Changes since v5:
- Follow reviewer's comments:
1. the license stuff is a separate change
2. split fix mt7628 pwm into a single patch
3. to ensure to not use mtk_pwm_clk_name[10]
(After dynamic allocate clock array patch,
this is no need to check)
4. Use clock-frequency property to replace
the use of has_clks
Changes since v4:
- Follow reviewer's comments (v3: pwm: mediatek: add a property "num-pwms")
Move the changes of droping the check for of_device_get_match_data
returning non-NULL to next patch
- Follow reviewers's comments
(v3: pwm: mediatek: allocate the clks array dynamically)
1. use pc->soc->has_clks to check clocks exist or not.
2. Add error message when probe() unable to get clks
- Fixes bug when SoC is old mips which has no complex clock tree.
if clocks not exist, use the new property from DT to apply period
calculation; otherwise, use clk_get_rate to get clock frequency and
apply period calculation.
Changes since v3:
- add a new property "clock-frequency" and fix mt7628 pwm
- add mt7629 pwm support
Changes since v2:
- use num-pwms instead of mediatek,num-pwms.
- rename the member from num_pwms to fallback_num_pwms to make it
more obvious that it doesn't represent the actually used value.
- add a dev_warn and a expressive comment to help other developers
to not start adding num_pwms in the compatible_data.
Changes since v1:
- add some checks for backwards compatibility.
Ryder Lee (5):
pwm: mediatek: add a property "num-pwms"
dt-bindings: pwm: add a property "num-pwms"
arm64: dts: mt7622: add a property "num-pwms" for PWM
arm: dts: mt7623: add a property "num-pwms" for PWM
dt-bindings: pwm: update bindings for MT7629 SoC
Sam Shih (8):
pwm: mediatek: droping the check for of_device_get_match_data
pwm: mediatek: add a property "clock-frequency"
pwm: mediatek: allocate the clks array dynamically
pwm: mediatek: use pwm_mediatek as common prefix
pwm: mediatek: update license and switch to SPDX tag
dt-bindings: pwm: update bindings for MT7628 SoC
pwm: mediatek: remove a property "has-clock"
arm: dts: mediatek: add mt7629 pwm support
.../devicetree/bindings/pwm/pwm-mediatek.txt | 12 +-
arch/arm/boot/dts/mt7623.dtsi | 1 +
arch/arm64/boot/dts/mediatek/mt7622.dtsi | 1 +
drivers/pwm/pwm-mediatek.c | 257 ++++++++++--------
arch/arm/boot/dts/mt7629.dtsi | 16 ++++++++++++++++
5 files changed, 168 insertions(+), 119 deletions(-)
--
2.17.1
This patch drop the check for of_device_get_match_data.
Due to the only way call driver probe is compatible match.
The .data pointer which point to the SoC specify data is
directly set by driver, and it should not be NULL in our case.
We can safety remove the check for of_device_get_match_data.
Signed-off-by: Ryder Lee <[email protected]>
Signed-off-by: Sam Shih <[email protected]>
---
Used:
https://patchwork.kernel.org/patch/11096905/
Changes since v4:
Follow reviewer's comments:
Move the changes of droping the check for of_device_get_match_data
returning non-NULL to this patch
Change-Id: Ibcba903d5b26159051adfc5ef2e601ee2f78729b
---
drivers/pwm/pwm-mediatek.c | 6 +-----
1 file changed, 1 insertion(+), 5 deletions(-)
diff --git a/drivers/pwm/pwm-mediatek.c b/drivers/pwm/pwm-mediatek.c
index e214f4f57107..ebd62629e3fe 100644
--- a/drivers/pwm/pwm-mediatek.c
+++ b/drivers/pwm/pwm-mediatek.c
@@ -226,7 +226,6 @@ static const struct pwm_ops mtk_pwm_ops = {
static int mtk_pwm_probe(struct platform_device *pdev)
{
- const struct mtk_pwm_platform_data *data;
struct device_node *np = pdev->dev.of_node;
struct mtk_pwm_chip *pc;
struct resource *res;
@@ -237,10 +236,7 @@ static int mtk_pwm_probe(struct platform_device *pdev)
if (!pc)
return -ENOMEM;
- data = of_device_get_match_data(&pdev->dev);
- if (data == NULL)
- return -EINVAL;
- pc->soc = data;
+ pc->soc = of_device_get_match_data(&pdev->dev);
res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
pc->regs = devm_ioremap_resource(&pdev->dev, res);
--
2.17.1
This fix mt7628 pwm during configure from userspace. The SoC
is legacy MIPS and has no complex clock tree. This patch add property
clock-frequency to the SoC specific data and legacy MIPS SoC need to
configure it in DT. This property is use for period calculation.
We will improve this fix by droping has-clks attribute and using
clock-frequency to do the same thing in a new patch.
Signed-off-by: Ryder Lee <[email protected]>
Signed-off-by: Sam Shih <[email protected]>
---
Changes since v5:
1. Follow reviewer's comments
Make the changes of fix mt7628 pwm as a single patch
Changes since v4:
- Follow reviewers's comments
1. use pc->soc->has_clks to check clocks exist or not.
2. Add error message when probe() unable to get clks
- Fixes bug when SoC is old mips which has no complex clock tree.
if clocks not exist, use the new property from DT to apply period caculation;
otherwise, use clk_get_rate to get clock frequency and apply period caculation.
Change-Id: Ibbe6d7a4f80b30f60725bcbeca1d02ce7834d28c
---
drivers/pwm/pwm-mediatek.c | 39 ++++++++++++++++++++++++++++++--------
1 file changed, 31 insertions(+), 8 deletions(-)
diff --git a/drivers/pwm/pwm-mediatek.c b/drivers/pwm/pwm-mediatek.c
index ebd62629e3fe..1f18bff4800c 100644
--- a/drivers/pwm/pwm-mediatek.c
+++ b/drivers/pwm/pwm-mediatek.c
@@ -65,11 +65,13 @@ struct mtk_pwm_platform_data {
* @chip: linux PWM chip representation
* @regs: base address of PWM chip
* @clks: list of clocks
+ * @clk_freq: the fix clock frequency of legacy MIPS SoC
*/
struct mtk_pwm_chip {
struct pwm_chip chip;
void __iomem *regs;
struct clk *clks[MTK_CLK_MAX];
+ unsigned int clk_freq;
const struct mtk_pwm_platform_data *soc;
};
@@ -141,19 +143,27 @@ static int mtk_pwm_config(struct pwm_chip *chip, struct pwm_device *pwm,
int duty_ns, int period_ns)
{
struct mtk_pwm_chip *pc = to_mtk_pwm_chip(chip);
- struct clk *clk = pc->clks[MTK_CLK_PWM1 + pwm->hwpwm];
+ unsigned int clk_freq;
u32 clkdiv = 0, cnt_period, cnt_duty, reg_width = PWMDWIDTH,
reg_thres = PWMTHRES;
u64 resolution;
int ret;
+ if (pc->soc->has_clks) {
+ struct clk *clk = pc->clks[MTK_CLK_PWM1 + pwm->hwpwm];
+
+ clk_freq = clk_get_rate(clk);
+ } else {
+ clk_freq = pc->clk_freq;
+ }
+
ret = mtk_pwm_clk_enable(chip, pwm);
if (ret < 0)
return ret;
/* Using resolution in picosecond gets accuracy higher */
resolution = (u64)NSEC_PER_SEC * 1000;
- do_div(resolution, clk_get_rate(clk));
+ do_div(resolution, clk_freq);
cnt_period = DIV_ROUND_CLOSEST_ULL((u64)period_ns * 1000, resolution);
while (cnt_period > 8191) {
@@ -262,13 +272,26 @@ static int mtk_pwm_probe(struct platform_device *pdev)
npwms = MTK_CLK_MAX - 2;
}
- for (i = 0; i < npwms + 2 && pc->soc->has_clks; i++) {
- pc->clks[i] = devm_clk_get(&pdev->dev, mtk_pwm_clk_name[i]);
- if (IS_ERR(pc->clks[i])) {
- dev_err(&pdev->dev, "clock: %s fail: %ld\n",
- mtk_pwm_clk_name[i], PTR_ERR(pc->clks[i]));
- return PTR_ERR(pc->clks[i]);
+ if (pc->soc->has_clks) {
+ for (i = 0; i < npwms + 2 ; i++) {
+ pc->clks[i] = devm_clk_get(&pdev->dev,
+ mtk_pwm_clk_name[i]);
+ if (IS_ERR(pc->clks[i])) {
+ dev_err(&pdev->dev, "clock: %s fail: %ld\n",
+ mtk_pwm_clk_name[i],
+ PTR_ERR(pc->clks[i]));
+ return PTR_ERR(pc->clks[i]);
+ }
+ }
+ } else {
+ unsigned int clk_freq;
+
+ ret = of_property_read_u32(np, "clock-frequency", &clk_freq);
+ if (ret < 0) {
+ dev_err(&pdev->dev, "failed to get clock_frequency\n");
+ return ret;
}
+ pc->clk_freq = clk_freq;
}
platform_set_drvdata(pdev, pc);
--
2.17.1
This adds pwm support for MT7629.
Signed-off-by: Sam Shih <[email protected]>
---
arch/arm/boot/dts/mt7629.dtsi | 16 ++++++++++++++++
1 file changed, 16 insertions(+)
diff --git a/arch/arm/boot/dts/mt7629.dtsi b/arch/arm/boot/dts/mt7629.dtsi
index 9608bc2ccb3f..493be9a9453b 100644
--- a/arch/arm/boot/dts/mt7629.dtsi
+++ b/arch/arm/boot/dts/mt7629.dtsi
@@ -241,6 +241,22 @@
status = "disabled";
};
+ pwm: pwm@11006000 {
+ compatible = "mediatek,mt7629-pwm",
+ "mediatek,mt7622-pwm";
+ reg = <0 0x11006000 0 0x1000>;
+ interrupts = <GIC_SPI 77 IRQ_TYPE_LEVEL_LOW>;
+ clocks = <&topckgen CLK_TOP_PWM_SEL>,
+ <&pericfg CLK_PERI_PWM_PD>,
+ <&pericfg CLK_PERI_PWM1_PD>;
+ clock-names = "top", "main", "pwm1";
+ assigned-clocks = <&topckgen CLK_TOP_PWM_SEL>;
+ assigned-clock-parents =
+ <&topckgen CLK_TOP_UNIVPLL2_D4>;
+ num-pwms = <1>;
+ status = "disabled";
+ };
+
i2c: i2c@11007000 {
compatible = "mediatek,mt7629-i2c",
"mediatek,mt2712-i2c";
--
2.17.1
Use pwm_mediatek as common prefix to match the filename.
No functional change intended.
Signed-off-by: Ryder Lee <[email protected]>
Signed-off-by: Sam Shih <[email protected]>
---
Changes since v5:
- Follow reviewers's comments
The license stuff is a separate change
Change-Id: I32f421d3eb26438e25edfd6e4f6da0e85c58526c
---
drivers/pwm/pwm-mediatek.c | 112 +++++++++++++++++++------------------
1 file changed, 57 insertions(+), 55 deletions(-)
diff --git a/drivers/pwm/pwm-mediatek.c b/drivers/pwm/pwm-mediatek.c
index 8d1fd8163f15..0938beb1a3cb 100644
--- a/drivers/pwm/pwm-mediatek.c
+++ b/drivers/pwm/pwm-mediatek.c
@@ -34,15 +34,14 @@
#define PWM45THRES_FIXUP 0x34
#define PWM_CLK_DIV_MAX 7
-
-struct mtk_pwm_platform_data {
+struct pwm_mediatek_of_data {
unsigned int fallback_npwms;
bool pwm45_fixup;
bool has_clks;
};
/**
- * struct mtk_pwm_chip - struct representing PWM chip
+ * struct pwm_mediatek_chip - struct representing PWM chip
* @chip: linux PWM chip representation
* @regs: base address of PWM chip
* @clk_top: the top clock generator
@@ -50,28 +49,30 @@ struct mtk_pwm_platform_data {
* @clk_pwms: the clock used by each PWM channel
* @clk_freq: the fix clock frequency of legacy MIPS SoC
*/
-struct mtk_pwm_chip {
+struct pwm_mediatek_chip {
struct pwm_chip chip;
void __iomem *regs;
struct clk *clk_top;
struct clk *clk_main;
struct clk **clk_pwms;
unsigned int clk_freq;
- const struct mtk_pwm_platform_data *soc;
+ const struct pwm_mediatek_of_data *soc;
};
-static const unsigned int mtk_pwm_reg_offset[] = {
+static const unsigned int pwm_mediatek_reg_offset[] = {
0x0010, 0x0050, 0x0090, 0x00d0, 0x0110, 0x0150, 0x0190, 0x0220
};
-static inline struct mtk_pwm_chip *to_mtk_pwm_chip(struct pwm_chip *chip)
+static inline struct pwm_mediatek_chip *
+to_pwm_mediatek_chip(struct pwm_chip *chip)
{
- return container_of(chip, struct mtk_pwm_chip, chip);
+ return container_of(chip, struct pwm_mediatek_chip, chip);
}
-static int mtk_pwm_clk_enable(struct pwm_chip *chip, struct pwm_device *pwm)
+static int pwm_mediatek_clk_enable(struct pwm_chip *chip,
+ struct pwm_device *pwm)
{
- struct mtk_pwm_chip *pc = to_mtk_pwm_chip(chip);
+ struct pwm_mediatek_chip *pc = to_pwm_mediatek_chip(chip);
int ret;
if (!pc->soc->has_clks)
@@ -99,9 +100,10 @@ static int mtk_pwm_clk_enable(struct pwm_chip *chip, struct pwm_device *pwm)
return ret;
}
-static void mtk_pwm_clk_disable(struct pwm_chip *chip, struct pwm_device *pwm)
+static void pwm_mediatek_clk_disable(struct pwm_chip *chip,
+ struct pwm_device *pwm)
{
- struct mtk_pwm_chip *pc = to_mtk_pwm_chip(chip);
+ struct pwm_mediatek_chip *pc = to_pwm_mediatek_chip(chip);
if (!pc->soc->has_clks)
return;
@@ -111,23 +113,23 @@ static void mtk_pwm_clk_disable(struct pwm_chip *chip, struct pwm_device *pwm)
clk_disable_unprepare(pc->clk_top);
}
-static inline u32 mtk_pwm_readl(struct mtk_pwm_chip *chip, unsigned int num,
- unsigned int offset)
+static inline u32 pwm_mediatek_readl(struct pwm_mediatek_chip *chip,
+ unsigned int num, unsigned int offset)
{
- return readl(chip->regs + mtk_pwm_reg_offset[num] + offset);
+ return readl(chip->regs + pwm_mediatek_reg_offset[num] + offset);
}
-static inline void mtk_pwm_writel(struct mtk_pwm_chip *chip,
- unsigned int num, unsigned int offset,
- u32 value)
+static inline void pwm_mediatek_writel(struct pwm_mediatek_chip *chip,
+ unsigned int num, unsigned int offset,
+ u32 value)
{
- writel(value, chip->regs + mtk_pwm_reg_offset[num] + offset);
+ writel(value, chip->regs + pwm_mediatek_reg_offset[num] + offset);
}
-static int mtk_pwm_config(struct pwm_chip *chip, struct pwm_device *pwm,
- int duty_ns, int period_ns)
+static int pwm_mediatek_config(struct pwm_chip *chip, struct pwm_device *pwm,
+ int duty_ns, int period_ns)
{
- struct mtk_pwm_chip *pc = to_mtk_pwm_chip(chip);
+ struct pwm_mediatek_chip *pc = to_pwm_mediatek_chip(chip);
unsigned int clk_freq;
u32 clkdiv = 0, cnt_period, cnt_duty, reg_width = PWMDWIDTH,
reg_thres = PWMTHRES;
@@ -139,7 +141,7 @@ static int mtk_pwm_config(struct pwm_chip *chip, struct pwm_device *pwm,
else
clk_freq = pc->clk_freq;
- ret = mtk_pwm_clk_enable(chip, pwm);
+ ret = pwm_mediatek_clk_enable(chip, pwm);
if (ret < 0)
return ret;
@@ -156,7 +158,7 @@ static int mtk_pwm_config(struct pwm_chip *chip, struct pwm_device *pwm,
}
if (clkdiv > PWM_CLK_DIV_MAX) {
- mtk_pwm_clk_disable(chip, pwm);
+ pwm_mediatek_clk_disable(chip, pwm);
dev_err(chip->dev, "period %d not supported\n", period_ns);
return -EINVAL;
}
@@ -171,22 +173,22 @@ static int mtk_pwm_config(struct pwm_chip *chip, struct pwm_device *pwm,
}
cnt_duty = DIV_ROUND_CLOSEST_ULL((u64)duty_ns * 1000, resolution);
- mtk_pwm_writel(pc, pwm->hwpwm, PWMCON, BIT(15) | clkdiv);
- mtk_pwm_writel(pc, pwm->hwpwm, reg_width, cnt_period);
- mtk_pwm_writel(pc, pwm->hwpwm, reg_thres, cnt_duty);
+ pwm_mediatek_writel(pc, pwm->hwpwm, PWMCON, BIT(15) | clkdiv);
+ pwm_mediatek_writel(pc, pwm->hwpwm, reg_width, cnt_period);
+ pwm_mediatek_writel(pc, pwm->hwpwm, reg_thres, cnt_duty);
- mtk_pwm_clk_disable(chip, pwm);
+ pwm_mediatek_clk_disable(chip, pwm);
return 0;
}
-static int mtk_pwm_enable(struct pwm_chip *chip, struct pwm_device *pwm)
+static int pwm_mediatek_enable(struct pwm_chip *chip, struct pwm_device *pwm)
{
- struct mtk_pwm_chip *pc = to_mtk_pwm_chip(chip);
+ struct pwm_mediatek_chip *pc = to_pwm_mediatek_chip(chip);
u32 value;
int ret;
- ret = mtk_pwm_clk_enable(chip, pwm);
+ ret = pwm_mediatek_clk_enable(chip, pwm);
if (ret < 0)
return ret;
@@ -197,29 +199,29 @@ static int mtk_pwm_enable(struct pwm_chip *chip, struct pwm_device *pwm)
return 0;
}
-static void mtk_pwm_disable(struct pwm_chip *chip, struct pwm_device *pwm)
+static void pwm_mediatek_disable(struct pwm_chip *chip, struct pwm_device *pwm)
{
- struct mtk_pwm_chip *pc = to_mtk_pwm_chip(chip);
+ struct pwm_mediatek_chip *pc = to_pwm_mediatek_chip(chip);
u32 value;
value = readl(pc->regs);
value &= ~BIT(pwm->hwpwm);
writel(value, pc->regs);
- mtk_pwm_clk_disable(chip, pwm);
+ pwm_mediatek_clk_disable(chip, pwm);
}
-static const struct pwm_ops mtk_pwm_ops = {
- .config = mtk_pwm_config,
- .enable = mtk_pwm_enable,
- .disable = mtk_pwm_disable,
+static const struct pwm_ops pwm_mediatek_ops = {
+ .config = pwm_mediatek_config,
+ .enable = pwm_mediatek_enable,
+ .disable = pwm_mediatek_disable,
.owner = THIS_MODULE,
};
-static int mtk_pwm_probe(struct platform_device *pdev)
+static int pwm_mediatek_probe(struct platform_device *pdev)
{
struct device_node *np = pdev->dev.of_node;
- struct mtk_pwm_chip *pc;
+ struct pwm_mediatek_chip *pc;
struct resource *res;
unsigned int npwms;
int ret;
@@ -293,7 +295,7 @@ static int mtk_pwm_probe(struct platform_device *pdev)
platform_set_drvdata(pdev, pc);
pc->chip.dev = &pdev->dev;
- pc->chip.ops = &mtk_pwm_ops;
+ pc->chip.ops = &pwm_mediatek_ops;
pc->chip.base = -1;
pc->chip.npwm = npwms;
@@ -306,55 +308,55 @@ static int mtk_pwm_probe(struct platform_device *pdev)
return 0;
}
-static int mtk_pwm_remove(struct platform_device *pdev)
+static int pwm_mediatek_remove(struct platform_device *pdev)
{
- struct mtk_pwm_chip *pc = platform_get_drvdata(pdev);
+ struct pwm_mediatek_chip *pc = platform_get_drvdata(pdev);
return pwmchip_remove(&pc->chip);
}
-static const struct mtk_pwm_platform_data mt2712_pwm_data = {
+static const struct pwm_mediatek_of_data mt2712_pwm_data = {
.fallback_npwms = 8,
.pwm45_fixup = false,
.has_clks = true,
};
-static const struct mtk_pwm_platform_data mt7622_pwm_data = {
+static const struct pwm_mediatek_of_data mt7622_pwm_data = {
.fallback_npwms = 6,
.pwm45_fixup = false,
.has_clks = true,
};
-static const struct mtk_pwm_platform_data mt7623_pwm_data = {
+static const struct pwm_mediatek_of_data mt7623_pwm_data = {
.fallback_npwms = 5,
.pwm45_fixup = true,
.has_clks = true,
};
-static const struct mtk_pwm_platform_data mt7628_pwm_data = {
+static const struct pwm_mediatek_of_data mt7628_pwm_data = {
.fallback_npwms = 4,
.pwm45_fixup = true,
.has_clks = false,
};
-static const struct of_device_id mtk_pwm_of_match[] = {
+static const struct of_device_id pwm_mediatek_of_match[] = {
{ .compatible = "mediatek,mt2712-pwm", .data = &mt2712_pwm_data },
{ .compatible = "mediatek,mt7622-pwm", .data = &mt7622_pwm_data },
{ .compatible = "mediatek,mt7623-pwm", .data = &mt7623_pwm_data },
{ .compatible = "mediatek,mt7628-pwm", .data = &mt7628_pwm_data },
{ },
};
-MODULE_DEVICE_TABLE(of, mtk_pwm_of_match);
+MODULE_DEVICE_TABLE(of, pwm_mediatek_of_match);
-static struct platform_driver mtk_pwm_driver = {
+static struct platform_driver pwm_mediatek_driver = {
.driver = {
- .name = "mtk-pwm",
- .of_match_table = mtk_pwm_of_match,
+ .name = "pwm-mediatek",
+ .of_match_table = pwm_mediatek_of_match,
},
- .probe = mtk_pwm_probe,
- .remove = mtk_pwm_remove,
+ .probe = pwm_mediatek_probe,
+ .remove = pwm_mediatek_remove,
};
-module_platform_driver(mtk_pwm_driver);
+module_platform_driver(pwm_mediatek_driver);
MODULE_AUTHOR("John Crispin <[email protected]>");
MODULE_LICENSE("GPL");
--
2.17.1
From: Ryder Lee <[email protected]>
This adds a property "num-pwms" in example so that we could
specify the number of PWM channels via device tree.
Signed-off-by: Ryder Lee <[email protected]>
Signed-off-by: Sam Shih <[email protected]>
Reviewed-by: Matthias Brugger <[email protected]>
Acked-by: Uwe Kleine-König <[email protected]>
---
Changes since v5:
- Add an Acked-by tag
- This file is original v4 patch 5/10
(https://patchwork.kernel.org/patch/11102577/)
Change-Id: I429048afeffa96f3f14533910efe242f88776043
---
Documentation/devicetree/bindings/pwm/pwm-mediatek.txt | 7 ++++---
1 file changed, 4 insertions(+), 3 deletions(-)
diff --git a/Documentation/devicetree/bindings/pwm/pwm-mediatek.txt b/Documentation/devicetree/bindings/pwm/pwm-mediatek.txt
index 991728cb46cb..ea95b490a913 100644
--- a/Documentation/devicetree/bindings/pwm/pwm-mediatek.txt
+++ b/Documentation/devicetree/bindings/pwm/pwm-mediatek.txt
@@ -14,12 +14,12 @@ Required properties:
has no clocks
- "top": the top clock generator
- "main": clock used by the PWM core
- - "pwm1-8": the eight per PWM clocks for mt2712
- - "pwm1-6": the six per PWM clocks for mt7622
- - "pwm1-5": the five per PWM clocks for mt7623
+ - "pwm1-N": the PWM clocks for each channel
+ where N starting from 1 to the maximum number of PWM channels
- pinctrl-names: Must contain a "default" entry.
- pinctrl-0: One property must exist for each entry in pinctrl-names.
See pinctrl/pinctrl-bindings.txt for details of the property values.
+ - num-pwms: the number of PWM channels.
Example:
pwm0: pwm@11006000 {
@@ -37,4 +37,5 @@ Example:
"pwm3", "pwm4", "pwm5";
pinctrl-names = "default";
pinctrl-0 = <&pwm0_pins>;
+ num-pwms = <5>;
};
--
2.17.1
This updates bindings for MT7628 pwm controller.
Signed-off-by: Sam Shih <[email protected]>
---
Documentation/devicetree/bindings/pwm/pwm-mediatek.txt | 4 ++++
1 file changed, 4 insertions(+)
diff --git a/Documentation/devicetree/bindings/pwm/pwm-mediatek.txt b/Documentation/devicetree/bindings/pwm/pwm-mediatek.txt
index ea95b490a913..93980e3da261 100644
--- a/Documentation/devicetree/bindings/pwm/pwm-mediatek.txt
+++ b/Documentation/devicetree/bindings/pwm/pwm-mediatek.txt
@@ -21,6 +21,10 @@ Required properties:
See pinctrl/pinctrl-bindings.txt for details of the property values.
- num-pwms: the number of PWM channels.
+
+ Optional properties:
+ - clock-frequency: fix clock frequency, this is only used in MT7628 SoC
+ for period calculation. This SoC has no complex clock tree.
Example:
pwm0: pwm@11006000 {
compatible = "mediatek,mt7623-pwm";
--
2.17.1
Due to we added clock-frequency property to fix
mt7628 pwm during configure from userspace.
We can alos use this property to determine whether
the complex clock tree exists in the SoC or not.
So we can safety remove has-clock property in the
driver specific data.
Signed-off-by: Sam Shih <[email protected]>
---
drivers/pwm/pwm-mediatek.c | 26 ++++++++------------------
1 file changed, 8 insertions(+), 18 deletions(-)
diff --git a/drivers/pwm/pwm-mediatek.c b/drivers/pwm/pwm-mediatek.c
index 290536a92a80..96f592595063 100644
--- a/drivers/pwm/pwm-mediatek.c
+++ b/drivers/pwm/pwm-mediatek.c
@@ -35,7 +35,6 @@
struct pwm_mediatek_of_data {
unsigned int fallback_npwms;
bool pwm45_fixup;
- bool has_clks;
};
/**
@@ -73,7 +72,7 @@ static int pwm_mediatek_clk_enable(struct pwm_chip *chip,
struct pwm_mediatek_chip *pc = to_pwm_mediatek_chip(chip);
int ret;
- if (!pc->soc->has_clks)
+ if (pc->clk_freq)
return 0;
ret = clk_prepare_enable(pc->clk_top);
@@ -103,7 +102,7 @@ static void pwm_mediatek_clk_disable(struct pwm_chip *chip,
{
struct pwm_mediatek_chip *pc = to_pwm_mediatek_chip(chip);
- if (!pc->soc->has_clks)
+ if (pc->clk_freq)
return;
clk_disable_unprepare(pc->clk_pwms[pwm->hwpwm]);
@@ -134,10 +133,10 @@ static int pwm_mediatek_config(struct pwm_chip *chip, struct pwm_device *pwm,
u64 resolution;
int ret;
- if (pc->soc->has_clks)
- clk_freq = clk_get_rate(pc->clk_pwms[pwm->hwpwm]);
- else
+ if (pc->clk_freq)
clk_freq = pc->clk_freq;
+ else
+ clk_freq = clk_get_rate(pc->clk_pwms[pwm->hwpwm]);
ret = pwm_mediatek_clk_enable(chip, pwm);
if (ret < 0)
@@ -222,6 +221,7 @@ static int pwm_mediatek_probe(struct platform_device *pdev)
struct pwm_mediatek_chip *pc;
struct resource *res;
unsigned int npwms;
+ unsigned int clk_freq;
int ret;
pc = devm_kzalloc(&pdev->dev, sizeof(*pc), GFP_KERNEL);
@@ -247,7 +247,8 @@ static int pwm_mediatek_probe(struct platform_device *pdev)
}
}
- if (pc->soc->has_clks) {
+ ret = of_property_read_u32(np, "clock-frequency", &clk_freq);
+ if (ret < 0) {
int i;
pc->clk_pwms = devm_kcalloc(&pdev->dev, npwms,
@@ -280,13 +281,6 @@ static int pwm_mediatek_probe(struct platform_device *pdev)
}
}
} else {
- unsigned int clk_freq;
-
- ret = of_property_read_u32(np, "clock-frequency", &clk_freq);
- if (ret < 0) {
- dev_err(&pdev->dev, "failed to get clock_frequency\n");
- return ret;
- }
pc->clk_freq = clk_freq;
}
@@ -316,25 +310,21 @@ static int pwm_mediatek_remove(struct platform_device *pdev)
static const struct pwm_mediatek_of_data mt2712_pwm_data = {
.fallback_npwms = 8,
.pwm45_fixup = false,
- .has_clks = true,
};
static const struct pwm_mediatek_of_data mt7622_pwm_data = {
.fallback_npwms = 6,
.pwm45_fixup = false,
- .has_clks = true,
};
static const struct pwm_mediatek_of_data mt7623_pwm_data = {
.fallback_npwms = 5,
.pwm45_fixup = true,
- .has_clks = true,
};
static const struct pwm_mediatek_of_data mt7628_pwm_data = {
.fallback_npwms = 4,
.pwm45_fixup = true,
- .has_clks = false,
};
static const struct of_device_id pwm_mediatek_of_match[] = {
--
2.17.1
Instead of using fixed size of arrays, allocate the memory for them
based on the information we get from the DT.
Also remove the check for num_pwms, due to dynamically allocate pwm
should not cause array index out of bound.
Signed-off-by: Ryder Lee <[email protected]>
Signed-off-by: Sam Shih <[email protected]>
---
Changes since v5:
- Follow reviewers's comments
Make the changes of allocate the clks array dynamically as a single patch
Changes since v4:
- Follow reviewers's comments
1. use pc->soc->has_clks to check clocks exist or not.
2. Add error message when probe() unable to get clks
- Fixes bug when SoC is old mips which has no complex clock tree.
if clocks not exist, use the new property from DT to apply period caculation;
otherwise, use clk_get_rate to get clock frequency and apply period caculation.
Change-Id: I42edf07548fd604a5a4f689b45e945809057e3e3
---
drivers/pwm/pwm-mediatek.c | 96 ++++++++++++++++++--------------------
1 file changed, 46 insertions(+), 50 deletions(-)
diff --git a/drivers/pwm/pwm-mediatek.c b/drivers/pwm/pwm-mediatek.c
index 1f18bff4800c..8d1fd8163f15 100644
--- a/drivers/pwm/pwm-mediatek.c
+++ b/drivers/pwm/pwm-mediatek.c
@@ -35,25 +35,6 @@
#define PWM_CLK_DIV_MAX 7
-enum {
- MTK_CLK_MAIN = 0,
- MTK_CLK_TOP,
- MTK_CLK_PWM1,
- MTK_CLK_PWM2,
- MTK_CLK_PWM3,
- MTK_CLK_PWM4,
- MTK_CLK_PWM5,
- MTK_CLK_PWM6,
- MTK_CLK_PWM7,
- MTK_CLK_PWM8,
- MTK_CLK_MAX,
-};
-
-static const char * const mtk_pwm_clk_name[MTK_CLK_MAX] = {
- "main", "top", "pwm1", "pwm2", "pwm3", "pwm4", "pwm5", "pwm6", "pwm7",
- "pwm8"
-};
-
struct mtk_pwm_platform_data {
unsigned int fallback_npwms;
bool pwm45_fixup;
@@ -64,13 +45,17 @@ struct mtk_pwm_platform_data {
* struct mtk_pwm_chip - struct representing PWM chip
* @chip: linux PWM chip representation
* @regs: base address of PWM chip
- * @clks: list of clocks
+ * @clk_top: the top clock generator
+ * @clk_main: the clock used by PWM core
+ * @clk_pwms: the clock used by each PWM channel
* @clk_freq: the fix clock frequency of legacy MIPS SoC
*/
struct mtk_pwm_chip {
struct pwm_chip chip;
void __iomem *regs;
- struct clk *clks[MTK_CLK_MAX];
+ struct clk *clk_top;
+ struct clk *clk_main;
+ struct clk **clk_pwms;
unsigned int clk_freq;
const struct mtk_pwm_platform_data *soc;
};
@@ -92,24 +77,24 @@ static int mtk_pwm_clk_enable(struct pwm_chip *chip, struct pwm_device *pwm)
if (!pc->soc->has_clks)
return 0;
- ret = clk_prepare_enable(pc->clks[MTK_CLK_TOP]);
+ ret = clk_prepare_enable(pc->clk_top);
if (ret < 0)
return ret;
- ret = clk_prepare_enable(pc->clks[MTK_CLK_MAIN]);
+ ret = clk_prepare_enable(pc->clk_main);
if (ret < 0)
goto disable_clk_top;
- ret = clk_prepare_enable(pc->clks[MTK_CLK_PWM1 + pwm->hwpwm]);
+ ret = clk_prepare_enable(pc->clk_pwms[pwm->hwpwm]);
if (ret < 0)
goto disable_clk_main;
return 0;
disable_clk_main:
- clk_disable_unprepare(pc->clks[MTK_CLK_MAIN]);
+ clk_disable_unprepare(pc->clk_main);
disable_clk_top:
- clk_disable_unprepare(pc->clks[MTK_CLK_TOP]);
+ clk_disable_unprepare(pc->clk_top);
return ret;
}
@@ -121,9 +106,9 @@ static void mtk_pwm_clk_disable(struct pwm_chip *chip, struct pwm_device *pwm)
if (!pc->soc->has_clks)
return;
- clk_disable_unprepare(pc->clks[MTK_CLK_PWM1 + pwm->hwpwm]);
- clk_disable_unprepare(pc->clks[MTK_CLK_MAIN]);
- clk_disable_unprepare(pc->clks[MTK_CLK_TOP]);
+ clk_disable_unprepare(pc->clk_pwms[pwm->hwpwm]);
+ clk_disable_unprepare(pc->clk_main);
+ clk_disable_unprepare(pc->clk_top);
}
static inline u32 mtk_pwm_readl(struct mtk_pwm_chip *chip, unsigned int num,
@@ -149,13 +134,10 @@ static int mtk_pwm_config(struct pwm_chip *chip, struct pwm_device *pwm,
u64 resolution;
int ret;
- if (pc->soc->has_clks) {
- struct clk *clk = pc->clks[MTK_CLK_PWM1 + pwm->hwpwm];
-
- clk_freq = clk_get_rate(clk);
- } else {
+ if (pc->soc->has_clks)
+ clk_freq = clk_get_rate(pc->clk_pwms[pwm->hwpwm]);
+ else
clk_freq = pc->clk_freq;
- }
ret = mtk_pwm_clk_enable(chip, pwm);
if (ret < 0)
@@ -239,7 +221,7 @@ static int mtk_pwm_probe(struct platform_device *pdev)
struct device_node *np = pdev->dev.of_node;
struct mtk_pwm_chip *pc;
struct resource *res;
- unsigned int i, npwms;
+ unsigned int npwms;
int ret;
pc = devm_kzalloc(&pdev->dev, sizeof(*pc), GFP_KERNEL);
@@ -265,22 +247,36 @@ static int mtk_pwm_probe(struct platform_device *pdev)
}
}
- /* MAIN + TOP + NPWM < MTK_CLK_MAX */
- if ((npwms + 2) > MTK_CLK_MAX) {
- dev_warn(&pdev->dev, "number of PWMs is larger than %d\n",
- MTK_CLK_MAX - 2);
- npwms = MTK_CLK_MAX - 2;
- }
-
if (pc->soc->has_clks) {
- for (i = 0; i < npwms + 2 ; i++) {
- pc->clks[i] = devm_clk_get(&pdev->dev,
- mtk_pwm_clk_name[i]);
- if (IS_ERR(pc->clks[i])) {
+ int i;
+
+ pc->clk_pwms = devm_kcalloc(&pdev->dev, npwms,
+ sizeof(*pc->clk_pwms), GFP_KERNEL);
+ if (!pc->clk_pwms)
+ return -ENOMEM;
+
+ pc->clk_top = devm_clk_get(&pdev->dev, "top");
+ if (IS_ERR(pc->clk_top)) {
+ dev_err(&pdev->dev, "clock: top fail: %ld\n",
+ PTR_ERR(pc->clk_top));
+ return PTR_ERR(pc->clk_top);
+ }
+
+ pc->clk_main = devm_clk_get(&pdev->dev, "main");
+ if (IS_ERR(pc->clk_main)) {
+ dev_err(&pdev->dev, "clock: main fail: %ld\n",
+ PTR_ERR(pc->clk_main));
+ return PTR_ERR(pc->clk_main);
+ }
+ for (i = 0; i < npwms; i++) {
+ char name[8];
+
+ snprintf(name, sizeof(name), "pwm%d", i + 1);
+ pc->clk_pwms[i] = devm_clk_get(&pdev->dev, name);
+ if (IS_ERR(pc->clk_pwms[i])) {
dev_err(&pdev->dev, "clock: %s fail: %ld\n",
- mtk_pwm_clk_name[i],
- PTR_ERR(pc->clks[i]));
- return PTR_ERR(pc->clks[i]);
+ name, PTR_ERR(pc->clk_pwms[i]));
+ return PTR_ERR(pc->clk_pwms[i]);
}
}
} else {
--
2.17.1
From: Ryder Lee <[email protected]>
This adds a property "num-pwms" for PWM controller.
Signed-off-by: Ryder Lee <[email protected]>
Signed-off-by: Sam Shih <[email protected]>
---
arch/arm64/boot/dts/mediatek/mt7622.dtsi | 1 +
1 file changed, 1 insertion(+)
diff --git a/arch/arm64/boot/dts/mediatek/mt7622.dtsi b/arch/arm64/boot/dts/mediatek/mt7622.dtsi
index d1e13d340e26..9a043938881f 100644
--- a/arch/arm64/boot/dts/mediatek/mt7622.dtsi
+++ b/arch/arm64/boot/dts/mediatek/mt7622.dtsi
@@ -439,6 +439,7 @@
<&pericfg CLK_PERI_PWM6_PD>;
clock-names = "top", "main", "pwm1", "pwm2", "pwm3", "pwm4",
"pwm5", "pwm6";
+ num-pwms = <6>;
status = "disabled";
};
--
2.17.1
From: Ryder Lee <[email protected]>
This updates bindings for MT7629 pwm controller.
This patch is the same as
https://patchwork.kernel.org/patch/10769381/
and it has a Reviewed-by tag in v1
Signed-off-by: Ryder Lee <[email protected]>
Signed-off-by: Sam Shih <[email protected]>
Reviewed-by: Matthias Brugger <[email protected]>
---
Changes since v1:
- add a Reviewed-by tag
Change-Id: Id9f0bd0b9aaa8748fcf6b0bfb2c6214d50b5ffa4
---
Documentation/devicetree/bindings/pwm/pwm-mediatek.txt | 1 +
1 file changed, 1 insertion(+)
diff --git a/Documentation/devicetree/bindings/pwm/pwm-mediatek.txt b/Documentation/devicetree/bindings/pwm/pwm-mediatek.txt
index 93980e3da261..e11cb0d93929 100644
--- a/Documentation/devicetree/bindings/pwm/pwm-mediatek.txt
+++ b/Documentation/devicetree/bindings/pwm/pwm-mediatek.txt
@@ -6,6 +6,7 @@ Required properties:
- "mediatek,mt7622-pwm": found on mt7622 SoC.
- "mediatek,mt7623-pwm": found on mt7623 SoC.
- "mediatek,mt7628-pwm": found on mt7628 SoC.
+ - "mediatek,mt7629-pwm", "mediatek,mt7622-pwm": found on mt7629 SoC.
- reg: physical base address and length of the controller's registers.
- #pwm-cells: must be 2. See pwm.txt in this directory for a description of
the cell format.
--
2.17.1
Add SPDX identifiers to pwm-mediatek.c
Update license to GNU General Public License v2.0
Signed-off-by: Ryder Lee <[email protected]>
Signed-off-by: Sam Shih <[email protected]>
---
Changes since v5:
- Follow reviewers's comments
The license stuff is a separate change
Change-Id: I8f1ff6daa76727ed203db8dc608509f1cd3b1144
---
drivers/pwm/pwm-mediatek.c | 8 +++-----
1 file changed, 3 insertions(+), 5 deletions(-)
diff --git a/drivers/pwm/pwm-mediatek.c b/drivers/pwm/pwm-mediatek.c
index 0938beb1a3cb..290536a92a80 100644
--- a/drivers/pwm/pwm-mediatek.c
+++ b/drivers/pwm/pwm-mediatek.c
@@ -1,12 +1,10 @@
+// SPDX-License-Identifier: GPL-2.0
/*
- * Mediatek Pulse Width Modulator driver
+ * MediaTek Pulse Width Modulator driver
*
* Copyright (C) 2015 John Crispin <[email protected]>
* Copyright (C) 2017 Zhi Mao <[email protected]>
*
- * This file is licensed under the terms of the GNU General Public
- * License version 2. This program is licensed "as is" without any
- * warranty of any kind, whether express or implied.
*/
#include <linux/err.h>
@@ -359,4 +357,4 @@ static struct platform_driver pwm_mediatek_driver = {
module_platform_driver(pwm_mediatek_driver);
MODULE_AUTHOR("John Crispin <[email protected]>");
-MODULE_LICENSE("GPL");
+MODULE_LICENSE("GPL v2");
--
2.17.1
From: Ryder Lee <[email protected]>
This adds a property "num-pwms" for PWM controller.
Signed-off-by: Ryder Lee <[email protected]>
Signed-off-by: Sam Shih <[email protected]>
---
arch/arm/boot/dts/mt7623.dtsi | 1 +
1 file changed, 1 insertion(+)
diff --git a/arch/arm/boot/dts/mt7623.dtsi b/arch/arm/boot/dts/mt7623.dtsi
index a79f0b6c3429..208e0d19a575 100644
--- a/arch/arm/boot/dts/mt7623.dtsi
+++ b/arch/arm/boot/dts/mt7623.dtsi
@@ -452,6 +452,7 @@
<&pericfg CLK_PERI_PWM5>;
clock-names = "top", "main", "pwm1", "pwm2",
"pwm3", "pwm4", "pwm5";
+ num-pwms = <5>;
status = "disabled";
};
--
2.17.1
On Thu, 2019-08-22 at 14:58 +0800, Sam Shih wrote:
> This updates bindings for MT7628 pwm controller.
>
> Signed-off-by: Sam Shih <[email protected]>
> ---
> Documentation/devicetree/bindings/pwm/pwm-mediatek.txt | 4 ++++
> 1 file changed, 4 insertions(+)
>
> diff --git a/Documentation/devicetree/bindings/pwm/pwm-mediatek.txt b/Documentation/devicetree/bindings/pwm/pwm-mediatek.txt
> index ea95b490a913..93980e3da261 100644
> --- a/Documentation/devicetree/bindings/pwm/pwm-mediatek.txt
> +++ b/Documentation/devicetree/bindings/pwm/pwm-mediatek.txt
> @@ -21,6 +21,10 @@ Required properties:
> See pinctrl/pinctrl-bindings.txt for details of the property values.
> - num-pwms: the number of PWM channels.
> +
> + Optional properties:
> + - clock-frequency: fix clock frequency, this is only used in MT7628 SoC
> + for period calculation. This SoC has no complex clock tree.
I'm sorry if this has been discussed before.
Would it be simpler if you just provide a fixed-clock as clock in device
tree? This way you don't need this optional properties and don't need to
special handle it in driver code.
After all, the hw is still connected to a simple clock tree.
Joe.C
On Thu, Aug 22, 2019 at 02:58:32PM +0800, Sam Shih wrote:
> This patch drop the check for of_device_get_match_data.
> Due to the only way call driver probe is compatible match.
> The .data pointer which point to the SoC specify data is
> directly set by driver, and it should not be NULL in our case.
> We can safety remove the check for of_device_get_match_data.
>
> Signed-off-by: Ryder Lee <[email protected]>
> Signed-off-by: Sam Shih <[email protected]>
Acked-by: Uwe Kleine-K?nig <[email protected]>
Thanks
Uwe
--
Pengutronix e.K. | Uwe Kleine-K?nig |
Industrial Linux Solutions | http://www.pengutronix.de/ |
Hello Sam,
On Thu, Aug 22, 2019 at 02:58:33PM +0800, Sam Shih wrote:
> This fix mt7628 pwm during configure from userspace. The SoC
> is legacy MIPS and has no complex clock tree. This patch add property
> clock-frequency to the SoC specific data and legacy MIPS SoC need to
> configure it in DT. This property is use for period calculation.
>
> We will improve this fix by droping has-clks attribute and using
> clock-frequency to do the same thing in a new patch.
>
> Signed-off-by: Ryder Lee <[email protected]>
> Signed-off-by: Sam Shih <[email protected]>
I wonder if instead the platform could provide some dummy and optional
clocks.
You could add a fixed-clock for the clk that is used to determine the
clock rate and switch to devm_clk_get_optional for the others.
Best regards
Uwe
--
Pengutronix e.K. | Uwe Kleine-K?nig |
Industrial Linux Solutions | http://www.pengutronix.de/ |
On Thu, Aug 22, 2019 at 02:58:34PM +0800, Sam Shih wrote:
> Instead of using fixed size of arrays, allocate the memory for them
> based on the information we get from the DT.
>
> Also remove the check for num_pwms, due to dynamically allocate pwm
> should not cause array index out of bound.
>
> Signed-off-by: Ryder Lee <[email protected]>
> Signed-off-by: Sam Shih <[email protected]>
Reviewed-by: Uwe Kleine-K?nig <[email protected]>
Thanks
Uwe
--
Pengutronix e.K. | Uwe Kleine-K?nig |
Industrial Linux Solutions | http://www.pengutronix.de/ |
On Thu, Aug 22, 2019 at 02:58:35PM +0800, Sam Shih wrote:
> Use pwm_mediatek as common prefix to match the filename.
> No functional change intended.
>
> Signed-off-by: Ryder Lee <[email protected]>
> Signed-off-by: Sam Shih <[email protected]>
> ---
> Changes since v5:
> - Follow reviewers's comments
> The license stuff is a separate change
this is a nice cleanup, I like it.
Acked-by: Uwe Kleine-K?nig <[email protected]>
Thanks
Uwe
--
Pengutronix e.K. | Uwe Kleine-K?nig |
Industrial Linux Solutions | http://www.pengutronix.de/ |
On Thu, Aug 22, 2019 at 02:58:36PM +0800, Sam Shih wrote:
> Add SPDX identifiers to pwm-mediatek.c
> Update license to GNU General Public License v2.0
>
> Signed-off-by: Ryder Lee <[email protected]>
> Signed-off-by: Sam Shih <[email protected]>
Reviewed-by: Uwe Kleine-K?nig <[email protected]>
Thanks
Uwe
--
Pengutronix e.K. | Uwe Kleine-K?nig |
Industrial Linux Solutions | http://www.pengutronix.de/ |
On Thu, Aug 22, 2019 at 04:12:20PM +0800, Yingjoe Chen wrote:
> On Thu, 2019-08-22 at 14:58 +0800, Sam Shih wrote:
> > This updates bindings for MT7628 pwm controller.
> >
> > Signed-off-by: Sam Shih <[email protected]>
> > ---
> > Documentation/devicetree/bindings/pwm/pwm-mediatek.txt | 4 ++++
> > 1 file changed, 4 insertions(+)
> >
> > diff --git a/Documentation/devicetree/bindings/pwm/pwm-mediatek.txt b/Documentation/devicetree/bindings/pwm/pwm-mediatek.txt
> > index ea95b490a913..93980e3da261 100644
> > --- a/Documentation/devicetree/bindings/pwm/pwm-mediatek.txt
> > +++ b/Documentation/devicetree/bindings/pwm/pwm-mediatek.txt
> > @@ -21,6 +21,10 @@ Required properties:
> > See pinctrl/pinctrl-bindings.txt for details of the property values.
> > - num-pwms: the number of PWM channels.
> > +
> > + Optional properties:
> > + - clock-frequency: fix clock frequency, this is only used in MT7628 SoC
> > + for period calculation. This SoC has no complex clock tree.
>
> I'm sorry if this has been discussed before.
>
> Would it be simpler if you just provide a fixed-clock as clock in device
> tree? This way you don't need this optional properties and don't need to
> special handle it in driver code.
>
> After all, the hw is still connected to a simple clock tree.
This is what I just wrote in reply to patch 3 (which implements handling
the clock-frequency property) even before reading your feedback. So I
fully agree.
Best regards
Uwe
--
Pengutronix e.K. | Uwe Kleine-K?nig |
Industrial Linux Solutions | http://www.pengutronix.de/ |
On Thu, Aug 22, 2019 at 02:58:39PM +0800, Sam Shih wrote:
> From: Ryder Lee <[email protected]>
>
> This adds a property "num-pwms" for PWM controller.
>
> Signed-off-by: Ryder Lee <[email protected]>
> Signed-off-by: Sam Shih <[email protected]>
> ---
> arch/arm64/boot/dts/mediatek/mt7622.dtsi | 1 +
> 1 file changed, 1 insertion(+)
>
> diff --git a/arch/arm64/boot/dts/mediatek/mt7622.dtsi b/arch/arm64/boot/dts/mediatek/mt7622.dtsi
> index d1e13d340e26..9a043938881f 100644
> --- a/arch/arm64/boot/dts/mediatek/mt7622.dtsi
> +++ b/arch/arm64/boot/dts/mediatek/mt7622.dtsi
> @@ -439,6 +439,7 @@
> <&pericfg CLK_PERI_PWM6_PD>;
> clock-names = "top", "main", "pwm1", "pwm2", "pwm3", "pwm4",
> "pwm5", "pwm6";
> + num-pwms = <6>;
> status = "disabled";
> };
FTR: The matching change to the binding is patch 7 in this series and
didn't get an Ack from the dt people yet.
Best regards
Uwe
--
Pengutronix e.K. | Uwe Kleine-K?nig |
Industrial Linux Solutions | http://www.pengutronix.de/ |
On Thu, Aug 22, 2019 at 02:58:40PM +0800, Sam Shih wrote:
> From: Ryder Lee <[email protected]>
>
> This adds a property "num-pwms" for PWM controller.
>
> Signed-off-by: Ryder Lee <[email protected]>
> Signed-off-by: Sam Shih <[email protected]>
> ---
> arch/arm/boot/dts/mt7623.dtsi | 1 +
> 1 file changed, 1 insertion(+)
>
> diff --git a/arch/arm/boot/dts/mt7623.dtsi b/arch/arm/boot/dts/mt7623.dtsi
> index a79f0b6c3429..208e0d19a575 100644
> --- a/arch/arm/boot/dts/mt7623.dtsi
> +++ b/arch/arm/boot/dts/mt7623.dtsi
> @@ -452,6 +452,7 @@
> <&pericfg CLK_PERI_PWM5>;
> clock-names = "top", "main", "pwm1", "pwm2",
> "pwm3", "pwm4", "pwm5";
> + num-pwms = <5>;
> status = "disabled";
> };
FTR: The matching change to the binding is patch 7 in this series and
didn't get an Ack from the dt people yet.
Best regards
Uwe
--
Pengutronix e.K. | Uwe Kleine-K?nig |
Industrial Linux Solutions | http://www.pengutronix.de/ |
On Thu, Aug 22, 2019 at 02:58:41PM +0800, Sam Shih wrote:
> From: Ryder Lee <[email protected]>
>
> This updates bindings for MT7629 pwm controller.
>
> This patch is the same as
> https://patchwork.kernel.org/patch/10769381/
> and it has a Reviewed-by tag in v1
This paragraph doesn't belong in the commit log.
>
> Signed-off-by: Ryder Lee <[email protected]>
> Signed-off-by: Sam Shih <[email protected]>
> Reviewed-by: Matthias Brugger <[email protected]>
Thanks
Uwe
--
Pengutronix e.K. | Uwe Kleine-K?nig |
Industrial Linux Solutions | http://www.pengutronix.de/ |
On Thu, Aug 22, 2019 at 02:58:42PM +0800, Sam Shih wrote:
> Due to we added clock-frequency property to fix
> mt7628 pwm during configure from userspace.
> We can alos use this property to determine whether
> the complex clock tree exists in the SoC or not.
> So we can safety remove has-clock property in the
> driver specific data.
Some suggestions in short form:
s/Due/Since/
s/alos/also/
Also please use more horizontal space, up to 76 chars per line is fine.
Other than that I suggest to first address the feedback for the earlier
patches as the needed changes there has influence on this patch.
Best regards
Uwe
--
Pengutronix e.K. | Uwe Kleine-K?nig |
Industrial Linux Solutions | http://www.pengutronix.de/ |
On Thu, Aug 22, 2019 at 02:58:37PM +0800, Sam Shih wrote:
> From: Ryder Lee <[email protected]>
The subject should indicate this is for Mediatek.
>
> This adds a property "num-pwms" in example so that we could
> specify the number of PWM channels via device tree.
>
> Signed-off-by: Ryder Lee <[email protected]>
> Signed-off-by: Sam Shih <[email protected]>
> Reviewed-by: Matthias Brugger <[email protected]>
> Acked-by: Uwe Kleine-K?nig <[email protected]>
> ---
> Changes since v5:
> - Add an Acked-by tag
> - This file is original v4 patch 5/10
> (https://patchwork.kernel.org/patch/11102577/)
>
> Change-Id: I429048afeffa96f3f14533910efe242f88776043
> ---
> Documentation/devicetree/bindings/pwm/pwm-mediatek.txt | 7 ++++---
> 1 file changed, 4 insertions(+), 3 deletions(-)
>
> diff --git a/Documentation/devicetree/bindings/pwm/pwm-mediatek.txt b/Documentation/devicetree/bindings/pwm/pwm-mediatek.txt
> index 991728cb46cb..ea95b490a913 100644
> --- a/Documentation/devicetree/bindings/pwm/pwm-mediatek.txt
> +++ b/Documentation/devicetree/bindings/pwm/pwm-mediatek.txt
> @@ -14,12 +14,12 @@ Required properties:
> has no clocks
> - "top": the top clock generator
> - "main": clock used by the PWM core
> - - "pwm1-8": the eight per PWM clocks for mt2712
> - - "pwm1-6": the six per PWM clocks for mt7622
> - - "pwm1-5": the five per PWM clocks for mt7623
> + - "pwm1-N": the PWM clocks for each channel
> + where N starting from 1 to the maximum number of PWM channels
Once converted to schema, you are going to be back to listing them out.
> - pinctrl-names: Must contain a "default" entry.
> - pinctrl-0: One property must exist for each entry in pinctrl-names.
> See pinctrl/pinctrl-bindings.txt for details of the property values.
> + - num-pwms: the number of PWM channels.
You can't add new required properties without breaking compatibility.
You already have to imply the number of channels from the compatible (or
number of clocks) and you have to keep doing so to maintain
compatibility, so why not just keep doing that for new chips?
Rob
On Tue, Aug 27, 2019 at 01:39:24PM -0500, Rob Herring wrote:
> On Thu, Aug 22, 2019 at 02:58:37PM +0800, Sam Shih wrote:
> > From: Ryder Lee <[email protected]>
>
> The subject should indicate this is for Mediatek.
>
> >
> > This adds a property "num-pwms" in example so that we could
> > specify the number of PWM channels via device tree.
> >
> > Signed-off-by: Ryder Lee <[email protected]>
> > Signed-off-by: Sam Shih <[email protected]>
> > Reviewed-by: Matthias Brugger <[email protected]>
> > Acked-by: Uwe Kleine-K?nig <[email protected]>
> > ---
> > Changes since v5:
> > - Add an Acked-by tag
> > - This file is original v4 patch 5/10
> > (https://patchwork.kernel.org/patch/11102577/)
> >
> > Change-Id: I429048afeffa96f3f14533910efe242f88776043
> > ---
> > Documentation/devicetree/bindings/pwm/pwm-mediatek.txt | 7 ++++---
> > 1 file changed, 4 insertions(+), 3 deletions(-)
> >
> > diff --git a/Documentation/devicetree/bindings/pwm/pwm-mediatek.txt b/Documentation/devicetree/bindings/pwm/pwm-mediatek.txt
> > index 991728cb46cb..ea95b490a913 100644
> > --- a/Documentation/devicetree/bindings/pwm/pwm-mediatek.txt
> > +++ b/Documentation/devicetree/bindings/pwm/pwm-mediatek.txt
> > @@ -14,12 +14,12 @@ Required properties:
> > has no clocks
> > - "top": the top clock generator
> > - "main": clock used by the PWM core
> > - - "pwm1-8": the eight per PWM clocks for mt2712
> > - - "pwm1-6": the six per PWM clocks for mt7622
> > - - "pwm1-5": the five per PWM clocks for mt7623
> > + - "pwm1-N": the PWM clocks for each channel
> > + where N starting from 1 to the maximum number of PWM channels
>
> Once converted to schema, you are going to be back to listing them out.
>
> > - pinctrl-names: Must contain a "default" entry.
> > - pinctrl-0: One property must exist for each entry in pinctrl-names.
> > See pinctrl/pinctrl-bindings.txt for details of the property values.
> > + - num-pwms: the number of PWM channels.
>
> You can't add new required properties without breaking compatibility.
>
> You already have to imply the number of channels from the compatible (or
> number of clocks) and you have to keep doing so to maintain
> compatibility, so why not just keep doing that for new chips?
This was a suggestion by me. The driver still handles compatibility
(i.e. falls back to the number of PWMs that was implied by the
compatible before). Given that there are various drivers that all solve
the same problem (i.e. different variants with different number of PWMs)
I thought it would be a good idea to introduce a property in the device
tree that specifies this number.
Only for newly introduced compatibles the num-pwms property is really
required. Differentiating the ones that need it and the ones that don't
seems over-engineered to me.
(BTW, using the number of clks doesn't really work because there are
also some variants without clocks. It is still under discussion if in
this case dummy-clocks should be provided IIRC.)
Best regards
Uwe
--
Pengutronix e.K. | Uwe Kleine-K?nig |
Industrial Linux Solutions | http://www.pengutronix.de/ |
On Mon, 2019-09-02 at 18:04 +0200, Uwe Kleine-K?nig wrote:
> On Tue, Aug 27, 2019 at 01:39:24PM -0500, Rob Herring wrote:
> > On Thu, Aug 22, 2019 at 02:58:37PM +0800, Sam Shih wrote:
> > > From: Ryder Lee <[email protected]>
> >
> > The subject should indicate this is for Mediatek.
> >
> > >
> > > This adds a property "num-pwms" in example so that we could
> > > specify the number of PWM channels via device tree.
> > >
> > > Signed-off-by: Ryder Lee <[email protected]>
> > > Signed-off-by: Sam Shih <[email protected]>
> > > Reviewed-by: Matthias Brugger <[email protected]>
> > > Acked-by: Uwe Kleine-K?nig <[email protected]>
> > > ---
> > > Changes since v5:
> > > - Add an Acked-by tag
> > > - This file is original v4 patch 5/10
> > > (https://patchwork.kernel.org/patch/11102577/)
> > >
> > > Change-Id: I429048afeffa96f3f14533910efe242f88776043
> > > ---
> > > Documentation/devicetree/bindings/pwm/pwm-mediatek.txt | 7 ++++---
> > > 1 file changed, 4 insertions(+), 3 deletions(-)
> > >
> > > diff --git a/Documentation/devicetree/bindings/pwm/pwm-mediatek.txt b/Documentation/devicetree/bindings/pwm/pwm-mediatek.txt
> > > index 991728cb46cb..ea95b490a913 100644
> > > --- a/Documentation/devicetree/bindings/pwm/pwm-mediatek.txt
> > > +++ b/Documentation/devicetree/bindings/pwm/pwm-mediatek.txt
> > > @@ -14,12 +14,12 @@ Required properties:
> > > has no clocks
> > > - "top": the top clock generator
> > > - "main": clock used by the PWM core
> > > - - "pwm1-8": the eight per PWM clocks for mt2712
> > > - - "pwm1-6": the six per PWM clocks for mt7622
> > > - - "pwm1-5": the five per PWM clocks for mt7623
> > > + - "pwm1-N": the PWM clocks for each channel
> > > + where N starting from 1 to the maximum number of PWM channels
> >
> > Once converted to schema, you are going to be back to listing them out.
> >
> > > - pinctrl-names: Must contain a "default" entry.
> > > - pinctrl-0: One property must exist for each entry in pinctrl-names.
> > > See pinctrl/pinctrl-bindings.txt for details of the property values.
> > > + - num-pwms: the number of PWM channels.
> >
> > You can't add new required properties without breaking compatibility.
> >
> > You already have to imply the number of channels from the compatible (or
> > number of clocks) and you have to keep doing so to maintain
> > compatibility, so why not just keep doing that for new chips?
>
> This was a suggestion by me. The driver still handles compatibility
> (i.e. falls back to the number of PWMs that was implied by the
> compatible before). Given that there are various drivers that all solve
> the same problem (i.e. different variants with different number of PWMs)
> I thought it would be a good idea to introduce a property in the device
> tree that specifies this number.
> Only for newly introduced compatibles the num-pwms property is really
> required. Differentiating the ones that need it and the ones that don't
> seems over-engineered to me.
I'm fine with both.
num-pwms and pwm1-N is required properties for new chip, but it still
have backward compatibility for old dt without num-pwms properties.
> (BTW, using the number of clks doesn't really work because there are
> also some variants without clocks. It is still under discussion if in
> this case dummy-clocks should be provided IIRC.)
Yes, the dummy-clocks of "top", "main" is needed in old MIPS dt.
We also need fixed-clock for period calculation.
Best Regards,
Sam
On Mon, 2019-09-02 at 18:04 +0200, Uwe Kleine-K?nig wrote:
> On Tue, Aug 27, 2019 at 01:39:24PM -0500, Rob Herring wrote:
> > On Thu, Aug 22, 2019 at 02:58:37PM +0800, Sam Shih wrote:
> > > From: Ryder Lee <[email protected]>
> >
> > The subject should indicate this is for Mediatek.
> >
> > >
> > > This adds a property "num-pwms" in example so that we could
> > > specify the number of PWM channels via device tree.
> > >
> > > Signed-off-by: Ryder Lee <[email protected]>
> > > Signed-off-by: Sam Shih <[email protected]>
> > > Reviewed-by: Matthias Brugger <[email protected]>
> > > Acked-by: Uwe Kleine-K?nig <[email protected]>
> > > ---
> > > Changes since v5:
> > > - Add an Acked-by tag
> > > - This file is original v4 patch 5/10
> > > (https://patchwork.kernel.org/patch/11102577/)
> > >
> > > Change-Id: I429048afeffa96f3f14533910efe242f88776043
> > > ---
> > > Documentation/devicetree/bindings/pwm/pwm-mediatek.txt | 7 ++++---
> > > 1 file changed, 4 insertions(+), 3 deletions(-)
> > >
> > > diff --git a/Documentation/devicetree/bindings/pwm/pwm-mediatek.txt b/Documentation/devicetree/bindings/pwm/pwm-mediatek.txt
> > > index 991728cb46cb..ea95b490a913 100644
> > > --- a/Documentation/devicetree/bindings/pwm/pwm-mediatek.txt
> > > +++ b/Documentation/devicetree/bindings/pwm/pwm-mediatek.txt
> > > @@ -14,12 +14,12 @@ Required properties:
> > > has no clocks
> > > - "top": the top clock generator
> > > - "main": clock used by the PWM core
> > > - - "pwm1-8": the eight per PWM clocks for mt2712
> > > - - "pwm1-6": the six per PWM clocks for mt7622
> > > - - "pwm1-5": the five per PWM clocks for mt7623
> > > + - "pwm1-N": the PWM clocks for each channel
> > > + where N starting from 1 to the maximum number of PWM channels
> >
> > Once converted to schema, you are going to be back to listing them out.
> >
> > > - pinctrl-names: Must contain a "default" entry.
> > > - pinctrl-0: One property must exist for each entry in pinctrl-names.
> > > See pinctrl/pinctrl-bindings.txt for details of the property values.
> > > + - num-pwms: the number of PWM channels.
> >
> > You can't add new required properties without breaking compatibility.
> >
> > You already have to imply the number of channels from the compatible (or
> > number of clocks) and you have to keep doing so to maintain
> > compatibility, so why not just keep doing that for new chips?
>
> This was a suggestion by me. The driver still handles compatibility
> (i.e. falls back to the number of PWMs that was implied by the
> compatible before). Given that there are various drivers that all solve
> the same problem (i.e. different variants with different number of PWMs)
> I thought it would be a good idea to introduce a property in the device
> tree that specifies this number.
>
> Only for newly introduced compatibles the num-pwms property is really
> required. Differentiating the ones that need it and the ones that don't
> seems over-engineered to me.
>
> (BTW, using the number of clks doesn't really work because there are
> also some variants without clocks. It is still under discussion if in
> this case dummy-clocks should be provided IIRC.)
>
> Best regards
> Uwe
>
Any conclusions ?
just a friendly reminder :)
regards Sam