Currently, reserved register fields must be fully described using dummy
enum IDs (zeroes), one for each possible state (2^bits states).
Add support for describing reserved fields using negative field width
values as shorthands, thus removing the need for dummy values. Apart
from the obvious size reduction due to the removal of the dummy values,
this will also enable merging adjacent reserved fields into a single
field, reducing the number of fields to describe, and thus kernel size.
Update the checker accordingly.
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
Link: https://lore.kernel.org/r/cad7c92ef039d9a4d039807efc15886a7aa862be.1649865241.git.geert+renesas@glider.be
The checker code to iterate over all drive strength and bias register
description items is cumbersome, due to the repeated calculation of
indices, and the use of hardcoded array sizes. The latter was done
under the assumption they would never need to be changed, which turned
out to be false.
Increase readability by introducing helper macros to access drive
strength and bias register description items.
Increase maintainability by replacing hardcoded numbers by array sizes
calculated at compile-time.
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
Reviewed-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
Link: https://lore.kernel.org/r/5051ae56a1388ccf2d283dfc9624de2991cce914.1650355619.git.geert+renesas@glider.be
The checker failed to validate all enum IDs in the description of a
register with fixed-width register fields, due to a miscalculation of
the number of described states: each register field of n bits can have
"1 << n" possible states, not "1".
Increase SH_PFC_MAX_ENUMS accordingly, now more enum IDs are checked
(SH-Mobile AG5 has more than 4000 enum IDs defined).
Fixes: 12d057bad6 ("pinctrl: sh-pfc: checker: Add check for enum ID conflicts")
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
Link: https://lore.kernel.org/r/6d8a6a05564f38f9d20464c1c17f96e52740cf6a.1645460429.git.geert+renesas@glider.be
Validate consistency of the pin control tables for pins with pull-up
and/or pull-down capabilities. If a pin has bias register bits
declarations in bias_regs[] or through .pin_to_portcr(), the
corresponding SH_PFC_PIN_CFG_PULL* flags should be set in the pin's
configs, and vice versa, and the .get_bias() and .set_bias() callbacks
should be implemented.
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
Link: https://lore.kernel.org/r/19b3e0773fbb36c015a43db683f79e75b0fec3ee.1640270559.git.geert+renesas@glider.be
If some bits in a pin Pull-Up control register (PUPR) control pin
pull-down instead of pin pull-up, there are two pinmux_bias_reg entries:
a first one with the puen field filled in, listing pins with pull-up
functionality, and a second one with the pud field filled in, listing
pins with pull-down functionality. On encountering the second entry,
where puen is NULL, the for-loop terminates early, causing the remaining
bias registers not to be checked. In addition, sh_pfc_check_bias_reg()
does not handle such entries.
Fix this by treating pinmux_bias_reg.puen and pinmux_bias_reg.pud the
same.
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
Link: https://lore.kernel.org/r/29526d06fa223cffd785cdb264b756a202b11cea.1633615652.git.geert+renesas@glider.be
If some bits in a pin Pull-Up control register (PUPR) control pin
pull-down instead of pin pull-up, there are two pinmux_bias_reg entries:
a first one with the puen field filled in, listing pins with pull-up
functionality, and a second one with the pud field filled in, listing
pins with pull-down functionality. On encountering the second entry,
where puen is NULL, the for-loop terminates early, causing the remaining
bias registers not to be saved/restored during PSCI system suspend.
Fortunately this does not trigger on any supported system yet, as PSCI
is only used on R-Car Gen3 and RZ/G2 systems, which all have separate
pin Pull-Enable (PUEN) and pin Pull-Up/Down control (PUD) registers.
Avoid this ever becoming a problem by treating pinmux_bias_reg.puen and
pinmux_bias_reg.pud the same. Note that a register controlling both
pull-up and pull-down pins would be saved and restored twice, which is
harmless.
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
Link: https://lore.kernel.org/r/59d2fbddff685b6a7a82ff17d2b37633e30e8860.1633615652.git.geert+renesas@glider.be
As R-Car H3 ES1.x (R8A77950) and R-Car ES2.0+ (R8A77951) use the same
compatible value, the pin control driver relies on soc_device_match()
with soc_id = "r8a7795" and the (non)matching of revision = "ES1.*" to
match with and distinguish between the two SoC variants. The
corresponding entries in the normal of_match_table are present only to
make the optional sanity checks work.
The R-Car H3e-2G (R8A779M1) SoC is a different grading of the R-Car H3
ES3.0 (R8A77951) SoC. It uses the same compatible values for individual
devices, but has an additional compatible value for the root node.
When running on an R-Car H3e-2G SoC, soc_device_match() with soc_id =
"r8a7795" does not return a match. Hence the pin control driver falls
back to the normal of_match_table, and, as the R8A77950 entry is listed
first, incorrectly uses the sub-driver for R-Car H3 ES1.x.
Fix this by moving the entry for R8A77951 before the entry for R8A77950.
Simplify sh_pfc_quirk_match() to only handle R-Car H3 ES1,x, as R-Car H3
ES2.0+ can now be matched using the normal of_match_table as well.
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Yoshihiro Shimoda <yoshihiro.shimoda.uh@renesas.com>
Link: https://lore.kernel.org/r/6cdc5bfa424461105779b56f455387e03560cf66.1626707688.git.geert+renesas@glider.be
When DEBUG is defined (e.g. if CONFIG_DEBUG_PINCTRL=y), the Renesas pin
control driver runs sanity checks against the pin control tables. This
may cause lots of output on the console, and can be annoying in ARM
multi-platform kernels. Fix this by only running the checks when
running on SuperH, or on a DT platform supported by the Renesas pin
controller driver.
Suggested-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
Reviewed-by: Arnd Bergmann <arnd@arndb.de>
Link: https://lore.kernel.org/r/20210111165013.496897-1-geert+renesas@glider.be
On SuperH and ARM SH/R-Mobile SoCs, the pin control driver handles
GPIOs, too. To reduce code size when compiling a kernel supporting only
modern SoCs, most, but not all, of the GPIO functionality is protected
by checks for CONFIG_PINCTRL_SH_FUNC_GPIO.
Factor out the remaining parts when not needed:
1. sh_pfc_soc_info.{in,out}put describe GPIO pins that have input
resp. output capabilities (SuperH and SH/R-Mobile).
2. sh_pfc_soc_info.gpio_irq{,_size} describe the mapping from GPIO
pins to interrupt numbers (SH/R-Mobile).
3. sh_pfc_gpio_set_direction() configures GPIO direction, called from
the GPIO driver through pinctrl_gpio_direction_{in,out}put()
(SH/R-Mobile). Unfortunately this function cannot just be moved to
drivers/pinctrl/renesas/gpio.c, as it relies on knowledge of
sh_pfc_pinctrl, which is internal to
drivers/pinctrl/renesas/pinctrl.c.
While code size reduction is minimal, this does help in documenting
depencies.
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
Link: https://lore.kernel.org/r/20201028151637.1734130-9-geert+renesas@glider.be
The drivers/pinctrl/sh-pfc subdirectory was originally created to group
pin control drivers for various Renesas SuperH and SH-Mobile platforms.
However, the name "sh-pfc" no longer reflects its contents, as the
directory now contains pin control drivers for Renesas SuperH, ARM32,
and ARM64 SoCs.
Hence rename the subdirectory from drivers/pinctrl/sh-pfc to
drivers/pinctrl/renesas, and the related Kconfig symbol from
PINCTRL_SH_PFC to PINCTRL_RENESAS.
Rename the git branch in MAINTAINERS, too, for consistency.
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
Link: https://lore.kernel.org/r/20200909131534.12897-3-geert+renesas@glider.be
