* [PATCH v2 0/3] mfd: twl: Add clock for TWL6030
@ 2024-10-02 11:07 Andreas Kemnade
2024-10-02 11:07 ` [PATCH v2 1/3] mfd: twl-core: Add a clock subdevice for the TWL6030 Andreas Kemnade
` (3 more replies)
0 siblings, 4 replies; 7+ messages in thread
From: Andreas Kemnade @ 2024-10-02 11:07 UTC (permalink / raw)
To: Stephen Boyd, Tony Lindgren, Aaro Koskinen, Kevin Hilman,
linux-clk, linux-kernel, Lee Jones, linux-omap, Roger Quadros,
Michael Turquette
Cc: Andreas Kemnade
Previously the clock support for only implemented for TWL6032 so add
it also for the TWL6030. There are devices out there where especially
WLAN only works if these clocks are enabled by some patched U-Boot.
This allows to explicitly specify the clock requirements.
Andreas Kemnade (3):
mfd: twl-core: Add a clock subdevice for the TWL6030
clk: twl: remove is_prepared
clk: twl: add TWL6030 support
drivers/clk/Kconfig | 2 +-
drivers/clk/clk-twl.c | 71 +++++++++++++++++++++++++-----------------
drivers/mfd/twl-core.c | 32 +++++++++++++------
3 files changed, 65 insertions(+), 40 deletions(-)
--
2.39.5
^ permalink raw reply [flat|nested] 7+ messages in thread
* [PATCH v2 1/3] mfd: twl-core: Add a clock subdevice for the TWL6030
2024-10-02 11:07 [PATCH v2 0/3] mfd: twl: Add clock for TWL6030 Andreas Kemnade
@ 2024-10-02 11:07 ` Andreas Kemnade
2024-10-02 11:07 ` [PATCH v2 2/3] clk: twl: remove is_prepared Andreas Kemnade
` (2 subsequent siblings)
3 siblings, 0 replies; 7+ messages in thread
From: Andreas Kemnade @ 2024-10-02 11:07 UTC (permalink / raw)
To: Stephen Boyd, Tony Lindgren, Aaro Koskinen, Kevin Hilman,
linux-clk, linux-kernel, Lee Jones, linux-omap, Roger Quadros,
Michael Turquette
Cc: Andreas Kemnade
Also the TWL6030 has some clocks, so add a subdevice for that.
Signed-off-by: Andreas Kemnade <andreas@kemnade.info>
Reviewed-by: Roger Quadros <rogerq@kernel.org>
---
drivers/mfd/twl-core.c | 32 ++++++++++++++++++++++----------
1 file changed, 22 insertions(+), 10 deletions(-)
diff --git a/drivers/mfd/twl-core.c b/drivers/mfd/twl-core.c
index c130ffef182f..c981922f62d5 100644
--- a/drivers/mfd/twl-core.c
+++ b/drivers/mfd/twl-core.c
@@ -711,6 +711,10 @@ static struct of_dev_auxdata twl_auxdata_lookup[] = {
{ /* sentinel */ },
};
+static const struct mfd_cell twl6030_cells[] = {
+ { .name = "twl6030-clk" },
+};
+
static const struct mfd_cell twl6032_cells[] = {
{ .name = "twl6032-clk" },
};
@@ -861,17 +865,25 @@ twl_probe(struct i2c_client *client)
TWL4030_DCDC_GLOBAL_CFG);
}
- if (id->driver_data == (TWL6030_CLASS | TWL6032_SUBCLASS)) {
- status = devm_mfd_add_devices(&client->dev,
- PLATFORM_DEVID_NONE,
- twl6032_cells,
- ARRAY_SIZE(twl6032_cells),
- NULL, 0, NULL);
- if (status < 0)
- goto free;
- }
-
if (twl_class_is_6030()) {
+ if (id->driver_data & TWL6032_SUBCLASS) {
+ status = devm_mfd_add_devices(&client->dev,
+ PLATFORM_DEVID_NONE,
+ twl6032_cells,
+ ARRAY_SIZE(twl6032_cells),
+ NULL, 0, NULL);
+ if (status < 0)
+ goto free;
+ } else {
+ status = devm_mfd_add_devices(&client->dev,
+ PLATFORM_DEVID_NONE,
+ twl6030_cells,
+ ARRAY_SIZE(twl6030_cells),
+ NULL, 0, NULL);
+ if (status < 0)
+ goto free;
+ }
+
if (of_device_is_system_power_controller(node)) {
if (!pm_power_off)
pm_power_off = twl6030_power_off;
--
2.39.5
^ permalink raw reply related [flat|nested] 7+ messages in thread
* [PATCH v2 2/3] clk: twl: remove is_prepared
2024-10-02 11:07 [PATCH v2 0/3] mfd: twl: Add clock for TWL6030 Andreas Kemnade
2024-10-02 11:07 ` [PATCH v2 1/3] mfd: twl-core: Add a clock subdevice for the TWL6030 Andreas Kemnade
@ 2024-10-02 11:07 ` Andreas Kemnade
2024-10-02 12:09 ` Roger Quadros
2024-10-02 11:07 ` [PATCH v2 3/3] clk: twl: add TWL6030 support Andreas Kemnade
2024-10-02 13:13 ` [PATCH v2 0/3] mfd: twl: Add clock for TWL6030 Andreas Kemnade
3 siblings, 1 reply; 7+ messages in thread
From: Andreas Kemnade @ 2024-10-02 11:07 UTC (permalink / raw)
To: Stephen Boyd, Tony Lindgren, Aaro Koskinen, Kevin Hilman,
linux-clk, linux-kernel, Lee Jones, linux-omap, Roger Quadros,
Michael Turquette
Cc: Andreas Kemnade
Remave is_prepared to simplify adding of TWL6030 support.
The default implementation should be enough.
Signed-off-by: Andreas Kemnade <andreas@kemnade.info>
---
drivers/clk/clk-twl.c | 17 -----------------
1 file changed, 17 deletions(-)
diff --git a/drivers/clk/clk-twl.c b/drivers/clk/clk-twl.c
index eab9d3c8ed8a..1d684b358401 100644
--- a/drivers/clk/clk-twl.c
+++ b/drivers/clk/clk-twl.c
@@ -77,26 +77,9 @@ static void twl6032_clks_unprepare(struct clk_hw *hw)
dev_err(cinfo->dev, "clk unprepare failed\n");
}
-static int twl6032_clks_is_prepared(struct clk_hw *hw)
-{
- struct twl_clock_info *cinfo = to_twl_clks_info(hw);
- int val;
-
- val = twlclk_read(cinfo, TWL_MODULE_PM_RECEIVER, VREG_STATE);
- if (val < 0) {
- dev_err(cinfo->dev, "clk read failed\n");
- return val;
- }
-
- val &= TWL6030_CFG_STATE_MASK;
-
- return val == TWL6030_CFG_STATE_ON;
-}
-
static const struct clk_ops twl6032_clks_ops = {
.prepare = twl6032_clks_prepare,
.unprepare = twl6032_clks_unprepare,
- .is_prepared = twl6032_clks_is_prepared,
.recalc_rate = twl_clks_recalc_rate,
};
--
2.39.5
^ permalink raw reply related [flat|nested] 7+ messages in thread
* [PATCH v2 3/3] clk: twl: add TWL6030 support
2024-10-02 11:07 [PATCH v2 0/3] mfd: twl: Add clock for TWL6030 Andreas Kemnade
2024-10-02 11:07 ` [PATCH v2 1/3] mfd: twl-core: Add a clock subdevice for the TWL6030 Andreas Kemnade
2024-10-02 11:07 ` [PATCH v2 2/3] clk: twl: remove is_prepared Andreas Kemnade
@ 2024-10-02 11:07 ` Andreas Kemnade
2024-10-02 12:22 ` Roger Quadros
2024-10-02 13:13 ` [PATCH v2 0/3] mfd: twl: Add clock for TWL6030 Andreas Kemnade
3 siblings, 1 reply; 7+ messages in thread
From: Andreas Kemnade @ 2024-10-02 11:07 UTC (permalink / raw)
To: Stephen Boyd, Tony Lindgren, Aaro Koskinen, Kevin Hilman,
linux-clk, linux-kernel, Lee Jones, linux-omap, Roger Quadros,
Michael Turquette
Cc: Andreas Kemnade
The TWL6030 has similar clocks, so add support for it. Take care of the
resource grouping handling needed.
Signed-off-by: Andreas Kemnade <andreas@kemnade.info>
---
drivers/clk/Kconfig | 2 +-
drivers/clk/clk-twl.c | 54 +++++++++++++++++++++++++++++++++----------
2 files changed, 43 insertions(+), 13 deletions(-)
diff --git a/drivers/clk/Kconfig b/drivers/clk/Kconfig
index 299bc678ed1b..82ec12f9b82c 100644
--- a/drivers/clk/Kconfig
+++ b/drivers/clk/Kconfig
@@ -291,7 +291,7 @@ config CLK_TWL
help
Enable support for controlling the clock resources on TWL family
PMICs. These devices have some 32K clock outputs which can be
- controlled by software. For now, only the TWL6032 clocks are
+ controlled by software. For now, the TWL6032 and TWL6030 clocks are
supported.
config CLK_TWL6040
diff --git a/drivers/clk/clk-twl.c b/drivers/clk/clk-twl.c
index 1d684b358401..f3a52f568887 100644
--- a/drivers/clk/clk-twl.c
+++ b/drivers/clk/clk-twl.c
@@ -11,13 +11,26 @@
#include <linux/platform_device.h>
#include <linux/slab.h>
-#define VREG_STATE 2
+#define VREG_STATE 2
+#define VREG_GRP 0
#define TWL6030_CFG_STATE_OFF 0x00
#define TWL6030_CFG_STATE_ON 0x01
#define TWL6030_CFG_STATE_MASK 0x03
+#define TWL6030_CFG_STATE_GRP_SHIFT 5
+#define TWL6030_CFG_STATE_APP_SHIFT 2
+#define TWL6030_CFG_STATE_APP_MASK (0x03 << TWL6030_CFG_STATE_APP_SHIFT)
+#define TWL6030_CFG_STATE_APP(v) (((v) & TWL6030_CFG_STATE_APP_MASK) >>\
+ TWL6030_CFG_STATE_APP_SHIFT)
+#define P1_GRP BIT(0) /* processor power group */
+#define P2_GRP BIT(1)
+#define P3_GRP BIT(2)
+#define ALL_GRP (P1_GRP | P2_GRP | P3_GRP)
+
+#define DRIVER_DATA_TWL6030 0
+#define DRIVER_DATA_TWL6032 1
struct twl_clock_info {
- struct device *dev;
+ struct platform_device *pdev;
u8 base;
struct clk_hw hw;
};
@@ -56,14 +69,21 @@ static unsigned long twl_clks_recalc_rate(struct clk_hw *hw,
static int twl6032_clks_prepare(struct clk_hw *hw)
{
struct twl_clock_info *cinfo = to_twl_clks_info(hw);
- int ret;
- ret = twlclk_write(cinfo, TWL_MODULE_PM_RECEIVER, VREG_STATE,
- TWL6030_CFG_STATE_ON);
- if (ret < 0)
- dev_err(cinfo->dev, "clk prepare failed\n");
+ if (platform_get_device_id(cinfo->pdev)->driver_data == DRIVER_DATA_TWL6030) {
+ int grp;
- return ret;
+ grp = twlclk_read(cinfo, TWL_MODULE_PM_RECEIVER, VREG_GRP);
+ if (grp < 0)
+ return grp;
+
+ return twlclk_write(cinfo, TWL_MODULE_PM_RECEIVER, VREG_STATE,
+ grp << TWL6030_CFG_STATE_GRP_SHIFT |
+ TWL6030_CFG_STATE_ON);
+ }
+
+ return twlclk_write(cinfo, TWL_MODULE_PM_RECEIVER, VREG_STATE,
+ TWL6030_CFG_STATE_ON);
}
static void twl6032_clks_unprepare(struct clk_hw *hw)
@@ -71,10 +91,16 @@ static void twl6032_clks_unprepare(struct clk_hw *hw)
struct twl_clock_info *cinfo = to_twl_clks_info(hw);
int ret;
- ret = twlclk_write(cinfo, TWL_MODULE_PM_RECEIVER, VREG_STATE,
- TWL6030_CFG_STATE_OFF);
+ if (platform_get_device_id(cinfo->pdev)->driver_data == DRIVER_DATA_TWL6030)
+ ret = twlclk_write(cinfo, TWL_MODULE_PM_RECEIVER, VREG_STATE,
+ ALL_GRP << TWL6030_CFG_STATE_GRP_SHIFT |
+ TWL6030_CFG_STATE_OFF);
+ else
+ ret = twlclk_write(cinfo, TWL_MODULE_PM_RECEIVER, VREG_STATE,
+ TWL6030_CFG_STATE_OFF);
+
if (ret < 0)
- dev_err(cinfo->dev, "clk unprepare failed\n");
+ dev_err(&cinfo->pdev->dev, "clk unprepare failed\n");
}
static const struct clk_ops twl6032_clks_ops = {
@@ -137,7 +163,7 @@ static int twl_clks_probe(struct platform_device *pdev)
for (i = 0; i < count; i++) {
cinfo[i].base = hw_data[i].base;
- cinfo[i].dev = &pdev->dev;
+ cinfo[i].pdev = pdev;
cinfo[i].hw.init = &hw_data[i].init;
ret = devm_clk_hw_register(&pdev->dev, &cinfo[i].hw);
if (ret) {
@@ -159,7 +185,11 @@ static int twl_clks_probe(struct platform_device *pdev)
static const struct platform_device_id twl_clks_id[] = {
{
+ .name = "twl6030-clk",
+ .driver_data = DRIVER_DATA_TWL6030,
+ }, {
.name = "twl6032-clk",
+ .driver_data = DRIVER_DATA_TWL6032,
}, {
/* sentinel */
}
--
2.39.5
^ permalink raw reply related [flat|nested] 7+ messages in thread
* Re: [PATCH v2 2/3] clk: twl: remove is_prepared
2024-10-02 11:07 ` [PATCH v2 2/3] clk: twl: remove is_prepared Andreas Kemnade
@ 2024-10-02 12:09 ` Roger Quadros
0 siblings, 0 replies; 7+ messages in thread
From: Roger Quadros @ 2024-10-02 12:09 UTC (permalink / raw)
To: Andreas Kemnade, Stephen Boyd, Tony Lindgren, Aaro Koskinen,
Kevin Hilman, linux-clk, linux-kernel, Lee Jones, linux-omap,
Michael Turquette
On 02/10/2024 14:07, Andreas Kemnade wrote:
> Remave is_prepared to simplify adding of TWL6030 support.
Remave/Remove
> The default implementation should be enough.
>
> Signed-off-by: Andreas Kemnade <andreas@kemnade.info>
Reviewed-by: Roger Quadros <rogerq@kernel.org>
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [PATCH v2 3/3] clk: twl: add TWL6030 support
2024-10-02 11:07 ` [PATCH v2 3/3] clk: twl: add TWL6030 support Andreas Kemnade
@ 2024-10-02 12:22 ` Roger Quadros
0 siblings, 0 replies; 7+ messages in thread
From: Roger Quadros @ 2024-10-02 12:22 UTC (permalink / raw)
To: Andreas Kemnade, Stephen Boyd, Tony Lindgren, Aaro Koskinen,
Kevin Hilman, linux-clk, linux-kernel, Lee Jones, linux-omap,
Michael Turquette
On 02/10/2024 14:07, Andreas Kemnade wrote:
> The TWL6030 has similar clocks, so add support for it. Take care of the
> resource grouping handling needed.
>
> Signed-off-by: Andreas Kemnade <andreas@kemnade.info>
> ---
> drivers/clk/Kconfig | 2 +-
> drivers/clk/clk-twl.c | 54 +++++++++++++++++++++++++++++++++----------
> 2 files changed, 43 insertions(+), 13 deletions(-)
>
> diff --git a/drivers/clk/Kconfig b/drivers/clk/Kconfig
> index 299bc678ed1b..82ec12f9b82c 100644
> --- a/drivers/clk/Kconfig
> +++ b/drivers/clk/Kconfig
> @@ -291,7 +291,7 @@ config CLK_TWL
> help
> Enable support for controlling the clock resources on TWL family
> PMICs. These devices have some 32K clock outputs which can be
> - controlled by software. For now, only the TWL6032 clocks are
> + controlled by software. For now, the TWL6032 and TWL6030 clocks are
> supported.
>
> config CLK_TWL6040
> diff --git a/drivers/clk/clk-twl.c b/drivers/clk/clk-twl.c
> index 1d684b358401..f3a52f568887 100644
> --- a/drivers/clk/clk-twl.c
> +++ b/drivers/clk/clk-twl.c
> @@ -11,13 +11,26 @@
> #include <linux/platform_device.h>
> #include <linux/slab.h>
>
> -#define VREG_STATE 2
> +#define VREG_STATE 2
> +#define VREG_GRP 0
> #define TWL6030_CFG_STATE_OFF 0x00
> #define TWL6030_CFG_STATE_ON 0x01
> #define TWL6030_CFG_STATE_MASK 0x03
> +#define TWL6030_CFG_STATE_GRP_SHIFT 5
> +#define TWL6030_CFG_STATE_APP_SHIFT 2
> +#define TWL6030_CFG_STATE_APP_MASK (0x03 << TWL6030_CFG_STATE_APP_SHIFT)
> +#define TWL6030_CFG_STATE_APP(v) (((v) & TWL6030_CFG_STATE_APP_MASK) >>\
> + TWL6030_CFG_STATE_APP_SHIFT)
> +#define P1_GRP BIT(0) /* processor power group */
> +#define P2_GRP BIT(1)
> +#define P3_GRP BIT(2)
> +#define ALL_GRP (P1_GRP | P2_GRP | P3_GRP)
> +
> +#define DRIVER_DATA_TWL6030 0
> +#define DRIVER_DATA_TWL6032 1
how about using enum here?
enum twl_type {
TWL_TYPE_6030,
TWL_TYPE_6032,
};
>
> struct twl_clock_info {
> - struct device *dev;
> + struct platform_device *pdev;
I wouldn't change this.
Instead, you can add twl_type member here and initialize it in probe.
enum twl_type type;
> u8 base;
> struct clk_hw hw;
> };
> @@ -56,14 +69,21 @@ static unsigned long twl_clks_recalc_rate(struct clk_hw *hw,
> static int twl6032_clks_prepare(struct clk_hw *hw)
> {
> struct twl_clock_info *cinfo = to_twl_clks_info(hw);
> - int ret;
>
> - ret = twlclk_write(cinfo, TWL_MODULE_PM_RECEIVER, VREG_STATE,
> - TWL6030_CFG_STATE_ON);
> - if (ret < 0)
> - dev_err(cinfo->dev, "clk prepare failed\n");
> + if (platform_get_device_id(cinfo->pdev)->driver_data == DRIVER_DATA_TWL6030) {
this changes to
if (cinfo->type == TWL_TYPE_6030)
> + int grp;
>
> - return ret;
> + grp = twlclk_read(cinfo, TWL_MODULE_PM_RECEIVER, VREG_GRP);
> + if (grp < 0)
> + return grp;
> +
> + return twlclk_write(cinfo, TWL_MODULE_PM_RECEIVER, VREG_STATE,
> + grp << TWL6030_CFG_STATE_GRP_SHIFT |
> + TWL6030_CFG_STATE_ON);
> + }
> +
> + return twlclk_write(cinfo, TWL_MODULE_PM_RECEIVER, VREG_STATE,
> + TWL6030_CFG_STATE_ON);
> }
>
> static void twl6032_clks_unprepare(struct clk_hw *hw)
> @@ -71,10 +91,16 @@ static void twl6032_clks_unprepare(struct clk_hw *hw)
> struct twl_clock_info *cinfo = to_twl_clks_info(hw);
> int ret;
>
> - ret = twlclk_write(cinfo, TWL_MODULE_PM_RECEIVER, VREG_STATE,
> - TWL6030_CFG_STATE_OFF);
> + if (platform_get_device_id(cinfo->pdev)->driver_data == DRIVER_DATA_TWL6030)
here too.
> + ret = twlclk_write(cinfo, TWL_MODULE_PM_RECEIVER, VREG_STATE,
> + ALL_GRP << TWL6030_CFG_STATE_GRP_SHIFT |
> + TWL6030_CFG_STATE_OFF);
> + else
> + ret = twlclk_write(cinfo, TWL_MODULE_PM_RECEIVER, VREG_STATE,
> + TWL6030_CFG_STATE_OFF);
> +
> if (ret < 0)
> - dev_err(cinfo->dev, "clk unprepare failed\n");
> + dev_err(&cinfo->pdev->dev, "clk unprepare failed\n");
> }
>
> static const struct clk_ops twl6032_clks_ops = {
> @@ -137,7 +163,7 @@ static int twl_clks_probe(struct platform_device *pdev)
>
> for (i = 0; i < count; i++) {
> cinfo[i].base = hw_data[i].base;
> - cinfo[i].dev = &pdev->dev;
> + cinfo[i].pdev = pdev;
here we can initialize twl type like so.
cinfo[i].type = platform_get_device_id(pdev)->driver_data;
> cinfo[i].hw.init = &hw_data[i].init;
> ret = devm_clk_hw_register(&pdev->dev, &cinfo[i].hw);
> if (ret) {
> @@ -159,7 +185,11 @@ static int twl_clks_probe(struct platform_device *pdev)
>
> static const struct platform_device_id twl_clks_id[] = {
> {
> + .name = "twl6030-clk",
> + .driver_data = DRIVER_DATA_TWL6030,
TWL_TYPE_6030
> + }, {
> .name = "twl6032-clk",
> + .driver_data = DRIVER_DATA_TWL6032,
TWL_TYPE_6032
> }, {
> /* sentinel */
> }
--
cheers,
-roger
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [PATCH v2 0/3] mfd: twl: Add clock for TWL6030
2024-10-02 11:07 [PATCH v2 0/3] mfd: twl: Add clock for TWL6030 Andreas Kemnade
` (2 preceding siblings ...)
2024-10-02 11:07 ` [PATCH v2 3/3] clk: twl: add TWL6030 support Andreas Kemnade
@ 2024-10-02 13:13 ` Andreas Kemnade
3 siblings, 0 replies; 7+ messages in thread
From: Andreas Kemnade @ 2024-10-02 13:13 UTC (permalink / raw)
To: Stephen Boyd, Tony Lindgren, Aaro Koskinen, Kevin Hilman,
linux-clk, linux-kernel, Lee Jones, linux-omap, Roger Quadros,
Michael Turquette
Am Wed, 2 Oct 2024 13:07:15 +0200
schrieb Andreas Kemnade <andreas@kemnade.info>:
> Previously the clock support for only implemented for TWL6032 so add
> it also for the TWL6030. There are devices out there where especially
> WLAN only works if these clocks are enabled by some patched U-Boot.
> This allows to explicitly specify the clock requirements.
>
oh, forgotten the changelog:
Changes in V2:
- cleanup some defines
- no separate ops for 6030
- remove is_prepared()
- update Kconfig
Regards,
Andreas
^ permalink raw reply [flat|nested] 7+ messages in thread
end of thread, other threads:[~2024-10-02 13:13 UTC | newest]
Thread overview: 7+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2024-10-02 11:07 [PATCH v2 0/3] mfd: twl: Add clock for TWL6030 Andreas Kemnade
2024-10-02 11:07 ` [PATCH v2 1/3] mfd: twl-core: Add a clock subdevice for the TWL6030 Andreas Kemnade
2024-10-02 11:07 ` [PATCH v2 2/3] clk: twl: remove is_prepared Andreas Kemnade
2024-10-02 12:09 ` Roger Quadros
2024-10-02 11:07 ` [PATCH v2 3/3] clk: twl: add TWL6030 support Andreas Kemnade
2024-10-02 12:22 ` Roger Quadros
2024-10-02 13:13 ` [PATCH v2 0/3] mfd: twl: Add clock for TWL6030 Andreas Kemnade
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).