2023-12-07 13:00:30

by Martin Kurbanov

[permalink] [raw]
Subject: [PATCH v1 0/2] leds: aw200xx: support for hw pattern controllers

This led-controller supports 3 pattern controllers for auto breathing or
group dimming control. Each pattern controller can work in auto
breathing or manual control mode. All breathing parameters including
rising/falling slope, on/off time, repeat times, min/max brightness
and so on are configurable.

Martin Kurbanov (2):
leds: aw200xx: support for hw pattern controllers
dt-bindings: leds: aw200xx: add 'interrupts' property

.../testing/sysfs-class-led-driver-aw200xx | 108 +++
.../bindings/leds/awinic,aw200xx.yaml | 6 +
Documentation/leds/leds-aw200xx.rst | 274 ++++++++
drivers/leds/leds-aw200xx.c | 649 ++++++++++++++++++
4 files changed, 1037 insertions(+)
create mode 100644 Documentation/leds/leds-aw200xx.rst

--
2.40.0


2023-12-07 13:01:41

by Martin Kurbanov

[permalink] [raw]
Subject: [PATCH v1 2/2] dt-bindings: leds: aw200xx: add 'interrupts' property

Added support for interrupt to the LED controller driver, which
indicates the finished of the auto-breath pattern.

Signed-off-by: Martin Kurbanov <[email protected]>
---
Documentation/devicetree/bindings/leds/awinic,aw200xx.yaml | 6 ++++++
1 file changed, 6 insertions(+)

diff --git a/Documentation/devicetree/bindings/leds/awinic,aw200xx.yaml b/Documentation/devicetree/bindings/leds/awinic,aw200xx.yaml
index 54d6d1f08e24..7ab35b7a3fe6 100644
--- a/Documentation/devicetree/bindings/leds/awinic,aw200xx.yaml
+++ b/Documentation/devicetree/bindings/leds/awinic,aw200xx.yaml
@@ -44,6 +44,9 @@ properties:
enable-gpios:
maxItems: 1

+ interrupts:
+ maxItems: 1
+
patternProperties:
"^led@[0-9a-f]+$":
type: object
@@ -137,6 +140,7 @@ additionalProperties: false
examples:
- |
#include <dt-bindings/gpio/gpio.h>
+ #include <dt-bindings/interrupt-controller/irq.h>
#include <dt-bindings/leds/common.h>

i2c {
@@ -149,6 +153,8 @@ examples:
#address-cells = <1>;
#size-cells = <0>;
enable-gpios = <&gpio 3 GPIO_ACTIVE_HIGH>;
+ interrupt-parent = <&gpio_intc>;
+ interrupts = <13 IRQ_TYPE_LEVEL_LOW>;

led@0 {
reg = <0x0>;
--
2.40.0

2023-12-07 13:01:51

by Martin Kurbanov

[permalink] [raw]
Subject: [PATCH v1 1/2] leds: aw200xx: support for hw pattern controllers

This led-controller supports 3 pattern controllers for auto breathing or
group dimming control. Each pattern controller can work in auto
breathing or manual control mode. All breathing parameters including
rising/falling slope, on/off time, repeat times, min/max brightness
and so on are configurable.

Signed-off-by: Martin Kurbanov <[email protected]>
---
.../testing/sysfs-class-led-driver-aw200xx | 108 +++
Documentation/leds/leds-aw200xx.rst | 274 ++++++++
drivers/leds/leds-aw200xx.c | 649 ++++++++++++++++++
3 files changed, 1031 insertions(+)
create mode 100644 Documentation/leds/leds-aw200xx.rst

diff --git a/Documentation/ABI/testing/sysfs-class-led-driver-aw200xx b/Documentation/ABI/testing/sysfs-class-led-driver-aw200xx
index 6d4449cf9d71..1d3d74fa6f08 100644
--- a/Documentation/ABI/testing/sysfs-class-led-driver-aw200xx
+++ b/Documentation/ABI/testing/sysfs-class-led-driver-aw200xx
@@ -3,3 +3,111 @@ Date: May 2023
Description: 64-level DIM current. If you write a negative value or
"auto", the dim will be calculated according to the
brightness.
+
+What: /sys/class/leds/<led>/device/<pattern>/clear_leds
+What: /sys/class/leds/<led>/device/<pattern>/select_leds
+Date: December 2023
+Description: Bitmask for clear/select leds to pattern.
+
+What: /sys/class/leds/<led>/device/<pattern>/fall_time
+What: /sys/class/leds/<led>/device/<pattern>/off_time
+What: /sys/class/leds/<led>/device/<pattern>/on_time
+What: /sys/class/leds/<led>/device/<pattern>/rise_time
+Date: December 2023
+Description:
+ Set the pattern fall, off, on and rise times (0..15), where:
+
+ == ======= === ======
+ 0 0 ms 8 2.1 s
+ 1 130 ms 9 2.6 s
+ 2 260 ms 10 3.1 s
+ 3 380 ms 11 4.2 s
+ 4 510 ms 12 5.2 s
+ 5 770 ms 13 6.2 s
+ 6 1.04 s 14 7.3 s
+ 7 1.6 s 15 8.3 s
+ == ======= === ======
+
+What: /sys/class/leds/<led>/device/<pattern>/loop_begin
+Date: December 2023
+Description:
+ Choose where to start the loop (0..3), where:
+
+ == ==================
+ 0 from 'rise' state
+ 1 from 'on' state
+ 2 from 'fall' state
+ 3 from 'off' state
+ == ==================
+
+What: /sys/class/leds/<led>/device/<pattern>/loop_end_on
+Date: December 2023
+Description:
+ Choose where to end the loop (0..1), where:
+
+ == ========================
+ 0 loop end at 'off' state
+ 1 loop end at 'on' state
+ == ========================
+
+What: /sys/class/leds/<led>/device/<pattern>/max_breathing_level
+Date: December 2023
+Description: Maximum breathing Level.
+
+What: /sys/class/leds/<led>/device/<pattern>/min_breathing_level
+Date: December 2023
+Description: Minimum breathing Level
+
+What: /sys/class/leds/<led>/device/<pattern>/mode
+Date: December 2023
+Description: Pattern mode: manual or auto breathing (0, 1) where:
+
+ == ====================
+ 0 manual mode
+ 1 auto breathing mode
+ == ====================
+
+What: /sys/class/leds/<led>/device/<pattern>/ramp
+Date: December 2023
+Description:
+ The smooth ramp down/up function (0, 1), where (see ):
+
+ == ==================
+ 0 ramp mode disable
+ 1 ramp mode enable
+ == ==================
+
+ For more information, see Documentation/leds/leds-aw200xx.rst
+
+What: /sys/class/leds/<led>/device/<pattern>/repeat
+Date: December 2023
+Description: Loop times (0..4095). 0 - the loop is end-less.
+
+What: /sys/class/leds/<led>/device/<pattern>/running
+Date: December 2023
+Description: Reading this file will return the pattern state.
+ This file is read only and supports poll() to detect when the
+ pattern ended.
+ 1 means pattern is running and number 0 are finish or not run.
+
+What: /sys/class/leds/<led>/device/<pattern>/start
+Date: December 2023
+Description:
+ Start/stop pattern (0, 1), where:
+
+ == ==============
+ 0 stop pattern
+ 1 start pattern
+ == ==============
+
+What: /sys/class/leds/<led>/device/<pattern>/toggle
+Date: December 2023
+Description:
+ Manual on/off control (0, 1), where:
+
+ == =========
+ 0 LEDs off
+ 1 LEDs on
+ == =========
+
+ For more information, see Documentation/leds/leds-aw200xx.rst
diff --git a/Documentation/leds/leds-aw200xx.rst b/Documentation/leds/leds-aw200xx.rst
new file mode 100644
index 000000000000..52e777b04885
--- /dev/null
+++ b/Documentation/leds/leds-aw200xx.rst
@@ -0,0 +1,274 @@
+.. SPDX-License-Identifier: GPL-2.0
+
+=================================================
+Kernel driver for AW20036/AW20054/AW20072/AW20108
+=================================================
+
+Description
+-----------
+
+The AW20036/AW20054/AW20072/AW20108 is a 3x12/6x9/6x12/9x12 matrix LED driver
+programmed via an I2C interface. The brightness of each LED is independently
+controlled by FADE and DIM parameter.
+
+Three integrated pattern controllers provide auto breathing or group dimming
+control. Each pattern controller can work in auto breathing or manual control
+mode. All breathing parameters including rising/falling slope, on/off time,
+repeat times, min/max brightness and so on are configurable.
+
+Device attribute
+-----------------------------------
+
+**/sys/class/leds/<led>/dim** - 64-level DIM current. If write negative value
+or "auto", the dim will be calculated according to the brightness.
+
+The configuration files for each pattern are located::
+
+ /sys/bus/i2c/devices/xxxx/pattern0/
+ /sys/bus/i2c/devices/xxxx/pattern1/
+ /sys/bus/i2c/devices/xxxx/pattern2/
+
+Directory layout example for pattern
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+::
+
+ $ ls -l /sys/bus/i2c/devices/xxxx/pattern0/
+ -rw-r--r-- 1 root root 4096 Jan 1 00:00 clear_leds
+ -rw-r--r-- 1 root root 4096 Jan 1 00:00 fall_time
+ -rw-r--r-- 1 root root 4096 Jan 1 00:00 loop_begin
+ -rw-r--r-- 1 root root 4096 Jan 1 00:00 loop_end_on
+ -rw-r--r-- 1 root root 4096 Jan 1 00:00 max_breathing_level
+ -rw-r--r-- 1 root root 4096 Jan 1 00:00 min_breathing_level
+ -rw-r--r-- 1 root root 4096 Jan 1 00:00 mode
+ -rw-r--r-- 1 root root 4096 Jan 1 00:00 off_time
+ -rw-r--r-- 1 root root 4096 Jan 1 00:00 on_time
+ -rw-r--r-- 1 root root 4096 Jan 1 00:00 ramp
+ -rw-r--r-- 1 root root 4096 Jan 1 00:00 repeat
+ -rw-r--r-- 1 root root 4096 Jan 1 00:00 rise_time
+ -r--r--r-- 1 root root 4096 Jan 1 00:00 running
+ -rw-r--r-- 1 root root 4096 Jan 1 00:00 select_leds
+ -rw-r--r-- 1 root root 4096 Jan 1 00:00 start
+ -rw-r--r-- 1 root root 4096 Jan 1 00:00 toggle
+
+Timing parameters
+~~~~~~~~~~~~~~~~~
+
+- **on_time**
+
+- **rise_time**
+
+- **fall_time**
+
+- **off_time**
+
+See :ref:`auto_breath_mode`.
+
+Select from predefined times:
+
+.. flat-table::
+
+ * - Value
+ - Time (in seconds)
+ - Value
+ - Time (in seconds)
+
+ * - 0
+ - 0.00
+ - 8
+ - 2.1
+
+ * - 1
+ - 0.13
+ - 9
+ - 2.6
+
+ * - 2
+ - 0.26
+ - 10
+ - 3.1
+
+ * - 3
+ - 0.38
+ - 11
+ - 4.2
+
+ * - 4
+ - 0.51
+ - 12
+ - 5.2
+
+ * - 5
+ - 0.77
+ - 13
+ - 6.2
+
+ * - 6
+ - 1.04
+ - 14
+ - 7.3
+
+ * - 7
+ - 1.6
+ - 15
+ - 8.3
+
+Example set for rise-time=0.13s, on-time=0.26s,
+fall_time=6.2s, off_time=0.51s:
+
+.. code-block:: bash
+
+ echo 1 > rise_time
+ echo 2 > on_time
+ echo 13 > fall_time
+ echo 4 > off_time
+
+Maximum and minimum breathing Level
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+- **max_breathing_level**
+
+- **min_breathing_level**
+
+Loop
+~~~~~
+
+- **loop_begin** - choose where to start the loop:
+
+ .. flat-table::
+
+ * - Value
+ - Description
+
+ * - 0
+ - Begin from 'rise' state
+
+ * - 1
+ - Begin from 'on' state
+
+ * - 2
+ - Begin from 'fall' state
+
+ * - 3
+ - Begin from 'off' state
+
+- **loop_end_on** - write ``1`` loop end at 'on' state.
+
+- **repeat** - loop times. When write ``0``, the loop is end-less.
+
+Others
+~~~~~~
+
+- **clear_leds** - bitmask for clear leds to pattern.
+ For example clear leds 1, 2, 3, 4, 5, 7 (``10111110``):
+
+ .. code-block:: bash
+
+ echo be > clear_leds
+
+- **select_leds** - bitmask for set leds to pattern.
+ For example select leds 0, 3, 6, 7 (``11001001``):
+
+ .. code-block:: bash
+
+ echo c9 > select_leds
+
+- **mode** - pattern mode:
+ ``0`` - manual control, ``1`` - auto breathing
+
+- **start** - start/stop pattern:
+ ``0`` - to stop, ``1`` - to start
+
+- **toggle** - manual on/off control (see :ref:`manual_breath_mode`):
+ ``0`` - LEDs off, ``1`` - LEDs on
+
+- **ramp** - the smooth ramp up/down function (see :ref:`manual_breath_mode`):
+ ``0`` - disable, ``1`` - enable
+
+- **running** - Reading this file will return the pattern state:
+ ``1`` - is running, ``0`` - is finished (or not running)
+
+This file supports poll() to detect when the pattern finished.
+
+.. _auto_breath_mode:
+
+Auto breathing mode
+~~~~~~~~~~~~~~~~~~~
+
+::
+
+ breathing level
+ ^
+ max _ ________________
+ | /. .\
+ | / . . \
+ | / . . \
+ | / . . \
+ | / . . \
+ | / . . \
+ min _ ___/ . . \_______
+ | . . . .
+ | . . . .
+ | . rise . on . fall . off
+ |
+ -|------------------------------------------------> time
+
+Example:
+
+.. code-block:: bash
+
+ echo 10 > rise_time # 3.1 seconds
+ echo 4 > on_time # 0.51 seconds
+ echo 1 > off_time # 0.13 seconds
+ echo 10 > fall_time # 3.1 seconds
+ echo 0 > min_breathing_level
+ echo 255 > max_breathing_level
+ echo 0 > loop_begin # begin from 'rise'
+ echo 0 > loop_end_on # loop end at 'off' state
+ echo 1 > mode # auto breathing mode
+ echo 5 > repeat # 5 times repeat
+ echo 1249 > select_leds # select 0, 3, 6, 9 12 leds (1001001001001)
+ echo 1 > start # run
+
+
+.. _manual_breath_mode:
+
+Manual control mode
+~~~~~~~~~~~~~~~~~~~
+
+When 'ramp' enabled (echo 1 > ramp)::
+
+ breathing level
+ ^
+ max _ ____________________
+ | / .\
+ | / . \
+ | / . \
+ | / . \
+ | / . \
+ | / . \
+ min _ ______/ . \_______
+ | . .
+ | . .
+ | . .
+ | (echo 1 > toggle) (echo 0 > toggle)
+ -|---------------------------------------------------> time
+
+
+When 'ramp' disabled (echo 0 > ramp)::
+
+ breathing level
+ ^
+ max _ __________________________
+ | | |
+ | | |
+ | | |
+ | | |
+ | | |
+ | | |
+ min _ ______| |_______
+ | . .
+ | . .
+ | . .
+ | echo 1 > toggle echo 0 > toggle
+ -|---------------------------------------------------> time
diff --git a/drivers/leds/leds-aw200xx.c b/drivers/leds/leds-aw200xx.c
index 1d3943f86f7f..8350a1a85bab 100644
--- a/drivers/leds/leds-aw200xx.c
+++ b/drivers/leds/leds-aw200xx.c
@@ -20,6 +20,8 @@
#include <linux/time.h>
#include <linux/units.h>

+#define AW200XX_LEDS_MAX 108
+#define AW200XX_PATTERN_MAX 3
#define AW200XX_DIM_MAX (BIT(6) - 1)
#define AW200XX_FADE_MAX (BIT(8) - 1)
#define AW200XX_IMAX_DEFAULT_uA 60000
@@ -68,6 +70,88 @@
#define AW200XX_REG_FCD AW200XX_REG(AW200XX_PAGE0, 0x04)
#define AW200XX_FCD_CLEAR 0x01

+/* Interrupt status register */
+#define AW200XX_REG_ISRFLT AW200XX_REG(AW200XX_PAGE0, 0x0B)
+#define AW200XX_ISRFLT_PATIS_MASK GENMASK(6, 4)
+
+/* Pattern enable control register */
+#define AW200XX_REG_PATCR AW200XX_REG(AW200XX_PAGE0, 0x43)
+#define AW200XX_PATCR_PAT_IE_MASK GENMASK(6, 4)
+#define AW200XX_PATCR_PAT_IE_ALL AW200XX_PATCR_PAT_IE_MASK
+#define AW200XX_PATCR_PAT_ENABLE(x) BIT(x)
+
+/*
+ * Maximum breathing level registers
+ * For patterns 0 - 0x44, 1 - 0x45, 2 - 0x46 (step 1)
+ */
+#define AW200XX_REG_PAT0_MAX_BREATH AW200XX_REG(AW200XX_PAGE0, 0x44)
+
+/*
+ * Minimum breathing level registers
+ * For patterns 0 - 0x47, 1 - 0x48, 2 - 0x49 (step 1)
+ */
+#define AW200XX_REG_PAT0_MIN_BREATH AW200XX_REG(AW200XX_PAGE0, 0x47)
+
+/*
+ * Template 1 (rise-time) & template 2 (on-time) configuration register
+ * For patterns 0 - 0x4A, 1 - 0x4E, 2 - 0x52 (step 4)
+ */
+#define AW200XX_REG_PAT0_T0 AW200XX_REG(AW200XX_PAGE0, 0x4A)
+
+/*
+ * Template 3 (fall-time) & template 4 (off-time) configuration register
+ * For patterns 0 - 0x4B, 1 - 0x4F, 2 - 0x53 (step 4)
+ */
+#define AW200XX_REG_PAT0_T1 AW200XX_REG(AW200XX_PAGE0, 0x4B)
+
+/*
+ * Loop configuration registers:
+ * loop end point setting (LE)
+ * loop beginning point setting (LB)
+ * MSB of loop times (LT)
+ * For patterns 0 - 0x4C, 1 - 0x50, 2 - 0x54 (step 4)
+ */
+#define AW200XX_REG_PAT0_T2 AW200XX_REG(AW200XX_PAGE0, 0x4C)
+#define AW200XX_REG_PATX_T2(x) (AW200XX_REG_PAT0_T2 + (x))
+
+/*
+ * Loop configuration registers:
+ * LSB of loop times (LT)
+ * For patterns 0 - 0x4D, 1 - 0x51, 2 - 0x55 (step 4)
+ */
+#define AW200XX_REG_PAT0_T3 AW200XX_REG(AW200XX_PAGE0, 0x4D)
+#define AW200XX_REG_PATX_T3(x) (AW200XX_REG_PAT0_T3 + (x))
+
+#define AW200XX_PAT_T2_LE_MASK GENMASK(7, 6)
+#define AW200XX_PAT_T2_LB_MASK GENMASK(5, 4)
+#define AW200XX_PAT_T2_LT_MASK GENMASK(3, 0)
+#define AW200XX_PAT_T3_LT_MASK GENMASK(7, 0)
+#define AW200XX_PAT0_T2_LT_MSB(x) ((x) >> 8)
+#define AW200XX_PAT0_T3_LT_LSB(x) ((x) & 0xFF)
+#define AW200XX_PAT0_T_LT(msb, lsb) ((msb) << 8 | (lsb))
+#define AW200XX_PAT0_T_LT_MAX (BIT(12) - 1)
+
+#define AW200XX_PAT_T_STEP 4
+
+#define AW200XX_PAT_T1_T3_MASK GENMASK(7, 4)
+#define AW200XX_PAT_T2_T4_MASK GENMASK(3, 0)
+#define AW200XX_TEMPLATE_TIME_MAX (BIT(4) - 1)
+
+/*
+ * Pattern mode configuration register
+ * For patterns 0 - 0x56, 1 - 0x57, 2 - 0x58 (step 1)
+ */
+#define AW200XX_REG_PAT0_CFG AW200XX_REG(AW200XX_PAGE0, 0x56)
+#define AW200XX_PAT_CFG_MODE_MASK BIT(0)
+#define AW200XX_PAT_CFG_RAMP_MASK BIT(1)
+#define AW200XX_PAT_CFG_SWITCH_MASK BIT(2)
+
+/* Start pattern register */
+#define AW200XX_REG_PATGO AW200XX_REG(AW200XX_PAGE0, 0x59)
+#define AW200XX_PATGO(x) BIT(x)
+#define AW200XX_PATGO_RUN(x, run) ((run) << (x))
+#define AW200XX_PATGO_STATE(x) BIT((x) + 4)
+
/* Display size configuration */
#define AW200XX_REG_DSIZE AW200XX_REG(AW200XX_PAGE0, 0x80)
#define AW200XX_DSIZE_COLUMNS_MAX 12
@@ -79,6 +163,11 @@
#define AW200XX_REG_DIM_PAGE1(x, columns) \
AW200XX_REG(AW200XX_PAGE1, AW200XX_LED2REG(x, columns))

+/* Pattern selection register*/
+#define AW200XX_REG_PAT_SELECT(x, columns) \
+ AW200XX_REG(AW200XX_PAGE3, AW200XX_LED2REG(x, columns))
+#define AW200XX_PATX_SELECT(x) ((x) + 1)
+
/*
* DIM current configuration register (page 4).
* The even address for current DIM configuration.
@@ -117,12 +206,500 @@ struct aw200xx {
struct i2c_client *client;
struct regmap *regmap;
struct mutex mutex;
+ DECLARE_BITMAP(pattern_leds[AW200XX_PATTERN_MAX], AW200XX_LEDS_MAX);
u32 num_leds;
u32 display_rows;
struct gpio_desc *hwen;
struct aw200xx_led leds[] __counted_by(num_leds);
};

+struct aw200xx_attribute {
+ struct device_attribute dev_attr;
+ u32 reg;
+ u32 mask;
+ u32 max;
+};
+
+#define to_aw200xx_attr(attr) \
+ container_of(attr, struct aw200xx_attribute, dev_attr)
+
+#define AW200XX_ATTR(_n, _m, _sh, _st, _r, _msk, _max) { \
+ .dev_attr = __ATTR(_n, _m, _sh, _st), \
+ .reg = _r, \
+ .mask = _msk, \
+ .max = _max, \
+}
+
+#define AW200XX_DEVICE_ATTR_RW(_v, _n, _sh, _st, _r, _msk, _max) \
+struct aw200xx_attribute _v##_attr = \
+ AW200XX_ATTR(_n, 0644, _sh, _st, \
+ _r, _msk, _max)
+
+#define AW200XX_DEVICE_ATTR_RO(_v, _n, _sh, _r, _msk, _max) \
+struct aw200xx_attribute _v##_attr = \
+ AW200XX_ATTR(_n, 0444, _sh, NULL, \
+ _r, _msk, _max)
+
+static ssize_t aw200xx_store_internal(struct device *dev,
+ struct device_attribute *devattr,
+ const char *buf, size_t count)
+{
+ const struct aw200xx_attribute *attr = to_aw200xx_attr(devattr);
+ struct aw200xx *chip = i2c_get_clientdata(to_i2c_client(dev));
+ u32 val;
+ int ret;
+
+ ret = kstrtouint(buf, 0, &val);
+ if (ret < 0 || val > attr->max)
+ return -EINVAL;
+
+ val <<= __ffs(attr->mask);
+
+ mutex_lock(&chip->mutex);
+ ret = regmap_update_bits(chip->regmap, attr->reg, attr->mask, val);
+ mutex_unlock(&chip->mutex);
+
+ if (ret)
+ return ret;
+
+ return count;
+}
+
+static ssize_t aw200xx_show_internal(struct device *dev,
+ struct device_attribute *devattr,
+ char *buf)
+{
+ const struct aw200xx_attribute *attr = to_aw200xx_attr(devattr);
+ struct aw200xx *chip = i2c_get_clientdata(to_i2c_client(dev));
+ u32 val;
+ int ret;
+
+ mutex_lock(&chip->mutex);
+ ret = regmap_read(chip->regmap, attr->reg, &val);
+ mutex_unlock(&chip->mutex);
+
+ if (ret)
+ return ret;
+
+ val = (val & attr->mask) >> __ffs(attr->mask);
+
+ return sysfs_emit(buf, "%u\n", val);
+}
+
+static ssize_t aw200xx_template_time_show(struct device *dev,
+ struct device_attribute *devattr,
+ char *buf)
+{
+ static const u32 ttimes_ms[] = {
+ 0, 130, 260, 380, 510, 770, 1040, 1600,
+ 2100, 2600, 3100, 4200, 5200, 6200, 7300, 8300,
+ };
+ const struct aw200xx_attribute *attr = to_aw200xx_attr(devattr);
+ struct aw200xx *chip = i2c_get_clientdata(to_i2c_client(dev));
+ u32 ttime;
+ int ret;
+
+ mutex_lock(&chip->mutex);
+ ret = regmap_read(chip->regmap, attr->reg, &ttime);
+ mutex_unlock(&chip->mutex);
+
+ if (ret)
+ return ret;
+
+ ttime = (ttime & attr->mask) >> __ffs(attr->mask);
+ if (ttime >= ARRAY_SIZE(ttimes_ms))
+ return -EIO;
+
+ ttime = ttimes_ms[ttime];
+
+ /* For On & Off time minimum is 40ms */
+ if (ttime == 0 && attr->mask == AW200XX_PAT_T2_T4_MASK)
+ ttime = 40;
+
+ return sysfs_emit(buf, "%ums\n", ttime);
+}
+
+static ssize_t aw200xx_pattern_leds_store(struct device *dev,
+ struct device_attribute *devattr,
+ const char *buf, size_t count,
+ bool clear)
+{
+ const struct aw200xx_attribute *attr = to_aw200xx_attr(devattr);
+ struct aw200xx *chip = i2c_get_clientdata(to_i2c_client(dev));
+ unsigned long *pattern_leds = chip->pattern_leds[attr->reg];
+ u32 columns = chip->cdef->display_size_columns;
+ DECLARE_BITMAP(leds, AW200XX_LEDS_MAX);
+ u32 val = clear ? 0 : AW200XX_PATX_SELECT(attr->reg);
+ u32 i;
+ ssize_t ret;
+
+ ret = bitmap_parse(buf, count, leds, chip->num_leds);
+ if (ret)
+ return -EINVAL;
+
+ mutex_lock(&chip->mutex);
+
+ for_each_set_bit(i, leds, chip->num_leds) {
+ u32 num = chip->leds[i].num;
+
+ ret = regmap_write(chip->regmap,
+ AW200XX_REG_PAT_SELECT(num, columns), val);
+ if (ret)
+ goto out_unlock;
+
+ if (clear)
+ __clear_bit(i, pattern_leds);
+ else
+ __set_bit(i, pattern_leds);
+ }
+
+ ret = count;
+
+out_unlock:
+ mutex_unlock(&chip->mutex);
+ return ret;
+}
+
+static ssize_t aw200xx_pattern_select_leds_show(struct device *dev,
+ struct device_attribute *devattr,
+ char *buf)
+{
+ const struct aw200xx_attribute *attr = to_aw200xx_attr(devattr);
+ struct aw200xx *chip = i2c_get_clientdata(to_i2c_client(dev));
+ int ret;
+
+ mutex_lock(&chip->mutex);
+ ret = sysfs_emit(buf, "%*pb\n",
+ chip->num_leds, chip->pattern_leds[attr->reg]);
+ mutex_unlock(&chip->mutex);
+
+ return ret;
+}
+
+static ssize_t aw200xx_pattern_select_leds_store(struct device *dev,
+ struct device_attribute *devattr,
+ const char *buf, size_t count)
+{
+ return aw200xx_pattern_leds_store(dev, devattr, buf, count, false);
+}
+
+static ssize_t aw200xx_pattern_clear_leds_show(struct device *dev,
+ struct device_attribute *devattr,
+ char *buf)
+{
+ const struct aw200xx_attribute *attr = to_aw200xx_attr(devattr);
+ struct aw200xx *chip = i2c_get_clientdata(to_i2c_client(dev));
+ DECLARE_BITMAP(leds, AW200XX_LEDS_MAX);
+ int ret;
+
+ mutex_lock(&chip->mutex);
+ bitmap_fill(leds, chip->num_leds);
+ bitmap_xor(leds, leds, chip->pattern_leds[attr->reg], chip->num_leds);
+ ret = sysfs_emit(buf, "%*pb\n", chip->num_leds, leds);
+ mutex_unlock(&chip->mutex);
+
+ return ret;
+}
+
+static ssize_t aw200xx_pattern_clear_leds_store(struct device *dev,
+ struct device_attribute *devattr,
+ const char *buf, size_t count)
+{
+ return aw200xx_pattern_leds_store(dev, devattr, buf, count, true);
+}
+
+static ssize_t aw200xx_pattern_start_show(struct device *dev,
+ struct device_attribute *devattr,
+ char *buf)
+{
+ const struct aw200xx_attribute *attr = to_aw200xx_attr(devattr);
+ struct aw200xx *chip = i2c_get_clientdata(to_i2c_client(dev));
+ u32 start = 0;
+ u32 val;
+ int ret;
+
+ mutex_lock(&chip->mutex);
+
+ ret = regmap_read(chip->regmap, AW200XX_REG_PATCR, &val);
+ if (ret)
+ goto out_unlock;
+
+ if (val & AW200XX_PATCR_PAT_ENABLE(attr->reg)) {
+ ret = regmap_read(chip->regmap, AW200XX_REG_PATGO, &val);
+ if (ret)
+ goto out_unlock;
+
+ start = !!(val & AW200XX_PATGO(attr->reg));
+ }
+
+ ret = sysfs_emit(buf, "%u\n", start);
+
+out_unlock:
+ mutex_unlock(&chip->mutex);
+ return ret;
+}
+
+static ssize_t aw200xx_pattern_start_store(struct device *dev,
+ struct device_attribute *devattr,
+ const char *buf, size_t count)
+{
+ const struct aw200xx_attribute *attr = to_aw200xx_attr(devattr);
+ struct aw200xx *chip = i2c_get_clientdata(to_i2c_client(dev));
+ u32 start;
+ ssize_t ret;
+
+ ret = kstrtouint(buf, 0, &start);
+ if (ret < 0 || start > attr->max)
+ return -EINVAL;
+
+ start = AW200XX_PATGO_RUN(attr->reg, start);
+
+ mutex_lock(&chip->mutex);
+
+ ret = regmap_update_bits(chip->regmap, AW200XX_REG_PATCR,
+ AW200XX_PATCR_PAT_ENABLE(attr->reg), start);
+ if (ret)
+ goto out_unlock;
+
+ ret = regmap_update_bits(chip->regmap, AW200XX_REG_PATGO,
+ AW200XX_PATGO(attr->reg), start);
+ if (ret)
+ goto out_unlock;
+
+ ret = count;
+
+out_unlock:
+ mutex_unlock(&chip->mutex);
+ return ret;
+}
+
+static ssize_t aw200xx_pattern_running_show(struct device *dev,
+ struct device_attribute *devattr,
+ char *buf)
+{
+ const struct aw200xx_attribute *attr = to_aw200xx_attr(devattr);
+ struct aw200xx *chip = i2c_get_clientdata(to_i2c_client(dev));
+ u32 running;
+ int ret;
+
+ mutex_lock(&chip->mutex);
+
+ ret = regmap_read(chip->regmap, AW200XX_REG_PATGO, &running);
+ if (ret)
+ goto out_unlock;
+
+ running &= AW200XX_PATGO_STATE(attr->reg);
+ ret = sysfs_emit(buf, "%u\n", !!running);
+
+out_unlock:
+ mutex_unlock(&chip->mutex);
+ return ret;
+}
+
+static ssize_t aw200xx_pattern_repeat_show(struct device *dev,
+ struct device_attribute *devattr,
+ char *buf)
+{
+ const struct aw200xx_attribute *attr = to_aw200xx_attr(devattr);
+ struct aw200xx *chip = i2c_get_clientdata(to_i2c_client(dev));
+ u32 repeat_msb;
+ u32 repeat_lsb;
+ int ret;
+
+ mutex_lock(&chip->mutex);
+
+ ret = regmap_read(chip->regmap,
+ AW200XX_REG_PATX_T2(attr->reg), &repeat_msb);
+ if (ret)
+ goto out_unlock;
+
+ ret = regmap_read(chip->regmap,
+ AW200XX_REG_PATX_T3(attr->reg), &repeat_lsb);
+ if (ret)
+ goto out_unlock;
+
+ repeat_msb &= AW200XX_PAT_T2_LT_MASK;
+ repeat_lsb &= AW200XX_PAT_T3_LT_MASK;
+
+ ret = sysfs_emit(buf, "%u\n",
+ AW200XX_PAT0_T_LT(repeat_msb, repeat_lsb));
+
+out_unlock:
+ mutex_unlock(&chip->mutex);
+ return ret;
+}
+
+static ssize_t aw200xx_pattern_repeat_store(struct device *dev,
+ struct device_attribute *devattr,
+ const char *buf, size_t count)
+{
+ const struct aw200xx_attribute *attr = to_aw200xx_attr(devattr);
+ struct aw200xx *chip = i2c_get_clientdata(to_i2c_client(dev));
+ u32 repeat;
+ ssize_t ret;
+
+ ret = kstrtouint(buf, 0, &repeat);
+ if (ret < 0 || repeat > attr->max)
+ return -EINVAL;
+
+ mutex_lock(&chip->mutex);
+
+ ret = regmap_update_bits(chip->regmap,
+ AW200XX_REG_PATX_T2(attr->reg),
+ AW200XX_PAT_T2_LT_MASK,
+ AW200XX_PAT0_T2_LT_MSB(repeat));
+ if (ret)
+ goto out_unlock;
+
+ ret = regmap_update_bits(chip->regmap,
+ AW200XX_REG_PATX_T3(attr->reg),
+ AW200XX_PAT_T3_LT_MASK,
+ AW200XX_PAT0_T3_LT_LSB(repeat));
+ if (ret)
+ goto out_unlock;
+
+ ret = count;
+
+out_unlock:
+ mutex_unlock(&chip->mutex);
+ return ret;
+}
+
+#define AW200XX_DEVICE_ATTR_PAT_RW(_n, _sh, _st, _r, _step, _msk, _max) \
+static AW200XX_DEVICE_ATTR_RW(_n##0, _n, _sh, _st, \
+ _r, _msk, _max); \
+static AW200XX_DEVICE_ATTR_RW(_n##1, _n, _sh, _st, \
+ _r + (1 * (_step)), _msk, _max); \
+static AW200XX_DEVICE_ATTR_RW(_n##2, _n, _sh, _st, \
+ _r + (2 * (_step)), _msk, _max)
+
+#define AW200XX_DEVICE_ATTR_PAT_RO(_n, _sh, _r, _step, _msk, _max) \
+static AW200XX_DEVICE_ATTR_RO(_n##0, _n, _sh, \
+ _r, _msk, _max); \
+static AW200XX_DEVICE_ATTR_RO(_n##1, _n, _sh, \
+ _r + (1 * (_step)), _msk, _max); \
+static AW200XX_DEVICE_ATTR_RO(_n##2, _n, _sh, \
+ _r + (2 * (_step)), _msk, _max)
+
+#define AW200XX_DEFINE_ATTR_GROUP(_idx, _a0, _a1, _a2, _a3, _a4, _a5, \
+ _a6, _a7, _a8, _a9, _a10, _a11, _a12, _a13, _a14, _a15) \
+static struct attribute *aw200xx_pattern##_idx##_attributes[] = { \
+ &_a0##_idx##_attr.dev_attr.attr, \
+ &_a1##_idx##_attr.dev_attr.attr, \
+ &_a2##_idx##_attr.dev_attr.attr, \
+ &_a3##_idx##_attr.dev_attr.attr, \
+ &_a4##_idx##_attr.dev_attr.attr, \
+ &_a5##_idx##_attr.dev_attr.attr, \
+ &_a6##_idx##_attr.dev_attr.attr, \
+ &_a7##_idx##_attr.dev_attr.attr, \
+ &_a8##_idx##_attr.dev_attr.attr, \
+ &_a9##_idx##_attr.dev_attr.attr, \
+ &_a10##_idx##_attr.dev_attr.attr, \
+ &_a11##_idx##_attr.dev_attr.attr, \
+ &_a12##_idx##_attr.dev_attr.attr, \
+ &_a13##_idx##_attr.dev_attr.attr, \
+ &_a14##_idx##_attr.dev_attr.attr, \
+ &_a15##_idx##_attr.dev_attr.attr, \
+ NULL}; \
+static const struct attribute_group aw200xx_pattern##_idx##_group = { \
+ .attrs = aw200xx_pattern##_idx##_attributes, \
+ .name = "pattern"#_idx, \
+}
+
+#define AW200XX_DEFINE_ATTR_GROUPS(...) \
+AW200XX_DEFINE_ATTR_GROUP(0, __VA_ARGS__); \
+AW200XX_DEFINE_ATTR_GROUP(1, __VA_ARGS__); \
+AW200XX_DEFINE_ATTR_GROUP(2, __VA_ARGS__); \
+static const struct attribute_group *aw200xx_pattern_groups[] = { \
+ &aw200xx_pattern0_group, \
+ &aw200xx_pattern1_group, \
+ &aw200xx_pattern2_group, \
+ NULL}
+
+AW200XX_DEVICE_ATTR_PAT_RW(rise_time,
+ aw200xx_template_time_show, aw200xx_store_internal,
+ AW200XX_REG_PAT0_T0, AW200XX_PAT_T_STEP,
+ AW200XX_PAT_T1_T3_MASK, AW200XX_TEMPLATE_TIME_MAX);
+
+AW200XX_DEVICE_ATTR_PAT_RW(on_time,
+ aw200xx_template_time_show, aw200xx_store_internal,
+ AW200XX_REG_PAT0_T0, AW200XX_PAT_T_STEP,
+ AW200XX_PAT_T2_T4_MASK, AW200XX_TEMPLATE_TIME_MAX);
+
+AW200XX_DEVICE_ATTR_PAT_RW(fall_time,
+ aw200xx_template_time_show, aw200xx_store_internal,
+ AW200XX_REG_PAT0_T1, AW200XX_PAT_T_STEP,
+ AW200XX_PAT_T1_T3_MASK, AW200XX_TEMPLATE_TIME_MAX);
+
+AW200XX_DEVICE_ATTR_PAT_RW(off_time,
+ aw200xx_template_time_show, aw200xx_store_internal,
+ AW200XX_REG_PAT0_T1, AW200XX_PAT_T_STEP,
+ AW200XX_PAT_T2_T4_MASK, AW200XX_TEMPLATE_TIME_MAX);
+
+AW200XX_DEVICE_ATTR_PAT_RW(mode,
+ aw200xx_show_internal, aw200xx_store_internal,
+ AW200XX_REG_PAT0_CFG, 1,
+ AW200XX_PAT_CFG_MODE_MASK, 1);
+
+AW200XX_DEVICE_ATTR_PAT_RW(ramp,
+ aw200xx_show_internal, aw200xx_store_internal,
+ AW200XX_REG_PAT0_CFG, 1,
+ AW200XX_PAT_CFG_RAMP_MASK, 1);
+
+AW200XX_DEVICE_ATTR_PAT_RW(toggle,
+ aw200xx_show_internal, aw200xx_store_internal,
+ AW200XX_REG_PAT0_CFG, 1,
+ AW200XX_PAT_CFG_SWITCH_MASK, 1);
+
+AW200XX_DEVICE_ATTR_PAT_RW(loop_end_on,
+ aw200xx_show_internal, aw200xx_store_internal,
+ AW200XX_REG_PAT0_T2, AW200XX_PAT_T_STEP,
+ AW200XX_PAT_T2_LE_MASK, 1);
+
+AW200XX_DEVICE_ATTR_PAT_RW(loop_begin,
+ aw200xx_show_internal, aw200xx_store_internal,
+ AW200XX_REG_PAT0_T2, AW200XX_PAT_T_STEP,
+ AW200XX_PAT_T2_LB_MASK, 3);
+
+AW200XX_DEVICE_ATTR_PAT_RW(max_breathing_level,
+ aw200xx_show_internal, aw200xx_store_internal,
+ AW200XX_REG_PAT0_MAX_BREATH, 1,
+ 0xFF, AW200XX_FADE_MAX);
+
+AW200XX_DEVICE_ATTR_PAT_RW(min_breathing_level,
+ aw200xx_show_internal, aw200xx_store_internal,
+ AW200XX_REG_PAT0_MIN_BREATH, 1,
+ 0xFF, AW200XX_FADE_MAX);
+
+AW200XX_DEVICE_ATTR_PAT_RW(start,
+ aw200xx_pattern_start_show,
+ aw200xx_pattern_start_store,
+ 0, 1, 1, 1);
+
+AW200XX_DEVICE_ATTR_PAT_RO(running,
+ aw200xx_pattern_running_show, 0, 1, 0, 0);
+
+AW200XX_DEVICE_ATTR_PAT_RW(repeat,
+ aw200xx_pattern_repeat_show,
+ aw200xx_pattern_repeat_store,
+ 0, AW200XX_PAT_T_STEP,
+ 0, AW200XX_PAT0_T_LT_MAX);
+
+AW200XX_DEVICE_ATTR_PAT_RW(select_leds,
+ aw200xx_pattern_select_leds_show,
+ aw200xx_pattern_select_leds_store,
+ 0, 1, 0, 0);
+
+AW200XX_DEVICE_ATTR_PAT_RW(clear_leds,
+ aw200xx_pattern_clear_leds_show,
+ aw200xx_pattern_clear_leds_store,
+ 0, 1, 0, 0);
+
+AW200XX_DEFINE_ATTR_GROUPS(start, running, mode, ramp, toggle, repeat,
+ loop_end_on, loop_begin, select_leds, clear_leds,
+ max_breathing_level, min_breathing_level,
+ rise_time, on_time, fall_time, off_time);
+
static ssize_t dim_show(struct device *dev, struct device_attribute *devattr,
char *buf)
{
@@ -212,6 +789,62 @@ static int aw200xx_brightness_set(struct led_classdev *cdev,
return ret;
}

+static irqreturn_t aw200xx_irq_thread(int irq, void *dev_id)
+{
+ struct aw200xx *chip = dev_id;
+ unsigned long pattern_state;
+ u32 interrupt_state;
+ int i;
+ int ret;
+
+ mutex_lock(&chip->mutex);
+ ret = regmap_read(chip->regmap, AW200XX_REG_ISRFLT, &interrupt_state);
+ mutex_unlock(&chip->mutex);
+
+ if (ret) {
+ dev_err(&chip->client->dev,
+ "Failed to get interrupt status: %d\n", ret);
+ return IRQ_HANDLED;
+ }
+
+ pattern_state = FIELD_GET(AW200XX_ISRFLT_PATIS_MASK, interrupt_state);
+
+ for_each_set_bit(i, &pattern_state, AW200XX_PATTERN_MAX) {
+ char dir[sizeof("patternx")];
+
+ snprintf(dir, sizeof(dir), "pattern%d", i);
+ sysfs_notify(&chip->client->dev.kobj, dir, "running");
+ }
+
+ return IRQ_HANDLED;
+}
+
+static int aw200xx_setup_interrupts(struct aw200xx *chip)
+{
+ struct i2c_client *i2c = chip->client;
+ int ret;
+
+ if (i2c->irq <= 0)
+ return 0;
+
+ ret = devm_request_threaded_irq(&i2c->dev, i2c->irq,
+ NULL,
+ aw200xx_irq_thread,
+ IRQF_ONESHOT,
+ i2c->name,
+ chip);
+ if (ret)
+ return dev_err_probe(&i2c->dev, ret, "Failed to request irq\n");
+
+ ret = regmap_update_bits(chip->regmap, AW200XX_REG_PATCR,
+ AW200XX_PATCR_PAT_IE_MASK,
+ AW200XX_PATCR_PAT_IE_ALL);
+ if (ret)
+ dev_err_probe(&i2c->dev, ret, "Failed to enable interrupts\n");
+
+ return ret;
+}
+
static u32 aw200xx_imax_from_global(const struct aw200xx *const chip,
u32 global_imax_uA)
{
@@ -511,6 +1144,7 @@ static const struct regmap_access_table aw200xx_readable_table = {

static const struct regmap_range aw200xx_readonly_ranges[] = {
regmap_reg_range(AW200XX_REG_IDR, AW200XX_REG_IDR),
+ regmap_reg_range(AW200XX_REG_ISRFLT, AW200XX_REG_ISRFLT),
};

static const struct regmap_access_table aw200xx_writeable_table = {
@@ -518,6 +1152,15 @@ static const struct regmap_access_table aw200xx_writeable_table = {
.n_no_ranges = ARRAY_SIZE(aw200xx_readonly_ranges),
};

+static const struct regmap_range aw200xx_volatile_registers[] = {
+ regmap_reg_range(AW200XX_REG_ISRFLT, AW200XX_REG_ISRFLT),
+};
+
+static const struct regmap_access_table aw200xx_volatile_table = {
+ .yes_ranges = aw200xx_volatile_registers,
+ .n_yes_ranges = ARRAY_SIZE(aw200xx_volatile_registers),
+};
+
static const struct regmap_config aw200xx_regmap_config = {
.reg_bits = 8,
.val_bits = 8,
@@ -526,6 +1169,7 @@ static const struct regmap_config aw200xx_regmap_config = {
.num_ranges = ARRAY_SIZE(aw200xx_ranges),
.rd_table = &aw200xx_readable_table,
.wr_table = &aw200xx_writeable_table,
+ .volatile_table = &aw200xx_volatile_table,
.cache_type = REGCACHE_MAPLE,
.disable_locking = true,
};
@@ -586,6 +1230,10 @@ static int aw200xx_probe(struct i2c_client *client)
goto out_unlock;

ret = aw200xx_chip_init(chip);
+ if (ret)
+ goto out_unlock;
+
+ ret = aw200xx_setup_interrupts(chip);

out_unlock:
if (ret)
@@ -650,6 +1298,7 @@ static struct i2c_driver aw200xx_driver = {
.driver = {
.name = "aw200xx",
.of_match_table = aw200xx_match_table,
+ .dev_groups = aw200xx_pattern_groups,
},
.probe_new = aw200xx_probe,
.remove = aw200xx_remove,
--
2.40.0

2023-12-07 17:18:51

by Conor Dooley

[permalink] [raw]
Subject: Re: [PATCH v1 2/2] dt-bindings: leds: aw200xx: add 'interrupts' property

On Thu, Dec 07, 2023 at 03:59:38PM +0300, Martin Kurbanov wrote:
> Added support for interrupt to the LED controller driver, which
> indicates the finished of the auto-breath pattern.
>
> Signed-off-by: Martin Kurbanov <[email protected]>

Acked-by: Conor Dooley <[email protected]>

Cheers,
Conor.

> ---
> Documentation/devicetree/bindings/leds/awinic,aw200xx.yaml | 6 ++++++
> 1 file changed, 6 insertions(+)
>
> diff --git a/Documentation/devicetree/bindings/leds/awinic,aw200xx.yaml b/Documentation/devicetree/bindings/leds/awinic,aw200xx.yaml
> index 54d6d1f08e24..7ab35b7a3fe6 100644
> --- a/Documentation/devicetree/bindings/leds/awinic,aw200xx.yaml
> +++ b/Documentation/devicetree/bindings/leds/awinic,aw200xx.yaml
> @@ -44,6 +44,9 @@ properties:
> enable-gpios:
> maxItems: 1
>
> + interrupts:
> + maxItems: 1
> +
> patternProperties:
> "^led@[0-9a-f]+$":
> type: object
> @@ -137,6 +140,7 @@ additionalProperties: false
> examples:
> - |
> #include <dt-bindings/gpio/gpio.h>
> + #include <dt-bindings/interrupt-controller/irq.h>
> #include <dt-bindings/leds/common.h>
>
> i2c {
> @@ -149,6 +153,8 @@ examples:
> #address-cells = <1>;
> #size-cells = <0>;
> enable-gpios = <&gpio 3 GPIO_ACTIVE_HIGH>;
> + interrupt-parent = <&gpio_intc>;
> + interrupts = <13 IRQ_TYPE_LEVEL_LOW>;
>
> led@0 {
> reg = <0x0>;
> --
> 2.40.0
>


Attachments:
(No filename) (1.51 kB)
signature.asc (235.00 B)
Download all attachments

2023-12-21 16:14:37

by Lee Jones

[permalink] [raw]
Subject: Re: [PATCH v1 1/2] leds: aw200xx: support for hw pattern controllers

On Thu, 07 Dec 2023, Martin Kurbanov wrote:

> This led-controller supports 3 pattern controllers for auto breathing or
> group dimming control. Each pattern controller can work in auto
> breathing or manual control mode. All breathing parameters including
> rising/falling slope, on/off time, repeat times, min/max brightness
> and so on are configurable.
>
> Signed-off-by: Martin Kurbanov <[email protected]>
> ---
> .../testing/sysfs-class-led-driver-aw200xx | 108 +++
> Documentation/leds/leds-aw200xx.rst | 274 ++++++++
> drivers/leds/leds-aw200xx.c | 649 ++++++++++++++++++
> 3 files changed, 1031 insertions(+)
> create mode 100644 Documentation/leds/leds-aw200xx.rst

This interface is bananas. Exposing an entire register interface to
sysfs does not sit will with me at all. When we add support to a sysfs
class, we usually require it to be generic and work across all devices.
Adding device specific interfaces is generally decried and to be
avoided. Don't forget, once we commit something to sysfs, it becomes
ABI and we have to support it forever.

A far better approach would be to add support for this in userspace
instead You can use the standard I2C character device API to achieve
the same result. That way we don't have the same level of commitment
and is generally a much more flexible/future-proof.

--
Lee Jones [李琼斯]

2024-01-12 11:43:41

by Martin Kurbanov

[permalink] [raw]
Subject: Re: [PATCH v1 1/2] leds: aw200xx: support for hw pattern controllers

Hello Lee,

On 21.12.2023 19:10, Lee Jones wrote:
> On Thu, 07 Dec 2023, Martin Kurbanov wrote:
>
>> This led-controller supports 3 pattern controllers for auto breathing or
>> group dimming control. Each pattern controller can work in auto
>> breathing or manual control mode. All breathing parameters including
>> rising/falling slope, on/off time, repeat times, min/max brightness
>> and so on are configurable.
>>
>> Signed-off-by: Martin Kurbanov <[email protected]>
>> ---
>> .../testing/sysfs-class-led-driver-aw200xx | 108 +++
>> Documentation/leds/leds-aw200xx.rst | 274 ++++++++
>> drivers/leds/leds-aw200xx.c | 649 ++++++++++++++++++
>> 3 files changed, 1031 insertions(+)
>> create mode 100644 Documentation/leds/leds-aw200xx.rst
>
> This interface is bananas. Exposing an entire register interface to
> sysfs does not sit will with me at all. When we add support to a sysfs
> class, we usually require it to be generic and work across all devices.
> Adding device specific interfaces is generally decried and to be
> avoided. Don't forget, once we commit something to sysfs, it becomes
> ABI and we have to support it forever.
>
> A far better approach would be to add support for this in userspace
> instead You can use the standard I2C character device API to achieve
> the same result. That way we don't have the same level of commitment
> and is generally a much more flexible/future-proof.
>

I used sysfs similarly to other LED drivers (for example, leds-lm3533).
Additionally, the controller has interrupts about the completion of the pattern,
which is best to handle in the kernel. In the case of implementation in user
mode, there may be synchronization problems, as the controller has several
memory pages that can be switched by writing the page number to register 0xF0.

--
Best Regards,
Martin Kurbanov

2024-01-25 13:01:04

by Lee Jones

[permalink] [raw]
Subject: Re: [PATCH v1 1/2] leds: aw200xx: support for hw pattern controllers

Looping in Jacek (LEDS) and Greg (SYFS) for some knowledgable input.

On Fri, 12 Jan 2024, Martin Kurbanov wrote:
> On 21.12.2023 19:10, Lee Jones wrote:
> > On Thu, 07 Dec 2023, Martin Kurbanov wrote:
> >
> >> This led-controller supports 3 pattern controllers for auto breathing or
> >> group dimming control. Each pattern controller can work in auto
> >> breathing or manual control mode. All breathing parameters including
> >> rising/falling slope, on/off time, repeat times, min/max brightness
> >> and so on are configurable.
> >>
> >> Signed-off-by: Martin Kurbanov <[email protected]>
> >> ---
> >> .../testing/sysfs-class-led-driver-aw200xx | 108 +++
> >> Documentation/leds/leds-aw200xx.rst | 274 ++++++++
> >> drivers/leds/leds-aw200xx.c | 649 ++++++++++++++++++
> >> 3 files changed, 1031 insertions(+)
> >> create mode 100644 Documentation/leds/leds-aw200xx.rst
> >
> > This interface is bananas. Exposing an entire register interface to
> > sysfs does not sit will with me at all. When we add support to a sysfs
> > class, we usually require it to be generic and work across all devices.
> > Adding device specific interfaces is generally decried and to be
> > avoided. Don't forget, once we commit something to sysfs, it becomes
> > ABI and we have to support it forever.
> >
> > A far better approach would be to add support for this in userspace
> > instead You can use the standard I2C character device API to achieve
> > the same result. That way we don't have the same level of commitment
> > and is generally a much more flexible/future-proof.
> >
>
> I used sysfs similarly to other LED drivers (for example, leds-lm3533).
> Additionally, the controller has interrupts about the completion of the pattern,
> which is best to handle in the kernel. In the case of implementation in user
> mode, there may be synchronization problems, as the controller has several
> memory pages that can be switched by writing the page number to register 0xF0.

leds-lm3533 is a 12 year old legacy exception AND has less than half of
the sysfs exports proposed here. What makes aw200xx so different it
needs to an incomparable interface to any other that we currently
support?

--
Lee Jones [李琼斯]

2024-01-26 12:23:29

by Dmitry Rokosov

[permalink] [raw]
Subject: Re: [PATCH v1 1/2] leds: aw200xx: support for hw pattern controllers

Hello Lee,

On Thu, Jan 25, 2024 at 01:00:49PM +0000, Lee Jones wrote:
> Looping in Jacek (LEDS) and Greg (SYFS) for some knowledgable input.
>
> On Fri, 12 Jan 2024, Martin Kurbanov wrote:
> > On 21.12.2023 19:10, Lee Jones wrote:
> > > On Thu, 07 Dec 2023, Martin Kurbanov wrote:
> > >
> > >> This led-controller supports 3 pattern controllers for auto breathing or
> > >> group dimming control. Each pattern controller can work in auto
> > >> breathing or manual control mode. All breathing parameters including
> > >> rising/falling slope, on/off time, repeat times, min/max brightness
> > >> and so on are configurable.
> > >>
> > >> Signed-off-by: Martin Kurbanov <[email protected]>
> > >> ---
> > >> .../testing/sysfs-class-led-driver-aw200xx | 108 +++
> > >> Documentation/leds/leds-aw200xx.rst | 274 ++++++++
> > >> drivers/leds/leds-aw200xx.c | 649 ++++++++++++++++++
> > >> 3 files changed, 1031 insertions(+)
> > >> create mode 100644 Documentation/leds/leds-aw200xx.rst
> > >
> > > This interface is bananas. Exposing an entire register interface to
> > > sysfs does not sit will with me at all. When we add support to a sysfs
> > > class, we usually require it to be generic and work across all devices.
> > > Adding device specific interfaces is generally decried and to be
> > > avoided. Don't forget, once we commit something to sysfs, it becomes
> > > ABI and we have to support it forever.
> > >
> > > A far better approach would be to add support for this in userspace
> > > instead You can use the standard I2C character device API to achieve
> > > the same result. That way we don't have the same level of commitment
> > > and is generally a much more flexible/future-proof.
> > >
> >
> > I used sysfs similarly to other LED drivers (for example, leds-lm3533).
> > Additionally, the controller has interrupts about the completion of the pattern,
> > which is best to handle in the kernel. In the case of implementation in user
> > mode, there may be synchronization problems, as the controller has several
> > memory pages that can be switched by writing the page number to register 0xF0.
>
> leds-lm3533 is a 12 year old legacy exception AND has less than half of
> the sysfs exports proposed here. What makes aw200xx so different it
> needs to an incomparable interface to any other that we currently
> support?

From my point of view, direct I2C raw requests from userspace are not a
good solution as well due to synchronization problems, as Martin
mentioned in the previous message.

We have honestly been attempting to integrate this functionality into
the official LED pattern interface, but it cannot be achieved due to the
absence of this interface's functionality:
1) Page-based access
2) Interrupts

HW patterns are very useful mechanism to draw animation faster without
any interactions with CPU, so I think we need to find the best architect
approach for its integration.

What is an alternative way to access such a hardware pattern interface?
Debugfs? Or perhaps we should consider extending the LED pattern
interface?

--
Thank you,
Dmitry

2024-01-26 14:09:53

by Greg Kroah-Hartman

[permalink] [raw]
Subject: Re: [PATCH v1 1/2] leds: aw200xx: support for hw pattern controllers

On Fri, Jan 26, 2024 at 03:23:10PM +0300, Dmitry Rokosov wrote:
> Hello Lee,
>
> On Thu, Jan 25, 2024 at 01:00:49PM +0000, Lee Jones wrote:
> > Looping in Jacek (LEDS) and Greg (SYFS) for some knowledgable input.
> >
> > On Fri, 12 Jan 2024, Martin Kurbanov wrote:
> > > On 21.12.2023 19:10, Lee Jones wrote:
> > > > On Thu, 07 Dec 2023, Martin Kurbanov wrote:
> > > >
> > > >> This led-controller supports 3 pattern controllers for auto breathing or
> > > >> group dimming control. Each pattern controller can work in auto
> > > >> breathing or manual control mode. All breathing parameters including
> > > >> rising/falling slope, on/off time, repeat times, min/max brightness
> > > >> and so on are configurable.
> > > >>
> > > >> Signed-off-by: Martin Kurbanov <[email protected]>
> > > >> ---
> > > >> .../testing/sysfs-class-led-driver-aw200xx | 108 +++
> > > >> Documentation/leds/leds-aw200xx.rst | 274 ++++++++
> > > >> drivers/leds/leds-aw200xx.c | 649 ++++++++++++++++++
> > > >> 3 files changed, 1031 insertions(+)
> > > >> create mode 100644 Documentation/leds/leds-aw200xx.rst
> > > >
> > > > This interface is bananas. Exposing an entire register interface to
> > > > sysfs does not sit will with me at all. When we add support to a sysfs
> > > > class, we usually require it to be generic and work across all devices.
> > > > Adding device specific interfaces is generally decried and to be
> > > > avoided. Don't forget, once we commit something to sysfs, it becomes
> > > > ABI and we have to support it forever.
> > > >
> > > > A far better approach would be to add support for this in userspace
> > > > instead You can use the standard I2C character device API to achieve
> > > > the same result. That way we don't have the same level of commitment
> > > > and is generally a much more flexible/future-proof.
> > > >
> > >
> > > I used sysfs similarly to other LED drivers (for example, leds-lm3533).
> > > Additionally, the controller has interrupts about the completion of the pattern,
> > > which is best to handle in the kernel. In the case of implementation in user
> > > mode, there may be synchronization problems, as the controller has several
> > > memory pages that can be switched by writing the page number to register 0xF0.
> >
> > leds-lm3533 is a 12 year old legacy exception AND has less than half of
> > the sysfs exports proposed here. What makes aw200xx so different it
> > needs to an incomparable interface to any other that we currently
> > support?
>
> >From my point of view, direct I2C raw requests from userspace are not a
> good solution as well due to synchronization problems, as Martin
> mentioned in the previous message.

Sorry, I missed this, what is the synchronization problem? This is an
led, shouldn't have any real specific performance issues.

> We have honestly been attempting to integrate this functionality into
> the official LED pattern interface, but it cannot be achieved due to the
> absence of this interface's functionality:
> 1) Page-based access
> 2) Interrupts

I don't understand this, sorry.

> HW patterns are very useful mechanism to draw animation faster without
> any interactions with CPU, so I think we need to find the best architect
> approach for its integration.

The CPU is totally involved here, that should be identical.

> What is an alternative way to access such a hardware pattern interface?
> Debugfs? Or perhaps we should consider extending the LED pattern
> interface?

Or again, userspace.

debugfs is for debugging.

thanks,

greg k-h

2024-01-26 23:33:38

by Pavel Machek

[permalink] [raw]
Subject: Re: [PATCH v1 1/2] leds: aw200xx: support for hw pattern controllers

Hi!

> > This led-controller supports 3 pattern controllers for auto breathing or
> > group dimming control. Each pattern controller can work in auto
> > breathing or manual control mode. All breathing parameters including
> > rising/falling slope, on/off time, repeat times, min/max brightness
> > and so on are configurable.
> >
> > Signed-off-by: Martin Kurbanov <[email protected]>
> > ---
> > .../testing/sysfs-class-led-driver-aw200xx | 108 +++
> > Documentation/leds/leds-aw200xx.rst | 274 ++++++++
> > drivers/leds/leds-aw200xx.c | 649 ++++++++++++++++++
> > 3 files changed, 1031 insertions(+)
> > create mode 100644 Documentation/leds/leds-aw200xx.rst
>
> This interface is bananas. Exposing an entire register interface to
> sysfs does not sit will with me at all. When we add support to a sysfs
> class, we usually require it to be generic and work across all devices.
> Adding device specific interfaces is generally decried and to be
> avoided. Don't forget, once we commit something to sysfs, it becomes
> ABI and we have to support it forever.

If you do git grep hw_pattern, you should get pointers to qcom-lpg
driver that solves similar problem, with interface that should be
acceptable.

Best regards,
Pavel

2024-01-29 14:21:30

by Dmitry Rokosov

[permalink] [raw]
Subject: Re: [PATCH v1 1/2] leds: aw200xx: support for hw pattern controllers

Hello Pavel,

On Sat, Jan 27, 2024 at 12:21:48AM +0100, Pavel Machek wrote:
> Hi!
>
> > > This led-controller supports 3 pattern controllers for auto breathing or
> > > group dimming control. Each pattern controller can work in auto
> > > breathing or manual control mode. All breathing parameters including
> > > rising/falling slope, on/off time, repeat times, min/max brightness
> > > and so on are configurable.
> > >
> > > Signed-off-by: Martin Kurbanov <[email protected]>
> > > ---
> > > .../testing/sysfs-class-led-driver-aw200xx | 108 +++
> > > Documentation/leds/leds-aw200xx.rst | 274 ++++++++
> > > drivers/leds/leds-aw200xx.c | 649 ++++++++++++++++++
> > > 3 files changed, 1031 insertions(+)
> > > create mode 100644 Documentation/leds/leds-aw200xx.rst
> >
> > This interface is bananas. Exposing an entire register interface to
> > sysfs does not sit will with me at all. When we add support to a sysfs
> > class, we usually require it to be generic and work across all devices.
> > Adding device specific interfaces is generally decried and to be
> > avoided. Don't forget, once we commit something to sysfs, it becomes
> > ABI and we have to support it forever.
>
> If you do git grep hw_pattern, you should get pointers to qcom-lpg
> driver that solves similar problem, with interface that should be
> acceptable.

Thank you for pointing that out. Yes, it's a very similar situation to
ours.

But I haven't observed the merging of this driver. Was it encountering
similar issues with the sysfs interface?

--
Thank you,
Dmitry

2024-01-29 14:23:06

by Dmitry Rokosov

[permalink] [raw]
Subject: Re: [PATCH v1 1/2] leds: aw200xx: support for hw pattern controllers

Hello Greg,

On Fri, Jan 26, 2024 at 06:09:43AM -0800, Greg Kroah-Hartman wrote:
> On Fri, Jan 26, 2024 at 03:23:10PM +0300, Dmitry Rokosov wrote:
> > Hello Lee,
> >
> > On Thu, Jan 25, 2024 at 01:00:49PM +0000, Lee Jones wrote:
> > > Looping in Jacek (LEDS) and Greg (SYFS) for some knowledgable input.
> > >
> > > On Fri, 12 Jan 2024, Martin Kurbanov wrote:
> > > > On 21.12.2023 19:10, Lee Jones wrote:
> > > > > On Thu, 07 Dec 2023, Martin Kurbanov wrote:
> > > > >
> > > > >> This led-controller supports 3 pattern controllers for auto breathing or
> > > > >> group dimming control. Each pattern controller can work in auto
> > > > >> breathing or manual control mode. All breathing parameters including
> > > > >> rising/falling slope, on/off time, repeat times, min/max brightness
> > > > >> and so on are configurable.
> > > > >>
> > > > >> Signed-off-by: Martin Kurbanov <[email protected]>
> > > > >> ---
> > > > >> .../testing/sysfs-class-led-driver-aw200xx | 108 +++
> > > > >> Documentation/leds/leds-aw200xx.rst | 274 ++++++++
> > > > >> drivers/leds/leds-aw200xx.c | 649 ++++++++++++++++++
> > > > >> 3 files changed, 1031 insertions(+)
> > > > >> create mode 100644 Documentation/leds/leds-aw200xx.rst
> > > > >
> > > > > This interface is bananas. Exposing an entire register interface to
> > > > > sysfs does not sit will with me at all. When we add support to a sysfs
> > > > > class, we usually require it to be generic and work across all devices.
> > > > > Adding device specific interfaces is generally decried and to be
> > > > > avoided. Don't forget, once we commit something to sysfs, it becomes
> > > > > ABI and we have to support it forever.
> > > > >
> > > > > A far better approach would be to add support for this in userspace
> > > > > instead You can use the standard I2C character device API to achieve
> > > > > the same result. That way we don't have the same level of commitment
> > > > > and is generally a much more flexible/future-proof.
> > > > >
> > > >
> > > > I used sysfs similarly to other LED drivers (for example, leds-lm3533).
> > > > Additionally, the controller has interrupts about the completion of the pattern,
> > > > which is best to handle in the kernel. In the case of implementation in user
> > > > mode, there may be synchronization problems, as the controller has several
> > > > memory pages that can be switched by writing the page number to register 0xF0.
> > >
> > > leds-lm3533 is a 12 year old legacy exception AND has less than half of
> > > the sysfs exports proposed here. What makes aw200xx so different it
> > > needs to an incomparable interface to any other that we currently
> > > support?
> >
> > >From my point of view, direct I2C raw requests from userspace are not a
> > good solution as well due to synchronization problems, as Martin
> > mentioned in the previous message.
>
> Sorry, I missed this, what is the synchronization problem? This is an
> led, shouldn't have any real specific performance issues.
>

By referring to the "synchronization problem," I am highlighting the
following issue:

The LED controller incorporates PAGE registers for accessing multiple
registers, which can be explained with the analogy of a "window." When
the driver needs to update the brightness value or perform any other
action, it must select the appropriate PAGE window to access the
register offset. It is challenging to believe that the driver,
responsible for adjusting brightness values, can be synchronized with
raw userspace I2C transactions in any manner. However, synchronization
is necessary because the hardware pattern employs the same PAGE
registers used for brightness and other parameter setups.

In the kernel driver, this issue is addressed through the simple mutex
usage.

> > We have honestly been attempting to integrate this functionality into
> > the official LED pattern interface, but it cannot be achieved due to the
> > absence of this interface's functionality:
> > 1) Page-based access
> > 2) Interrupts
>
> I don't understand this, sorry.
>

The PAGE window registers, which I described a few lines earlier, are
relevant. Regarding interrupts, the AW200XX LED controller incorporates
a hardware pattern event based on a straightforward GPIO-connected
interrupt line. It's advisable to asynchronously wait for the hardware
pattern ending event using methods like poll() and to notify userspace
using sysfs_notify().

> > HW patterns are very useful mechanism to draw animation faster without
> > any interactions with CPU, so I think we need to find the best architect
> > approach for its integration.
>
> The CPU is totally involved here, that should be identical.
>

Currently, we have two methods for playing LED animations:
1) Direct sysfs access to the appropriate brightness node.
2) LED-triggered pattern with hrtimer.

The first method always runs a separate kworker on each access,
resulting in slow performance and complete CPU involvement. The second
method is straightforward in terms of hardware pattern mechanism
synchronization, but it lacks specific support for the appropriate
hardware (as mentioned earlier). In this option, the CPU is not entirely
engaged because we can free the CPU from the waiting task until the LED
pattern is executed.

> > What is an alternative way to access such a hardware pattern interface?
> > Debugfs? Or perhaps we should consider extending the LED pattern
> > interface?
>
> Or again, userspace.
>
> debugfs is for debugging.
>
> thanks,
>
> greg k-h

--
Thank you,
Dmitry

2024-02-05 10:57:42

by Dmitry Rokosov

[permalink] [raw]
Subject: Re: [PATCH v1 1/2] leds: aw200xx: support for hw pattern controllers

Hello Greg, Lee and Pavel,

Apologies for the ping, but I would appreciate it if you could spare a
couple of minutes to decide on the next steps. From my perspective, the
problems I previously described persist, and we need to discuss the
possible solutions.

On Mon, Jan 29, 2024 at 05:13:39PM +0300, Dmitry Rokosov wrote:
> Hello Greg,
>
> On Fri, Jan 26, 2024 at 06:09:43AM -0800, Greg Kroah-Hartman wrote:
> > On Fri, Jan 26, 2024 at 03:23:10PM +0300, Dmitry Rokosov wrote:
> > > Hello Lee,
> > >
> > > On Thu, Jan 25, 2024 at 01:00:49PM +0000, Lee Jones wrote:
> > > > Looping in Jacek (LEDS) and Greg (SYFS) for some knowledgable input.
> > > >
> > > > On Fri, 12 Jan 2024, Martin Kurbanov wrote:
> > > > > On 21.12.2023 19:10, Lee Jones wrote:
> > > > > > On Thu, 07 Dec 2023, Martin Kurbanov wrote:
> > > > > >
> > > > > >> This led-controller supports 3 pattern controllers for auto breathing or
> > > > > >> group dimming control. Each pattern controller can work in auto
> > > > > >> breathing or manual control mode. All breathing parameters including
> > > > > >> rising/falling slope, on/off time, repeat times, min/max brightness
> > > > > >> and so on are configurable.
> > > > > >>
> > > > > >> Signed-off-by: Martin Kurbanov <[email protected]>
> > > > > >> ---
> > > > > >> .../testing/sysfs-class-led-driver-aw200xx | 108 +++
> > > > > >> Documentation/leds/leds-aw200xx.rst | 274 ++++++++
> > > > > >> drivers/leds/leds-aw200xx.c | 649 ++++++++++++++++++
> > > > > >> 3 files changed, 1031 insertions(+)
> > > > > >> create mode 100644 Documentation/leds/leds-aw200xx.rst
> > > > > >
> > > > > > This interface is bananas. Exposing an entire register interface to
> > > > > > sysfs does not sit will with me at all. When we add support to a sysfs
> > > > > > class, we usually require it to be generic and work across all devices.
> > > > > > Adding device specific interfaces is generally decried and to be
> > > > > > avoided. Don't forget, once we commit something to sysfs, it becomes
> > > > > > ABI and we have to support it forever.
> > > > > >
> > > > > > A far better approach would be to add support for this in userspace
> > > > > > instead You can use the standard I2C character device API to achieve
> > > > > > the same result. That way we don't have the same level of commitment
> > > > > > and is generally a much more flexible/future-proof.
> > > > > >
> > > > >
> > > > > I used sysfs similarly to other LED drivers (for example, leds-lm3533).
> > > > > Additionally, the controller has interrupts about the completion of the pattern,
> > > > > which is best to handle in the kernel. In the case of implementation in user
> > > > > mode, there may be synchronization problems, as the controller has several
> > > > > memory pages that can be switched by writing the page number to register 0xF0.
> > > >
> > > > leds-lm3533 is a 12 year old legacy exception AND has less than half of
> > > > the sysfs exports proposed here. What makes aw200xx so different it
> > > > needs to an incomparable interface to any other that we currently
> > > > support?
> > >
> > > >From my point of view, direct I2C raw requests from userspace are not a
> > > good solution as well due to synchronization problems, as Martin
> > > mentioned in the previous message.
> >
> > Sorry, I missed this, what is the synchronization problem? This is an
> > led, shouldn't have any real specific performance issues.
> >
>
> By referring to the "synchronization problem," I am highlighting the
> following issue:
>
> The LED controller incorporates PAGE registers for accessing multiple
> registers, which can be explained with the analogy of a "window." When
> the driver needs to update the brightness value or perform any other
> action, it must select the appropriate PAGE window to access the
> register offset. It is challenging to believe that the driver,
> responsible for adjusting brightness values, can be synchronized with
> raw userspace I2C transactions in any manner. However, synchronization
> is necessary because the hardware pattern employs the same PAGE
> registers used for brightness and other parameter setups.
>
> In the kernel driver, this issue is addressed through the simple mutex
> usage.
>
> > > We have honestly been attempting to integrate this functionality into
> > > the official LED pattern interface, but it cannot be achieved due to the
> > > absence of this interface's functionality:
> > > 1) Page-based access
> > > 2) Interrupts
> >
> > I don't understand this, sorry.
> >
>
> The PAGE window registers, which I described a few lines earlier, are
> relevant. Regarding interrupts, the AW200XX LED controller incorporates
> a hardware pattern event based on a straightforward GPIO-connected
> interrupt line. It's advisable to asynchronously wait for the hardware
> pattern ending event using methods like poll() and to notify userspace
> using sysfs_notify().
>
> > > HW patterns are very useful mechanism to draw animation faster without
> > > any interactions with CPU, so I think we need to find the best architect
> > > approach for its integration.
> >
> > The CPU is totally involved here, that should be identical.
> >
>
> Currently, we have two methods for playing LED animations:
> 1) Direct sysfs access to the appropriate brightness node.
> 2) LED-triggered pattern with hrtimer.
>
> The first method always runs a separate kworker on each access,
> resulting in slow performance and complete CPU involvement. The second
> method is straightforward in terms of hardware pattern mechanism
> synchronization, but it lacks specific support for the appropriate
> hardware (as mentioned earlier). In this option, the CPU is not entirely
> engaged because we can free the CPU from the waiting task until the LED
> pattern is executed.
>
> > > What is an alternative way to access such a hardware pattern interface?
> > > Debugfs? Or perhaps we should consider extending the LED pattern
> > > interface?
> >
> > Or again, userspace.
> >
> > debugfs is for debugging.
> >
> > thanks,
> >
> > greg k-h

--
Thank you,
Dmitry

2024-02-08 13:59:37

by Lee Jones

[permalink] [raw]
Subject: Re: [PATCH v1 1/2] leds: aw200xx: support for hw pattern controllers

On Mon, 29 Jan 2024, Dmitry Rokosov wrote:

> Hello Pavel,
>
> On Sat, Jan 27, 2024 at 12:21:48AM +0100, Pavel Machek wrote:
> > Hi!
> >
> > > > This led-controller supports 3 pattern controllers for auto breathing or
> > > > group dimming control. Each pattern controller can work in auto
> > > > breathing or manual control mode. All breathing parameters including
> > > > rising/falling slope, on/off time, repeat times, min/max brightness
> > > > and so on are configurable.
> > > >
> > > > Signed-off-by: Martin Kurbanov <[email protected]>
> > > > ---
> > > > .../testing/sysfs-class-led-driver-aw200xx | 108 +++
> > > > Documentation/leds/leds-aw200xx.rst | 274 ++++++++
> > > > drivers/leds/leds-aw200xx.c | 649 ++++++++++++++++++
> > > > 3 files changed, 1031 insertions(+)
> > > > create mode 100644 Documentation/leds/leds-aw200xx.rst
> > >
> > > This interface is bananas. Exposing an entire register interface to
> > > sysfs does not sit will with me at all. When we add support to a sysfs
> > > class, we usually require it to be generic and work across all devices.
> > > Adding device specific interfaces is generally decried and to be
> > > avoided. Don't forget, once we commit something to sysfs, it becomes
> > > ABI and we have to support it forever.
> >
> > If you do git grep hw_pattern, you should get pointers to qcom-lpg
> > driver that solves similar problem, with interface that should be
> > acceptable.
>
> Thank you for pointing that out. Yes, it's a very similar situation to
> ours.
>
> But I haven't observed the merging of this driver. Was it encountering
> similar issues with the sysfs interface?

`git grep` kinda implies that it's merged.

--
Lee Jones [李琼斯]

2024-02-08 14:14:23

by Lee Jones

[permalink] [raw]
Subject: Re: [PATCH v1 1/2] leds: aw200xx: support for hw pattern controllers

On Mon, 05 Feb 2024, Dmitry Rokosov wrote:

> Hello Greg, Lee and Pavel,
>
> Apologies for the ping, but I would appreciate it if you could spare a
> couple of minutes to decide on the next steps. From my perspective, the
> problems I previously described persist, and we need to discuss the
> possible solutions.

I thought you were going to use hw_pattern?

--
Lee Jones [李琼斯]