[PATCH v6 0/9] ASPEED sgpio driver enhancement.
Steven Lee
steven_lee at aspeedtech.com
Fri Jul 23 13:16:16 AEST 2021
The 07/21/2021 21:27, Bartosz Golaszewski wrote:
> On Mon, Jul 12, 2021 at 12:03 PM Steven Lee <steven_lee at aspeedtech.com> wrote:
> >
> > AST2600 SoC has 2 SGPIO master interfaces one with 128 pins another one
> > with 80 pins, AST2500/AST2400 SoC has 1 SGPIO master interface that
> > supports up to 80 pins.
> > In the current driver design, the max number of sgpio pins is hardcoded
> > in macro MAX_NR_HW_SGPIO and the value is 80.
> >
> > For supporting sgpio master interfaces of AST2600 SoC, the patch series
> > contains the following enhancement:
> > - Convert txt dt-bindings to yaml.
> > - Update aspeed-g6 dtsi to support the enhanced sgpio.
> > - Support muiltiple SGPIO master interfaces.
> > - Support up to 128 pins by dts ngpios property.
> > - Pair input/output GPIOs instead of using 0 as GPIO input pin base and
> > MAX_NR_HW_SGPIO as GPIO output pin base.
> > - Support wdt reset tolerance.
> > - Fix irq_chip issues which causes multiple sgpio devices use the same
> > irq_chip data.
> > - Replace all of_*() APIs with device_*().
> >
> > Changes from v5:
> > * Squash v5 patch-05 and patch-06 to one patch.
> > * Remove MAX_NR_HW_SGPIO and corresponding design to make the gpio
> > input/output pin base are determined by ngpios.
> > For example, if MAX_NR_HW_SGPIO is 80 and ngpios is 10, the original
> > pin order is as follows:
> > Input:
> > 0 1 2 3 ... 9
> > Output:
> > 80 81 82 ... 89
> >
> > With the new design, pin order is changed as follows:
> > Input:
> > 0 2 4 6 ... 18(ngpios * 2 - 2)
> > Output:
> > 1 3 5 7 ... 19(ngpios * 2 - 1)
> > * Replace ast2600-sgpiom-128 and ast2600-sgpiom-80 compatibles by
> > ast2600-sgpiom.
> > * Fix coding style issues.
> >
> > Changes from v4:
> > * Remove ngpios from dtsi
> > * Add ast2400 and ast2500 platform data.
> > * Remove unused macros.
> > * Add ngpios check in a separate patch.
> > * Fix coding style issues.
> >
> > Changes from v3:
> > * Split dt-bindings patch to 2 patches
> > * Rename ast2600-sgpiom1 compatible with ast2600-sgiom-128
> > * Rename ast2600-sgpiom2 compatible with ast2600-sgiom-80
> > * Correct the typo in commit messages.
> > * Fix coding style issues.
> > * Replace all of_*() APIs with device_*().
> >
> > Changes from v2:
> > * Remove maximum/minimum of ngpios from bindings.
> > * Remove max-ngpios from bindings and dtsi.
> > * Remove ast2400-sgpiom and ast2500-sgpiom compatibles from dts and
> > driver.
> > * Add ast2600-sgpiom1 and ast2600-sgpiom2 compatibles as their max
> > number of available gpio pins are different.
> > * Modify functions to pass aspeed_sgpio struct instead of passing
> > max_ngpios.
> > * Split sgpio driver patch to 3 patches
> >
> > Changes from v1:
> > * Fix yaml format issues.
> > * Fix issues reported by kernel test robot.
> >
> > Please help to review.
> >
> > Thanks,
> > Steven
> >
> > Steven Lee (9):
> > dt-bindings: aspeed-sgpio: Convert txt bindings to yaml.
> > dt-bindings: aspeed-sgpio: Add ast2600 sgpio
> > ARM: dts: aspeed-g6: Add SGPIO node.
> > ARM: dts: aspeed-g5: Remove ngpios from sgpio node.
> > gpio: gpio-aspeed-sgpio: Add AST2600 sgpio support
> > gpio: gpio-aspeed-sgpio: Add set_config function
> > gpio: gpio-aspeed-sgpio: Move irq_chip to aspeed-sgpio struct
> > gpio: gpio-aspeed-sgpio: Use generic device property APIs
> > gpio: gpio-aspeed-sgpio: Return error if ngpios is not multiple of 8.
> >
> > .../bindings/gpio/aspeed,sgpio.yaml | 77 ++++++++
> > .../devicetree/bindings/gpio/sgpio-aspeed.txt | 46 -----
> > arch/arm/boot/dts/aspeed-g5.dtsi | 1 -
> > arch/arm/boot/dts/aspeed-g6.dtsi | 28 +++
> > drivers/gpio/gpio-aspeed-sgpio.c | 178 +++++++++++-------
> > 5 files changed, 215 insertions(+), 115 deletions(-)
> > create mode 100644 Documentation/devicetree/bindings/gpio/aspeed,sgpio.yaml
> > delete mode 100644 Documentation/devicetree/bindings/gpio/sgpio-aspeed.txt
> >
> > --
> > 2.17.1
> >
>
> The series looks good to me. Can the DTS and GPIO patches go into
> v5.15 separately?
>
Hi Bart,
Thanks for the review.
Shall we do anything to make the patches go into v5.15 or wait for picking-up?
Steven
> Bart
More information about the Linux-aspeed
mailing list