[v4,0/9] serial: sc16is7xx: fix GPIO regression and rs485 improvements

  From: Hugo Villeneuve <hvilleneuve@dimonoff.com>

Hello,
this patch series mainly fixes a GPIO regression and improve RS485 flags and
properties detection from DT.

It now also includes various small fixes and improvements that were previously
sent as separate patches, but that made testing everything difficult.

Patch 1 fixes an issue when debugging IOcontrol register. After testing the GPIO
regression patches (patches 6 and 7, tests done by Lech Perczak), it appers that
this patch is also necessary for having the correct IOcontrol register values.

Patch 2 introduces a delay after a reset operation to respect datasheet
timing recommandations.

Patch 3 fixes an issue with init of first port during probing.

Patch 4 fixes a bug with the output value when first setting the GPIO direction.

Patch 5 is a refactor of GPIO registration code.

Patches 6 and 7 fix a GPIO regression by (re)allowing to choose GPIO function
for GPIO pins shared with modem status lines.

Patch 8 allows to read common rs485 device-tree flags and properties.

Patch 9 improves comments about chip variants.

I have tested the changes on a custom board with two SC16IS752 DUART using a
Variscite IMX8MN NANO SOM.

Thank you.

Link: [v1] https://lkml.org/lkml/2023/5/17/967
      [v1] https://lkml.org/lkml/2023/5/17/777
      [v1] https://lkml.org/lkml/2023/5/17/780
      [v1] https://lkml.org/lkml/2023/5/17/785
      [v1] https://lkml.org/lkml/2023/5/17/1311
      [v2] https://lkml.org/lkml/2023/5/18/516
      [v3] https://lkml.org/lkml/2023/5/25/7

Changes for V3:
- Integrated all patches into single serie to facilitate debugging and tests.
- Reduce number of exported GPIOs depending on new property
  nxp,modem-control-line-ports
- Added additional example in DT bindings

Changes for V4:
- Increase reset post delay to relax scheduler.
- Put comments patches at the end.
- Remove Fixes tag for patch "mark IOCONTROL register as volatile".
- Improve commit messages after reviews.
- Fix coding style issues after reviews.
- Change GPIO registration to always register the maximum number of GPIOs
  supported by the chip, but maks-out GPIOs declared as modem control lines.
- Add patch to refactor GPIO registration.
- Remove patch "serial: sc16is7xx: fix syntax error in comments".
- Remove patch "add dump registers function"

Hugo Villeneuve (9):
  serial: sc16is7xx: mark IOCONTROL register as volatile
  serial: sc16is7xx: add post reset delay
  serial: sc16is7xx: fix broken port 0 uart init
  serial: sc16is7xx: fix bug when first setting GPIO direction
  serial: sc16is7xx: refactor GPIO controller registration
  dt-bindings: sc16is7xx: Add property to change GPIO function
  serial: sc16is7xx: fix regression with GPIO configuration
  serial: sc16is7xx: add call to get rs485 DT flags and properties
  serial: sc16is7xx: improve comments about variants

 .../bindings/serial/nxp,sc16is7xx.txt         |  46 ++++++
 drivers/tty/serial/sc16is7xx.c                | 150 +++++++++++++-----
 2 files changed, 156 insertions(+), 40 deletions(-)


base-commit: 8b817fded42d8fe3a0eb47b1149d907851a3c942
  

Mon, May 29, 2023 at 10:07:02AM -0400, Hugo Villeneuve kirjoitti:
> From: Hugo Villeneuve <hvilleneuve@dimonoff.com>
> 
> Hello,
> this patch series mainly fixes a GPIO regression and improve RS485 flags and
> properties detection from DT.
> 
> It now also includes various small fixes and improvements that were previously
> sent as separate patches, but that made testing everything difficult.
> 
> Patch 1 fixes an issue when debugging IOcontrol register. After testing the GPIO
> regression patches (patches 6 and 7, tests done by Lech Perczak), it appers that
> this patch is also necessary for having the correct IOcontrol register values.
> 
> Patch 2 introduces a delay after a reset operation to respect datasheet
> timing recommandations.

These two patches are w/o Fixes tag, they should be moved in the series further
as I explained before.

> Patch 3 fixes an issue with init of first port during probing.
> 
> Patch 4 fixes a bug with the output value when first setting the GPIO direction.
> 
> Patch 5 is a refactor of GPIO registration code.
> 
> Patches 6 and 7 fix a GPIO regression by (re)allowing to choose GPIO function
> for GPIO pins shared with modem status lines.
> 
> Patch 8 allows to read common rs485 device-tree flags and properties.
> 
> Patch 9 improves comments about chip variants.
> 
> I have tested the changes on a custom board with two SC16IS752 DUART using a
> Variscite IMX8MN NANO SOM.
  

Mon, May 29, 2023 at 10:07:02AM -0400, Hugo Villeneuve kirjoitti:
> From: Hugo Villeneuve <hvilleneuve@dimonoff.com>
> 
> Hello,
> this patch series mainly fixes a GPIO regression and improve RS485 flags and
> properties detection from DT.
> 
> It now also includes various small fixes and improvements that were previously
> sent as separate patches, but that made testing everything difficult.
> 
> Patch 1 fixes an issue when debugging IOcontrol register. After testing the GPIO
> regression patches (patches 6 and 7, tests done by Lech Perczak), it appers that
> this patch is also necessary for having the correct IOcontrol register values.
> 
> Patch 2 introduces a delay after a reset operation to respect datasheet
> timing recommandations.
> 
> Patch 3 fixes an issue with init of first port during probing.
> 
> Patch 4 fixes a bug with the output value when first setting the GPIO direction.
> 
> Patch 5 is a refactor of GPIO registration code.
> 
> Patches 6 and 7 fix a GPIO regression by (re)allowing to choose GPIO function
> for GPIO pins shared with modem status lines.
> 
> Patch 8 allows to read common rs485 device-tree flags and properties.
> 
> Patch 9 improves comments about chip variants.
> 
> I have tested the changes on a custom board with two SC16IS752 DUART using a
> Variscite IMX8MN NANO SOM.

In general looks good to me, just some minor things and patch ordering issues.
Thank you for doing this!
  

On Tue, 30 May 2023 01:31:28 +0300
andy.shevchenko@gmail.com wrote:

> Mon, May 29, 2023 at 10:07:02AM -0400, Hugo Villeneuve kirjoitti:
> > From: Hugo Villeneuve <hvilleneuve@dimonoff.com>
> > 
> > Hello,
> > this patch series mainly fixes a GPIO regression and improve RS485 flags and
> > properties detection from DT.
> > 
> > It now also includes various small fixes and improvements that were previously
> > sent as separate patches, but that made testing everything difficult.
> > 
> > Patch 1 fixes an issue when debugging IOcontrol register. After testing the GPIO
> > regression patches (patches 6 and 7, tests done by Lech Perczak), it appers that
> > this patch is also necessary for having the correct IOcontrol register values.
> > 
> > Patch 2 introduces a delay after a reset operation to respect datasheet
> > timing recommandations.
> 
> These two patches are w/o Fixes tag, they should be moved in the series further
> as I explained before.

Your explanation was not clear.

Anyway, I moved them in position 7 and 8.

 
> > Patch 3 fixes an issue with init of first port during probing.
> > 
> > Patch 4 fixes a bug with the output value when first setting the GPIO direction.
> > 
> > Patch 5 is a refactor of GPIO registration code.
> > 
> > Patches 6 and 7 fix a GPIO regression by (re)allowing to choose GPIO function
> > for GPIO pins shared with modem status lines.
> > 
> > Patch 8 allows to read common rs485 device-tree flags and properties.
> > 
> > Patch 9 improves comments about chip variants.
> > 
> > I have tested the changes on a custom board with two SC16IS752 DUART using a
> > Variscite IMX8MN NANO SOM.
> 
> -- 
> With Best Regards,
> Andy Shevchenko
  

W dniu 29.05.2023 o 16:07, Hugo Villeneuve pisze:
> From: Hugo Villeneuve <hvilleneuve@dimonoff.com>
>
> Hello,
> this patch series mainly fixes a GPIO regression and improve RS485 flags and
> properties detection from DT.
>
> It now also includes various small fixes and improvements that were previously
> sent as separate patches, but that made testing everything difficult.
>
> Patch 1 fixes an issue when debugging IOcontrol register. After testing the GPIO
> regression patches (patches 6 and 7, tests done by Lech Perczak), it appers that
> this patch is also necessary for having the correct IOcontrol register values.
>
> Patch 2 introduces a delay after a reset operation to respect datasheet
> timing recommandations.
>
> Patch 3 fixes an issue with init of first port during probing.
>
> Patch 4 fixes a bug with the output value when first setting the GPIO direction.
>
> Patch 5 is a refactor of GPIO registration code.
>
> Patches 6 and 7 fix a GPIO regression by (re)allowing to choose GPIO function
> for GPIO pins shared with modem status lines.
>
> Patch 8 allows to read common rs485 device-tree flags and properties.
>
> Patch 9 improves comments about chip variants.
>
> I have tested the changes on a custom board with two SC16IS752 DUART using a
> Variscite IMX8MN NANO SOM.
>
> Thank you.
>
> Link: [v1] https://lkml.org/lkml/2023/5/17/967 <https://lkml.org/lkml/2023/5/17/967>
> [v1] https://lkml.org/lkml/2023/5/17/777 <https://lkml.org/lkml/2023/5/17/777>
> [v1] https://lkml.org/lkml/2023/5/17/780 <https://lkml.org/lkml/2023/5/17/780>
> [v1] https://lkml.org/lkml/2023/5/17/785 <https://lkml.org/lkml/2023/5/17/785>
> [v1] https://lkml.org/lkml/2023/5/17/1311 <https://lkml.org/lkml/2023/5/17/1311>
> [v2] https://lkml.org/lkml/2023/5/18/516 <https://lkml.org/lkml/2023/5/18/516>
> [v3] https://lkml.org/lkml/2023/5/25/7 <https://lkml.org/lkml/2023/5/25/7>
>
> Changes for V3:
> - Integrated all patches into single serie to facilitate debugging and tests.
> - Reduce number of exported GPIOs depending on new property
> nxp,modem-control-line-ports
> - Added additional example in DT bindings
>
> Changes for V4:
> - Increase reset post delay to relax scheduler.
> - Put comments patches at the end.
> - Remove Fixes tag for patch "mark IOCONTROL register as volatile".
> - Improve commit messages after reviews.
> - Fix coding style issues after reviews.
> - Change GPIO registration to always register the maximum number of GPIOs
> supported by the chip, but maks-out GPIOs declared as modem control lines.
> - Add patch to refactor GPIO registration.
> - Remove patch "serial: sc16is7xx: fix syntax error in comments".
> - Remove patch "add dump registers function"
>
> Hugo Villeneuve (9):
> serial: sc16is7xx: mark IOCONTROL register as volatile
> serial: sc16is7xx: add post reset delay
> serial: sc16is7xx: fix broken port 0 uart init
> serial: sc16is7xx: fix bug when first setting GPIO direction
> serial: sc16is7xx: refactor GPIO controller registration
> dt-bindings: sc16is7xx: Add property to change GPIO function
> serial: sc16is7xx: fix regression with GPIO configuration
> serial: sc16is7xx: add call to get rs485 DT flags and properties
> serial: sc16is7xx: improve comments about variants
>
> .../bindings/serial/nxp,sc16is7xx.txt | 46 ++++++
> drivers/tty/serial/sc16is7xx.c | 150 +++++++++++++-----
> 2 files changed, 156 insertions(+), 40 deletions(-)
>
>
> base-commit: 8b817fded42d8fe3a0eb47b1149d907851a3c942

It would be a lot of sending, to do that for every patch separately, so for whole series:
Reviewed-by: Lech Perczak <lech.perczak@camlingroup.com>

And where applicable - for code patches:
Tested-by: Lech Perczak <lech.perczak@camlingroup.com>

I tested whole series at the same time.
I did my tests on an i.MX6 board with SC16IS760 over SPI, which differs a tiny bit from SC16IS752,
and everything works as it should.
Thank you for fixing this!

> -- 
> 2.30.2
>
  

Mon, May 29, 2023 at 10:07:08PM -0400, Hugo Villeneuve kirjoitti:
> On Tue, 30 May 2023 01:31:28 +0300
> andy.shevchenko@gmail.com wrote:
> > Mon, May 29, 2023 at 10:07:02AM -0400, Hugo Villeneuve kirjoitti:
> > > From: Hugo Villeneuve <hvilleneuve@dimonoff.com>
> > > 
> > > Hello,
> > > this patch series mainly fixes a GPIO regression and improve RS485 flags and
> > > properties detection from DT.
> > > 
> > > It now also includes various small fixes and improvements that were previously
> > > sent as separate patches, but that made testing everything difficult.
> > > 
> > > Patch 1 fixes an issue when debugging IOcontrol register. After testing the GPIO
> > > regression patches (patches 6 and 7, tests done by Lech Perczak), it appers that
> > > this patch is also necessary for having the correct IOcontrol register values.
> > > 
> > > Patch 2 introduces a delay after a reset operation to respect datasheet
> > > timing recommandations.
> > 
> > These two patches are w/o Fixes tag, they should be moved in the series further
> > as I explained before.
> 
> Your explanation was not clear.

Sorry if it feels like this. The documentation should have more clarity
on the matter.

> Anyway, I moved them in position 7 and 8.

Thank you, but take also what Greg KH replied to you into consideration.
He is the maintainer and seems other patches needs some additional work
in the scope of Fixes / backport (see stable kernel patches flow in the
kernel documentation, which I also mentioned earlier).
  

On Tue, 30 May 2023 11:30:07 +0200
Lech Perczak <lech.perczak@camlingroup.com> wrote:

> W dniu 29.05.2023 o 16:07, Hugo Villeneuve pisze:
> > From: Hugo Villeneuve <hvilleneuve@dimonoff.com>
> >
> > Hello,
> > this patch series mainly fixes a GPIO regression and improve RS485 flags and
> > properties detection from DT.
> >
> > It now also includes various small fixes and improvements that were previously
> > sent as separate patches, but that made testing everything difficult.
> >
> > Patch 1 fixes an issue when debugging IOcontrol register. After testing the GPIO
> > regression patches (patches 6 and 7, tests done by Lech Perczak), it appers that
> > this patch is also necessary for having the correct IOcontrol register values.
> >
> > Patch 2 introduces a delay after a reset operation to respect datasheet
> > timing recommandations.
> >
> > Patch 3 fixes an issue with init of first port during probing.
> >
> > Patch 4 fixes a bug with the output value when first setting the GPIO direction.
> >
> > Patch 5 is a refactor of GPIO registration code.
> >
> > Patches 6 and 7 fix a GPIO regression by (re)allowing to choose GPIO function
> > for GPIO pins shared with modem status lines.
> >
> > Patch 8 allows to read common rs485 device-tree flags and properties.
> >
> > Patch 9 improves comments about chip variants.
> >
> > I have tested the changes on a custom board with two SC16IS752 DUART using a
> > Variscite IMX8MN NANO SOM.
> >
> > Thank you.
> >
> > Link: [v1] https://lkml.org/lkml/2023/5/17/967 <https://lkml.org/lkml/2023/5/17/967>
> > [v1] https://lkml.org/lkml/2023/5/17/777 <https://lkml.org/lkml/2023/5/17/777>
> > [v1] https://lkml.org/lkml/2023/5/17/780 <https://lkml.org/lkml/2023/5/17/780>
> > [v1] https://lkml.org/lkml/2023/5/17/785 <https://lkml.org/lkml/2023/5/17/785>
> > [v1] https://lkml.org/lkml/2023/5/17/1311 <https://lkml.org/lkml/2023/5/17/1311>
> > [v2] https://lkml.org/lkml/2023/5/18/516 <https://lkml.org/lkml/2023/5/18/516>
> > [v3] https://lkml.org/lkml/2023/5/25/7 <https://lkml.org/lkml/2023/5/25/7>
> >
> > Changes for V3:
> > - Integrated all patches into single serie to facilitate debugging and tests.
> > - Reduce number of exported GPIOs depending on new property
> > nxp,modem-control-line-ports
> > - Added additional example in DT bindings
> >
> > Changes for V4:
> > - Increase reset post delay to relax scheduler.
> > - Put comments patches at the end.
> > - Remove Fixes tag for patch "mark IOCONTROL register as volatile".
> > - Improve commit messages after reviews.
> > - Fix coding style issues after reviews.
> > - Change GPIO registration to always register the maximum number of GPIOs
> > supported by the chip, but maks-out GPIOs declared as modem control lines.
> > - Add patch to refactor GPIO registration.
> > - Remove patch "serial: sc16is7xx: fix syntax error in comments".
> > - Remove patch "add dump registers function"
> >
> > Hugo Villeneuve (9):
> > serial: sc16is7xx: mark IOCONTROL register as volatile
> > serial: sc16is7xx: add post reset delay
> > serial: sc16is7xx: fix broken port 0 uart init
> > serial: sc16is7xx: fix bug when first setting GPIO direction
> > serial: sc16is7xx: refactor GPIO controller registration
> > dt-bindings: sc16is7xx: Add property to change GPIO function
> > serial: sc16is7xx: fix regression with GPIO configuration
> > serial: sc16is7xx: add call to get rs485 DT flags and properties
> > serial: sc16is7xx: improve comments about variants
> >
> > .../bindings/serial/nxp,sc16is7xx.txt | 46 ++++++
> > drivers/tty/serial/sc16is7xx.c | 150 +++++++++++++-----
> > 2 files changed, 156 insertions(+), 40 deletions(-)
> >
> >
> > base-commit: 8b817fded42d8fe3a0eb47b1149d907851a3c942
> 
> It would be a lot of sending, to do that for every patch separately, so for whole series:
> Reviewed-by: Lech Perczak <lech.perczak@camlingroup.com>
> 
> And where applicable - for code patches:
> Tested-by: Lech Perczak <lech.perczak@camlingroup.com>
> 
> I tested whole series at the same time.
> I did my tests on an i.MX6 board with SC16IS760 over SPI, which differs a tiny bit from SC16IS752,
> and everything works as it should.
> Thank you for fixing this!

Hi Lech,
thank for your feedback.

You mentioned before that without the patch "mark IOCONTROL register as volatile", things were not working properly for you. Could you retest by removing this patch and see if things are still working?

Thank you, Hugo.
  

On Tue, 30 May 2023 14:15:39 +0300
andy.shevchenko@gmail.com wrote:

> Mon, May 29, 2023 at 10:07:08PM -0400, Hugo Villeneuve kirjoitti:
> > On Tue, 30 May 2023 01:31:28 +0300
> > andy.shevchenko@gmail.com wrote:
> > > Mon, May 29, 2023 at 10:07:02AM -0400, Hugo Villeneuve kirjoitti:
> > > > From: Hugo Villeneuve <hvilleneuve@dimonoff.com>
> > > > 
> > > > Hello,
> > > > this patch series mainly fixes a GPIO regression and improve RS485 flags and
> > > > properties detection from DT.
> > > > 
> > > > It now also includes various small fixes and improvements that were previously
> > > > sent as separate patches, but that made testing everything difficult.
> > > > 
> > > > Patch 1 fixes an issue when debugging IOcontrol register. After testing the GPIO
> > > > regression patches (patches 6 and 7, tests done by Lech Perczak), it appers that
> > > > this patch is also necessary for having the correct IOcontrol register values.
> > > > 
> > > > Patch 2 introduces a delay after a reset operation to respect datasheet
> > > > timing recommandations.
> > > 
> > > These two patches are w/o Fixes tag, they should be moved in the series further
> > > as I explained before.
> > 
> > Your explanation was not clear.
> 
> Sorry if it feels like this. The documentation should have more clarity
> on the matter.
> 
> > Anyway, I moved them in position 7 and 8.
> 
> Thank you, but take also what Greg KH replied to you into consideration.
> He is the maintainer and seems other patches needs some additional work
> in the scope of Fixes / backport (see stable kernel patches flow in the
> kernel documentation, which I also mentioned earlier).

Hi,
will do.

Thank you,.
Hugo.
  

W dniu 30.05.2023 o 15:08, Hugo Villeneuve pisze:
> On Tue, 30 May 2023 11:30:07 +0200
> Lech Perczak <lech.perczak@camlingroup.com> wrote:
>
> > W dniu 29.05.2023 o 16:07, Hugo Villeneuve pisze:
> > > From: Hugo Villeneuve <hvilleneuve@dimonoff.com>
> > >
> > > Hello,
> > > this patch series mainly fixes a GPIO regression and improve RS485 flags and
> > > properties detection from DT.
> > >
> > > It now also includes various small fixes and improvements that were previously
> > > sent as separate patches, but that made testing everything difficult.
> > >
> > > Patch 1 fixes an issue when debugging IOcontrol register. After testing the GPIO
> > > regression patches (patches 6 and 7, tests done by Lech Perczak), it appers that
> > > this patch is also necessary for having the correct IOcontrol register values.
> > >
> > > Patch 2 introduces a delay after a reset operation to respect datasheet
> > > timing recommandations.
> > >
> > > Patch 3 fixes an issue with init of first port during probing.
> > >
> > > Patch 4 fixes a bug with the output value when first setting the GPIO direction.
> > >
> > > Patch 5 is a refactor of GPIO registration code.
> > >
> > > Patches 6 and 7 fix a GPIO regression by (re)allowing to choose GPIO function
> > > for GPIO pins shared with modem status lines.
> > >
> > > Patch 8 allows to read common rs485 device-tree flags and properties.
> > >
> > > Patch 9 improves comments about chip variants.
> > >
> > > I have tested the changes on a custom board with two SC16IS752 DUART using a
> > > Variscite IMX8MN NANO SOM.
> > >
> > > Thank you.
> > >
> > > Link: [v1] https://lkml.org/lkml/2023/5/17/967 <https://lkml.org/lkml/2023/5/17/967> <https://lkml.org/lkml/2023/5/17/967 <https://lkml.org/lkml/2023/5/17/967>>
> > > [v1] https://lkml.org/lkml/2023/5/17/777 <https://lkml.org/lkml/2023/5/17/777> <https://lkml.org/lkml/2023/5/17/777 <https://lkml.org/lkml/2023/5/17/777>>
> > > [v1] https://lkml.org/lkml/2023/5/17/780 <https://lkml.org/lkml/2023/5/17/780> <https://lkml.org/lkml/2023/5/17/780 <https://lkml.org/lkml/2023/5/17/780>>
> > > [v1] https://lkml.org/lkml/2023/5/17/785 <https://lkml.org/lkml/2023/5/17/785> <https://lkml.org/lkml/2023/5/17/785 <https://lkml.org/lkml/2023/5/17/785>>
> > > [v1] https://lkml.org/lkml/2023/5/17/1311 <https://lkml.org/lkml/2023/5/17/1311> <https://lkml.org/lkml/2023/5/17/1311 <https://lkml.org/lkml/2023/5/17/1311>>
> > > [v2] https://lkml.org/lkml/2023/5/18/516 <https://lkml.org/lkml/2023/5/18/516> <https://lkml.org/lkml/2023/5/18/516 <https://lkml.org/lkml/2023/5/18/516>>
> > > [v3] https://lkml.org/lkml/2023/5/25/7 <https://lkml.org/lkml/2023/5/25/7> <https://lkml.org/lkml/2023/5/25/7 <https://lkml.org/lkml/2023/5/25/7>>
> > >
> > > Changes for V3:
> > > - Integrated all patches into single serie to facilitate debugging and tests.
> > > - Reduce number of exported GPIOs depending on new property
> > > nxp,modem-control-line-ports
> > > - Added additional example in DT bindings
> > >
> > > Changes for V4:
> > > - Increase reset post delay to relax scheduler.
> > > - Put comments patches at the end.
> > > - Remove Fixes tag for patch "mark IOCONTROL register as volatile".
> > > - Improve commit messages after reviews.
> > > - Fix coding style issues after reviews.
> > > - Change GPIO registration to always register the maximum number of GPIOs
> > > supported by the chip, but maks-out GPIOs declared as modem control lines.
> > > - Add patch to refactor GPIO registration.
> > > - Remove patch "serial: sc16is7xx: fix syntax error in comments".
> > > - Remove patch "add dump registers function"
> > >
> > > Hugo Villeneuve (9):
> > > serial: sc16is7xx: mark IOCONTROL register as volatile
> > > serial: sc16is7xx: add post reset delay
> > > serial: sc16is7xx: fix broken port 0 uart init
> > > serial: sc16is7xx: fix bug when first setting GPIO direction
> > > serial: sc16is7xx: refactor GPIO controller registration
> > > dt-bindings: sc16is7xx: Add property to change GPIO function
> > > serial: sc16is7xx: fix regression with GPIO configuration
> > > serial: sc16is7xx: add call to get rs485 DT flags and properties
> > > serial: sc16is7xx: improve comments about variants
> > >
> > > .../bindings/serial/nxp,sc16is7xx.txt | 46 ++++++
> > > drivers/tty/serial/sc16is7xx.c | 150 +++++++++++++-----
> > > 2 files changed, 156 insertions(+), 40 deletions(-)
> > >
> > >
> > > base-commit: 8b817fded42d8fe3a0eb47b1149d907851a3c942
> >
> > It would be a lot of sending, to do that for every patch separately, so for whole series:
> > Reviewed-by: Lech Perczak <lech.perczak@camlingroup.com>
> >
> > And where applicable - for code patches:
> > Tested-by: Lech Perczak <lech.perczak@camlingroup.com>
> >
> > I tested whole series at the same time.
> > I did my tests on an i.MX6 board with SC16IS760 over SPI, which differs a tiny bit from SC16IS752,
> > and everything works as it should.
> > Thank you for fixing this!
>
> Hi Lech,
> thank for your feedback.
>
> You mentioned before that without the patch "mark IOCONTROL register as volatile", things were not working properly for you. Could you retest by removing this patch and see if things are still working?
>
> Thank you, Hugo.

Hello Hugo,

Just checked - this patch is required, reverting it causes things to fail, so this patch should be marked as a pre-requisite for the actual fix and included in backports.
Perhaps using direct write to this register made it work, but it was likely by accident.
  

On Wed, 31 May 2023 12:43:48 +0200
Lech Perczak <lech.perczak@camlingroup.com> wrote:

> W dniu 30.05.2023 o 15:08, Hugo Villeneuve pisze:
> > On Tue, 30 May 2023 11:30:07 +0200
> > Lech Perczak <lech.perczak@camlingroup.com> wrote:
> >
> > > W dniu 29.05.2023 o 16:07, Hugo Villeneuve pisze:
> > > > From: Hugo Villeneuve <hvilleneuve@dimonoff.com>
> > > >
> > > > Hello,
> > > > this patch series mainly fixes a GPIO regression and improve RS485 flags and
> > > > properties detection from DT.
> > > >
> > > > It now also includes various small fixes and improvements that were previously
> > > > sent as separate patches, but that made testing everything difficult.
> > > >
> > > > Patch 1 fixes an issue when debugging IOcontrol register. After testing the GPIO
> > > > regression patches (patches 6 and 7, tests done by Lech Perczak), it appers that
> > > > this patch is also necessary for having the correct IOcontrol register values.
> > > >
> > > > Patch 2 introduces a delay after a reset operation to respect datasheet
> > > > timing recommandations.
> > > >
> > > > Patch 3 fixes an issue with init of first port during probing.
> > > >
> > > > Patch 4 fixes a bug with the output value when first setting the GPIO direction.
> > > >
> > > > Patch 5 is a refactor of GPIO registration code.
> > > >
> > > > Patches 6 and 7 fix a GPIO regression by (re)allowing to choose GPIO function
> > > > for GPIO pins shared with modem status lines.
> > > >
> > > > Patch 8 allows to read common rs485 device-tree flags and properties.
> > > >
> > > > Patch 9 improves comments about chip variants.
> > > >
> > > > I have tested the changes on a custom board with two SC16IS752 DUART using a
> > > > Variscite IMX8MN NANO SOM.
> > > >
> > > > Thank you.
> > > >
> > > > Link: [v1] https://lkml.org/lkml/2023/5/17/967 <https://lkml.org/lkml/2023/5/17/967> <https://lkml.org/lkml/2023/5/17/967 <https://lkml.org/lkml/2023/5/17/967>>
> > > > [v1] https://lkml.org/lkml/2023/5/17/777 <https://lkml.org/lkml/2023/5/17/777> <https://lkml.org/lkml/2023/5/17/777 <https://lkml.org/lkml/2023/5/17/777>>
> > > > [v1] https://lkml.org/lkml/2023/5/17/780 <https://lkml.org/lkml/2023/5/17/780> <https://lkml.org/lkml/2023/5/17/780 <https://lkml.org/lkml/2023/5/17/780>>
> > > > [v1] https://lkml.org/lkml/2023/5/17/785 <https://lkml.org/lkml/2023/5/17/785> <https://lkml.org/lkml/2023/5/17/785 <https://lkml.org/lkml/2023/5/17/785>>
> > > > [v1] https://lkml.org/lkml/2023/5/17/1311 <https://lkml.org/lkml/2023/5/17/1311> <https://lkml.org/lkml/2023/5/17/1311 <https://lkml.org/lkml/2023/5/17/1311>>
> > > > [v2] https://lkml.org/lkml/2023/5/18/516 <https://lkml.org/lkml/2023/5/18/516> <https://lkml.org/lkml/2023/5/18/516 <https://lkml.org/lkml/2023/5/18/516>>
> > > > [v3] https://lkml.org/lkml/2023/5/25/7 <https://lkml.org/lkml/2023/5/25/7> <https://lkml.org/lkml/2023/5/25/7 <https://lkml.org/lkml/2023/5/25/7>>
> > > >
> > > > Changes for V3:
> > > > - Integrated all patches into single serie to facilitate debugging and tests.
> > > > - Reduce number of exported GPIOs depending on new property
> > > > nxp,modem-control-line-ports
> > > > - Added additional example in DT bindings
> > > >
> > > > Changes for V4:
> > > > - Increase reset post delay to relax scheduler.
> > > > - Put comments patches at the end.
> > > > - Remove Fixes tag for patch "mark IOCONTROL register as volatile".
> > > > - Improve commit messages after reviews.
> > > > - Fix coding style issues after reviews.
> > > > - Change GPIO registration to always register the maximum number of GPIOs
> > > > supported by the chip, but maks-out GPIOs declared as modem control lines.
> > > > - Add patch to refactor GPIO registration.
> > > > - Remove patch "serial: sc16is7xx: fix syntax error in comments".
> > > > - Remove patch "add dump registers function"
> > > >
> > > > Hugo Villeneuve (9):
> > > > serial: sc16is7xx: mark IOCONTROL register as volatile
> > > > serial: sc16is7xx: add post reset delay
> > > > serial: sc16is7xx: fix broken port 0 uart init
> > > > serial: sc16is7xx: fix bug when first setting GPIO direction
> > > > serial: sc16is7xx: refactor GPIO controller registration
> > > > dt-bindings: sc16is7xx: Add property to change GPIO function
> > > > serial: sc16is7xx: fix regression with GPIO configuration
> > > > serial: sc16is7xx: add call to get rs485 DT flags and properties
> > > > serial: sc16is7xx: improve comments about variants
> > > >
> > > > .../bindings/serial/nxp,sc16is7xx.txt | 46 ++++++
> > > > drivers/tty/serial/sc16is7xx.c | 150 +++++++++++++-----
> > > > 2 files changed, 156 insertions(+), 40 deletions(-)
> > > >
> > > >
> > > > base-commit: 8b817fded42d8fe3a0eb47b1149d907851a3c942
> > >
> > > It would be a lot of sending, to do that for every patch separately, so for whole series:
> > > Reviewed-by: Lech Perczak <lech.perczak@camlingroup.com>
> > >
> > > And where applicable - for code patches:
> > > Tested-by: Lech Perczak <lech.perczak@camlingroup.com>
> > >
> > > I tested whole series at the same time.
> > > I did my tests on an i.MX6 board with SC16IS760 over SPI, which differs a tiny bit from SC16IS752,
> > > and everything works as it should.
> > > Thank you for fixing this!
> >
> > Hi Lech,
> > thank for your feedback.
> >
> > You mentioned before that without the patch "mark IOCONTROL register as volatile", things were not working properly for you. Could you retest by removing this patch and see if things are still working?
> >
> > Thank you, Hugo.
> 
> Hello Hugo,
> 
> Just checked - this patch is required, reverting it causes things to fail, so this patch should be marked as a pre-requisite for the actual fix and included in backports.
> Perhaps using direct write to this register made it work, but it was likely by accident.

Hi Lech,
thank you for the test, I will mark it as such in upcoming series V5.

Hugo.
  

On Wed, 31 May 2023 09:56:08 -0400
Hugo Villeneuve <hugo@hugovil.com> wrote:

> On Wed, 31 May 2023 12:43:48 +0200
> Lech Perczak <lech.perczak@camlingroup.com> wrote:
> 
> > W dniu 30.05.2023 o 15:08, Hugo Villeneuve pisze:
> > > On Tue, 30 May 2023 11:30:07 +0200
> > > Lech Perczak <lech.perczak@camlingroup.com> wrote:
> > >
> > > > W dniu 29.05.2023 o 16:07, Hugo Villeneuve pisze:
> > > > > From: Hugo Villeneuve <hvilleneuve@dimonoff.com>
> > > > >
> > > > > Hello,
> > > > > this patch series mainly fixes a GPIO regression and improve RS485 flags and
> > > > > properties detection from DT.
> > > > >
> > > > > It now also includes various small fixes and improvements that were previously
> > > > > sent as separate patches, but that made testing everything difficult.
> > > > >
> > > > > Patch 1 fixes an issue when debugging IOcontrol register. After testing the GPIO
> > > > > regression patches (patches 6 and 7, tests done by Lech Perczak), it appers that
> > > > > this patch is also necessary for having the correct IOcontrol register values.
> > > > >
> > > > > Patch 2 introduces a delay after a reset operation to respect datasheet
> > > > > timing recommandations.
> > > > >
> > > > > Patch 3 fixes an issue with init of first port during probing.
> > > > >
> > > > > Patch 4 fixes a bug with the output value when first setting the GPIO direction.
> > > > >
> > > > > Patch 5 is a refactor of GPIO registration code.
> > > > >
> > > > > Patches 6 and 7 fix a GPIO regression by (re)allowing to choose GPIO function
> > > > > for GPIO pins shared with modem status lines.
> > > > >
> > > > > Patch 8 allows to read common rs485 device-tree flags and properties.
> > > > >
> > > > > Patch 9 improves comments about chip variants.
> > > > >
> > > > > I have tested the changes on a custom board with two SC16IS752 DUART using a
> > > > > Variscite IMX8MN NANO SOM.
> > > > >
> > > > > Thank you.
> > > > >
> > > > > Link: [v1] https://lkml.org/lkml/2023/5/17/967 <https://lkml.org/lkml/2023/5/17/967> <https://lkml.org/lkml/2023/5/17/967 <https://lkml.org/lkml/2023/5/17/967>>
> > > > > [v1] https://lkml.org/lkml/2023/5/17/777 <https://lkml.org/lkml/2023/5/17/777> <https://lkml.org/lkml/2023/5/17/777 <https://lkml.org/lkml/2023/5/17/777>>
> > > > > [v1] https://lkml.org/lkml/2023/5/17/780 <https://lkml.org/lkml/2023/5/17/780> <https://lkml.org/lkml/2023/5/17/780 <https://lkml.org/lkml/2023/5/17/780>>
> > > > > [v1] https://lkml.org/lkml/2023/5/17/785 <https://lkml.org/lkml/2023/5/17/785> <https://lkml.org/lkml/2023/5/17/785 <https://lkml.org/lkml/2023/5/17/785>>
> > > > > [v1] https://lkml.org/lkml/2023/5/17/1311 <https://lkml.org/lkml/2023/5/17/1311> <https://lkml.org/lkml/2023/5/17/1311 <https://lkml.org/lkml/2023/5/17/1311>>
> > > > > [v2] https://lkml.org/lkml/2023/5/18/516 <https://lkml.org/lkml/2023/5/18/516> <https://lkml.org/lkml/2023/5/18/516 <https://lkml.org/lkml/2023/5/18/516>>
> > > > > [v3] https://lkml.org/lkml/2023/5/25/7 <https://lkml.org/lkml/2023/5/25/7> <https://lkml.org/lkml/2023/5/25/7 <https://lkml.org/lkml/2023/5/25/7>>
> > > > >
> > > > > Changes for V3:
> > > > > - Integrated all patches into single serie to facilitate debugging and tests.
> > > > > - Reduce number of exported GPIOs depending on new property
> > > > > nxp,modem-control-line-ports
> > > > > - Added additional example in DT bindings
> > > > >
> > > > > Changes for V4:
> > > > > - Increase reset post delay to relax scheduler.
> > > > > - Put comments patches at the end.
> > > > > - Remove Fixes tag for patch "mark IOCONTROL register as volatile".
> > > > > - Improve commit messages after reviews.
> > > > > - Fix coding style issues after reviews.
> > > > > - Change GPIO registration to always register the maximum number of GPIOs
> > > > > supported by the chip, but maks-out GPIOs declared as modem control lines.
> > > > > - Add patch to refactor GPIO registration.
> > > > > - Remove patch "serial: sc16is7xx: fix syntax error in comments".
> > > > > - Remove patch "add dump registers function"
> > > > >
> > > > > Hugo Villeneuve (9):
> > > > > serial: sc16is7xx: mark IOCONTROL register as volatile
> > > > > serial: sc16is7xx: add post reset delay
> > > > > serial: sc16is7xx: fix broken port 0 uart init
> > > > > serial: sc16is7xx: fix bug when first setting GPIO direction
> > > > > serial: sc16is7xx: refactor GPIO controller registration
> > > > > dt-bindings: sc16is7xx: Add property to change GPIO function
> > > > > serial: sc16is7xx: fix regression with GPIO configuration
> > > > > serial: sc16is7xx: add call to get rs485 DT flags and properties
> > > > > serial: sc16is7xx: improve comments about variants
> > > > >
> > > > > .../bindings/serial/nxp,sc16is7xx.txt | 46 ++++++
> > > > > drivers/tty/serial/sc16is7xx.c | 150 +++++++++++++-----
> > > > > 2 files changed, 156 insertions(+), 40 deletions(-)
> > > > >
> > > > >
> > > > > base-commit: 8b817fded42d8fe3a0eb47b1149d907851a3c942
> > > >
> > > > It would be a lot of sending, to do that for every patch separately, so for whole series:
> > > > Reviewed-by: Lech Perczak <lech.perczak@camlingroup.com>
> > > >
> > > > And where applicable - for code patches:
> > > > Tested-by: Lech Perczak <lech.perczak@camlingroup.com>
> > > >
> > > > I tested whole series at the same time.
> > > > I did my tests on an i.MX6 board with SC16IS760 over SPI, which differs a tiny bit from SC16IS752,
> > > > and everything works as it should.
> > > > Thank you for fixing this!
> > >
> > > Hi Lech,
> > > thank for your feedback.
> > >
> > > You mentioned before that without the patch "mark IOCONTROL register as volatile", things were not working properly for you. Could you retest by removing this patch and see if things are still working?
> > >
> > > Thank you, Hugo.
> > 
> > Hello Hugo,
> > 
> > Just checked - this patch is required, reverting it causes things to fail, so this patch should be marked as a pre-requisite for the actual fix and included in backports.
> > Perhaps using direct write to this register made it work, but it was likely by accident.
> 
> Hi Lech,
> thank you for the test, I will mark it as such in upcoming series V5.
> 
> Hugo.

Since I reworked a bit patch 5/9 in series 5, I removed your "Tested-by" and "Reviewed-by" tags only for this patch. Please reconfirm these tags when you have tested series 5.

Thank you,
Hugo.