Add PM handlers for System suspend/resume.
As DMA driver doesn't yet support suspend/resume we free up
the DMA channels at suspend and acquire and initialize them
at resume.
In this revised approach we do not free the TX/RX IRQs at
am65_cpsw_nuss_common_stop() as it causes problems.
We will now free them only on .suspend() as we need to release
the DMA channels (as DMA looses context) and re-acquiring
them on .resume() may not necessarily give us the same
IRQs.
To make this easier:
- introduce am65_cpsw_nuss_remove_rx_chns() which is
similar to am65_cpsw_nuss_remove_tx_chns(). These will
be invoked in pm.suspend() to release the DMA channels
and free up the IRQs.
- move napi_add() and request_irq() calls to
am65_cpsw_nuss_init_rx/tx_chns() so we can invoke them
in pm.resume() to acquire the DMA channels and IRQs.
As CPTS looses contect during suspend/resume, invoke the
necessary CPTS suspend/resume helpers.
ALE_CLEAR command is issued in cpsw_ale_start() so no need
to issue it before the call to cpsw_ale_start().
Signed-off-by: Roger Quadros <rogerq@kernel.org>
---
drivers/net/ethernet/ti/am65-cpsw-nuss.c | 222 ++++++++++++++++++-----
1 file changed, 173 insertions(+), 49 deletions(-)
diff --git a/drivers/net/ethernet/ti/am65-cpsw-nuss.c b/drivers/net/ethernet/ti/am65-cpsw-nuss.c
index 4836960b0dd8..1c5a51439427 100644
--- a/drivers/net/ethernet/ti/am65-cpsw-nuss.c
+++ b/drivers/net/ethernet/ti/am65-cpsw-nuss.c
@@ -24,6 +24,7 @@
#include <linux/platform_device.h>
#include <linux/pm_runtime.h>
#include <linux/regmap.h>
+#include <linux/rtnetlink.h>
#include <linux/mfd/syscon.h>
#include <linux/sys_soc.h>
#include <linux/dma/ti-cppi5.h>
@@ -555,11 +556,26 @@ static int am65_cpsw_nuss_ndo_slave_open(struct net_device *ndev)
struct am65_cpsw_common *common = am65_ndev_to_common(ndev);
struct am65_cpsw_port *port = am65_ndev_to_port(ndev);
int ret, i;
+ u32 reg;
ret = pm_runtime_resume_and_get(common->dev);
if (ret < 0)
return ret;
+ /* Idle MAC port */
+ cpsw_sl_ctl_set(port->slave.mac_sl, CPSW_SL_CTL_CMD_IDLE);
+ cpsw_sl_wait_for_idle(port->slave.mac_sl, 100);
+ cpsw_sl_ctl_reset(port->slave.mac_sl);
+
+ /* soft reset MAC */
+ cpsw_sl_reg_write(port->slave.mac_sl, CPSW_SL_SOFT_RESET, 1);
+ mdelay(1);
+ reg = cpsw_sl_reg_read(port->slave.mac_sl, CPSW_SL_SOFT_RESET);
+ if (reg) {
+ dev_err(common->dev, "soft RESET didn't complete\n");
+ return -EBUSY;
+ }
+
/* Notify the stack of the actual queue counts. */
ret = netif_set_real_num_tx_queues(ndev, common->tx_ch_num);
if (ret) {
@@ -1533,6 +1549,32 @@ void am65_cpsw_nuss_remove_tx_chns(struct am65_cpsw_common *common)
}
}
+static int am65_cpsw_nuss_ndev_add_tx_napi(struct am65_cpsw_common *common)
+{
+ struct device *dev = common->dev;
+ int i, ret = 0;
+
+ for (i = 0; i < common->tx_ch_num; i++) {
+ struct am65_cpsw_tx_chn *tx_chn = &common->tx_chns[i];
+
+ netif_napi_add_tx(common->dma_ndev, &tx_chn->napi_tx,
+ am65_cpsw_nuss_tx_poll);
+
+ ret = devm_request_irq(dev, tx_chn->irq,
+ am65_cpsw_nuss_tx_irq,
+ IRQF_TRIGGER_HIGH,
+ tx_chn->tx_chn_name, tx_chn);
+ if (ret) {
+ dev_err(dev, "failure requesting tx%u irq %u, %d\n",
+ tx_chn->id, tx_chn->irq, ret);
+ goto err;
+ }
+ }
+
+err:
+ return ret;
+}
+
static int am65_cpsw_nuss_init_tx_chns(struct am65_cpsw_common *common)
{
u32 max_desc_num = ALIGN(AM65_CPSW_MAX_TX_DESC, MAX_SKB_FRAGS);
@@ -1599,6 +1641,12 @@ static int am65_cpsw_nuss_init_tx_chns(struct am65_cpsw_common *common)
dev_name(dev), tx_chn->id);
}
+ ret = am65_cpsw_nuss_ndev_add_tx_napi(common);
+ if (ret) {
+ dev_err(dev, "Failed to add tx NAPI %d\n", ret);
+ goto err;
+ }
+
err:
i = devm_add_action(dev, am65_cpsw_nuss_free_tx_chns, common);
if (i) {
@@ -1623,6 +1671,29 @@ static void am65_cpsw_nuss_free_rx_chns(void *data)
k3_udma_glue_release_rx_chn(rx_chn->rx_chn);
}
+static void am65_cpsw_nuss_remove_rx_chns(void *data)
+{
+ struct am65_cpsw_common *common = data;
+ struct am65_cpsw_rx_chn *rx_chn;
+ struct device *dev = common->dev;
+
+ rx_chn = &common->rx_chns;
+ devm_remove_action(dev, am65_cpsw_nuss_free_rx_chns, common);
+
+ if (!(rx_chn->irq < 0))
+ devm_free_irq(dev, rx_chn->irq, common);
+
+ netif_napi_del(&common->napi_rx);
+
+ if (!IS_ERR_OR_NULL(rx_chn->desc_pool))
+ k3_cppi_desc_pool_destroy(rx_chn->desc_pool);
+
+ if (!IS_ERR_OR_NULL(rx_chn->rx_chn))
+ k3_udma_glue_release_rx_chn(rx_chn->rx_chn);
+
+ common->rx_flow_id_base = -1;
+}
+
static int am65_cpsw_nuss_init_rx_chns(struct am65_cpsw_common *common)
{
struct am65_cpsw_rx_chn *rx_chn = &common->rx_chns;
@@ -1710,6 +1781,18 @@ static int am65_cpsw_nuss_init_rx_chns(struct am65_cpsw_common *common)
}
}
+ netif_napi_add(common->dma_ndev, &common->napi_rx,
+ am65_cpsw_nuss_rx_poll);
+
+ ret = devm_request_irq(dev, rx_chn->irq,
+ am65_cpsw_nuss_rx_irq,
+ IRQF_TRIGGER_HIGH, dev_name(dev), common);
+ if (ret) {
+ dev_err(dev, "failure requesting rx irq %u, %d\n",
+ rx_chn->irq, ret);
+ goto err;
+ }
+
err:
i = devm_add_action(dev, am65_cpsw_nuss_free_rx_chns, common);
if (i) {
@@ -1981,6 +2064,7 @@ am65_cpsw_nuss_init_port_ndev(struct am65_cpsw_common *common, u32 port_idx)
port->slave.phylink_config.dev = &port->ndev->dev;
port->slave.phylink_config.type = PHYLINK_NETDEV;
port->slave.phylink_config.mac_capabilities = MAC_SYM_PAUSE | MAC_10 | MAC_100 | MAC_1000FD;
+ port->slave.phylink_config.mac_managed_pm = true; /* MAC does PM */
if (phy_interface_mode_is_rgmii(port->slave.phy_if)) {
phy_interface_set_rgmii(port->slave.phylink_config.supported_interfaces);
@@ -2034,35 +2118,6 @@ static int am65_cpsw_nuss_init_ndevs(struct am65_cpsw_common *common)
return ret;
}
- netif_napi_add(common->dma_ndev, &common->napi_rx,
- am65_cpsw_nuss_rx_poll);
-
- return ret;
-}
-
-static int am65_cpsw_nuss_ndev_add_tx_napi(struct am65_cpsw_common *common)
-{
- struct device *dev = common->dev;
- int i, ret = 0;
-
- for (i = 0; i < common->tx_ch_num; i++) {
- struct am65_cpsw_tx_chn *tx_chn = &common->tx_chns[i];
-
- netif_napi_add_tx(common->dma_ndev, &tx_chn->napi_tx,
- am65_cpsw_nuss_tx_poll);
-
- ret = devm_request_irq(dev, tx_chn->irq,
- am65_cpsw_nuss_tx_irq,
- IRQF_TRIGGER_HIGH,
- tx_chn->tx_chn_name, tx_chn);
- if (ret) {
- dev_err(dev, "failure requesting tx%u irq %u, %d\n",
- tx_chn->id, tx_chn->irq, ret);
- goto err;
- }
- }
-
-err:
return ret;
}
@@ -2528,18 +2583,13 @@ static int am65_cpsw_nuss_register_ndevs(struct am65_cpsw_common *common)
struct am65_cpsw_port *port;
int ret = 0, i;
- ret = am65_cpsw_nuss_ndev_add_tx_napi(common);
+ /* init tx channels */
+ ret = am65_cpsw_nuss_init_tx_chns(common);
if (ret)
return ret;
-
- ret = devm_request_irq(dev, common->rx_chns.irq,
- am65_cpsw_nuss_rx_irq,
- IRQF_TRIGGER_HIGH, dev_name(dev), common);
- if (ret) {
- dev_err(dev, "failure requesting rx irq %u, %d\n",
- common->rx_chns.irq, ret);
+ ret = am65_cpsw_nuss_init_rx_chns(common);
+ if (ret)
return ret;
- }
ret = am65_cpsw_nuss_register_devlink(common);
if (ret)
@@ -2584,10 +2634,8 @@ int am65_cpsw_nuss_update_tx_chns(struct am65_cpsw_common *common, int num_tx)
common->tx_ch_num = num_tx;
ret = am65_cpsw_nuss_init_tx_chns(common);
- if (ret)
- return ret;
- return am65_cpsw_nuss_ndev_add_tx_napi(common);
+ return ret;
}
struct am65_cpsw_soc_pdata {
@@ -2736,14 +2784,6 @@ static int am65_cpsw_nuss_probe(struct platform_device *pdev)
am65_cpsw_nuss_get_ver(common);
- /* init tx channels */
- ret = am65_cpsw_nuss_init_tx_chns(common);
- if (ret)
- goto err_of_clear;
- ret = am65_cpsw_nuss_init_rx_chns(common);
- if (ret)
- goto err_of_clear;
-
ret = am65_cpsw_nuss_init_host_p(common);
if (ret)
goto err_of_clear;
@@ -2828,10 +2868,94 @@ static int am65_cpsw_nuss_remove(struct platform_device *pdev)
return 0;
}
+#ifdef CONFIG_PM_SLEEP
+static int am65_cpsw_nuss_suspend(struct device *dev)
+{
+ struct am65_cpsw_common *common = dev_get_drvdata(dev);
+ struct am65_cpsw_port *port;
+ struct net_device *ndev;
+ int i, ret;
+
+ for (i = 0; i < common->port_num; i++) {
+ port = &common->ports[i];
+ ndev = port->ndev;
+
+ if (!ndev)
+ continue;
+
+ netif_device_detach(ndev);
+ if (netif_running(ndev)) {
+ rtnl_lock();
+ ret = am65_cpsw_nuss_ndo_slave_stop(ndev);
+ rtnl_unlock();
+ if (ret < 0) {
+ netdev_err(ndev, "failed to stop: %d", ret);
+ return ret;
+ }
+ }
+ }
+
+ am65_cpts_suspend(common->cpts);
+
+ am65_cpsw_nuss_remove_rx_chns(common);
+ am65_cpsw_nuss_remove_tx_chns(common);
+
+ return 0;
+}
+
+static int am65_cpsw_nuss_resume(struct device *dev)
+{
+ struct am65_cpsw_common *common = dev_get_drvdata(dev);
+ struct am65_cpsw_port *port;
+ struct net_device *ndev;
+ int i, ret;
+
+ ret = am65_cpsw_nuss_init_tx_chns(common);
+ if (ret)
+ return ret;
+ ret = am65_cpsw_nuss_init_rx_chns(common);
+ if (ret)
+ return ret;
+
+ /* If RX IRQ was disabled before suspend, keep it disabled */
+ if (common->rx_irq_disabled)
+ disable_irq(common->rx_chns.irq);
+
+ am65_cpts_resume(common->cpts);
+
+ for (i = 0; i < common->port_num; i++) {
+ port = &common->ports[i];
+ ndev = port->ndev;
+
+ if (!ndev)
+ continue;
+
+ if (netif_running(ndev)) {
+ rtnl_lock();
+ ret = am65_cpsw_nuss_ndo_slave_open(ndev);
+ rtnl_unlock();
+ if (ret < 0) {
+ netdev_err(ndev, "failed to start: %d", ret);
+ return ret;
+ }
+ }
+
+ netif_device_attach(ndev);
+ }
+
+ return 0;
+}
+#endif /* CONFIG_PM_SLEEP */
+
+static const struct dev_pm_ops am65_cpsw_nuss_dev_pm_ops = {
+ SET_SYSTEM_SLEEP_PM_OPS(am65_cpsw_nuss_suspend, am65_cpsw_nuss_resume)
+};
+
static struct platform_driver am65_cpsw_nuss_driver = {
.driver = {
.name = AM65_CPSW_DRV_NAME,
.of_match_table = am65_cpsw_nuss_of_mtable,
+ .pm = &am65_cpsw_nuss_dev_pm_ops,
},
.probe = am65_cpsw_nuss_probe,
.remove = am65_cpsw_nuss_remove,
--
2.17.1On Tue, 2022-11-29 at 15:34 +0200, Roger Quadros wrote:
> @@ -555,11 +556,26 @@ static int am65_cpsw_nuss_ndo_slave_open(struct net_device *ndev)
> struct am65_cpsw_common *common = am65_ndev_to_common(ndev);
> struct am65_cpsw_port *port = am65_ndev_to_port(ndev);
> int ret, i;
> + u32 reg;
>
> ret = pm_runtime_resume_and_get(common->dev);
> if (ret < 0)
> return ret;
>
> + /* Idle MAC port */
> + cpsw_sl_ctl_set(port->slave.mac_sl, CPSW_SL_CTL_CMD_IDLE);
> + cpsw_sl_wait_for_idle(port->slave.mac_sl, 100);
> + cpsw_sl_ctl_reset(port->slave.mac_sl);
> +
> + /* soft reset MAC */
> + cpsw_sl_reg_write(port->slave.mac_sl, CPSW_SL_SOFT_RESET, 1);
> + mdelay(1);
> + reg = cpsw_sl_reg_read(port->slave.mac_sl, CPSW_SL_SOFT_RESET);
> + if (reg) {
> + dev_err(common->dev, "soft RESET didn't complete\n");
I *think* Andrew was asking for dev_dbg() here, but let's see what he
has to say :)
Cheers,
Paolo
Hi,
On 01/12/2022 13:40, Paolo Abeni wrote:
> On Tue, 2022-11-29 at 15:34 +0200, Roger Quadros wrote:
>> @@ -555,11 +556,26 @@ static int am65_cpsw_nuss_ndo_slave_open(struct net_device *ndev)
>> struct am65_cpsw_common *common = am65_ndev_to_common(ndev);
>> struct am65_cpsw_port *port = am65_ndev_to_port(ndev);
>> int ret, i;
>> + u32 reg;
>>
>> ret = pm_runtime_resume_and_get(common->dev);
>> if (ret < 0)
>> return ret;
>>
>> + /* Idle MAC port */
>> + cpsw_sl_ctl_set(port->slave.mac_sl, CPSW_SL_CTL_CMD_IDLE);
>> + cpsw_sl_wait_for_idle(port->slave.mac_sl, 100);
>> + cpsw_sl_ctl_reset(port->slave.mac_sl);
>> +
>> + /* soft reset MAC */
>> + cpsw_sl_reg_write(port->slave.mac_sl, CPSW_SL_SOFT_RESET, 1);
>> + mdelay(1);
>> + reg = cpsw_sl_reg_read(port->slave.mac_sl, CPSW_SL_SOFT_RESET);
>> + if (reg) {
>> + dev_err(common->dev, "soft RESET didn't complete\n");
>
> I *think* Andrew was asking for dev_dbg() here, but let's see what he
> has to say :)
In the earlier revision we were not exiting with error, so dev_dbg()
was more appropriate there.
In this revision we error out so I thought dev_err() was ok.
--
cheers,
-roger
On Thu, Dec 01, 2022 at 01:44:28PM +0200, Roger Quadros wrote:
> Hi,
>
> On 01/12/2022 13:40, Paolo Abeni wrote:
> > On Tue, 2022-11-29 at 15:34 +0200, Roger Quadros wrote:
> >> @@ -555,11 +556,26 @@ static int am65_cpsw_nuss_ndo_slave_open(struct net_device *ndev)
> >> struct am65_cpsw_common *common = am65_ndev_to_common(ndev);
> >> struct am65_cpsw_port *port = am65_ndev_to_port(ndev);
> >> int ret, i;
> >> + u32 reg;
> >>
> >> ret = pm_runtime_resume_and_get(common->dev);
> >> if (ret < 0)
> >> return ret;
> >>
> >> + /* Idle MAC port */
> >> + cpsw_sl_ctl_set(port->slave.mac_sl, CPSW_SL_CTL_CMD_IDLE);
> >> + cpsw_sl_wait_for_idle(port->slave.mac_sl, 100);
> >> + cpsw_sl_ctl_reset(port->slave.mac_sl);
> >> +
> >> + /* soft reset MAC */
> >> + cpsw_sl_reg_write(port->slave.mac_sl, CPSW_SL_SOFT_RESET, 1);
> >> + mdelay(1);
> >> + reg = cpsw_sl_reg_read(port->slave.mac_sl, CPSW_SL_SOFT_RESET);
> >> + if (reg) {
> >> + dev_err(common->dev, "soft RESET didn't complete\n");
> >
> > I *think* Andrew was asking for dev_dbg() here, but let's see what he
> > has to say :)
>
> In the earlier revision we were not exiting with error, so dev_dbg()
> was more appropriate there.
> In this revision we error out so I thought dev_err() was ok.
Yes, i would agree. It is fatal, so dev_err() is appropriate.
What is not shown here is the return value. I think it is -EBUSY? I'm
wondering if -ETIMEDOUT is better?
Andrew
On 01/12/2022 17:19, Andrew Lunn wrote:
> On Thu, Dec 01, 2022 at 01:44:28PM +0200, Roger Quadros wrote:
>> Hi,
>>
>> On 01/12/2022 13:40, Paolo Abeni wrote:
>>> On Tue, 2022-11-29 at 15:34 +0200, Roger Quadros wrote:
>>>> @@ -555,11 +556,26 @@ static int am65_cpsw_nuss_ndo_slave_open(struct net_device *ndev)
>>>> struct am65_cpsw_common *common = am65_ndev_to_common(ndev);
>>>> struct am65_cpsw_port *port = am65_ndev_to_port(ndev);
>>>> int ret, i;
>>>> + u32 reg;
>>>>
>>>> ret = pm_runtime_resume_and_get(common->dev);
>>>> if (ret < 0)
>>>> return ret;
>>>>
>>>> + /* Idle MAC port */
>>>> + cpsw_sl_ctl_set(port->slave.mac_sl, CPSW_SL_CTL_CMD_IDLE);
>>>> + cpsw_sl_wait_for_idle(port->slave.mac_sl, 100);
>>>> + cpsw_sl_ctl_reset(port->slave.mac_sl);
>>>> +
>>>> + /* soft reset MAC */
>>>> + cpsw_sl_reg_write(port->slave.mac_sl, CPSW_SL_SOFT_RESET, 1);
>>>> + mdelay(1);
>>>> + reg = cpsw_sl_reg_read(port->slave.mac_sl, CPSW_SL_SOFT_RESET);
>>>> + if (reg) {
>>>> + dev_err(common->dev, "soft RESET didn't complete\n");
>>>
>>> I *think* Andrew was asking for dev_dbg() here, but let's see what he
>>> has to say :)
>>
>> In the earlier revision we were not exiting with error, so dev_dbg()
>> was more appropriate there.
>> In this revision we error out so I thought dev_err() was ok.
>
> Yes, i would agree. It is fatal, so dev_err() is appropriate.
>
> What is not shown here is the return value. I think it is -EBUSY? I'm
> wondering if -ETIMEDOUT is better?
Yes it is -EBUSY. -ETIMEDOUT is better though so I'll re-spin this series.
cheers,
-roger
© 2016 - 2025 Red Hat, Inc.