Skip to contentSkip to conversation

omap4 ehci sporadic resume issue

DISCUSSION Michael Trimarchi 6 participants 23 messages Jun 27, 2013 – Sep 2, 2013 6 messages not in the archive Arm (32-bit)

lore
Share & tools
lore
#0 lore raw

Hi

I'm working on omap4460 with two ulpi connected to (SMSC3320 -> HUB SMSC2514) or (TUSB1210 -> HUB SMSC2514). The problem only happen when both port are used and after few suspend resume are triggered. If I use just one port there is no issue on suspend resume. I already covered all TI errata that I know and I'm working on TI kernel.

The problem is here

[   77.701934] ehci-omap ehci-omap.0: irq status a004 Async Recl PCD

Both ports change status from 001005 to 001009 (you have a log just after).
So from host point of view both hub connected are not working in HS mode.
After that the omap ehci has gone because the bus can not work in FS and LS and I can not recover from here. Status of transceivers are dumped and they are ok after resume.

Do you have any suggestion?

Normal case

SUSPEND

DIR HIGH
CLOCK is OFF

AFTER RESUME

DIR has signal

CLOCK is 60 Mhz

Disconnect case

SUSPEND

DIR HIGH
CLOCK is OFF

AFTER RESUME
DIR is LOW
CLOCK is 60Mhz

[   77.593566] ehci-omap ehci-omap.0: ehci_omap_bus_resume
[   77.593658] ehci-omap ehci-omap.0: resume root hub
[   77.615814] hub 2-0:1.0: hub_resume
[   77.639251] hub 1-0:1.0: hub_resume
[   77.639282] ehci-omap ehci-omap.0: GetStatus port:1 status 001885 0  ACK POWER sig=j SUSPEND PE CONNECT
[   77.639282] hub 1-0:1.0: port 1: status 0507 change 0000
[   77.639312] ehci-omap ehci-omap.0: GetStatus port:2 status 001885 0  ACK POWER sig=j SUSPEND PE CONNECT
[   77.639312] hub 1-0:1.0: port 2: status 0507 change 0000
[   77.639312] ehci-omap ehci-omap.0: GetStatus port:3 status 001000 0  ACK POWER sig=se0
[   77.639373] ehci-omap ehci-omap.0: GetStatus port:1 status 001885 0  ACK POWER sig=j SUSPEND PE CONNECT
[   77.639404] usb 1-1: usb resume
[   77.639434] ehci-omap ehci-omap.0: GetStatus port:2 status 001885 0  ACK POWER sig=j SUSPEND PE CONNECT
[   77.639434] usb 1-2: usb resume
[   77.678314] dump_tranceiver -----------> on port 1
[   77.678314] ULPI_FUNC_CTRL 0x40
[   77.678344] ULPI_IFC_CTRL 0x00
[   77.678344] ULPI_OTG_CTRL 0x06
[   77.678344] ULPI_DEBUG 0x00
[   77.678344] ULPI_USB_INT_STS 0x14
[   77.678344] dump_tranceiver <---------- on port 1
[   77.678375] ehci-omap ehci-omap.0: GetStatus port:1 status 001005 0  ACK POWER sig=se0 PE CONNECT
[   77.678405] dump_tranceiver -----------> on port 2
[   77.678405] ULPI_FUNC_CTRL 0x40
[   77.678405] ULPI_IFC_CTRL 0x00
[   77.678436] ULPI_OTG_CTRL 0x06
[   77.678436] ULPI_DEBUG 0x00
[   77.678436] ULPI_USB_INT_STS 0x14
[   77.678436] dump_tranceiver <---------- on port 2
[   77.678466] ehci-omap ehci-omap.0: GetStatus port:2 status 001005 0  ACK POWER sig=se0 PE CONNECT
[   77.701782] usb 1-1: finish resume
[   77.701812] usb 1-2: finish resume
[   77.701934] ehci-omap ehci-omap.0: irq status a004 Async Recl PCD
[   77.701934] Port 2 Status 0x1000
[   77.701934] Port 1 Status 0x1009
[   77.701965] Port 0 Status 0x1009
[   82.717529] ehci-omap ehci-omap.0: IAA watchdog: status a008 cmd 10065
[   82.717590] usb 1-2: kworker/u:12 timed out on ep0in len=0/2
[   82.717590] usb 1-2: gone after usb resume? status -110
[   82.717590] usb 1-2: can't resume, status -110
[   82.717620] hub 1-0:1.0: logical disconnect on port 2
[   82.717620] Disable PORT_PE
[   82.717651] pm_op(): usb_dev_resume+0x0/0x18 returns -110
[   82.717651] PM: Device 1-2 failed to resume async: error -110
[   82.733154] ehci-omap ehci-omap.0: IAA watchdog: status a008 cmd 10065
[   82.733184] usb 1-1: kworker/u:15 timed out on ep0in len=0/2
[   82.733215] usb 1-1: gone after usb resume? status -110
[   82.733215] usb 1-1: can't resume, status -110
[   82.733215] hub 1-0:1.0: logical disconnect on port 1
[   82.733215] Disable PORT_PE
[   82.733245] pm_op(): usb_dev_resume+0x0/0x18 returns -110
[   82.733245] PM: Device 1-1 failed to resume async: error -110
[   82.734741] PM: resume of devices complete after 5229.652 msecs
[   83.303863] PM: Finishing wakeup.
[   83.307525] Restarting tasks ... 
[   83.311157] hub 2-0:1.0: state 7 ports 1 chg 0000 evt 0000
[   83.317504] hub 2-0:1.0: hub_resume
[   83.321716] done.
[   83.323974] suspend: exit suspend, ret = 0 (2000-01-02 21:07:12.171992466 UTC)
[   83.323974] hub 2-0:1.0: activate --> -22
[   83.324005] hub 1-0:1.0: state 7 ports 3 chg 0006 evt 0006
[   83.324035] ehci-omap ehci-omap.0: GetStatus port:1 status 001801 0  ACK POWER sig=j CONNECT
[   83.324035] hub 1-0:1.0: port 1, status 0501, change 0000, 480 Mb/s
[   83.324035] usb 1-1: USB disconnect, device number 2
[   83.324066] usb 1-1.2: USB disconnect, device number 4
[   83.324066] usb 1-1.2: unregistering device
[   83.324066] usb 1-1.2: unregistering interface 1-1.2:1.0
[   83.530700] usb 1-1.2: usb_disable_device nuking all URBs
[   83.537200] usb 1-1.3: USB disconnect, device number 5
[   83.542907] usb 1-1.3: unregistering device
[   83.547546] usb 1-1.3: unregistering interface 1-1.3:1.0
[   83.554107] usb 1-1.3: usb_disable_device nuking all URBs
[   83.560577] usb 1-1: unregistering device
[   83.565063] usb 1-1: unregistering interface 1-1:1.0
[   83.570800] usb 1-1: usb_disable_device nuking all URBs

Michael
--
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

#1 lore raw

Hi Michael,

Quoted 9 lines · Michael Trimarchi

Hi

I'm working on omap4460 with two ulpi connected to (SMSC3320 -> HUB SMSC2514) or (TUSB1210 -> HUB SMSC2514). The problem only happen when both port are used and after few suspend resume are triggered. If I use just one port there is no issue on suspend resume. I already covered all TI errata that I know and I'm working on TI kernel.

The problem is here

Quoted 8 lines
[   77.701934] ehci-omap ehci-omap.0: irq status a004 Async Recl PCD

Both ports change status from 001005 to 001009 (you have a log just after).
So from host point of view both hub connected are not working in HS mode.
After that the omap ehci has gone because the bus can not work in FS and LS and I can not recover from here. Status of transceivers are dumped and they are ok after resume.

Do you have any suggestion?

I'm not very sure but both ports suddenly changing state like that look like a hardware issue. Also, it is strange that you can reproduce it only when two ports are simultaneously in use. Unfortunately, I can't match your setup with 2 ULPI ports.

I have a OMAP5 uEVM that uses 2 ports but it won't be identical to your setup as they are on HSIC and not ULPI.

Did you try errata i693?

Also, are you suspending and resuming only the USB or the entire system?

cheers,
-roger

A message that is not in the archive (sent off-list or not yet received): [email protected]

#3 lore raw

Hi Roger

Quoted 19 lines · Roger Quadros

Hi Michael,

Quoted 16 lines · Michael Trimarchi

Hi

I'm working on omap4460 with two ulpi connected to (SMSC3320 -> HUB SMSC2514) or (TUSB1210 -> HUB SMSC2514). The problem only happen when both port are used and after few suspend resume are triggered. If I use just one port there is no issue on suspend resume. I already covered all TI errata that I know and I'm working on TI kernel.

The problem is here

[   77.701934] ehci-omap ehci-omap.0: irq status a004 Async Recl PCD

Both ports change status from 001005 to 001009 (you have a log just after).
So from host point of view both hub connected are not working in HS mode.
After that the omap ehci has gone because the bus can not work in FS and LS and I can not recover from here. Status of transceivers are dumped and they are ok after resume.

Quoted 5 lines
Quoted 1 line

Do you have any suggestion?

I'm not very sure but both ports suddenly changing state like that look like a hardware issue. Also, it is strange that you can reproduce it only when two ports are simultaneously in use. Unfortunately, I can't match your setup with 2 ULPI

Yes I know that TI doesn't have any setup like that.

Quoted 6 lines

ports.

I have a OMAP5 uEVM that uses 2 ports but it won't be identical to your setup as they are on HSIC and not ULPI.

Did you try errata i693?

Yes I have it. It's not clear if I need to wait after
ehci_writel(ehci, temp | PORT_SUSPEND, status_reg);
polling the suspendM of the SMSC or the suspend status of the PORT or I can
switch just after this instruction. Because TI kernel use an msleep of 4 mseconds and then switch. It could be a timing issue on how errata is implemented when I have two ports? How this internal count works?

First time is 18, and then?

Quoted 3 lines

Also, are you suspending and resuming only the USB or the entire system?

Whole system. Right the only susbsytem that doens't go to suspend is the FSUSB but this depends on the bootloader.

Michael

Quoted 2 lines

cheers,
-roger

Signature · 7 lines
| Michael Nazzareno Trimarchi                     Amarula Solutions BV |
| COO  -  Founder                                      Cruquiuskade 47 |
| +31(0)851119172                                 Amsterdam 1018 AM NL |
--
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
#4 lore raw

+Ruslan

Quoted 35 lines · Michael Trimarchi

Hi Roger

Quoted 19 lines · Roger Quadros

Hi Michael,

Quoted 16 lines · Michael Trimarchi

Hi

I'm working on omap4460 with two ulpi connected to (SMSC3320 -> HUB SMSC2514) or (TUSB1210 -> HUB SMSC2514). The problem only happen when both port are used and after few suspend resume are triggered. If I use just one port there is no issue on suspend resume. I already covered all TI errata that I know and I'm working on TI kernel.

The problem is here

[   77.701934] ehci-omap ehci-omap.0: irq status a004 Async Recl PCD

Both ports change status from 001005 to 001009 (you have a log just after).
So from host point of view both hub connected are not working in HS mode.
After that the omap ehci has gone because the bus can not work in FS and LS and I can not recover from here. Status of transceivers are dumped and they are ok after resume.

Quoted 5 lines
Quoted 1 line

Do you have any suggestion?

I'm not very sure but both ports suddenly changing state like that look like a hardware issue. Also, it is strange that you can reproduce it only when two ports are simultaneously in use. Unfortunately, I can't match your setup with 2 ULPI

Yes I know that TI doesn't have any setup like that.

Quoted 5 lines

ports.

I have a OMAP5 uEVM that uses 2 ports but it won't be identical to your setup as they are on HSIC and not ULPI.

Quoted 6 lines
Quoted 1 line

Did you try errata i693?

Yes I have it. It's not clear if I need to wait after
ehci_writel(ehci, temp | PORT_SUSPEND, status_reg);
polling the suspendM of the SMSC or the suspend status of the PORT or I can
switch just after this instruction. Because TI kernel use an msleep of 4 mseconds and then switch. It could be a timing issue on how errata is implemented when I have two ports? How this internal count works?

The OMAP EHCI controller transparently sets the suspendM bit of the PHY when you set the PORT_SUSPEND feature on the EHCI port. So you don't need to poll for anything.

What the errata says is that once software sets the PORT_SUSPEND feature, the PHY will got into suspend and cut the PHY clock sooner than required for the EHCI controller to complete its suspend operations. (NOTE: this is only applicable when the PHY is providing the ulpi_clk).

What the workaround does is to just wait for a while (don't know why 4ms), and remux the ulpi_clock to an internal 60MHz clock for a while so that it can complete its suspend operations.

Quoted 3 lines

First time is 18, and then?

I think it is 18 for every port suspend.

Quoted 6 lines
Quoted 3 lines

Also, are you suspending and resuming only the USB or the entire system?

Whole system. Right the only susbsytem that doens't go to suspend is the FSUSB but this depends on the bootloader.

OK.

cheers,
-roger

#5 lore raw

Hi Roger

Quoted 40 lines · Roger Quadros

+Ruslan

Quoted 37 lines · Michael Trimarchi

Hi Roger

Quoted 19 lines · Roger Quadros

Hi Michael,

Quoted 16 lines · Michael Trimarchi

Hi

I'm working on omap4460 with two ulpi connected to (SMSC3320 -> HUB SMSC2514) or (TUSB1210 -> HUB SMSC2514). The problem only happen when both port are used and after few suspend resume are triggered. If I use just one port there is no issue on suspend resume. I already covered all TI errata that I know and I'm working on TI kernel.

The problem is here

[   77.701934] ehci-omap ehci-omap.0: irq status a004 Async Recl PCD

Both ports change status from 001005 to 001009 (you have a log just after).
So from host point of view both hub connected are not working in HS mode.
After that the omap ehci has gone because the bus can not work in FS and LS and I can not recover from here. Status of transceivers are dumped and they are ok after resume.

Quoted 5 lines
Quoted 1 line

Do you have any suggestion?

I'm not very sure but both ports suddenly changing state like that look like a hardware issue. Also, it is strange that you can reproduce it only when two ports are simultaneously in use. Unfortunately, I can't match your setup with 2 ULPI

Yes I know that TI doesn't have any setup like that.

Quoted 6 lines

ports.

I have a OMAP5 uEVM that uses 2 ports but it won't be identical to your setup as they are on HSIC and not ULPI.

Did you try errata i693?

Quoted 9 lines
Quoted 4 lines

Yes I have it. It's not clear if I need to wait after
ehci_writel(ehci, temp | PORT_SUSPEND, status_reg);
polling the suspendM of the SMSC or the suspend status of the PORT or I can
switch just after this instruction. Because TI kernel use an msleep of 4 mseconds and then switch. It could be a timing issue on how errata is implemented when I have two ports? How this internal count works?

The OMAP EHCI controller transparently sets the suspendM bit of the PHY when you set the PORT_SUSPEND feature on the EHCI port. So you don't need to poll for anything.

A delay is necessary. If we apply the errata to the port before a delay the errata doesn't work.

/* Use internal 60Mhz clock ULPIBx */
temp_reg = omap_readl(L3INIT_HSUSBHOST_CLKCTRL);
temp_reg |= 1 << (8 + port);
temp_reg &= ~(1 << (24 + port));
omap_writel(temp_reg, L3INIT_HSUSBHOST_CLKCTRL);

/* Wait 2ms to have transceiver transaction */
mdelay(2);

/* Use external clock ULPIBx */
temp_reg &= ~(1 << (8 + port));
temp_reg |= 1 << (24 + port);
omap_writel(temp_reg, L3INIT_HSUSBHOST_CLKCTRL);

Now it's not clear to me what happen if I apply this clock too early (transeceiver still driver the clock) or too late. Any clue?

Quoted 5 lines

What the errata says is that once software sets the PORT_SUSPEND feature, the PHY will got into suspend and cut the PHY clock sooner than required for the EHCI controller to complete its suspend operations. (NOTE: this is only applicable when the PHY is providing the ulpi_clk).

Yes this is the case

Quoted 2 lines

What the workaround does is to just wait for a while (don't know why 4ms), and remux the ulpi_clock to an internal 60MHz clock for a while so that it can complete its suspend operations.

What happen if I apply a big delay after PORT_SUSPEND. Will the internal state machine of the omap continue to wait the clock?

Michael

Quoted 17 lines
Quoted 4 lines

First time is 18, and then?

I think it is 18 for every port suspend.

Quoted 6 lines
Quoted 3 lines

Also, are you suspending and resuming only the USB or the entire system?

Whole system. Right the only susbsytem that doens't go to suspend is the FSUSB but this depends on the bootloader.

OK.

cheers,
-roger

#6 lore raw

Hello guys,

Quoted 65 lines · Michael Trimarchi

Hi Roger

Quoted 40 lines · Roger Quadros

+Ruslan

Quoted 37 lines · Michael Trimarchi

Hi Roger

Quoted 19 lines · Roger Quadros

Hi Michael,

Quoted 16 lines · Michael Trimarchi

Hi

I'm working on omap4460 with two ulpi connected to (SMSC3320 -> HUB SMSC2514) or (TUSB1210 -> HUB SMSC2514). The problem only happen when both port are used and after few suspend resume are triggered. If I use just one port there is no issue on suspend resume. I already covered all TI errata that I know and I'm working on TI kernel.

The problem is here

[   77.701934] ehci-omap ehci-omap.0: irq status a004 Async Recl PCD

Both ports change status from 001005 to 001009 (you have a log just after).
So from host point of view both hub connected are not working in HS mode.
After that the omap ehci has gone because the bus can not work in FS and LS and I can not recover from here. Status of transceivers are dumped and they are ok after resume.

Quoted 5 lines
Quoted 1 line

Do you have any suggestion?

I'm not very sure but both ports suddenly changing state like that look like a hardware issue. Also, it is strange that you can reproduce it only when two ports are simultaneously in use. Unfortunately, I can't match your setup with 2 ULPI

Yes I know that TI doesn't have any setup like that.

Quoted 6 lines

ports.

I have a OMAP5 uEVM that uses 2 ports but it won't be identical to your setup as they are on HSIC and not ULPI.

Did you try errata i693?

Quoted 9 lines
Quoted 4 lines

Yes I have it. It's not clear if I need to wait after
ehci_writel(ehci, temp | PORT_SUSPEND, status_reg);
polling the suspendM of the SMSC or the suspend status of the PORT or I can
switch just after this instruction. Because TI kernel use an msleep of 4 mseconds and then switch. It could be a timing issue on how errata is implemented when I have two ports? How this internal count works?

The OMAP EHCI controller transparently sets the suspendM bit of the PHY when you set the PORT_SUSPEND feature on the EHCI port. So you don't need to poll for anything.

A delay is necessary. If we apply the errata to the port before a delay the errata doesn't work.

/* Use internal 60Mhz clock ULPIBx */
temp_reg = omap_readl(L3INIT_HSUSBHOST_CLKCTRL);
temp_reg |= 1 << (8 + port);
temp_reg &= ~(1 << (24 + port));
omap_writel(temp_reg, L3INIT_HSUSBHOST_CLKCTRL);

/* Wait 2ms to have transceiver transaction */
mdelay(2);

Quoted 7 lines

/* Use external clock ULPIBx */
temp_reg &= ~(1 << (8 + port));
temp_reg |= 1 << (24 + port);
omap_writel(temp_reg, L3INIT_HSUSBHOST_CLKCTRL);

Now it's not clear to me what happen if I apply this clock too early (transeceiver still driver the clock) or too late. Any clue?

We need to wait 3ms for entire USB bus to go into suspend after setting the PORT_SUSPEND bit. During this time, internal OMAP EHCI logic will communicate with PHY so it is not safe to switch the clocks to internal source.
That's why with 2ms delay it fails. 4ms delay (3ms + 1ms for safety) is enough here
and is successfully used last few years for production devices.

Signature · 2 lines
Best regards,
Ruslan Bilvol
Quoted 39 lines
Quoted 7 lines

What the errata says is that once software sets the PORT_SUSPEND feature, the PHY will got into suspend and cut the PHY clock sooner than required for the EHCI controller to complete its suspend operations. (NOTE: this is only applicable when the PHY is providing the ulpi_clk).

Yes this is the case

Quoted 2 lines

What the workaround does is to just wait for a while (don't know why 4ms), and remux the ulpi_clock to an internal 60MHz clock for a while so that it can complete its suspend operations.

What happen if I apply a big delay after PORT_SUSPEND. Will the internal state machine of the omap continue to wait the clock?

Michael

Quoted 10 lines
Quoted 4 lines

First time is 18, and then?

I think it is 18 for every port suspend.

Quoted 4 lines
Quoted 3 lines

Also, are you suspending and resuming only the USB or the entire system?

Quoted 7 lines
Quoted 2 lines

Whole system. Right the only susbsytem that doens't go to suspend is the FSUSB but this depends on the bootloader.

OK.

cheers,
-roger

--
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

#7 lore raw

Hi

Quoted 77 lines · Ruslan Bilovol

Hello guys,

Quoted 65 lines · Michael Trimarchi

Hi Roger

Quoted 40 lines · Roger Quadros

+Ruslan

Quoted 37 lines · Michael Trimarchi

Hi Roger

Quoted 19 lines · Roger Quadros

Hi Michael,

Quoted 16 lines · Michael Trimarchi

Hi

I'm working on omap4460 with two ulpi connected to (SMSC3320 -> HUB SMSC2514) or (TUSB1210 -> HUB SMSC2514). The problem only happen when both port are used and after few suspend resume are triggered. If I use just one port there is no issue on suspend resume. I already covered all TI errata that I know and I'm working on TI kernel.

The problem is here

[   77.701934] ehci-omap ehci-omap.0: irq status a004 Async Recl PCD

Both ports change status from 001005 to 001009 (you have a log just after).
So from host point of view both hub connected are not working in HS mode.
After that the omap ehci has gone because the bus can not work in FS and LS and I can not recover from here. Status of transceivers are dumped and they are ok after resume.

Quoted 5 lines
Quoted 1 line

Do you have any suggestion?

I'm not very sure but both ports suddenly changing state like that look like a hardware issue. Also, it is strange that you can reproduce it only when two ports are simultaneously in use. Unfortunately, I can't match your setup with 2 ULPI

Yes I know that TI doesn't have any setup like that.

Quoted 6 lines

ports.

I have a OMAP5 uEVM that uses 2 ports but it won't be identical to your setup as they are on HSIC and not ULPI.

Did you try errata i693?

Quoted 9 lines
Quoted 4 lines

Yes I have it. It's not clear if I need to wait after
ehci_writel(ehci, temp | PORT_SUSPEND, status_reg);
polling the suspendM of the SMSC or the suspend status of the PORT or I can
switch just after this instruction. Because TI kernel use an msleep of 4 mseconds and then switch. It could be a timing issue on how errata is implemented when I have two ports? How this internal count works?

The OMAP EHCI controller transparently sets the suspendM bit of the PHY when you set the PORT_SUSPEND feature on the EHCI port. So you don't need to poll for anything.

A delay is necessary. If we apply the errata to the port before a delay the errata doesn't work.

/* Use internal 60Mhz clock ULPIBx */
temp_reg = omap_readl(L3INIT_HSUSBHOST_CLKCTRL);
temp_reg |= 1 << (8 + port);
temp_reg &= ~(1 << (24 + port));
omap_writel(temp_reg, L3INIT_HSUSBHOST_CLKCTRL);

/* Wait 2ms to have transceiver transaction */
mdelay(2);

Quoted 7 lines

/* Use external clock ULPIBx */
temp_reg &= ~(1 << (8 + port));
temp_reg |= 1 << (24 + port);
omap_writel(temp_reg, L3INIT_HSUSBHOST_CLKCTRL);

Now it's not clear to me what happen if I apply this clock too early (transeceiver still driver the clock) or too late. Any clue?

Quoted 7 lines

We need to wait 3ms for entire USB bus to go into suspend after setting the PORT_SUSPEND bit. During this time, internal OMAP EHCI logic will communicate with PHY so it is not safe to switch the clocks to internal source.
That's why with 2ms delay it fails. 4ms delay (3ms + 1ms for safety) is enough here
and is successfully used last few years for production devices.

Well this mdelay is the switch of the clock and not the delay after PORT_SUSPEND. So after writing PORT_SUSPEND I need to wait
4ms and then in 2ms you have a lot of clock to reach the 18 count. Correct?

Anyway I understand, but why both hub connected to the smsc3320 move from HS to FS? So it's not important if there is a drift of delay but it must be >= 4ms. Correct?

The code works if I just use one port or remove one hub ;)

Now the code is like this:

temp &= ~(PORT_WKCONN_E | PORT_RWC_BITS); temp |= PORT_WKDISC_E | PORT_WKOC_E; ehci_writel(ehci, temp | PORT_SUSPEND, status_reg);

mdelay(4);
omap_ehci_erratum_i693(ehci, ((wIndex & 0xff) - 1));

The code of the function is up on this email.

Michael

Quoted 44 lines
Signature · 3 lines
-- 
Best regards,
Ruslan Bilvol
Quoted 39 lines
Quoted 7 lines

What the errata says is that once software sets the PORT_SUSPEND feature, the PHY will got into suspend and cut the PHY clock sooner than required for the EHCI controller to complete its suspend operations. (NOTE: this is only applicable when the PHY is providing the ulpi_clk).

Yes this is the case

Quoted 2 lines

What the workaround does is to just wait for a while (don't know why 4ms), and remux the ulpi_clock to an internal 60MHz clock for a while so that it can complete its suspend operations.

What happen if I apply a big delay after PORT_SUSPEND. Will the internal state machine of the omap continue to wait the clock?

Michael

Quoted 10 lines
Quoted 4 lines

First time is 18, and then?

I think it is 18 for every port suspend.

Quoted 4 lines
Quoted 3 lines

Also, are you suspending and resuming only the USB or the entire system?

Quoted 7 lines
Quoted 2 lines

Whole system. Right the only susbsytem that doens't go to suspend is the FSUSB but this depends on the bootloader.

OK.

cheers,
-roger

--
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

#8 lore raw
Quoted 88 lines · Michael Trimarchi

Hi

Quoted 77 lines · Ruslan Bilovol

Hello guys,

Quoted 65 lines · Michael Trimarchi

Hi Roger

Quoted 40 lines · Roger Quadros

+Ruslan

Quoted 37 lines · Michael Trimarchi

Hi Roger

Quoted 19 lines · Roger Quadros

Hi Michael,

Quoted 16 lines · Michael Trimarchi

Hi

I'm working on omap4460 with two ulpi connected to (SMSC3320 -> HUB SMSC2514) or (TUSB1210 -> HUB SMSC2514). The problem only happen when both port are used and after few suspend resume are triggered. If I use just one port there is no issue on suspend resume. I already covered all TI errata that I know and I'm working on TI kernel.

The problem is here

[   77.701934] ehci-omap ehci-omap.0: irq status a004 Async Recl PCD

Both ports change status from 001005 to 001009 (you have a log just after).
So from host point of view both hub connected are not working in HS mode.
After that the omap ehci has gone because the bus can not work in FS and LS and I can not recover from here. Status of transceivers are dumped and they are ok after resume.

Quoted 5 lines
Quoted 1 line

Do you have any suggestion?

I'm not very sure but both ports suddenly changing state like that look like a hardware issue. Also, it is strange that you can reproduce it only when two ports are simultaneously in use. Unfortunately, I can't match your setup with 2 ULPI

Yes I know that TI doesn't have any setup like that.

Quoted 6 lines

ports.

I have a OMAP5 uEVM that uses 2 ports but it won't be identical to your setup as they are on HSIC and not ULPI.

Did you try errata i693?

Quoted 9 lines
Quoted 4 lines

Yes I have it. It's not clear if I need to wait after
ehci_writel(ehci, temp | PORT_SUSPEND, status_reg);
polling the suspendM of the SMSC or the suspend status of the PORT or I can
switch just after this instruction. Because TI kernel use an msleep of 4 mseconds and then switch. It could be a timing issue on how errata is implemented when I have two ports? How this internal count works?

The OMAP EHCI controller transparently sets the suspendM bit of the PHY when you set the PORT_SUSPEND feature on the EHCI port. So you don't need to poll for anything.

A delay is necessary. If we apply the errata to the port before a delay the errata doesn't work.

/* Use internal 60Mhz clock ULPIBx */
temp_reg = omap_readl(L3INIT_HSUSBHOST_CLKCTRL);
temp_reg |= 1 << (8 + port);
temp_reg &= ~(1 << (24 + port));
omap_writel(temp_reg, L3INIT_HSUSBHOST_CLKCTRL);

/* Wait 2ms to have transceiver transaction */
mdelay(2);

Quoted 7 lines

/* Use external clock ULPIBx */
temp_reg &= ~(1 << (8 + port));
temp_reg |= 1 << (24 + port);
omap_writel(temp_reg, L3INIT_HSUSBHOST_CLKCTRL);

Now it's not clear to me what happen if I apply this clock too early (transeceiver still driver the clock) or too late. Any clue?

Quoted 7 lines

We need to wait 3ms for entire USB bus to go into suspend after setting the PORT_SUSPEND bit. During this time, internal OMAP EHCI logic will communicate with PHY so it is not safe to switch the clocks to internal source.
That's why with 2ms delay it fails. 4ms delay (3ms + 1ms for safety) is enough here
and is successfully used last few years for production devices.

Quoted 3 lines

Well this mdelay is the switch of the clock and not the delay after PORT_SUSPEND. So after writing PORT_SUSPEND I need to wait
4ms and then in 2ms you have a lot of clock to reach the 18 count. Correct?

Yes, sorry for confusing :)
Moreover, 2ms is more than enough, errata document says about 1ms delay (and probably may be decreased up to few useconds)

Quoted 9 lines

Anyway I understand, but why both hub connected to the smsc3320 move from HS to FS? So it's not important if there is a drift of delay but it must be >= 4ms. Correct?

The code works if I just use one port or remove one hub ;)

Now the code is like this:

Quoted 8 lines

temp &= ~(PORT_WKCONN_E | PORT_RWC_BITS); temp |= PORT_WKDISC_E | PORT_WKOC_E; ehci_writel(ehci, temp | PORT_SUSPEND, status_reg);

mdelay(4);
omap_ehci_erratum_i693(ehci, ((wIndex & 0xff) - 1));

The code of the function is up on this email.

Do you have locks around this software workaround? The patch I did against 3.4 linux kernel may be helpful for you in such case: http://review.omapzoom.org/28515
Another patch extends this WA for all OMAP4 SoCs:
http://review.omapzoom.org/31108

Regards
Ruslan

Quoted 51 lines

Michael

Quoted 44 lines

--
Best regards,
Ruslan Bilvol

Quoted 39 lines
Quoted 7 lines

What the errata says is that once software sets the PORT_SUSPEND feature, the PHY will got into suspend and cut the PHY clock sooner than required for the EHCI controller to complete its suspend operations. (NOTE: this is only applicable when the PHY is providing the ulpi_clk).

Yes this is the case

Quoted 2 lines

What the workaround does is to just wait for a while (don't know why 4ms), and remux the ulpi_clock to an internal 60MHz clock for a while so that it can complete its suspend operations.

What happen if I apply a big delay after PORT_SUSPEND. Will the internal state machine of the omap continue to wait the clock?

Michael

Quoted 10 lines
Quoted 4 lines

First time is 18, and then?

I think it is 18 for every port suspend.

Quoted 4 lines
Quoted 3 lines

Also, are you suspending and resuming only the USB or the entire system?

Quoted 7 lines
Quoted 2 lines

Whole system. Right the only susbsytem that doens't go to suspend is the FSUSB but this depends on the bootloader.

OK.

cheers,
-roger

--
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

--
To unsubscribe from this list: send the line "unsubscribe linux-omap" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html

#9 lore raw

Hi

Quoted 114 lines · Ruslan Bilovol
Quoted 88 lines · Michael Trimarchi

Hi

Quoted 77 lines · Ruslan Bilovol

Hello guys,

Quoted 65 lines · Michael Trimarchi

Hi Roger

Quoted 40 lines · Roger Quadros

+Ruslan

Quoted 37 lines · Michael Trimarchi

Hi Roger

Quoted 19 lines · Roger Quadros

Hi Michael,

Quoted 16 lines · Michael Trimarchi

Hi

I'm working on omap4460 with two ulpi connected to (SMSC3320 -> HUB SMSC2514) or (TUSB1210 -> HUB SMSC2514). The problem only happen when both port are used and after few suspend resume are triggered. If I use just one port there is no issue on suspend resume. I already covered all TI errata that I know and I'm working on TI kernel.

The problem is here

[   77.701934] ehci-omap ehci-omap.0: irq status a004 Async Recl PCD

Both ports change status from 001005 to 001009 (you have a log just after).
So from host point of view both hub connected are not working in HS mode.
After that the omap ehci has gone because the bus can not work in FS and LS and I can not recover from here. Status of transceivers are dumped and they are ok after resume.

Quoted 5 lines
Quoted 1 line

Do you have any suggestion?

I'm not very sure but both ports suddenly changing state like that look like a hardware issue. Also, it is strange that you can reproduce it only when two ports are simultaneously in use. Unfortunately, I can't match your setup with 2 ULPI

Yes I know that TI doesn't have any setup like that.

Quoted 6 lines

ports.

I have a OMAP5 uEVM that uses 2 ports but it won't be identical to your setup as they are on HSIC and not ULPI.

Did you try errata i693?

Quoted 9 lines
Quoted 4 lines

Yes I have it. It's not clear if I need to wait after
ehci_writel(ehci, temp | PORT_SUSPEND, status_reg);
polling the suspendM of the SMSC or the suspend status of the PORT or I can
switch just after this instruction. Because TI kernel use an msleep of 4 mseconds and then switch. It could be a timing issue on how errata is implemented when I have two ports? How this internal count works?

The OMAP EHCI controller transparently sets the suspendM bit of the PHY when you set the PORT_SUSPEND feature on the EHCI port. So you don't need to poll for anything.

A delay is necessary. If we apply the errata to the port before a delay the errata doesn't work.

/* Use internal 60Mhz clock ULPIBx */
temp_reg = omap_readl(L3INIT_HSUSBHOST_CLKCTRL);
temp_reg |= 1 << (8 + port);
temp_reg &= ~(1 << (24 + port));
omap_writel(temp_reg, L3INIT_HSUSBHOST_CLKCTRL);

/* Wait 2ms to have transceiver transaction */
mdelay(2);

Quoted 7 lines

/* Use external clock ULPIBx */
temp_reg &= ~(1 << (8 + port));
temp_reg |= 1 << (24 + port);
omap_writel(temp_reg, L3INIT_HSUSBHOST_CLKCTRL);

Now it's not clear to me what happen if I apply this clock too early (transeceiver still driver the clock) or too late. Any clue?

Quoted 7 lines

We need to wait 3ms for entire USB bus to go into suspend after setting the PORT_SUSPEND bit. During this time, internal OMAP EHCI logic will communicate with PHY so it is not safe to switch the clocks to internal source.
That's why with 2ms delay it fails. 4ms delay (3ms + 1ms for safety) is enough here
and is successfully used last few years for production devices.

Quoted 3 lines

Well this mdelay is the switch of the clock and not the delay after PORT_SUSPEND. So after writing PORT_SUSPEND I need to wait
4ms and then in 2ms you have a lot of clock to reach the 18 count. Correct?

Yes, sorry for confusing :)
Moreover, 2ms is more than enough, errata document says about 1ms delay (and probably may be decreased up to few useconds)

Quoted 16 lines

Anyway I understand, but why both hub connected to the smsc3320 move from HS to FS? So it's not important if there is a drift of delay but it must be >= 4ms. Correct?

The code works if I just use one port or remove one hub ;)

Now the code is like this:

temp &= ~(PORT_WKCONN_E | PORT_RWC_BITS); temp |= PORT_WKDISC_E | PORT_WKOC_E; ehci_writel(ehci, temp | PORT_SUSPEND, status_reg);

mdelay(4);
omap_ehci_erratum_i693(ehci, ((wIndex & 0xff) - 1));

Quoted 7 lines
Quoted 1 line

The code of the function is up on this email.

Do you have locks around this software workaround? The patch I did against 3.4 linux kernel may be helpful for you in such case: http://review.omapzoom.org/28515
Another patch extends this WA for all OMAP4 SoCs:
http://review.omapzoom.org/31108

Yes, I'm using this code already applied to p-android-3.4 tree. So the code is fine because it works when I use just one port but for some reason I have problem when both ulpibx are active.

The scenario is this one:

GOOD ONE

[   23.389892] ehci-omap ehci-omap.0: GetStatus port:1 status 001885 0  ACK POWER sig=j SUSPEND PE CONNECT
[   23.389923] usb 1-1: usb resume
[   23.390014] ehci-omap ehci-omap.0: GetStatus port:2 status 001885 0  ACK POWER sig=j SUSPEND PE CONNECT
[   23.390045] usb 1-2: usb resume
[   23.428497] dump_tranceiver -----------> on port 1
[   23.428527] ULPI_FUNC_CTRL 0x40
[   23.428558] ULPI_IFC_CTRL 0x00
[   23.428558] ULPI_OTG_CTRL 0x06
[   23.428588] ULPI_DEBUG 0x00
[   23.428588] ULPI_USB_INT_STS 0x14
[   23.428619] dump_tranceiver <---------- on port 1
[   23.428649] ehci-omap ehci-omap.0: GetStatus port:1 status 001005 0  ACK POWER sig=se0 PE CONNECT
[   23.428710] dump_tranceiver -----------> on port 2
[   23.428741] ULPI_FUNC_CTRL 0x40
[   23.428771] ULPI_IFC_CTRL 0x00
[   23.428771] ULPI_OTG_CTRL 0x06
[   23.428802] ULPI_DEBUG 0x00
[   23.428802] ULPI_USB_INT_STS 0x14
[   23.428802] dump_tranceiver <---------- on port 2
[   23.428863] ehci-omap ehci-omap.0: GetStatus port:2 status 001005 0  ACK POWER sig=se0 PE CONNECT

BAD ONE

[   32.741638] ehci-omap ehci-omap.0: GetStatus port:3 status 001000 0  ACK POWER sig=se0
[   32.741851] ehci-omap ehci-omap.0: GetStatus port:1 status 001885 0  ACK POWER sig=j SUSPEND PE CONNECT
[   32.741912] usb 1-1: usb resume
[   32.741973] ehci-omap ehci-omap.0: GetStatus port:2 status 001885 0  ACK POWER sig=j SUSPEND PE CONNECT
[   32.742034] usb 1-2: usb resume
[   32.780395] dump_tranceiver -----------> on port 1
[   32.780426] ULPI_FUNC_CTRL 0x40
[   32.780456] ULPI_IFC_CTRL 0x00
[   32.780456] ULPI_OTG_CTRL 0x06
[   32.780487] ULPI_DEBUG 0x00
[   32.780487] ULPI_USB_INT_STS 0x14
[   32.780517] dump_tranceiver <---------- on port 1
[   32.780578] ehci-omap ehci-omap.0: GetStatus port:1 status 001005 0  ACK POWER sig=se0 PE CONNECT
[   32.780639] dump_tranceiver -----------> on port 2
[   32.780670] ULPI_FUNC_CTRL 0x40
[   32.780670] ULPI_IFC_CTRL 0x00
[   32.780700] ULPI_OTG_CTRL 0x06
[   32.780700] ULPI_DEBUG 0x00
[   32.780731] ULPI_USB_INT_STS 0x14
[   32.780731] dump_tranceiver <---------- on port 2
[   32.780792] ehci-omap ehci-omap.0: GetStatus port:2 status 001005 0  ACK POWER sig=se0 PE CONNECT
[   32.803894] usb 1-2: finish resume
[   32.803985] usb 1-1: finish resume
[   32.804046] ehci-omap ehci-omap.0: irq status a004 Async Recl PCD
[   32.804046] Port 2 Status 0x1000
[   32.804077] Port 1 Status 0x1009
[   32.804077] Port 0 Status 0x1009

So status 1005 was good HS
1009 switch move to FS/LS device

[   37.819274] ehci-omap ehci-omap.0: IAA watchdog: status a008 cmd 10065
[   37.819427] usb 1-2: kworker/u:0 timed out on ep0in len=0/2
[   37.819458] usb 1-2: retry with reset-resume
[   37.819519] ehci-omap ehci-omap.0: port 2 reset
[   37.834899] ehci-omap ehci-omap.0: IAA watchdog: status a008 cmd 10065
[   37.835021] usb 1-1: kworker/u:3 timed out on ep0in len=0/2
[   37.835021] usb 1-1: retry with reset-resume
[   37.882049] ehci-omap ehci-omap.0: port 2 full speed --> companion
[   37.882110] ehci-omap ehci-omap.0: GetStatus port:2 status 003001 0  ACK POWER OWNER sig=se0 CONNECT
[   37.882141] ehci-omap ehci-omap.0: irq status 200c Recl FLR PCD
[   37.882171] Port 2 Status 0x1000
[   37.882202] Port 1 Status 0x300a
[   37.882202] Port 0 Status 0x1809
[   37.882263] hub 1-0:1.0: port 2 not reset yet, waiting 50ms
[   37.944396] ehci-omap ehci-omap.0: GetStatus port:2 status 00300a 0  ACK POWER OWNER sig=se0 PEC CSC
[   37.944488] hub 1-0:1.0: logical disconnect on port 2
[   37.944549] usb 1-2: gone after usb resume? status -19
[   37.944580] usb 1-2: can't resume, status -19
[   37.944580] hub 1-0:1.0: logical disconnect on port 2
[   37.944763] ehci-omap ehci-omap.0: port 1 reset
[   38.007171] ehci-omap ehci-omap.0: port 1 full speed --> companion
[   38.007232] ehci-omap ehci-omap.0: GetStatus port:1 status 003001 0  ACK POWER OWNER sig=se0 CONNECT
[   38.007263] ehci-omap ehci-omap.0: irq status 2004 Recl PCD

Michael

Quoted 55 lines

Regards
Ruslan

Quoted 51 lines

Michael

Quoted 44 lines

--
Best regards,
Ruslan Bilvol

Quoted 39 lines
Quoted 7 lines

What the errata says is that once software sets the PORT_SUSPEND feature, the PHY will got into suspend and cut the PHY clock sooner than required for the EHCI controller to complete its suspend operations. (NOTE: this is only applicable when the PHY is providing the ulpi_clk).

Yes this is the case

Quoted 2 lines

What the workaround does is to just wait for a while (don't know why 4ms), and remux the ulpi_clock to an internal 60MHz clock for a while so that it can complete its suspend operations.

What happen if I apply a big delay after PORT_SUSPEND. Will the internal state machine of the omap continue to wait the clock?

Michael

Quoted 10 lines
Quoted 4 lines

First time is 18, and then?

I think it is 18 for every port suspend.

Quoted 4 lines
Quoted 3 lines

Also, are you suspending and resuming only the USB or the entire system?

Quoted 7 lines
Quoted 2 lines

Whole system. Right the only susbsytem that doens't go to suspend is the FSUSB but this depends on the bootloader.

OK.

cheers,
-roger

--
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

--
To unsubscribe from this list: send the line "unsubscribe linux-omap" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html

#10 lore raw
In reply to Ruslan Bilovol

Hi Roger

Quoted 114 lines · Ruslan Bilovol
Quoted 88 lines · Michael Trimarchi

Hi

Quoted 77 lines · Ruslan Bilovol

Hello guys,

Quoted 65 lines · Michael Trimarchi

Hi Roger

Quoted 40 lines · Roger Quadros

+Ruslan

Quoted 37 lines · Michael Trimarchi

Hi Roger

Quoted 19 lines · Roger Quadros

Hi Michael,

Quoted 16 lines · Michael Trimarchi

Hi

I'm working on omap4460 with two ulpi connected to (SMSC3320 -> HUB SMSC2514) or (TUSB1210 -> HUB SMSC2514). The problem only happen when both port are used and after few suspend resume are triggered. If I use just one port there is no issue on suspend resume. I already covered all TI errata that I know and I'm working on TI kernel.

The problem is here

[   77.701934] ehci-omap ehci-omap.0: irq status a004 Async Recl PCD

Both ports change status from 001005 to 001009 (you have a log just after).
So from host point of view both hub connected are not working in HS mode.
After that the omap ehci has gone because the bus can not work in FS and LS and I can not recover from here. Status of transceivers are dumped and they are ok after resume.

Quoted 5 lines
Quoted 1 line

Do you have any suggestion?

I'm not very sure but both ports suddenly changing state like that look like a hardware issue. Also, it is strange that you can reproduce it only when two ports are simultaneously in use. Unfortunately, I can't match your setup with 2 ULPI

Yes I know that TI doesn't have any setup like that.

Quoted 6 lines

ports.

I have a OMAP5 uEVM that uses 2 ports but it won't be identical to your setup as they are on HSIC and not ULPI.

Did you try errata i693?

Quoted 9 lines
Quoted 4 lines

Yes I have it. It's not clear if I need to wait after
ehci_writel(ehci, temp | PORT_SUSPEND, status_reg);
polling the suspendM of the SMSC or the suspend status of the PORT or I can
switch just after this instruction. Because TI kernel use an msleep of 4 mseconds and then switch. It could be a timing issue on how errata is implemented when I have two ports? How this internal count works?

The OMAP EHCI controller transparently sets the suspendM bit of the PHY when you set the PORT_SUSPEND feature on the EHCI port. So you don't need to poll for anything.

A delay is necessary. If we apply the errata to the port before a delay the errata doesn't work.

/* Use internal 60Mhz clock ULPIBx */
temp_reg = omap_readl(L3INIT_HSUSBHOST_CLKCTRL);
temp_reg |= 1 << (8 + port);
temp_reg &= ~(1 << (24 + port));
omap_writel(temp_reg, L3INIT_HSUSBHOST_CLKCTRL);

/* Wait 2ms to have transceiver transaction */
mdelay(2);

Quoted 7 lines

/* Use external clock ULPIBx */
temp_reg &= ~(1 << (8 + port));
temp_reg |= 1 << (24 + port);
omap_writel(temp_reg, L3INIT_HSUSBHOST_CLKCTRL);

Now it's not clear to me what happen if I apply this clock too early (transeceiver still driver the clock) or too late. Any clue?

Quoted 7 lines

We need to wait 3ms for entire USB bus to go into suspend after setting the PORT_SUSPEND bit. During this time, internal OMAP EHCI logic will communicate with PHY so it is not safe to switch the clocks to internal source.
That's why with 2ms delay it fails. 4ms delay (3ms + 1ms for safety) is enough here
and is successfully used last few years for production devices.

Quoted 3 lines

Well this mdelay is the switch of the clock and not the delay after PORT_SUSPEND. So after writing PORT_SUSPEND I need to wait
4ms and then in 2ms you have a lot of clock to reach the 18 count. Correct?

Yes, sorry for confusing :)
Moreover, 2ms is more than enough, errata document says about 1ms delay (and probably may be decreased up to few useconds)

Quoted 16 lines

Anyway I understand, but why both hub connected to the smsc3320 move from HS to FS? So it's not important if there is a drift of delay but it must be >= 4ms. Correct?

The code works if I just use one port or remove one hub ;)

Now the code is like this:

temp &= ~(PORT_WKCONN_E | PORT_RWC_BITS); temp |= PORT_WKDISC_E | PORT_WKOC_E; ehci_writel(ehci, temp | PORT_SUSPEND, status_reg);

mdelay(4);
omap_ehci_erratum_i693(ehci, ((wIndex & 0xff) - 1));

Quoted 7 lines
Quoted 1 line

The code of the function is up on this email.

Do you have locks around this software workaround? The patch I did against 3.4 linux kernel may be helpful for you in such case: http://review.omapzoom.org/28515
Another patch extends this WA for all OMAP4 SoCs:
http://review.omapzoom.org/31108

I'm testing using pm_test and stop to core (5ms and not 5 seconds) (usb suspend cycle are done correctly) so the problem could be:

1) SAR usb context restore. I have applied the SAR workaround but the core doesn't go in full retantion could be it a problem?

2) idle status of ulpis pins?

Michael

Quoted 55 lines

Regards
Ruslan

Quoted 51 lines

Michael

Quoted 44 lines

--
Best regards,
Ruslan Bilvol

Quoted 39 lines
Quoted 7 lines

What the errata says is that once software sets the PORT_SUSPEND feature, the PHY will got into suspend and cut the PHY clock sooner than required for the EHCI controller to complete its suspend operations. (NOTE: this is only applicable when the PHY is providing the ulpi_clk).

Yes this is the case

Quoted 2 lines

What the workaround does is to just wait for a while (don't know why 4ms), and remux the ulpi_clock to an internal 60MHz clock for a while so that it can complete its suspend operations.

What happen if I apply a big delay after PORT_SUSPEND. Will the internal state machine of the omap continue to wait the clock?

Michael

Quoted 10 lines
Quoted 4 lines

First time is 18, and then?

I think it is 18 for every port suspend.

Quoted 4 lines
Quoted 3 lines

Also, are you suspending and resuming only the USB or the entire system?

Quoted 7 lines
Quoted 2 lines

Whole system. Right the only susbsytem that doens't go to suspend is the FSUSB but this depends on the bootloader.

OK.

cheers,
-roger

--
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

--
To unsubscribe from this list: send the line "unsubscribe linux-omap" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html

#11 lore raw
Quoted 12 lines · Michael Trimarchi

Hi Roger

Quoted 8 lines · Ruslan Bilovol

On Thu, Jun 27, 2013 at 10:24 PM, Michael Trimarchi <michael-dyjBcgdgk7Pe9wHmmfpqLFaTQe2KTcn/@public.gmane.org> wrote:

Do you have locks around this software workaround? The patch I did against 3.4 linux kernel may be helpful for you in such case: http://review.omapzoom.org/28515
Another patch extends this WA for all OMAP4 SoCs:
http://review.omapzoom.org/31108

Quoted 5 lines

I'm testing using pm_test and stop to core (5ms and not 5 seconds) (usb suspend cycle are done correctly) so the problem could be:

1) SAR usb context restore. I have applied the SAR workaround but the core doesn't go in full retantion could be it a problem?

If core doesn't go in to OFF then SAR will not come into play. Are you still affected by the issue if OFF mode is disabled? If yes then it probably is not related to SAR.

Quoted 2 lines

2) idle status of ulpis pins?

Yes this can be possible.

When you said earlier that the problem doesn't happen when one of the ULPIs is used did you try both of them individually. e.g. case 1: port 1 only enabled, case 2: port 2 only enabled.

Did it work in both the cases?

cheers,
-roger

--
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): [email protected]

#13 lore raw

Hi

Quoted 13 lines · Roger Quadros
Quoted 12 lines · Michael Trimarchi

Hi Roger

Quoted 8 lines · Ruslan Bilovol

On Thu, Jun 27, 2013 at 10:24 PM, Michael Trimarchi <michael-dyjBcgdgk7Pe9wHmmfpqLFaTQe2KTcn/@public.gmane.org> wrote:

Do you have locks around this software workaround? The patch I did against 3.4 linux kernel may be helpful for you in such case: http://review.omapzoom.org/28515
Another patch extends this WA for all OMAP4 SoCs:
http://review.omapzoom.org/31108

Quoted 9 lines
Quoted 5 lines

I'm testing using pm_test and stop to core (5ms and not 5 seconds) (usb suspend cycle are done correctly) so the problem could be:

1) SAR usb context restore. I have applied the SAR workaround but the core doesn't go in full retantion could be it a problem?

If core doesn't go in to OFF then SAR will not come into play. Are you still affected by the issue if OFF mode is disabled? If yes then it probably is not related to SAR.

I don't go in full retantion for FSUSB now

[   36.317413]    PD_CORE curr=ON prev=ON logic=ON
[   36.317413]    PD_L3_INIT curr=ON prev=ON logic=ON
[   36.317413]       CD_L3_INIT mode=SW_SLEEP activity=0x2100
[   36.317413]          FSUSB mode=DISABLED stbyst=STBY idlest=TRANSITION
[   36.317413] Powerdomain (core_pwrdm) didn't enter target state 1 Vs achieved state 3. current state 3
[   36.317413] Powerdomain (l3init_pwrdm) didn't enter target state 1 Vs achieved state 3. current state 3

This with enable_off_mode equal to 1 and usb stuck

but if I switch to enable_off_mode 0 (debug file), it works. So the problem seems somewhere there.

Michael

Quoted 5 lines
Quoted 2 lines

2) idle status of ulpis pins?

Yes this can be possible.

Quoted 9 lines

When you said earlier that the problem doesn't happen when one of the ULPIs is used did you try both of them individually. e.g. case 1: port 1 only enabled, case 2: port 2 only enabled.

Did it work in both the cases?

cheers,
-roger

--
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

#14 lore raw
Quoted 35 lines · Michael Trimarchi

Hi

Quoted 13 lines · Roger Quadros
Quoted 12 lines · Michael Trimarchi

Hi Roger

Quoted 8 lines · Ruslan Bilovol

On Thu, Jun 27, 2013 at 10:24 PM, Michael Trimarchi <[email protected]> wrote:

Do you have locks around this software workaround? The patch I did against 3.4 linux kernel may be helpful for you in such case: http://review.omapzoom.org/28515
Another patch extends this WA for all OMAP4 SoCs:
http://review.omapzoom.org/31108

Quoted 9 lines
Quoted 5 lines

I'm testing using pm_test and stop to core (5ms and not 5 seconds) (usb suspend cycle are done correctly) so the problem could be:

1) SAR usb context restore. I have applied the SAR workaround but the core doesn't go in full retantion could be it a problem?

If core doesn't go in to OFF then SAR will not come into play. Are you still affected by the issue if OFF mode is disabled? If yes then it probably is not related to SAR.

I don't go in full retantion for FSUSB now

[   36.317413]    PD_CORE curr=ON prev=ON logic=ON
[   36.317413]    PD_L3_INIT curr=ON prev=ON logic=ON
[   36.317413]       CD_L3_INIT mode=SW_SLEEP activity=0x2100
[   36.317413]          FSUSB mode=DISABLED stbyst=STBY idlest=TRANSITION
[   36.317413] Powerdomain (core_pwrdm) didn't enter target state 1 Vs achieved state 3. current state 3
[   36.317413] Powerdomain (l3init_pwrdm) didn't enter target state 1 Vs achieved state 3. current state 3
Quoted 3 lines

This with enable_off_mode equal to 1 and usb stuck

but if I switch to enable_off_mode 0 (debug file), it works. So the problem seems somewhere there.

OK. Looks like the FSUSB module is stuck in transition. This can happen if the bootloader hasn't improperly managed the FSUSB clock.

Can you apply the below patch to your bootloader and make sure FSUSB module is OFF at boot.

Let's see how it behaves with FSUSB out of the picture.

cheers,
-roger

Quoted 1 line

From 0e42d3643d531daabd086f7ee451fdda8f22e72a Mon Sep 17 00:00:00 2001

From: Tero Kristo <[email protected]>
Date: Wed, 25 Apr 2012 06:05:20 +0000
Subject: [PATCH] omap4: do not enable fs-usb module

If this is done in the bootloader, the FS-USB will later be stuck into intransition state, which will prevent the device from entering idle.

---
 arch/arm/cpu/armv7/omap4/clocks.c |    2 --
 1 files changed, 0 insertions(+), 2 deletions(-)
arch/arm/cpu/armv7/omap4/clocks.c +0 -2
index e2189f7..2802559 100644
@@ -355,7 +355,6 @@ void enable_basic_clocks(void)
 		&prcm->cm_l4per_gptimer2_clkctrl,
 		&prcm->cm_wkup_wdtimer2_clkctrl,
 		&prcm->cm_l4per_uart3_clkctrl,
-		&prcm->cm_l3init_fsusb_clkctrl,
 		&prcm->cm_l3init_hsusbhost_clkctrl,
 		0
 	};
@@ -482,7 +481,6 @@ void enable_non_essential_clocks(void)
 		&prcm->cm_dss_dss_clkctrl,
 		&prcm->cm_sgx_sgx_clkctrl,
 		&prcm->cm_l3init_hsusbhost_clkctrl,
-		&prcm->cm_l3init_fsusb_clkctrl,
 		0
 	};
 
Signature · 1 line
1.7.4.1

A message that is not in the archive (sent off-list or not yet received): [email protected]

#16 lore raw

Hi

Quoted 38 lines · Roger Quadros
Quoted 37 lines · Michael Trimarchi

Hi

Quoted 13 lines · Roger Quadros
Quoted 12 lines · Michael Trimarchi

Hi Roger

Quoted 8 lines · Ruslan Bilovol

On Thu, Jun 27, 2013 at 10:24 PM, Michael Trimarchi <michael-dyjBcgdgk7Pe9wHmmfpqLFaTQe2KTcn/@public.gmane.org> wrote:

Do you have locks around this software workaround? The patch I did against 3.4 linux kernel may be helpful for you in such case: http://review.omapzoom.org/28515
Another patch extends this WA for all OMAP4 SoCs:
http://review.omapzoom.org/31108

Quoted 9 lines
Quoted 5 lines

I'm testing using pm_test and stop to core (5ms and not 5 seconds) (usb suspend cycle are done correctly) so the problem could be:

1) SAR usb context restore. I have applied the SAR workaround but the core doesn't go in full retantion could be it a problem?

If core doesn't go in to OFF then SAR will not come into play. Are you still affected by the issue if OFF mode is disabled? If yes then it probably is not related to SAR.

I don't go in full retantion for FSUSB now

[   36.317413]    PD_CORE curr=ON prev=ON logic=ON
[   36.317413]    PD_L3_INIT curr=ON prev=ON logic=ON
[   36.317413]       CD_L3_INIT mode=SW_SLEEP activity=0x2100
[   36.317413]          FSUSB mode=DISABLED stbyst=STBY idlest=TRANSITION
[   36.317413] Powerdomain (core_pwrdm) didn't enter target state 1 Vs achieved state 3. current state 3
[   36.317413] Powerdomain (l3init_pwrdm) didn't enter target state 1 Vs achieved state 3. current state 3

This with enable_off_mode equal to 1 and usb stuck

Quoted 8 lines
Quoted 1 line

but if I switch to enable_off_mode 0 (debug file), it works. So the problem seems somewhere there.

OK. Looks like the FSUSB module is stuck in transition. This can happen if the bootloader hasn't improperly managed the FSUSB clock.

Can you apply the below patch to your bootloader and make sure FSUSB module is OFF at boot.

Let's see how it behaves with FSUSB out of the picture.

I was having already in the queue. Not was testing because the system is very complicated with boot security. BTW now I have fixed the retantion and change suspend state of REGEN1 for a shutdown problem. So now the system
can suspend perfectly but the problem is still there. I have done a very
good step forward ;) anyway (was planned to test the new version of bootloader)

Michael

Quoted 39 lines

cheers,
-roger

From 0e42d3643d531daabd086f7ee451fdda8f22e72a Mon Sep 17 00:00:00 2001
From: Tero Kristo <[email protected]>
Date: Wed, 25 Apr 2012 06:05:20 +0000
Subject: [PATCH] omap4: do not enable fs-usb module

If this is done in the bootloader, the FS-USB will later be stuck into intransition state, which will prevent the device from entering idle.

---
 arch/arm/cpu/armv7/omap4/clocks.c |    2 --
 1 files changed, 0 insertions(+), 2 deletions(-)
arch/arm/cpu/armv7/omap4/clocks.c +0 -2
index e2189f7..2802559 100644
@@ -355,7 +355,6 @@ void enable_basic_clocks(void)
 		&prcm->cm_l4per_gptimer2_clkctrl,
 		&prcm->cm_wkup_wdtimer2_clkctrl,
 		&prcm->cm_l4per_uart3_clkctrl,
-		&prcm->cm_l3init_fsusb_clkctrl,
 		&prcm->cm_l3init_hsusbhost_clkctrl,
 		0
 	};
@@ -482,7 +481,6 @@ void enable_non_essential_clocks(void)
 		&prcm->cm_dss_dss_clkctrl,
 		&prcm->cm_sgx_sgx_clkctrl,
 		&prcm->cm_l3init_hsusbhost_clkctrl,
-		&prcm->cm_l3init_fsusb_clkctrl,
 		0
 	};
 
Signature · 1 line
1.7.4.1

--
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

#17 lore raw
In reply to a message not in the archive

Hi

Quoted 25 lines · Roger Quadros
Quoted 12 lines · Michael Trimarchi

Hi Roger

Quoted 8 lines · Ruslan Bilovol

On Thu, Jun 27, 2013 at 10:24 PM, Michael Trimarchi <michael-dyjBcgdgk7Pe9wHmmfpqLFaTQe2KTcn/@public.gmane.org> wrote:

Do you have locks around this software workaround? The patch I did against 3.4 linux kernel may be helpful for you in such case: http://review.omapzoom.org/28515
Another patch extends this WA for all OMAP4 SoCs:
http://review.omapzoom.org/31108

Quoted 5 lines

I'm testing using pm_test and stop to core (5ms and not 5 seconds) (usb suspend cycle are done correctly) so the problem could be:

1) SAR usb context restore. I have applied the SAR workaround but the core doesn't go in full retantion could be it a problem?

If core doesn't go in to OFF then SAR will not come into play. Are you still affected by the issue if OFF mode is disabled? If yes then it probably is not related to SAR.

Quoted 2 lines

2) idle status of ulpis pins?

Quoted 7 lines

Yes this can be possible.

When you said earlier that the problem doesn't happen when one of the ULPIs is used did you try both of them individually. e.g. case 1: port 1 only enabled, case 2: port 2 only enabled.

Did it work in both the cases?

Yes, so I don't think could be a problem of ulpi pins and why this happen on both at the same time? Seems more connected to somenthing else.

Michael

Quoted 4 lines

cheers,
-roger

--
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

#18 lore raw
Quoted 30 lines · Michael Trimarchi

Hi

Quoted 27 lines · Roger Quadros
Quoted 12 lines · Michael Trimarchi

Hi Roger

Quoted 8 lines · Ruslan Bilovol

On Thu, Jun 27, 2013 at 10:24 PM, Michael Trimarchi <[email protected]> wrote:

Do you have locks around this software workaround? The patch I did against 3.4 linux kernel may be helpful for you in such case: http://review.omapzoom.org/28515
Another patch extends this WA for all OMAP4 SoCs:
http://review.omapzoom.org/31108

Quoted 5 lines

I'm testing using pm_test and stop to core (5ms and not 5 seconds) (usb suspend cycle are done correctly) so the problem could be:

1) SAR usb context restore. I have applied the SAR workaround but the core doesn't go in full retantion could be it a problem?

If core doesn't go in to OFF then SAR will not come into play. Are you still affected by the issue if OFF mode is disabled? If yes then it probably is not related to SAR.

Quoted 2 lines

2) idle status of ulpis pins?

Yes this can be possible.

Quoted 9 lines
Quoted 5 lines

When you said earlier that the problem doesn't happen when one of the ULPIs is used did you try both of them individually. e.g. case 1: port 1 only enabled, case 2: port 2 only enabled.

Did it work in both the cases?

Yes, so I don't think could be a problem of ulpi pins and why this happen on both at the same time? Seems more connected to somenthing else.

Right. Seems to be related to two things. OFF Mode and 2 ports being used simultaneously.

I'm not sure how to go about fixing this. How important is OFF Mode for your application. Can you keep it always disabled?

cheers,
-roger

A message that is not in the archive (sent off-list or not yet received): [email protected]

#20 lore raw

Hi Roger

Quoted 37 lines · Roger Quadros
Quoted 36 lines · Michael Trimarchi

Hi

Quoted 32 lines · Roger Quadros
Quoted 12 lines · Michael Trimarchi

Hi Roger

Quoted 8 lines · Ruslan Bilovol

On Thu, Jun 27, 2013 at 10:24 PM, Michael Trimarchi <michael-dyjBcgdgk7Pe9wHmmfpqLFaTQe2KTcn/@public.gmane.org> wrote:

Do you have locks around this software workaround? The patch I did against 3.4 linux kernel may be helpful for you in such case: http://review.omapzoom.org/28515
Another patch extends this WA for all OMAP4 SoCs:
http://review.omapzoom.org/31108

Quoted 5 lines

I'm testing using pm_test and stop to core (5ms and not 5 seconds) (usb suspend cycle are done correctly) so the problem could be:

1) SAR usb context restore. I have applied the SAR workaround but the core doesn't go in full retantion could be it a problem?

If core doesn't go in to OFF then SAR will not come into play. Are you still affected by the issue if OFF mode is disabled? If yes then it probably is not related to SAR.

Quoted 2 lines

2) idle status of ulpis pins?

Yes this can be possible.

When you said earlier that the problem doesn't happen when one of the ULPIs is used did you try both of them individually. e.g. case 1: port 1 only enabled, case 2: port 2 only enabled.

Did it work in both the cases?

Quoted 9 lines
Quoted 3 lines

Yes, so I don't think could be a problem of ulpi pins and why this happen on both at the same time? Seems more connected to somenthing else.

Right. Seems to be related to two things. OFF Mode and 2 ports being used simultaneously.

I'm not sure how to go about fixing this. How important is OFF Mode for your application. Can you keep it always disabled?

Right now I delivery without off mode. I will try to fix in long run the usb and I think that it could be a good platform to test omap4 mainline.

Last question:
If one domain is in RET mode and not OFF mode what happen during SAR restore in OFF mode?

Michael

Quoted 3 lines

cheers,
-roger

--
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

#21 lore raw
Quoted 2 lines · Michael Trimarchi

Last question:
If one domain is in RET mode and not OFF mode what happen during SAR restore in OFF mode?

Without going to the depth of what TRM says already:
SAR comes into play only if device-off sequence was triggered. Depending on which domain and what level of retention state was programmed, device-off sequence may not even start. Note: Generically speaking, not achieving device OFF may not mean other powerdomains cannot achieve OFF and loose context. Specific example of USB tied on core domain, the behavior could be different.

Signature · 2 lines
Regards,
Nishanth Menon
#22 lore raw
In reply to Michael Trimarchi
Quoted 20 lines · Michael Trimarchi

Hi Roger

Quoted 17 lines · Roger Quadros
Quoted 4 lines · Michael Trimarchi

Hi

Quoted 1 line
Quoted 9 lines
Quoted 5 lines

When you said earlier that the problem doesn't happen when one of the ULPIs is used did you try both of them individually. e.g. case 1: port 1 only enabled, case 2: port 2 only enabled.

Did it work in both the cases?

Yes, so I don't think could be a problem of ulpi pins and why this happen on both at the same time? Seems more connected to somenthing else.

Right. Seems to be related to two things. OFF Mode and 2 ports being used simultaneously.

Quoted 8 lines
Quoted 3 lines

I'm not sure how to go about fixing this. How important is OFF Mode for your application. Can you keep it always disabled?

Right now I delivery without off mode. I will try to fix in long run the usb and I think that it could be a good platform to test omap4 mainline.

Yes, our aim is to get it working with mainline as well.

Quoted 3 lines

Last question:
If one domain is in RET mode and not OFF mode what happen during SAR restore in OFF mode?

SAR restore will only happen when the Device enters OFF mode.

Device OFF mode can only be reached when all voltage domains are switched OFF and that would depend if all power domains entered OFF or not. Just a simplistic explanation. You can refer to chapter "3.9.3 Device OFF State Management" in the TRM.

cheers,
-roger

A message that is not in the archive (sent off-list or not yet received): [email protected]

#24 lore raw

Hi

Quoted 37 lines · Roger Quadros
Quoted 20 lines · Michael Trimarchi

Hi Roger

Quoted 17 lines · Roger Quadros
Quoted 4 lines · Michael Trimarchi

Hi

Quoted 1 line
Quoted 9 lines
Quoted 5 lines

When you said earlier that the problem doesn't happen when one of the ULPIs is used did you try both of them individually. e.g. case 1: port 1 only enabled, case 2: port 2 only enabled.

Did it work in both the cases?

Yes, so I don't think could be a problem of ulpi pins and why this happen on both at the same time? Seems more connected to somenthing else.

Right. Seems to be related to two things. OFF Mode and 2 ports being used simultaneously.

Quoted 8 lines
Quoted 3 lines

I'm not sure how to go about fixing this. How important is OFF Mode for your application. Can you keep it always disabled?

Right now I delivery without off mode. I will try to fix in long run the usb and I think that it could be a good platform to test omap4 mainline.

Yes, our aim is to get it working with mainline as well.

Quoted 3 lines

Last question:
If one domain is in RET mode and not OFF mode what happen during SAR restore in OFF mode?

Quoted 5 lines

SAR restore will only happen when the Device enters OFF mode.

Device OFF mode can only be reached when all voltage domains are switched OFF and that would depend if all power domains entered OFF or not. Just a simplistic explanation. You can refer to chapter "3.9.3 Device OFF State Management" in the TRM.

What happen if we ask to go in off mode for all domains but one doesn't go in off mode so the device will not go in off mode and the sar will not be used? How can work the restore?

Quoted 1 line

I have added a check of CM_RESTORE_ST. This register need to be clear before
going in OFF mode and then show if the status of phase1 2a and 2b is completed.
So before proceed with system resume and after resetting OFF_MODE bit I have tried to wait on this register, but without success.

Michael

Quoted 3 lines

cheers,
-roger

--
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

#25 lore raw

Hi Roger

Quoted 12 lines · Michael Trimarchi

Hi

Quoted 9 lines · Roger Quadros
Quoted 6 lines · Michael Trimarchi

Hi Roger

Quoted 3 lines · Roger Quadros
Quoted 2 lines · Michael Trimarchi

Hi

I'm working on PM consumption of other subsystem because it's a very complex device.
Right now the pm consumption is sleep mode is 6mA just for (off mode disabled)

omap4 + LPDDR2 and TWL6032

I don't exactly know if they have updated the last bootloader but I think so.

I have tried to work on STP signal and re-enable it just before resume but nothing change

Anyway my idea is the problem is releated on 18clk counter and an invalid state of the
hw. I will try to implement save & restore register by hand instead using the sar.

Michael

Quoted 51 lines
Quoted 28 lines
Quoted 13 lines
Quoted 13 lines
Quoted 10 lines
Quoted 6 lines

When you said earlier that the problem doesn't happen when one of the ULPIs is used did you try both of them individually. e.g. case 1: port 1 only enabled, case 2: port 2 only enabled.

Did it work in both the cases?

Yes, so I don't think could be a problem of ulpi pins and why this happen on both at the same time? Seems more connected to somenthing else.

Right. Seems to be related to two things. OFF Mode and 2 ports being used simultaneously.

Quoted 8 lines
Quoted 3 lines

I'm not sure how to go about fixing this. How important is OFF Mode for your application. Can you keep it always disabled?

Right now I delivery without off mode. I will try to fix in long run the usb and I think that it could be a good platform to test omap4 mainline.

Yes, our aim is to get it working with mainline as well.

Quoted 3 lines

Last question:
If one domain is in RET mode and not OFF mode what happen during SAR restore in OFF mode?

Quoted 5 lines

SAR restore will only happen when the Device enters OFF mode.

Device OFF mode can only be reached when all voltage domains are switched OFF and that would depend if all power domains entered OFF or not. Just a simplistic explanation. You can refer to chapter "3.9.3 Device OFF State Management" in the TRM.

What happen if we ask to go in off mode for all domains but one doesn't go in off mode so the device will not go in off mode and the sar will not be used? How can work the restore?

Quoted 1 line

I have added a check of CM_RESTORE_ST. This register need to be clear before
going in OFF mode and then show if the status of phase1 2a and 2b is completed.
So before proceed with system resume and after resetting OFF_MODE bit I have tried to wait on this register, but without success.

Michael

Quoted 3 lines

cheers,
-roger

A message that is not in the archive (sent off-list or not yet received): CAOf5uwk9U8CHWywfopCt-H5BJmYyuXrwPvpwaqqP+5-3g_qYhQ-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org

#27 lore raw

Hi Michael,

Quoted 25 lines · Michael Trimarchi

Hi Roger

Quoted 12 lines · Michael Trimarchi

Hi

Quoted 9 lines · Roger Quadros
Quoted 6 lines · Michael Trimarchi

Hi Roger

Quoted 3 lines · Roger Quadros
Quoted 2 lines · Michael Trimarchi

Hi

I'm working on PM consumption of other subsystem because it's a very complex device.
Right now the pm consumption is sleep mode is 6mA just for (off mode disabled)

omap4 + LPDDR2 and TWL6032

I don't exactly know if they have updated the last bootloader but I think so.

Quoted 7 lines

I have tried to work on STP signal and re-enable it just before resume but nothing change

Anyway my idea is the problem is releated on 18clk counter and an invalid state of the
hw. I will try to implement save & restore register by hand instead using the sar.

There are 2 erratas related to the issues you mentioned.

i693 - USB HOST EHCI - Port Resume Fails On Second Resume Iteration i719 - HS USB: Multiple OFF Mode Transitions Introduce Corruption

cheers,
-roger

Quoted 55 lines

Michael

Quoted 51 lines
Quoted 28 lines
Quoted 13 lines
Quoted 13 lines
Quoted 10 lines
Quoted 6 lines

When you said earlier that the problem doesn't happen when one of the ULPIs is used did you try both of them individually. e.g. case 1: port 1 only enabled, case 2: port 2 only enabled.

Did it work in both the cases?

Yes, so I don't think could be a problem of ulpi pins and why this happen on both at the same time? Seems more connected to somenthing else.

Right. Seems to be related to two things. OFF Mode and 2 ports being used simultaneously.

Quoted 8 lines
Quoted 3 lines

I'm not sure how to go about fixing this. How important is OFF Mode for your application. Can you keep it always disabled?

Right now I delivery without off mode. I will try to fix in long run the usb and I think that it could be a good platform to test omap4 mainline.

Yes, our aim is to get it working with mainline as well.

Quoted 3 lines

Last question:
If one domain is in RET mode and not OFF mode what happen during SAR restore in OFF mode?

Quoted 5 lines

SAR restore will only happen when the Device enters OFF mode.

Device OFF mode can only be reached when all voltage domains are switched OFF and that would depend if all power domains entered OFF or not. Just a simplistic explanation. You can refer to chapter "3.9.3 Device OFF State Management" in the TRM.

What happen if we ask to go in off mode for all domains but one doesn't go in off mode so the device will not go in off mode and the sar will not be used? How can work the restore?

Quoted 1 line

I have added a check of CM_RESTORE_ST. This register need to be clear before
going in OFF mode and then show if the status of phase1 2a and 2b is completed.
So before proceed with system resume and after resetting OFF_MODE bit I have tried to wait on this register, but without success.

Michael

Quoted 3 lines

cheers,
-roger

--
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

#28 lore raw
In reply to Roger Quadros

Hi

Quoted 25 lines · Roger Quadros
Quoted 12 lines · Michael Trimarchi

Hi Roger

Quoted 8 lines · Ruslan Bilovol

On Thu, Jun 27, 2013 at 10:24 PM, Michael Trimarchi <[email protected]> wrote:

Do you have locks around this software workaround? The patch I did against 3.4 linux kernel may be helpful for you in such case: http://review.omapzoom.org/28515
Another patch extends this WA for all OMAP4 SoCs:
http://review.omapzoom.org/31108

Quoted 5 lines

I'm testing using pm_test and stop to core (5ms and not 5 seconds) (usb suspend cycle are done correctly) so the problem could be:

1) SAR usb context restore. I have applied the SAR workaround but the core doesn't go in full retantion could be it a problem?

If core doesn't go in to OFF then SAR will not come into play. Are you still affected by the issue if OFF mode is disabled? If yes then it probably is not related to SAR.

Quoted 2 lines

2) idle status of ulpis pins?

Quoted 7 lines

Yes this can be possible.

When you said earlier that the problem doesn't happen when one of the ULPIs is used did you try both of them individually. e.g. case 1: port 1 only enabled, case 2: port 2 only enabled.

Did it work in both the cases?

off state of the line can make the situation worst ;). What are the idle/off state of the lines on your platforms?

I can use PAD_REMUX flag to change the datax and signal of each port. I think that when the core is in RET_OFF the signal lines are remuxed in off mode. Correct?

Michael

Quoted 4 lines

cheers,
-roger