| Message ID | 20170724225142.19975-4-fcooper@ti.com |
|---|---|
| State | Awaiting Upstream, archived |
| Delegated to: | David Miller |
| Headers | show |
+ OMAP mailing list On Tuesday 25 July 2017 04:21 AM, Franklin Cooper wrote: > Add support for PM Runtime which is the new way to handle managing clocks. > However, to avoid breaking SoCs not using PM_RUNTIME leave the old clk > management approach in place. Since hclk is the interface clock, I think at a minimum there needs to be an assumption that pm_runtime_get_sync() will enable that clock and make the module ready for access. The only in-kernel user of this driver seems to be an atmel SoC. I am ccing folks who added support for M_CAN on that SoC to see if hclk management can be left to pm_runtime*() on their SoC. If they are okay, I think its a good to atleast drop explicit hclk enable disable in the driver. That way, we avoid double enable/disable of interface clock (hclk). > > PM_RUNTIME is required by OMAP based devices to handle clock management. > Therefore, this allows future Texas Instruments SoCs that have the MCAN IP > to work with this driver. > > Signed-off-by: Franklin S Cooper Jr <fcooper@ti.com> Thanks, Sekhar > --- > drivers/net/can/m_can/m_can.c | 13 +++++++++++++ > 1 file changed, 13 insertions(+) > > diff --git a/drivers/net/can/m_can/m_can.c b/drivers/net/can/m_can/m_can.c > index ea48e59..93e23f5 100644 > --- a/drivers/net/can/m_can/m_can.c > +++ b/drivers/net/can/m_can/m_can.c > @@ -23,6 +23,7 @@ > #include <linux/of.h> > #include <linux/of_device.h> > #include <linux/platform_device.h> > +#include <linux/pm_runtime.h> > #include <linux/iopoll.h> > #include <linux/can/dev.h> > > @@ -633,11 +634,15 @@ static int m_can_clk_start(struct m_can_priv *priv) > if (err) > clk_disable_unprepare(priv->hclk); > > + pm_runtime_get_sync(priv->device); > + > return err; > } > > static void m_can_clk_stop(struct m_can_priv *priv) > { > + pm_runtime_put_sync(priv->device); > + > clk_disable_unprepare(priv->cclk); > clk_disable_unprepare(priv->hclk); > } > @@ -1582,6 +1587,8 @@ static int m_can_plat_probe(struct platform_device *pdev) > /* Enable clocks. Necessary to read Core Release in order to determine > * M_CAN version > */ > + pm_runtime_enable(&pdev->dev); > + > ret = clk_prepare_enable(hclk); > if (ret) > goto disable_hclk_ret; > @@ -1626,6 +1633,8 @@ static int m_can_plat_probe(struct platform_device *pdev) > */ > tx_fifo_size = mram_config_vals[7]; > > + pm_runtime_get_sync(&pdev->dev); > + > /* allocate the m_can device */ > dev = alloc_m_can_dev(pdev, addr, tx_fifo_size); > if (!dev) { > @@ -1670,6 +1679,7 @@ static int m_can_plat_probe(struct platform_device *pdev) > disable_hclk_ret: > clk_disable_unprepare(hclk); > failed_ret: > + pm_runtime_put_sync(&pdev->dev); > return ret; > } > > @@ -1726,6 +1736,9 @@ static int m_can_plat_remove(struct platform_device *pdev) > struct net_device *dev = platform_get_drvdata(pdev); > > unregister_m_can_dev(dev); > + > + pm_runtime_disable(&pdev->dev); > + > platform_set_drvdata(pdev, NULL); > > free_m_can_dev(dev); >
On 08/24/2017 03:30 AM, Sekhar Nori wrote: > + OMAP mailing list > > On Tuesday 25 July 2017 04:21 AM, Franklin Cooper wrote: >> Add support for PM Runtime which is the new way to handle managing clocks. >> However, to avoid breaking SoCs not using PM_RUNTIME leave the old clk >> management approach in place. > > Since hclk is the interface clock, I think at a minimum there needs to > be an assumption that pm_runtime_get_sync() will enable that clock and > make the module ready for access. > > The only in-kernel user of this driver seems to be an atmel SoC. I am > ccing folks who added support for M_CAN on that SoC to see if hclk > management can be left to pm_runtime*() on their SoC. > > If they are okay, I think its a good to atleast drop explicit hclk > enable disable in the driver. That way, we avoid double enable/disable > of interface clock (hclk). Wenyou, Do you foresee this being a problem on your board? If not I can send a v3 removing these clk_enable and clk_disable calls and it would be great if you can test it out and provide your tested by. > >> >> PM_RUNTIME is required by OMAP based devices to handle clock management. >> Therefore, this allows future Texas Instruments SoCs that have the MCAN IP >> to work with this driver. >> >> Signed-off-by: Franklin S Cooper Jr <fcooper@ti.com> > > Thanks, > Sekhar > >> --- >> drivers/net/can/m_can/m_can.c | 13 +++++++++++++ >> 1 file changed, 13 insertions(+) >> >> diff --git a/drivers/net/can/m_can/m_can.c b/drivers/net/can/m_can/m_can.c >> index ea48e59..93e23f5 100644 >> --- a/drivers/net/can/m_can/m_can.c >> +++ b/drivers/net/can/m_can/m_can.c >> @@ -23,6 +23,7 @@ >> #include <linux/of.h> >> #include <linux/of_device.h> >> #include <linux/platform_device.h> >> +#include <linux/pm_runtime.h> >> #include <linux/iopoll.h> >> #include <linux/can/dev.h> >> >> @@ -633,11 +634,15 @@ static int m_can_clk_start(struct m_can_priv *priv) >> if (err) >> clk_disable_unprepare(priv->hclk); >> >> + pm_runtime_get_sync(priv->device); >> + >> return err; >> } >> >> static void m_can_clk_stop(struct m_can_priv *priv) >> { >> + pm_runtime_put_sync(priv->device); >> + >> clk_disable_unprepare(priv->cclk); >> clk_disable_unprepare(priv->hclk); >> } >> @@ -1582,6 +1587,8 @@ static int m_can_plat_probe(struct platform_device *pdev) >> /* Enable clocks. Necessary to read Core Release in order to determine >> * M_CAN version >> */ >> + pm_runtime_enable(&pdev->dev); >> + >> ret = clk_prepare_enable(hclk); >> if (ret) >> goto disable_hclk_ret; >> @@ -1626,6 +1633,8 @@ static int m_can_plat_probe(struct platform_device *pdev) >> */ >> tx_fifo_size = mram_config_vals[7]; >> >> + pm_runtime_get_sync(&pdev->dev); >> + >> /* allocate the m_can device */ >> dev = alloc_m_can_dev(pdev, addr, tx_fifo_size); >> if (!dev) { >> @@ -1670,6 +1679,7 @@ static int m_can_plat_probe(struct platform_device *pdev) >> disable_hclk_ret: >> clk_disable_unprepare(hclk); >> failed_ret: >> + pm_runtime_put_sync(&pdev->dev); >> return ret; >> } >> >> @@ -1726,6 +1736,9 @@ static int m_can_plat_remove(struct platform_device *pdev) >> struct net_device *dev = platform_get_drvdata(pdev); >> >> unregister_m_can_dev(dev); >> + >> + pm_runtime_disable(&pdev->dev); >> + >> platform_set_drvdata(pdev, NULL); >> >> free_m_can_dev(dev); >> >
Hi Franklin, On 2017/9/21 8:36, Franklin S Cooper Jr wrote: > On 08/24/2017 03:30 AM, Sekhar Nori wrote: >> + OMAP mailing list >> >> On Tuesday 25 July 2017 04:21 AM, Franklin Cooper wrote: >>> Add support for PM Runtime which is the new way to handle managing clocks. >>> However, to avoid breaking SoCs not using PM_RUNTIME leave the old clk >>> management approach in place. >> Since hclk is the interface clock, I think at a minimum there needs to >> be an assumption that pm_runtime_get_sync() will enable that clock and >> make the module ready for access. >> >> The only in-kernel user of this driver seems to be an atmel SoC. I am >> ccing folks who added support for M_CAN on that SoC to see if hclk >> management can be left to pm_runtime*() on their SoC. >> >> If they are okay, I think its a good to atleast drop explicit hclk >> enable disable in the driver. That way, we avoid double enable/disable >> of interface clock (hclk). > Wenyou, > > Do you foresee this being a problem on your board? If not I can send a > v3 removing these clk_enable and clk_disable calls and it would be great > if you can test it out and provide your tested by. Please send a v3 patch. I would like test it on my side. > >>> PM_RUNTIME is required by OMAP based devices to handle clock management. >>> Therefore, this allows future Texas Instruments SoCs that have the MCAN IP >>> to work with this driver. >>> >>> Signed-off-by: Franklin S Cooper Jr <fcooper@ti.com> >> Thanks, >> Sekhar >> >>> --- >>> drivers/net/can/m_can/m_can.c | 13 +++++++++++++ >>> 1 file changed, 13 insertions(+) >>> >>> diff --git a/drivers/net/can/m_can/m_can.c b/drivers/net/can/m_can/m_can.c >>> index ea48e59..93e23f5 100644 >>> --- a/drivers/net/can/m_can/m_can.c >>> +++ b/drivers/net/can/m_can/m_can.c >>> @@ -23,6 +23,7 @@ >>> #include <linux/of.h> >>> #include <linux/of_device.h> >>> #include <linux/platform_device.h> >>> +#include <linux/pm_runtime.h> >>> #include <linux/iopoll.h> >>> #include <linux/can/dev.h> >>> >>> @@ -633,11 +634,15 @@ static int m_can_clk_start(struct m_can_priv *priv) >>> if (err) >>> clk_disable_unprepare(priv->hclk); >>> >>> + pm_runtime_get_sync(priv->device); >>> + >>> return err; >>> } >>> >>> static void m_can_clk_stop(struct m_can_priv *priv) >>> { >>> + pm_runtime_put_sync(priv->device); >>> + >>> clk_disable_unprepare(priv->cclk); >>> clk_disable_unprepare(priv->hclk); >>> } >>> @@ -1582,6 +1587,8 @@ static int m_can_plat_probe(struct platform_device *pdev) >>> /* Enable clocks. Necessary to read Core Release in order to determine >>> * M_CAN version >>> */ >>> + pm_runtime_enable(&pdev->dev); >>> + >>> ret = clk_prepare_enable(hclk); >>> if (ret) >>> goto disable_hclk_ret; >>> @@ -1626,6 +1633,8 @@ static int m_can_plat_probe(struct platform_device *pdev) >>> */ >>> tx_fifo_size = mram_config_vals[7]; >>> >>> + pm_runtime_get_sync(&pdev->dev); >>> + >>> /* allocate the m_can device */ >>> dev = alloc_m_can_dev(pdev, addr, tx_fifo_size); >>> if (!dev) { >>> @@ -1670,6 +1679,7 @@ static int m_can_plat_probe(struct platform_device *pdev) >>> disable_hclk_ret: >>> clk_disable_unprepare(hclk); >>> failed_ret: >>> + pm_runtime_put_sync(&pdev->dev); >>> return ret; >>> } >>> >>> @@ -1726,6 +1736,9 @@ static int m_can_plat_remove(struct platform_device *pdev) >>> struct net_device *dev = platform_get_drvdata(pdev); >>> >>> unregister_m_can_dev(dev); >>> + >>> + pm_runtime_disable(&pdev->dev); >>> + >>> platform_set_drvdata(pdev, NULL); >>> >>> free_m_can_dev(dev); >>> > -- > To unsubscribe from this list: send the line "unsubscribe linux-can" in > the body of a message to majordomo@vger.kernel.org > More majordomo info at http://vger.kernel.org/majordomo-info.html Best Regards, Wenyou Yang
diff --git a/drivers/net/can/m_can/m_can.c b/drivers/net/can/m_can/m_can.c index ea48e59..93e23f5 100644 --- a/drivers/net/can/m_can/m_can.c +++ b/drivers/net/can/m_can/m_can.c @@ -23,6 +23,7 @@ #include <linux/of.h> #include <linux/of_device.h> #include <linux/platform_device.h> +#include <linux/pm_runtime.h> #include <linux/iopoll.h> #include <linux/can/dev.h> @@ -633,11 +634,15 @@ static int m_can_clk_start(struct m_can_priv *priv) if (err) clk_disable_unprepare(priv->hclk); + pm_runtime_get_sync(priv->device); + return err; } static void m_can_clk_stop(struct m_can_priv *priv) { + pm_runtime_put_sync(priv->device); + clk_disable_unprepare(priv->cclk); clk_disable_unprepare(priv->hclk); } @@ -1582,6 +1587,8 @@ static int m_can_plat_probe(struct platform_device *pdev) /* Enable clocks. Necessary to read Core Release in order to determine * M_CAN version */ + pm_runtime_enable(&pdev->dev); + ret = clk_prepare_enable(hclk); if (ret) goto disable_hclk_ret; @@ -1626,6 +1633,8 @@ static int m_can_plat_probe(struct platform_device *pdev) */ tx_fifo_size = mram_config_vals[7]; + pm_runtime_get_sync(&pdev->dev); + /* allocate the m_can device */ dev = alloc_m_can_dev(pdev, addr, tx_fifo_size); if (!dev) { @@ -1670,6 +1679,7 @@ static int m_can_plat_probe(struct platform_device *pdev) disable_hclk_ret: clk_disable_unprepare(hclk); failed_ret: + pm_runtime_put_sync(&pdev->dev); return ret; } @@ -1726,6 +1736,9 @@ static int m_can_plat_remove(struct platform_device *pdev) struct net_device *dev = platform_get_drvdata(pdev); unregister_m_can_dev(dev); + + pm_runtime_disable(&pdev->dev); + platform_set_drvdata(pdev, NULL); free_m_can_dev(dev);
Add support for PM Runtime which is the new way to handle managing clocks. However, to avoid breaking SoCs not using PM_RUNTIME leave the old clk management approach in place. PM_RUNTIME is required by OMAP based devices to handle clock management. Therefore, this allows future Texas Instruments SoCs that have the MCAN IP to work with this driver. Signed-off-by: Franklin S Cooper Jr <fcooper@ti.com> --- drivers/net/can/m_can/m_can.c | 13 +++++++++++++ 1 file changed, 13 insertions(+)