Skip to contentSkip to conversation

usb: dwc2: reset dwc2 core before dwc2_get_hwparams()

PATCH Superseded Yunzhi Li 3 participants 7 messages Aug 18, 2015 – Oct 9, 2015

Subsystems (2)

v2 of 3 Superseded Latest is v3 →

lore
Share & tools
lore

Yunzhi Li

author
#0 lore raw

PATCH v2 1/1 usb: dwc2: reset dwc2 core before dwc2_get_hwparams()

We initiate dwc2 usb controller in BIOS, dwc2_core_reset() should be called before dwc2_get_hwparams() to reset core registers to default value. Without this the FIFO setting might be incorrect because calculating FIFO size need power-on value of GRXFSIZ/GNPTXFSIZ/HPTXFSIZ registers.

This patch could avoid warnning massage like in rk3288 platform:

[    2.074764] dwc2 ff580000.usb: 256 invalid for

host_perio_tx_fifo_size. Check HW configuration.

---

 drivers/usb/dwc2/core.c     | 2 +-
 drivers/usb/dwc2/core.h     | 1 +
 drivers/usb/dwc2/platform.c | 6 ++++++
 3 files changed, 8 insertions(+), 1 deletion(-)
drivers/usb/dwc2/core.c +1 -1
index c3cc1a7..86d1d65 100644
@@ -474,7 +474,7 @@ static void dwc2_init_fs_ls_pclk_sel(struct dwc2_hsotg *hsotg)
  * Do core a soft reset of the core.  Be careful with this because it
  * resets all the internal state machines of the core.
  */
-static int dwc2_core_reset(struct dwc2_hsotg *hsotg)
+int dwc2_core_reset(struct dwc2_hsotg *hsotg)
 {
 	u32 greset;
 	int count = 0;
drivers/usb/dwc2/core.h +1 -0
index 0ed87620..5d95aec 100644
@@ -846,6 +846,7 @@ enum dwc2_halt_status {
  * The following functions support initialization of the core driver component
  * and the DWC_otg controller
  */
+extern int dwc2_core_reset(struct dwc2_hsotg *hsotg);
 extern void dwc2_core_host_init(struct dwc2_hsotg *hsotg);
 extern int dwc2_enter_hibernation(struct dwc2_hsotg *hsotg);
 extern int dwc2_exit_hibernation(struct dwc2_hsotg *hsotg, bool restore);
drivers/usb/dwc2/platform.c +6 -0
index 9093530..8d3be4a 100644
@@ -243,6 +243,12 @@ static int dwc2_driver_probe(struct platform_device *dev)
 	spin_lock_init(&hsotg->lock);
 	mutex_init(&hsotg->init_mutex);
 
+	/*
+	 * Reset before dwc2_get_hwparams() then it could get power-on real
+	 * reset value form registers.
+	 */
+	dwc2_core_reset(hsotg);
+
 	/* Detect config values from hardware */
 	retval = dwc2_get_hwparams(hsotg);
 	if (retval)
Signature · 1 line
2.0.0
Reviewed-by on v2 1/1
#1 lore raw

lyz,

Quoted 10 lines · Yunzhi Li

We initiate dwc2 usb controller in BIOS, dwc2_core_reset() should be called before dwc2_get_hwparams() to reset core registers to default value. Without this the FIFO setting might be incorrect because calculating FIFO size need power-on value of GRXFSIZ/GNPTXFSIZ/HPTXFSIZ registers.

This patch could avoid warnning massage like in rk3288 platform:

[    2.074764] dwc2 ff580000.usb: 256 invalid for

host_perio_tx_fifo_size. Check HW configuration.

Quoted 8 lines

---

 drivers/usb/dwc2/core.c     | 2 +-
 drivers/usb/dwc2/core.h     | 1 +
 drivers/usb/dwc2/platform.c | 6 ++++++
 3 files changed, 8 insertions(+), 1 deletion(-)

This seems reasonable to me.

#2 lore raw
In reply to Yunzhi Li
Quoted 52 lines · Yunzhi Li

We initiate dwc2 usb controller in BIOS, dwc2_core_reset() should be called before dwc2_get_hwparams() to reset core registers to default value. Without this the FIFO setting might be incorrect because calculating FIFO size need power-on value of GRXFSIZ/GNPTXFSIZ/HPTXFSIZ registers.

This patch could avoid warnning massage like in rk3288 platform:

[    2.074764] dwc2 ff580000.usb: 256 invalid for

host_perio_tx_fifo_size. Check HW configuration.

---

 drivers/usb/dwc2/core.c     | 2 +-
 drivers/usb/dwc2/core.h     | 1 +
 drivers/usb/dwc2/platform.c | 6 ++++++
 3 files changed, 8 insertions(+), 1 deletion(-)
drivers/usb/dwc2/core.c +1 -1
index c3cc1a7..86d1d65 100644
@@ -474,7 +474,7 @@ static void dwc2_init_fs_ls_pclk_sel(struct dwc2_hsotg *hsotg)
  * Do core a soft reset of the core.  Be careful with this because it
  * resets all the internal state machines of the core.
  */
-static int dwc2_core_reset(struct dwc2_hsotg *hsotg)
+int dwc2_core_reset(struct dwc2_hsotg *hsotg)
 {
 	u32 greset;
 	int count = 0;
drivers/usb/dwc2/core.h +1 -0
index 0ed87620..5d95aec 100644
@@ -846,6 +846,7 @@ enum dwc2_halt_status {
  * The following functions support initialization of the core driver component
  * and the DWC_otg controller
  */
+extern int dwc2_core_reset(struct dwc2_hsotg *hsotg);
 extern void dwc2_core_host_init(struct dwc2_hsotg *hsotg);
 extern int dwc2_enter_hibernation(struct dwc2_hsotg *hsotg);
 extern int dwc2_exit_hibernation(struct dwc2_hsotg *hsotg, bool restore);
drivers/usb/dwc2/platform.c +0 -0
index 9093530..8d3be4a 100644
@@ -243,6 +243,12 @@ static int dwc2_driver_probe(struct platform_device *dev)
 	spin_lock_init(&hsotg->lock);
 	mutex_init(&hsotg->init_mutex);

drivers/usb/dwc2/platform.c static int dwc2_driver_probe(struct platform_device *dev) · line 254

Quoted 10 lines
drivers/usb/dwc2/platform.c +6 -0
 
+	/*
+	 * Reset before dwc2_get_hwparams() then it could get power-on real
+	 * reset value form registers.
+	 */
+	dwc2_core_reset(hsotg);
+
 	/* Detect config values from hardware */
 	retval = dwc2_get_hwparams(hsotg);
 	if (retval)

Hi Yunzhi,

My concern is with the delays due to calling the dwc2_core_reset during probe. You could factor out the assertion of the core soft reset from the dwc2_core_reset and just use that before calling dwc2_get_hwparams().

You had previously addressed the lengthy probe time issue here:
http://marc.info/?l=linux-usb&m=142357721304377

This reducing delays patch looks reasonable to me and I think it should get merged also. I'll do some testing with it later this week.

John

#3 lore raw

John,

Quoted 7 lines · John Youn

Hi Yunzhi,

My concern is with the delays due to calling the dwc2_core_reset during probe. You could factor out the assertion of the core soft reset from the dwc2_core_reset and just use that before calling dwc2_get_hwparams().

Quoted 6 lines

You had previously addressed the lengthy probe time issue here:
http://marc.info/?l=linux-usb&m=142357721304377

This reducing delays patch looks reasonable to me and I think it should get merged also. I'll do some testing with it later this week.

Note: you can also avoid the extra reset during probe with something like <https://chromium-review.googlesource.com/#/c/303152/>. I'm happy to post that up if you want, though it depends on lyz's patch.

...in Chrome OS we've also just landed lyz's patch to reduce delays. If there's any fallout I'll report on the list.

-Doug

#4 lore raw
Quoted 17 lines · Doug Anderson

John,

Quoted 7 lines · John Youn

Hi Yunzhi,

My concern is with the delays due to calling the dwc2_core_reset during probe. You could factor out the assertion of the core soft reset from the dwc2_core_reset and just use that before calling dwc2_get_hwparams().

Quoted 6 lines

You had previously addressed the lengthy probe time issue here:
http://marc.info/?l=linux-usb&m=142357721304377

This reducing delays patch looks reasonable to me and I think it should get merged also. I'll do some testing with it later this week.

Quoted 9 lines

Note: you can also avoid the extra reset during probe with something like <https://chromium-review.googlesource.com/#/c/303152/>. I'm happy to post that up if you want, though it depends on lyz's patch.

...in Chrome OS we've also just landed lyz's patch to reduce delays. If there's any fallout I'll report on the list.

-Doug

Yes, I appreciate if you could submit that. I'd like to merge lyz's reduce delays, lyz's bug fix for hwparams, and your fix for double reset.

The gadget side will also need something similar which I can do if needed. Do you guys run gadget mode?

Regards,
John

#5 lore raw

John,

Quoted 25 lines · John Youn
Quoted 24 lines · Doug Anderson

John,

Quoted 7 lines · John Youn

Hi Yunzhi,

My concern is with the delays due to calling the dwc2_core_reset during probe. You could factor out the assertion of the core soft reset from the dwc2_core_reset and just use that before calling dwc2_get_hwparams().

Quoted 6 lines

You had previously addressed the lengthy probe time issue here:
http://marc.info/?l=linux-usb&m=142357721304377

This reducing delays patch looks reasonable to me and I think it should get merged also. I'll do some testing with it later this week.

Note: you can also avoid the extra reset during probe with something like <https://chromium-review.googlesource.com/#/c/303152/>. I'm happy to post that up if you want, though it depends on lyz's patch.

...in Chrome OS we've also just landed lyz's patch to reduce delays. If there's any fallout I'll report on the list.

Quoted 7 lines
Quoted 2 lines

-Doug

Yes, I appreciate if you could submit that. I'd like to merge lyz's reduce delays, lyz's bug fix for hwparams, and your fix for double reset.

OK, I've posted things up. Let me know what you think. Note that I took v2 of lyz's patch (not v3) since I liked it better.

Quoted 2 lines

The gadget side will also need something similar which I can do if needed. Do you guys run gadget mode?

I don't personally run in gadget mode. I think it's possible to get rk3288 to do it on one of the two ports, but I don't have it setup.

-Doug

#6 lore raw
Quoted 40 lines · Doug Anderson

John,

Quoted 25 lines · John Youn
Quoted 24 lines · Doug Anderson

John,

Quoted 7 lines · John Youn

Hi Yunzhi,

My concern is with the delays due to calling the dwc2_core_reset during probe. You could factor out the assertion of the core soft reset from the dwc2_core_reset and just use that before calling dwc2_get_hwparams().

Quoted 6 lines

You had previously addressed the lengthy probe time issue here:
http://marc.info/?l=linux-usb&m=142357721304377

This reducing delays patch looks reasonable to me and I think it should get merged also. I'll do some testing with it later this week.

Note: you can also avoid the extra reset during probe with something like <https://chromium-review.googlesource.com/#/c/303152/>. I'm happy to post that up if you want, though it depends on lyz's patch.

...in Chrome OS we've also just landed lyz's patch to reduce delays. If there's any fallout I'll report on the list.

Quoted 7 lines
Quoted 2 lines

-Doug

Yes, I appreciate if you could submit that. I'd like to merge lyz's reduce delays, lyz's bug fix for hwparams, and your fix for double reset.

OK, I've posted things up. Let me know what you think. Note that I took v2 of lyz's patch (not v3) since I liked it better.

Quoted 6 lines
Quoted 2 lines

The gadget side will also need something similar which I can do if needed. Do you guys run gadget mode?

I don't personally run in gadget mode. I think it's possible to get rk3288 to do it on one of the two ports, but I don't have it setup.

No problem. I can make the gadget changes.

Thanks for submitting this.

Regards,
John