[U-Boot] Need help with u-boot problem with usb-keyboard / kvm switch

[U-Boot] Need help with u-boot problem with usb-keyboard / kvm switch

[Hans de Goede]

Hi Marek, et al,

I'm working on cleaning up Luc's hdmi out support patchset for
sunxi.

As part of this I want to also add support for usb keyboards,
so as to get a full console without needing to solder wires
to testpoints on some boards :)

So when I plug in the usb coming from my kvm I get this:

(Re)start USB...
USB0:   USB EHCI 1.00
scanning bus 0 for devices... EHCI timed out on TD - token=0x80008c80
3 USB Device(s) found
       scanning usb for storage devices... 0 Storage Device(s) found

And the usb keyboard does not work.

If I plug in a single usb-2 hub (no ohci support for sunxi in u-boot
yet), then things do work, but after a few minutes of inactivity the
usb code starts spamming the console with:

EHCI timed out on TD - token=0x80008c80
EHCI timed out on TD - token=0x80008c80
...

Could this be a problem with the phy settings (iow a sunxi specific
problem)?

When using the kvm on my pc, the tree structure looks like this:

/:  Bus 06.Port 1: Dev 1, Class=root_hub, Driver=xhci_hcd/2p, 480M
    |__ Port 1: Dev 2, If 0, Class=Hub, Driver=hub/4p, 480M
        |__ Port 4: Dev 4, If 0, Class=Mass Storage, Driver=usb-storage, 480M
    |__ Port 2: Dev 3, If 0, Class=Hub, Driver=hub/4p, 480M
        |__ Port 3: Dev 11, If 0, Class=Human Interface Device, Driver=usbhid, 1.5M
        |__ Port 4: Dev 9, If 0, Class=Hub, Driver=hub/4p, 12M
            |__ Port 1: Dev 10, If 0, Class=Human Interface Device, Driver=usbhid, 12M
            |__ Port 1: Dev 10, If 1, Class=Human Interface Device, Driver=usbhid, 12M
            |__ Port 1: Dev 10, If 2, Class=Human Interface Device, Driver=usbhid, 12M

Hmm, after a bit longer I also get these errors:

ERROR: v7_dcache_inval_range - start address is not aligned - 0xe59ff014

Could the problem be that the usb code is feeding unaligned buffers to the ehci
controller ?

Regards,

Hans

[U-Boot] Need help with u-boot problem with usb-keyboard / kvm switch

[Luc Verhaegen]

On Wed, Sep 17, 2014 at 09:11:38PM +0200, Hans de Goede wrote:
> Hi Marek, et al,
> 
> I'm working on cleaning up Luc's hdmi out support patchset for
> sunxi.

?

Luc Verhaegen.

[U-Boot] Need help with u-boot problem with usb-keyboard / kvm switch

[Marek Vasut]

On Wednesday, September 17, 2014 at 09:19:11 PM, Luc Verhaegen wrote:
> On Wed, Sep 17, 2014 at 09:11:38PM +0200, Hans de Goede wrote:
> > Hi Marek, et al,
> > 
> > I'm working on cleaning up Luc's hdmi out support patchset for
> > sunxi.
> 
> ?

Your email is missing it's contents, is that at typo ?

Best regards,
Marek Vasut

[U-Boot] Need help with u-boot problem with usb-keyboard / kvm switch

[Hans de Goede]

Hi,

On 09/17/2014 09:19 PM, Luc Verhaegen wrote:
> On Wed, Sep 17, 2014 at 09:11:38PM +0200, Hans de Goede wrote:
>> Hi Marek, et al,
>>
>> I'm working on cleaning up Luc's hdmi out support patchset for
>> sunxi.
> 
> ?

Cleaning up may not have been the best choice of words, I was
talking about the rebase on top of upstream u-boot + the various
fixes you already know about.

I plan to submit a patch-set to the kernel for the clock resource
stuff for simplefb to the proper lists for that discussion soon,
and assuming that gets accepted I would like to land the
hdmi support in upstream u-boot soon.

Regards,

Hans

[U-Boot] Need help with u-boot problem with usb-keyboard / kvm switch

[Luc Verhaegen]

On Thu, Sep 18, 2014 at 10:34:04AM +0200, Hans de Goede wrote:
> Hi,
> 
> Cleaning up may not have been the best choice of words, I was
> talking about the rebase on top of upstream u-boot + the various
> fixes you already know about.
> 
> I plan to submit a patch-set to the kernel for the clock resource
> stuff for simplefb to the proper lists for that discussion soon,
> and assuming that gets accepted I would like to land the
> hdmi support in upstream u-boot soon.

Ok, thanks for getting me up to date on this.

As for simplefb, i had given up on that, and wanted to cough up 
an alternate driver that simply wouldn't be called that. But as usual, i 
am all over the place. So let's see how the simplefb route goes. Good 
luck.

Thanks,

Luc Verhaegen.

[U-Boot] Need help with u-boot problem with usb-keyboard / kvm switch

[Eric Nelson]

Hi Hans,

On 09/17/2014 12:11 PM, Hans de Goede wrote:
> Hi Marek, et al,
> 
> I'm working on cleaning up Luc's hdmi out support patchset for
> sunxi.
> 
> As part of this I want to also add support for usb keyboards,
> so as to get a full console without needing to solder wires
> to testpoints on some boards :)
> 
> So when I plug in the usb coming from my kvm I get this:
> 
> (Re)start USB...
> USB0:   USB EHCI 1.00
> scanning bus 0 for devices... EHCI timed out on TD - token=0x80008c80
> 3 USB Device(s) found
>        scanning usb for storage devices... 0 Storage Device(s) found
> 
> And the usb keyboard does not work.
> 
> If I plug in a single usb-2 hub (no ohci support for sunxi in u-boot
> yet), then things do work, but after a few minutes of inactivity the
> usb code starts spamming the console with:
> 
> EHCI timed out on TD - token=0x80008c80
> EHCI timed out on TD - token=0x80008c80
> ...
> 
> Could this be a problem with the phy settings (iow a sunxi specific
> problem)?
> 

Probably not.

We've seen the same thing on SABRE Lite and Nitrogen6X boards.

U-Boot > usb start ; setenv stdin serial,usbkbd
(Re)start USB...
USB0:   USB EHCI 1.00
scanning bus 0 for devices... 2 USB Device(s) found
USB1:   USB EHCI 1.00
scanning bus 1 for devices... EHCI timed out on TD - token=0x80008c80
EHCI timed out on TD - token=0x80008c80
EHCI timed out on TD - token=0x80008c80
6 USB Device(s) found
       scanning usb for storage devices... 1 Storage Device(s) found
       scanning usb for ethernet devices... 0 Ethernet Device(s) found
Timeout poll on interrupt endpoint
Failed to get keyboard state from device 1c4f:0002
U-Boot > usb tree
USB device tree:
  1  Hub (480 Mb/s, 0mA)
  |  u-boot EHCI Host Controller
  |
  +-2  Mass Storage (480 Mb/s, 100mA)
       Generic Mass Storage 4A3709D5

  3  Hub (480 Mb/s, 0mA)
  |  u-boot EHCI Host Controller
  |
  +-4  Hub (480 Mb/s, 2mA)
    |
    +-5  Human Interface (1.5 Mb/s, 100mA)
    |    Logitech USB Optical Mouse
    |
    +-6  Hub (480 Mb/s, 100mA)
    | |   USB 2.0 Hub
    | |
    | +-7  Human Interface (1.5 Mb/s, 98mA)
    |      USB USB Keykoard
    |
    +-8   (12 Mb/s, 100mA)

Oddly, the issue seems to be specific to **some** USB keyboards.

We have some (mostly older) USB keyboards that just work and
have had investigation on our to-do list for quite some
time now.

I was going to start with a question to the list:

	Is anyone else using USB keyboard support?

USB mass storage seems to be pretty reliable. We have
folks using it as an alternate boot for emergency
situations.

Regards,


Eric

[U-Boot] Need help with u-boot problem with usb-keyboard / kvm switch

[Marek Vasut]

On Wednesday, September 17, 2014 at 09:29:07 PM, Eric Nelson wrote:
> Hi Hans,
> 
> On 09/17/2014 12:11 PM, Hans de Goede wrote:
> > Hi Marek, et al,
> > 
> > I'm working on cleaning up Luc's hdmi out support patchset for
> > sunxi.
> > 
> > As part of this I want to also add support for usb keyboards,
> > so as to get a full console without needing to solder wires
> > to testpoints on some boards :)
> > 
> > So when I plug in the usb coming from my kvm I get this:
> > 
> > (Re)start USB...
> > USB0:   USB EHCI 1.00
> > scanning bus 0 for devices... EHCI timed out on TD - token=0x80008c80
> > 3 USB Device(s) found
> > 
> >        scanning usb for storage devices... 0 Storage Device(s) found
> > 
> > And the usb keyboard does not work.
> > 
> > If I plug in a single usb-2 hub (no ohci support for sunxi in u-boot
> > yet), then things do work, but after a few minutes of inactivity the
> > usb code starts spamming the console with:
> > 
> > EHCI timed out on TD - token=0x80008c80
> > EHCI timed out on TD - token=0x80008c80
> > ...
> > 
> > Could this be a problem with the phy settings (iow a sunxi specific
> > problem)?
> 
> Probably not.
> 
> We've seen the same thing on SABRE Lite and Nitrogen6X boards.

I agree, this happens. Which controller and driver do you use on the Sunxi 
hardware?

btw you can check if this is a cache issue by disabling dcache (set 
CONFIG_CMD_CACHE in the config and use 'dcache off' before 'usb start'). You can 
also patch the arch/arm/lib/cache-cp15.c functions to do additional alignment 
checks (always a good idea) , just like I did that in 
arch/arm/cpu/arm926ejs/cache.c

Best regards,
Marek Vasut

[U-Boot] Need help with u-boot problem with usb-keyboard / kvm switch

[Bo Shen]

Hi All,

On 09/18/2014 03:29 AM, Eric Nelson wrote:
> Hi Hans,
>
> On 09/17/2014 12:11 PM, Hans de Goede wrote:
>> Hi Marek, et al,
>>
>> I'm working on cleaning up Luc's hdmi out support patchset for
>> sunxi.
>>
>> As part of this I want to also add support for usb keyboards,
>> so as to get a full console without needing to solder wires
>> to testpoints on some boards :)
>>
>> So when I plug in the usb coming from my kvm I get this:
>>
>> (Re)start USB...
>> USB0:   USB EHCI 1.00
>> scanning bus 0 for devices... EHCI timed out on TD - token=0x80008c80
>> 3 USB Device(s) found
>>         scanning usb for storage devices... 0 Storage Device(s) found
>>
>> And the usb keyboard does not work.
>>
>> If I plug in a single usb-2 hub (no ohci support for sunxi in u-boot
>> yet), then things do work, but after a few minutes of inactivity the
>> usb code starts spamming the console with:
>>
>> EHCI timed out on TD - token=0x80008c80
>> EHCI timed out on TD - token=0x80008c80
>> ...
>>
>> Could this be a problem with the phy settings (iow a sunxi specific
>> problem)?
>>
>
> Probably not.
>
> We've seen the same thing on SABRE Lite and Nitrogen6X boards.
>
> U-Boot > usb start ; setenv stdin serial,usbkbd
> (Re)start USB...
> USB0:   USB EHCI 1.00
> scanning bus 0 for devices... 2 USB Device(s) found
> USB1:   USB EHCI 1.00
> scanning bus 1 for devices... EHCI timed out on TD - token=0x80008c80
> EHCI timed out on TD - token=0x80008c80
> EHCI timed out on TD - token=0x80008c80
> 6 USB Device(s) found
>         scanning usb for storage devices... 1 Storage Device(s) found
>         scanning usb for ethernet devices... 0 Ethernet Device(s) found
> Timeout poll on interrupt endpoint
> Failed to get keyboard state from device 1c4f:0002
> U-Boot > usb tree
> USB device tree:
>    1  Hub (480 Mb/s, 0mA)
>    |  u-boot EHCI Host Controller
>    |
>    +-2  Mass Storage (480 Mb/s, 100mA)
>         Generic Mass Storage 4A3709D5
>
>    3  Hub (480 Mb/s, 0mA)
>    |  u-boot EHCI Host Controller
>    |
>    +-4  Hub (480 Mb/s, 2mA)
>      |
>      +-5  Human Interface (1.5 Mb/s, 100mA)
>      |    Logitech USB Optical Mouse
>      |
>      +-6  Hub (480 Mb/s, 100mA)
>      | |   USB 2.0 Hub
>      | |
>      | +-7  Human Interface (1.5 Mb/s, 98mA)
>      |      USB USB Keykoard
>      |
>      +-8   (12 Mb/s, 100mA)
>
> Oddly, the issue seems to be specific to **some** USB keyboards.
>
> We have some (mostly older) USB keyboards that just work and
> have had investigation on our to-do list for quite some
> time now.
>
> I was going to start with a question to the list:
>
> 	Is anyone else using USB keyboard support?

I use the USB keyboard with following hack patch (which based on 
v2013.04), then it works well on my side.

--->8---
diff --git a/drivers/usb/host/ohci-hcd.c b/drivers/usb/host/ohci-hcd.c
index bc17b85..b74bfd7 100644
--- a/drivers/usb/host/ohci-hcd.c
+++ b/drivers/usb/host/ohci-hcd.c
@@ -515,6 +515,7 @@ static inline int sohci_return_job(struct ohci *hc, 
urb_priv_t *urb)
                         ohci_readl(&regs->intrdisable); /* PCI posting 
flush */
                 }
                 urb->actual_length = 0;
+#if 0
                 td_submit_job(
                                 urb->dev,
                                 urb->pipe,
@@ -523,6 +524,8 @@ static inline int sohci_return_job(struct ohci *hc, 
urb_priv_t *urb)
                                 NULL,
                                 urb,
                                 urb->interval);
+#endif
+               urb_free_priv(urb);
                 break;
         case PIPE_CONTROL:
         case PIPE_BULK:
---8<---

> USB mass storage seems to be pretty reliable. We have
> folks using it as an alternate boot for emergency
> situations.
>
> Regards,
>
>
> Eric

Best Regards,
Bo Shen