usb: musb_dsps: support for suspend and resume
Subsystems (2)
v2 cover usb: musb_dsps: support for suspend and resume
I've been working on some patches that allow suspending and resuming the musb-dsps platform. This was tested for host mode only.
With these patches applied, I can successfully bring an AM335x board to suspend with a USB media connected, and access it again after resume.
This works for both PIO mode and CPPI41 enabled configurations, together with my cppi41 patches that I posted here:
http://marc.info/?l=linux-usb&m=138063429219730&w=2
More testing is certainly appreciated.
Thanks,
Daniel
Daniel Mack (5):
usb: musb: move port reset to worker
usb: musb: call musb_port_suspend from musb_bus_suspend
usb: musb: conditionally save and restore the context on suspend
usb: musb: dsps: add {tx,rx}_mode to wrapper
usb: musb: dsps: add support for suspend and resume
drivers/usb/musb/musb_core.c | 24 +++++++++++++++- drivers/usb/musb/musb_core.h | 3 ++ drivers/usb/musb/musb_dsps.c | 64 +++++++++++++++++++++++++++++++++++++++++ drivers/usb/musb/musb_host.c | 2 ++ drivers/usb/musb/musb_host.h | 4 +++ drivers/usb/musb/musb_virthub.c | 15 ++++++---- include/linux/usb/musb.h | 1 + 7 files changed, 106 insertions(+), 7 deletions(-)
Signature · 6 lines
1.8.3.1 -- To unsubscribe from this list: send the line "unsubscribe linux-usb" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html
A message that is not in the archive (sent off-list or not yet received): 1380634797-29541-1-git-send-email-zonque-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org
PATCH v2 2/5 usb: musb: call musb_port_suspend from musb_bus_suspend
Make musb_port_suspend() externally available, and call it when to host goes into suspend. This allows the core to go into suspend while a device is connected.
--- drivers/usb/musb/musb_host.c | 2 ++ drivers/usb/musb/musb_host.h | 2 ++ drivers/usb/musb/musb_virthub.c | 2 +- 3 files changed, 5 insertions(+), 1 deletion(-)
drivers/usb/musb/musb_host.c +2 -0
@@ -2433,6 +2433,8 @@ static int musb_bus_suspend(struct usb_hcd *hcd)struct musb *musb =hcd_to_musb (hcd);u8 devctl; +musb_port_suspend (musb, true); +if (!is_host_active (musb))return 0 ;
drivers/usb/musb/musb_host.h +2 -0
@@ -93,6 +93,7 @@ extern void musb_root_disconnect(struct musb *musb);extern void musb_host_resume_root_hub (struct musb *musb);extern void musb_host_poke_root_hub (struct musb *musb);extern void musb_port_reset (struct musb *musb); +extern void musb_port_suspend (struct musb *musb,bool do_suspend);#else static inline struct musb *hcd_to_musb (struct usb_hcd *hcd) {
@@ -123,6 +124,7 @@ static inline void musb_host_resume_root_hub(struct musb *musb) {}static inline void musb_host_poll_rh_status (struct musb *musb) {}static inline void musb_host_poke_root_hub (struct musb *musb) {}static inline void musb_port_reset (struct musb *musb) {} +static inline void musb_port_suspend (struct musb *musb,bool do_suspend) {}#endif struct usb_hcd ;
drivers/usb/musb/musb_virthub.c +1 -1
@@ -90,7 +90,7 @@ static void musb_start(struct musb *musb)musb_writeb (regs,MUSB_DEVCTL , devctl); } -static void musb_port_suspend (struct musb *musb,bool do_suspend) +void musb_port_suspend (struct musb *musb,bool do_suspend) {struct usb_otg *otg = musb->xceiv->otg;u8 power;
Signature · 6 lines
1.8.3.1 -- To unsubscribe from this list: send the line "unsubscribe linux-usb" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: [PATCH v2 2/5] usb: musb: call musb_port_suspend from musb_bus_suspend
Quoted 3 lines · * Daniel Mack | 2013-10-01 15:39:54 [+0200]:
Make
musb_port_suspend()externally available, and call it when to host goes into suspend. This allows the core to go into suspend while a device is connected.
Sorry for asking this stupid question but what is usally happening when the host goes to suspend and a device is connected? I think it is disconnected & re-enumerated and resume and this not what you want, right?
What happens if the device is unplugged while the host is suspended and not there on resume?
Quoted 1 line
Signed-off-by Daniel Mack <[email protected]>
Sebastian
Re: [PATCH v2 2/5] usb: musb: call musb_port_suspend from musb_bus_suspend
Quoted 10 lines · Sebastian Andrzej Siewior
Quoted 3 lines · * Daniel Mack | 2013-10-01 15:39:54 [+0200]:
Make
musb_port_suspend()externally available, and call it when to host goes into suspend. This allows the core to go into suspend while a device is connected.Sorry for asking this stupid question but what is usally happening when the host goes to suspend and a device is connected? I think it is disconnected & re-enumerated and resume and this not what you want, right?
It is usually put into USB suspend and the host preserves its state. After resume, the usb storage driver, for example, will not register new partitions but expect things to be as they were before. That is, your previously mounted devices remain functional.
However, the enumeration process has to be done again, because when the device looses power, it can't memorize its address.
Quoted 2 lines
What happens if the device is unplugged while the host is suspended and not there on resume?
That condition is detected and a full teardown of the connected drivers is conducted. Try what happens on your notebook when you do that. Embedded systems should behave just the same.
Daniel
A message that is not in the archive (sent off-list or not yet received): [email protected]
Re: [PATCH v2 2/5] usb: musb: call musb_port_suspend from musb_bus_suspend
Quoted 20 lines · Daniel Mack
Quoted 10 lines · Sebastian Andrzej Siewior
Quoted 3 lines · * Daniel Mack | 2013-10-01 15:39:54 [+0200]:
Make
musb_port_suspend()externally available, and call it when to host goes into suspend. This allows the core to go into suspend while a device is connected.Sorry for asking this stupid question but what is usally happening when the host goes to suspend and a device is connected? I think it is disconnected & re-enumerated and resume and this not what you want, right?
It is usually put into USB suspend and the host preserves its state. After resume, the usb storage driver, for example, will not register new partitions but expect things to be as they were before. That is, your previously mounted devices remain functional.
However, the enumeration process has to be done again, because when the device looses power, it can't memorize its address.
Quoted 6 lines
Quoted 2 lines
What happens if the device is unplugged while the host is suspended and not there on resume?
That condition is detected and a full teardown of the connected drivers is conducted. Try what happens on your notebook when you do that. Embedded systems should behave just the same.
I had the feeling that the USB device gets disconnected and re-enumerated on resume. But if you say that the ehci-hcd saves the state and keeps the device connected if it wasn't disconnected during suspend then there is no reason for musb to behave differently.
Quoted 1 line
Daniel
Sebastian
--
To unsubscribe from this list: send the line "unsubscribe linux-usb" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: [PATCH v2 2/5] usb: musb: call musb_port_suspend from musb_bus_suspend
Quoted 6 lines · Sebastian Andrzej Siewior
Quoted 5 lines · Daniel Mack
Quoted 2 lines · Sebastian Andrzej Siewior
What happens if the device is unplugged while the host is suspended and not there on resume?
Quoted 6 lines
Quoted 3 lines
That condition is detected and a full teardown of the connected drivers is conducted. Try what happens on your notebook when you do that. Embedded systems should behave just the same.
I had the feeling that the USB device gets disconnected and re-enumerated on resume.
Yes. The device looses its +5V power supply, so it has to be reset and re-enumerated on resume of course. It's just that USB drivers know about the state the device has to be put back into after resume, and don't assume the device was disconnected and reconnected. The latter would (for usb-storage) also cause paritions to be removed and added.
Quoted 3 lines
But if you say that the ehci-hcd saves the
state and keeps the device connected if it wasn't disconnected during suspend then there is no reason for musb to behave differently.
Which it doesn't with my patches applied. I can mount a USB volume, put the system into suspend and access the contents after resume. Just the way it should be :)
Daniel
A message that is not in the archive (sent off-list or not yet received): [email protected]
Re: [PATCH v2 2/5] usb: musb: call musb_port_suspend from musb_bus_suspend
Quoted 11 lines · Daniel Mack
Quoted 10 lines · Sebastian Andrzej Siewior
Quoted 8 lines · Daniel Mack
Quoted 2 lines · Sebastian Andrzej Siewior
What happens if the device is unplugged while the host is suspended and not there on resume?
That condition is detected and a full teardown of the connected drivers is conducted. Try what happens on your notebook when you do that. Embedded systems should behave just the same.
Quoted 8 lines
Quoted 2 lines
I had the feeling that the USB device gets disconnected and re-enumerated on resume.
Yes. The device looses its +5V power supply, so it has to be reset and re-enumerated on resume of course. It's just that USB drivers know about the state the device has to be put back into after resume, and don't assume the device was disconnected and reconnected. The latter would (for usb-storage) also cause paritions to be removed and added.
This description may be slightly misleading.
According to the USB spec, a device is not supposed to lose its +5V power supply during suspend. Therefore it does not need to be reset and re-enumerated upon resume. (Except that some devices don't handle suspend properly because of buggy firmware; they _do_ need to be reset.)
Of course, some systems don't follow the spec. If they can't supply suspend power to the host controller and the USB bus, then the device _will_ need to be reset and re-enumerated.
Some USB drivers (those which define a reset_resume method) are able to handle this -- they put the device back into the right state and then act as if nothing had happened. Others aren't; they get unbound and rebound just as though the device really had been disconnected and reconnected.
Alan Stern
--
To unsubscribe from this list: send the line "unsubscribe linux-usb" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html
PATCH v2 1/5 usb: musb: move port reset to worker
In reply to Daniel Mackmusb_port_reset() sleeps, so we can't call it from atomic context. It is, however, called from places inside musb_hub_control() while &musb->lock is held, which leads to a "scheduling while atomic" warning.
Fix this by moving the logic into a worker, and call it where the function was previously called directly.
--- drivers/usb/musb/musb_core.c | 7 +++++++ drivers/usb/musb/musb_core.h | 3 +++ drivers/usb/musb/musb_host.h | 2 ++ drivers/usb/musb/musb_virthub.c | 13 ++++++++----- 4 files changed, 20 insertions(+), 5 deletions(-)
drivers/usb/musb/musb_core.c +7 -0
@@ -1699,6 +1699,12 @@ static void musb_irq_work(struct work_struct *data) } } +static void musb_port_reset_work (struct work_struct *data) +{ + struct musb *musb = container_of(data, struct musb, port_reset_work); +musb_port_reset (musb); +} +/* -------------------------------------------------------------------------- * Init support */
@@ -1857,6 +1863,7 @@ musb_init_controller(struct device *dev, int nIrq, void __iomem *ctrl)/* Init IRQ workqueue before request_irq */ INIT_WORK (&musb->irq_work, musb_irq_work); +INIT_WORK (&musb->port_reset_work, musb_port_reset_work);/* attach to the IRQ */ if (request_irq (nIrq, musb->isr,0 ,dev_name (dev), musb)) {
drivers/usb/musb/musb_core.h +3 -0
@@ -294,6 +294,9 @@ struct musb {irqreturn_t (*isr)(int ,void *);struct work_struct irq_work; +struct work_struct port_reset_work; +bool port_reset_state; +u16 hwvers;u16 intrrxe;
drivers/usb/musb/musb_host.h +2 -0
@@ -92,6 +92,7 @@ extern void musb_host_rx(struct musb *, u8);extern void musb_root_disconnect (struct musb *musb);extern void musb_host_resume_root_hub (struct musb *musb);extern void musb_host_poke_root_hub (struct musb *musb); +extern void musb_port_reset (struct musb *musb);#else static inline struct musb *hcd_to_musb (struct usb_hcd *hcd) {
@@ -121,6 +122,7 @@ static inline void musb_root_disconnect(struct musb *musb) {}static inline void musb_host_resume_root_hub (struct musb *musb) {}static inline void musb_host_poll_rh_status (struct musb *musb) {}static inline void musb_host_poke_root_hub (struct musb *musb) {} +static inline void musb_port_reset (struct musb *musb) {}#endif struct usb_hcd ;
drivers/usb/musb/musb_virthub.c +8 -5
@@ -155,7 +155,7 @@ static void musb_port_suspend(struct musb *musb, bool do_suspend) } } -static void musb_port_reset (struct musb *musb,bool do_reset) +void musb_port_reset (struct musb *musb) {u8 power;void __iomem *mbase = musb->mregs;
@@ -173,7 +173,7 @@ static void musb_port_reset(struct musb *musb, bool do_reset)* the appropriate amount of time has passed */ power =musb_readb (mbase,MUSB_POWER ); -if (do_reset) { +if (musb->port_reset_state) {/* * If RESUME is set, we must make sure it stays minimum 20 ms.
@@ -356,8 +356,10 @@ int musb_hub_control(/* finish RESET signaling? */ if ((musb->port1_status &USB_PORT_STAT_RESET ) - &&time_after_eq (jiffies, musb->rh_timer)) -musb_port_reset (musb, false); + &&time_after_eq (jiffies, musb->rh_timer)) { + musb->port_reset_state = false; +schedule_work (&musb->port_reset_work); + }/* finish RESUME signaling? */ if ((musb->port1_status & MUSB_PORT_STAT_RESUME)
@@ -412,7 +414,8 @@ int musb_hub_control(musb_start (musb);break ;case USB_PORT_FEAT_RESET : -musb_port_reset (musb, true); + musb->port_reset_state = true; +schedule_work (&musb->port_reset_work);break ;case USB_PORT_FEAT_SUSPEND :musb_port_suspend (musb, true);
Signature · 1 line
1.8.3.1
Re: [PATCH v2 1/5] usb: musb: move port reset to worker
Quoted 3 lines · * Daniel Mack | 2013-10-01 15:39:53 [+0200]:
musb_port_reset()sleeps, so we can't call it from atomic context. It is, however, called from places insidemusb_hub_control()while&musb->lockis held, which leads to a "scheduling while atomic" warning.
I guess you are hit by the msleep(1) here.
Quoted 2 lines
Fix this by moving the logic into a worker, and call it where the function was previously called directly.
I *think* the better approach here would be to replace the msleep(1) by a busy loop. The thing is that you do now want to continue the operation and having the reset in progress. It should complete _right_ _now_ i.e. SET_PORT_FEATURE RESET is completed before the the next SET/GET request arrives and with a workqueue you can not guarantee this.
Quoted 1 line
Signed-off-by Daniel Mack <[email protected]>
Sebastian
Re: [PATCH v2 1/5] usb: musb: move port reset to worker
Quoted 7 lines · Sebastian Andrzej Siewior
Quoted 3 lines · * Daniel Mack | 2013-10-01 15:39:53 [+0200]:
musb_port_reset()sleeps, so we can't call it from atomic context. It is, however, called from places insidemusb_hub_control()while&musb->lockis held, which leads to a "scheduling while atomic" warning.I guess you are hit by the msleep(1) here.
Yes.
Quoted 8 lines
Quoted 2 lines
Fix this by moving the logic into a worker, and call it where the function was previously called directly.
I *think* the better approach here would be to replace the msleep(1) by a busy loop. The thing is that you do now want to continue the operation and having the reset in progress. It should complete _right_ _now_ i.e.
SET_PORT_FEATURERESET is completed before the the next SET/GET request arrives and with a workqueue you can not guarantee this.
Hmm, ok. I'm just not sure what time periods we're really sleeping for here. After all, we're blocking all interrupts from occuring while the lock is held.
So my question is rather whether it's sensible at all to hold the lock while waiting. We might change some of the execution pathes to mutexes and waitqueues as well. Felipe?
Thanks,
Daniel
PATCH v2 3/5 usb: musb: conditionally save and restore the context on suspend
In reply to Daniel MackIt appears not all platforms featuring a musb core need to save the musb core registers at suspend time and restore them on resume.
The dsps platform does, however. So add a bit in struct musb_hdrc_platform_data to let platforms specify their need of such action being taken.
--- drivers/usb/musb/musb_core.c | 17 ++++++++++++++++- include/linux/usb/musb.h | 1 + 2 files changed, 17 insertions(+), 1 deletion(-)
drivers/usb/musb/musb_core.c +16 -1
@@ -2152,6 +2152,7 @@ static int musb_suspend(struct device *dev) {struct musb *musb =dev_to_musb (dev);unsigned long flags; +struct musb_hdrc_platform_data *plat =dev_get_platdata (dev);spin_lock_irqsave (&musb->lock, flags);
@@ -2165,16 +2166,30 @@ static int musb_suspend(struct device *dev)*/ } +if (plat->restore_after_suspend) +musb_save_context (musb); +spin_unlock_irqrestore (&musb->lock, flags);return 0 ; }static int musb_resume_noirq (struct device *dev) { -/* for static cmos like DaVinci, register values were preserved +struct musb *musb =dev_to_musb (dev); +struct musb_hdrc_platform_data *plat =dev_get_platdata (dev); + +/* +* For static cmos like DaVinci, register values were preserved * unless for some reason the whole soc powered down or the USB * module got reset through the PSC (vs just being disabled). +* +* The plaform data tells us about the necessity of saving and +* restoring the context across a suspend cycle. */ + +if (plat->restore_after_suspend) +musb_restore_context (musb); +return 0 ; }
include/linux/usb/musb.h +1 -0
@@ -100,6 +100,7 @@ struct musb_hdrc_platform_data {u8 mode; u8 has_mailbox:1; + u8 restore_after_suspend:1;/* for clk_get() */ const char *clock;
Signature · 1 line
1.8.3.1
PATCH v2 4/5 usb: musb: dsps: add {tx,rx}_mode to wrapper
In reply to Daniel Mackrx_mode and tx_mode need to be read at suspend time and restored on resume for dsps platforms. So add it to the wrapper struct first, and initialize the values.
--- drivers/usb/musb/musb_dsps.c | 4 ++++ 1 file changed, 4 insertions(+)
drivers/usb/musb/musb_dsps.c +4 -0
@@ -83,6 +83,8 @@ struct dsps_musb_wrapper {u16 coreintr_status;u16 phy_utmi;u16 mode; +u16 tx_mode; +u16 rx_mode;/* bit positions for control */ unsigned reset:5;
@@ -600,6 +602,8 @@ static const struct dsps_musb_wrapper am33xx_driver_data = { .coreintr_status = 0x34, .phy_utmi = 0xe0, .mode = 0xe8, + .tx_mode = 0x70, + .rx_mode = 0x74, .reset = 0, .otg_disable = 21, .iddig = 8,
Signature · 1 line
1.8.3.1
PATCH v2 5/5 usb: musb: dsps: add support for suspend and resume
In reply to Daniel MackThe dsps platform needs to save save some registers at suspend time and restore them after resume. This patch adds a struct for these registers, and also lets the musb core know that the core registers need to be saved as well.
--- drivers/usb/musb/musb_dsps.c | 60 ++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 60 insertions(+)
drivers/usb/musb/musb_dsps.c +60 -0
@@ -112,6 +112,19 @@ struct dsps_musb_wrapper {u8 poll_seconds; }; +/* +* register shadow for suspend +*/ +struct dsps_context { +u32 control; +u32 epintr; +u32 coreintr; +u32 phy_utmi; +u32 mode; +u32 tx_mode; +u32 rx_mode; +}; +/** * DSPS glue structure. */
@@ -121,6 +134,8 @@ struct dsps_glue {const struct dsps_musb_wrapper *wrp;/* wrapper register offsets */ struct timer_list timer;/* otg_workaround timer */ unsigned long last_timer;/* last timer data for each instance */ + +struct dsps_context context; };/**
@@ -504,6 +519,7 @@ static int dsps_create_musb_pdev(struct dsps_glue *glue, } pdata.config = config; pdata.platform_ops = &dsps_ops; + pdata.restore_after_suspend =1 ; config->num_eps =get_int_prop (dn,"mentor,num-eps" ); config->ram_bits =get_int_prop (dn,"mentor,ram-bits" );
@@ -627,11 +643,55 @@ static const struct of_device_id musb_dsps_of_match[] = { };MODULE_DEVICE_TABLE (of, musb_dsps_of_match); +#ifdef CONFIG_PM +static int dsps_suspend (struct device *dev) +{ +struct dsps_glue *glue =dev_get_drvdata (dev); +const struct dsps_musb_wrapper *wrp = glue->wrp; +struct musb *musb =platform_get_drvdata (glue->musb); +void __iomem *mbase = musb->ctrl_base; + + glue->context.control =dsps_readl (mbase, wrp->control); + glue->context.epintr =dsps_readl (mbase, wrp->epintr_set); + glue->context.coreintr =dsps_readl (mbase, wrp->coreintr_set); + glue->context.phy_utmi =dsps_readl (mbase, wrp->phy_utmi); + glue->context.mode =dsps_readl (mbase, wrp->mode); + glue->context.tx_mode =dsps_readl (mbase, wrp->tx_mode); + glue->context.rx_mode =dsps_readl (mbase, wrp->rx_mode); + +return 0 ; +} + +static int dsps_resume (struct device *dev) +{ +struct dsps_glue *glue =dev_get_drvdata (dev); +const struct dsps_musb_wrapper *wrp = glue->wrp; +struct musb *musb =platform_get_drvdata (glue->musb); +void __iomem *mbase = musb->ctrl_base; + +dsps_writel (mbase, wrp->control, glue->context.control); +dsps_writel (mbase, wrp->epintr_set, glue->context.epintr); +dsps_writel (mbase, wrp->coreintr_set, glue->context.coreintr); +dsps_writel (mbase, wrp->phy_utmi, glue->context.phy_utmi); +dsps_writel (mbase, wrp->mode, glue->context.mode); +dsps_writel (mbase, wrp->tx_mode, glue->context.tx_mode); +dsps_writel (mbase, wrp->rx_mode, glue->context.rx_mode); + + musb->port_reset_state = false; +schedule_work (&musb->port_reset_work); + +return 0 ; +} +#endif + +static SIMPLE_DEV_PM_OPS(dsps_pm_ops, dsps_suspend, dsps_resume); +static struct platform_driver dsps_usbss_driver = { .probe = dsps_probe, .remove = dsps_remove, .driver = { .name ="musb-dsps" , + .pm = &dsps_pm_ops, .of_match_table =of_match_ptr (musb_dsps_of_match), }, };
Signature · 1 line
1.8.3.1
Re: [PATCH v2 5/5] usb: musb: dsps: add support for suspend and resume
Hi,
Quoted 6 lines · Daniel Mack
The dsps platform needs to save save some registers at suspend time and restore them after resume. This patch adds a struct for these registers, and also lets the musb core know that the core registers need to be saved as well.
Signed-off-by Daniel Mack <[email protected]>
this last patch didn't apply. can you rebase on my 'testing' branch ?
Signature · 1 line
balbi
Re: [PATCH v2 5/5] usb: musb: dsps: add support for suspend and resume
Quoted 9 lines · Felipe Balbi
Quoted 6 lines · Daniel Mack
The dsps platform needs to save save some registers at suspend time and restore them after resume. This patch adds a struct for these registers, and also lets the musb core know that the core registers need to be saved as well.
Signed-off-by Daniel Mack <[email protected]>this last patch didn't apply. can you rebase on my 'testing' branch ?
I've seen that you already applied the others, so I'm going to resend only that last one. Hope that's ok.
Daniel
A message that is not in the archive (sent off-list or not yet received): [email protected]
Re: [PATCH v2 5/5] usb: musb: dsps: add support for suspend and resume
Quoted 7 lines · Daniel Mack
Quoted 6 lines · Felipe Balbi
Quoted 5 lines · Daniel Mack
The dsps platform needs to save save some registers at suspend time and restore them after resume. This patch adds a struct for these registers, and also lets the musb core know that the core registers need to be saved as well.
Quoted 6 lines
Quoted 3 lines
Quoted 1 line
Signed-off-by Daniel Mack <[email protected]>this last patch didn't apply. can you rebase on my 'testing' branch ?
I've seen that you already applied the others, so I'm going to resend only that last one. Hope that's ok.
That's perfectly fine ;-) Thanks
Signature · 1 line
balbi