>
> > > Date:   Thu Jun 20 08:24:20 2019 +0200
> > >
> > >     tty/serial/8250: use mctrl_gpio helpers
> > >
> > >     This patch permits the usage for GPIOs to control
> > >     the CTS/RTS/DTR/DSR/DCD/RI signals.
> > >
> > >     Changed by Stefan:
> > >     Only call mctrl_gpio_init(), if the device has no ACPI companion device
> > >     to not break existing ACPI based systems. Also only use the mctrl_gpio_
> > >     functions when "gpios" is available.
> > >
> > >     Use MSR / MCR <-> TIOCM wrapper functions.
> > >
> > >     Signed-off-by: Yegor Yefremov <yegorslists@googlemail.com>
> > >     Signed-off-by: Stefan Roese <sr@denx.de>
> > >     Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
> > >     Reviewed-by: Mika Westerberg <mika.westerberg@linux.intel.com>
> > >     Tested-by: Yegor Yefremov <yegorslists@googlemail.com>
> > >     Cc: Mika Westerberg <mika.westerberg@linux.intel.com>
> > >     Cc: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
> > >     Cc: Giulio Benetti <giulio.benetti@micronovasrl.com>
> > >     Cc: Yegor Yefremov <yegorslists@googlemail.com>
> > >     Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
> > >     Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

^ permalink raw reply	[flat|nested] 10+ messages in thread
* Re: Regression since 4a96895f74c9633b51427fd080ab70fa62b65bc4
  2019-10-01 21:03     ` Nuno Gonçalves
@ 2019-10-17  9:19       ` Nuno Gonçalves
  2019-10-17 13:34         ` Yegor Yefremov
  0 siblings, 1 reply; 10+ messages in thread
From: Nuno Gonçalves @ 2019-10-17  9:19 UTC (permalink / raw)
  To: Yegor Yefremov
  Cc: linux-serial, Stefan Roese, Andy Shevchenko, Mika Westerberg,
	Greg KH, giulio.benetti

Hi Yegor,

I'm not sure if by your previous message you think I have some kind of
hardware or DT issue or if this is being handled as a regression.

Thanks,
Nuno

On Tue, Oct 1, 2019 at 11:03 PM Nuno Gonçalves <nunojpg@gmail.com> wrote:
>
> Yes, I remember seeing those warnings.
>
> On Tue, Oct 1, 2019 at 10:02 PM Yegor Yefremov
> <yegorslists@googlemail.com> wrote:
> >
> > Hi Nuno,
> >
> > On Sun, Sep 29, 2019 at 8:46 AM Nuno Gonçalves <nunojpg@gmail.com> wrote:
> > >
> > > I have since found that I am affected by two apparently different
> > > regressions since 5.3.
> > >
> > > 4a96895f74c9633b51427fd080ab70fa62b65bc4 is only confirmed to affect
> > > me on a omap8250 where the issue is that there are a few char errors
> > > every few seconds. I can't confirm if chars are added, removed or
> > > changed, for now I only can confirm my state machine looses tracking.
> >
> > Can you see kernel warnings about an overrun on OMAP UART?
> >
> > > I will make a separate report from the problem identified on sunxi.
> > >
> > > On Sat, Sep 28, 2019 at 7:59 AM Nuno Gonçalves <nunojpg@gmail.com> wrote:
> > > >
> > > > On 32 bit arm platforms (am335x and sunxi) I get unexpected bytes on
> > > > serial ports after this commit.
> > > >
> > > > I use CTS (so: for input flow control).
> > > >
> > > > I've observed two different issues, one where I get just a few missed
> > > > or added bytes about once a minute, and another where I get totally
> > > > unexpected chars until I reopen the port.
> > > >
> > > > I will have a hard time to analyse and give a better description of
> > > > how exactly the port is failing, so maybe you have a idea of what is
> > > > the issue with this patch.
> > > >
> > > > I've reverted this on top of 5.3.1 and consistently got the issue fixed.
> > > >
> > > > Thanks,
> > > > Nuno
> > > >
> > > > commit 4a96895f74c9633b51427fd080ab70fa62b65bc4
> > > > Author: Yegor Yefremov <yegorslists@googlemail.com>
> > > > Date:   Thu Jun 20 08:24:20 2019 +0200
> > > >
> > > >     tty/serial/8250: use mctrl_gpio helpers
> > > >
> > > >     This patch permits the usage for GPIOs to control
> > > >     the CTS/RTS/DTR/DSR/DCD/RI signals.
> > > >
> > > >     Changed by Stefan:
> > > >     Only call mctrl_gpio_init(), if the device has no ACPI companion device
> > > >     to not break existing ACPI based systems. Also only use the mctrl_gpio_
> > > >     functions when "gpios" is available.
> > > >
> > > >     Use MSR / MCR <-> TIOCM wrapper functions.
> > > >
> > > >     Signed-off-by: Yegor Yefremov <yegorslists@googlemail.com>
> > > >     Signed-off-by: Stefan Roese <sr@denx.de>
> > > >     Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
> > > >     Reviewed-by: Mika Westerberg <mika.westerberg@linux.intel.com>
> > > >     Tested-by: Yegor Yefremov <yegorslists@googlemail.com>
> > > >     Cc: Mika Westerberg <mika.westerberg@linux.intel.com>
> > > >     Cc: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
> > > >     Cc: Giulio Benetti <giulio.benetti@micronovasrl.com>
> > > >     Cc: Yegor Yefremov <yegorslists@googlemail.com>
> > > >     Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
> > > >     Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

^ permalink raw reply	[flat|nested] 10+ messages in thread
* Re: Regression since 4a96895f74c9633b51427fd080ab70fa62b65bc4
  2019-10-17  9:19       ` Nuno Gonçalves
@ 2019-10-17 13:34         ` Yegor Yefremov
  2019-10-28 11:09           ` Yegor Yefremov
  0 siblings, 1 reply; 10+ messages in thread
From: Yegor Yefremov @ 2019-10-17 13:34 UTC (permalink / raw)
  To: Nuno Gonçalves
  Cc: linux-serial, Stefan Roese, Andy Shevchenko, Mika Westerberg,
	Greg KH, Giulio Benetti

Hi Nuno,

On Thu, Oct 17, 2019 at 11:19 AM Nuno Gonçalves <nunojpg@gmail.com> wrote:
>
> Hi Yegor,
>
> I'm not sure if by your previous message you think I have some kind of
> hardware or DT issue or if this is being handled as a regression.

There were some upstreamed fixes recently. See commits [1], [2]. Also,
make sure DMA for OMAP-8250 driver is disabled. Let me know if RTS/CTS
is working as expected on am335x. See this discussion [3].

[1] https://marc.info/?l=linux-kernel&m=157037963426476&w=2
[2] https://marc.info/?l=linux-kernel&m=157037964726483&w=2
[3] https://marc.info/?l=linux-omap&m=157018333915769&w=2

Yegor

> Thanks,
> Nuno
>
> On Tue, Oct 1, 2019 at 11:03 PM Nuno Gonçalves <nunojpg@gmail.com> wrote:
> >
> > Yes, I remember seeing those warnings.
> >
> > On Tue, Oct 1, 2019 at 10:02 PM Yegor Yefremov
> > <yegorslists@googlemail.com> wrote:
> > >
> > > Hi Nuno,
> > >
> > > On Sun, Sep 29, 2019 at 8:46 AM Nuno Gonçalves <nunojpg@gmail.com> wrote:
> > > >
> > > > I have since found that I am affected by two apparently different
> > > > regressions since 5.3.
> > > >
> > > > 4a96895f74c9633b51427fd080ab70fa62b65bc4 is only confirmed to affect
> > > > me on a omap8250 where the issue is that there are a few char errors
> > > > every few seconds. I can't confirm if chars are added, removed or
> > > > changed, for now I only can confirm my state machine looses tracking.
> > >
> > > Can you see kernel warnings about an overrun on OMAP UART?
> > >
> > > > I will make a separate report from the problem identified on sunxi.
> > > >
> > > > On Sat, Sep 28, 2019 at 7:59 AM Nuno Gonçalves <nunojpg@gmail.com> wrote:
> > > > >
> > > > > On 32 bit arm platforms (am335x and sunxi) I get unexpected bytes on
> > > > > serial ports after this commit.
> > > > >
> > > > > I use CTS (so: for input flow control).
> > > > >
> > > > > I've observed two different issues, one where I get just a few missed
> > > > > or added bytes about once a minute, and another where I get totally
> > > > > unexpected chars until I reopen the port.
> > > > >
> > > > > I will have a hard time to analyse and give a better description of
> > > > > how exactly the port is failing, so maybe you have a idea of what is
> > > > > the issue with this patch.
> > > > >
> > > > > I've reverted this on top of 5.3.1 and consistently got the issue fixed.
> > > > >
> > > > > Thanks,
> > > > > Nuno
> > > > >
> > > > > commit 4a96895f74c9633b51427fd080ab70fa62b65bc4
> > > > > Author: Yegor Yefremov <yegorslists@googlemail.com>
> > > > > Date:   Thu Jun 20 08:24:20 2019 +0200
> > > > >
> > > > >     tty/serial/8250: use mctrl_gpio helpers
> > > > >
> > > > >     This patch permits the usage for GPIOs to control
> > > > >     the CTS/RTS/DTR/DSR/DCD/RI signals.
> > > > >
> > > > >     Changed by Stefan:
> > > > >     Only call mctrl_gpio_init(), if the device has no ACPI companion device
> > > > >     to not break existing ACPI based systems. Also only use the mctrl_gpio_
> > > > >     functions when "gpios" is available.
> > > > >
> > > > >     Use MSR / MCR <-> TIOCM wrapper functions.
> > > > >
> > > > >     Signed-off-by: Yegor Yefremov <yegorslists@googlemail.com>
> > > > >     Signed-off-by: Stefan Roese <sr@denx.de>
> > > > >     Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
> > > > >     Reviewed-by: Mika Westerberg <mika.westerberg@linux.intel.com>
> > > > >     Tested-by: Yegor Yefremov <yegorslists@googlemail.com>
> > > > >     Cc: Mika Westerberg <mika.westerberg@linux.intel.com>
> > > > >     Cc: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
> > > > >     Cc: Giulio Benetti <giulio.benetti@micronovasrl.com>
> > > > >     Cc: Yegor Yefremov <yegorslists@googlemail.com>
> > > > >     Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
> > > > >     Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

^ permalink raw reply	[flat|nested] 10+ messages in thread
* Re: Regression since 4a96895f74c9633b51427fd080ab70fa62b65bc4
  2019-10-17 13:34         ` Yegor Yefremov
@ 2019-10-28 11:09           ` Yegor Yefremov
  2019-10-28 11:11             ` Nuno Gonçalves
  0 siblings, 1 reply; 10+ messages in thread
From: Yegor Yefremov @ 2019-10-28 11:09 UTC (permalink / raw)
  To: Nuno Gonçalves
  Cc: linux-serial, Stefan Roese, Andy Shevchenko, Mika Westerberg,
	Greg KH, Giulio Benetti

Hi Nuno,

On Thu, Oct 17, 2019 at 3:34 PM Yegor Yefremov
<yegorslists@googlemail.com> wrote:
>
> Hi Nuno,
>
> On Thu, Oct 17, 2019 at 11:19 AM Nuno Gonçalves <nunojpg@gmail.com> wrote:
> >
> > Hi Yegor,
> >
> > I'm not sure if by your previous message you think I have some kind of
> > hardware or DT issue or if this is being handled as a regression.
>
> There were some upstreamed fixes recently. See commits [1], [2]. Also,
> make sure DMA for OMAP-8250 driver is disabled. Let me know if RTS/CTS
> is working as expected on am335x. See this discussion [3].
>
> [1] https://marc.info/?l=linux-kernel&m=157037963426476&w=2
> [2] https://marc.info/?l=linux-kernel&m=157037964726483&w=2
> [3] https://marc.info/?l=linux-omap&m=157018333915769&w=2

could you already try at least 5.4-rc4 and test whether the hardware
flow control is fixed for am335x and your application?

Regards,
Yegor

> > Thanks,
> > Nuno
> >
> > On Tue, Oct 1, 2019 at 11:03 PM Nuno Gonçalves <nunojpg@gmail.com> wrote:
> > >
> > > Yes, I remember seeing those warnings.
> > >
> > > On Tue, Oct 1, 2019 at 10:02 PM Yegor Yefremov
> > > <yegorslists@googlemail.com> wrote:
> > > >
> > > > Hi Nuno,
> > > >
> > > > On Sun, Sep 29, 2019 at 8:46 AM Nuno Gonçalves <nunojpg@gmail.com> wrote:
> > > > >
> > > > > I have since found that I am affected by two apparently different
> > > > > regressions since 5.3.
> > > > >
> > > > > 4a96895f74c9633b51427fd080ab70fa62b65bc4 is only confirmed to affect
> > > > > me on a omap8250 where the issue is that there are a few char errors
> > > > > every few seconds. I can't confirm if chars are added, removed or
> > > > > changed, for now I only can confirm my state machine looses tracking.
> > > >
> > > > Can you see kernel warnings about an overrun on OMAP UART?
> > > >
> > > > > I will make a separate report from the problem identified on sunxi.
> > > > >
> > > > > On Sat, Sep 28, 2019 at 7:59 AM Nuno Gonçalves <nunojpg@gmail.com> wrote:
> > > > > >
> > > > > > On 32 bit arm platforms (am335x and sunxi) I get unexpected bytes on
> > > > > > serial ports after this commit.
> > > > > >
> > > > > > I use CTS (so: for input flow control).
> > > > > >
> > > > > > I've observed two different issues, one where I get just a few missed
> > > > > > or added bytes about once a minute, and another where I get totally
> > > > > > unexpected chars until I reopen the port.
> > > > > >
> > > > > > I will have a hard time to analyse and give a better description of
> > > > > > how exactly the port is failing, so maybe you have a idea of what is
> > > > > > the issue with this patch.
> > > > > >
> > > > > > I've reverted this on top of 5.3.1 and consistently got the issue fixed.
> > > > > >
> > > > > > Thanks,
> > > > > > Nuno
> > > > > >
> > > > > > commit 4a96895f74c9633b51427fd080ab70fa62b65bc4
> > > > > > Author: Yegor Yefremov <yegorslists@googlemail.com>
> > > > > > Date:   Thu Jun 20 08:24:20 2019 +0200
> > > > > >
> > > > > >     tty/serial/8250: use mctrl_gpio helpers
> > > > > >
> > > > > >     This patch permits the usage for GPIOs to control
> > > > > >     the CTS/RTS/DTR/DSR/DCD/RI signals.
> > > > > >
> > > > > >     Changed by Stefan:
> > > > > >     Only call mctrl_gpio_init(), if the device has no ACPI companion device
> > > > > >     to not break existing ACPI based systems. Also only use the mctrl_gpio_
> > > > > >     functions when "gpios" is available.
> > > > > >
> > > > > >     Use MSR / MCR <-> TIOCM wrapper functions.
> > > > > >
> > > > > >     Signed-off-by: Yegor Yefremov <yegorslists@googlemail.com>
> > > > > >     Signed-off-by: Stefan Roese <sr@denx.de>
> > > > > >     Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
> > > > > >     Reviewed-by: Mika Westerberg <mika.westerberg@linux.intel.com>
> > > > > >     Tested-by: Yegor Yefremov <yegorslists@googlemail.com>
> > > > > >     Cc: Mika Westerberg <mika.westerberg@linux.intel.com>
> > > > > >     Cc: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
> > > > > >     Cc: Giulio Benetti <giulio.benetti@micronovasrl.com>
> > > > > >     Cc: Yegor Yefremov <yegorslists@googlemail.com>
> > > > > >     Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
> > > > > >     Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

^ permalink raw reply	[flat|nested] 10+ messages in thread
* Re: Regression since 4a96895f74c9633b51427fd080ab70fa62b65bc4
  2019-10-28 11:09           ` Yegor Yefremov
@ 2019-10-28 11:11             ` Nuno Gonçalves
  2019-10-28 11:12               ` Yegor Yefremov
  0 siblings, 1 reply; 10+ messages in thread
From: Nuno Gonçalves @ 2019-10-28 11:11 UTC (permalink / raw)
  To: Yegor Yefremov
  Cc: linux-serial, Stefan Roese, Andy Shevchenko, Mika Westerberg,
	Greg KH, Giulio Benetti

On Mon, Oct 28, 2019 at 12:09 PM Yegor Yefremov
<yegorslists@googlemail.com> wrote:
> could you already try at least 5.4-rc4 and test whether the hardware
> flow control is fixed for am335x and your application?

I did test last week. 5.4-rc4 is working fine for me.

Thanks,
Nuno

^ permalink raw reply	[flat|nested] 10+ messages in thread
* Re: Regression since 4a96895f74c9633b51427fd080ab70fa62b65bc4
  2019-10-28 11:11             ` Nuno Gonçalves
@ 2019-10-28 11:12               ` Yegor Yefremov
  2019-10-28 11:14                 ` Nuno Gonçalves
  0 siblings, 1 reply; 10+ messages in thread
From: Yegor Yefremov @ 2019-10-28 11:12 UTC (permalink / raw)
  To: Nuno Gonçalves
  Cc: linux-serial, Stefan Roese, Andy Shevchenko, Mika Westerberg,
	Greg KH, Giulio Benetti

On Mon, Oct 28, 2019 at 12:11 PM Nuno Gonçalves <nunojpg@gmail.com> wrote:
>
> On Mon, Oct 28, 2019 at 12:09 PM Yegor Yefremov
> <yegorslists@googlemail.com> wrote:
> > could you already try at least 5.4-rc4 and test whether the hardware
> > flow control is fixed for am335x and your application?
>
> I did test last week. 5.4-rc4 is working fine for me.

Just curious. What baud rate were you using?

Yegor

^ permalink raw reply	[flat|nested] 10+ messages in thread
* Re: Regression since 4a96895f74c9633b51427fd080ab70fa62b65bc4
  2019-10-28 11:12               ` Yegor Yefremov
@ 2019-10-28 11:14                 ` Nuno Gonçalves
  0 siblings, 0 replies; 10+ messages in thread
From: Nuno Gonçalves @ 2019-10-28 11:14 UTC (permalink / raw)
  To: Yegor Yefremov
  Cc: linux-serial, Stefan Roese, Andy Shevchenko, Mika Westerberg,
	Greg KH, Giulio Benetti

On Mon, Oct 28, 2019 at 12:12 PM Yegor Yefremov
<yegorslists@googlemail.com> wrote:
> Just curious. What baud rate were you using?

B3500000

(3.5Mbps)

Thanks,
Nuno

^ permalink raw reply	[flat|nested] 10+ messages in thread
end of thread, other threads:[~2019-10-28 11:14 UTC | newest]

Thread overview: 10+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2019-09-28  5:59 Regression since 4a96895f74c9633b51427fd080ab70fa62b65bc4 Nuno Gonçalves
2019-09-29  6:46 ` Nuno Gonçalves
2019-10-01 20:02   ` Yegor Yefremov
2019-10-01 21:03     ` Nuno Gonçalves
2019-10-17  9:19       ` Nuno Gonçalves
2019-10-17 13:34         ` Yegor Yefremov
2019-10-28 11:09           ` Yegor Yefremov
2019-10-28 11:11             ` Nuno Gonçalves
2019-10-28 11:12               ` Yegor Yefremov
2019-10-28 11:14                 ` Nuno Gonçalves

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.