[U-Boot] question regarding USB gadget driver (dwc2)
Hi Marek & Lukasz,
I have an issue running fastboot on my board (which is USB_SPEED_HIGH):
=> fastboot0
failed to enable in ep
failed to enable in ep
failed to enable in ep
Because of this, fastboot does not even initialize properly, and I need to "^C" to break out…
I have traced this the following code in "drivers/usb/gadget/dwc2_udc_otg.c":
559/* hardware _could_ do smaller, but driver doesn't */
560if ((desc->bmAttributes== USB_ENDPOINT_XFER_BULK
561&& le16_to_cpu(get_unaligned(&desc->wMaxPacketSize)) !=
562ep_maxpacket(ep)) || !get_unaligned(&desc->wMaxPacketSize)) {
563
564debug("%s: bad %s maxpacket\n", __func__, _ep->name);
565return -ERANGE;
566}
567
568dev= ep->dev;
569if (!dev->driver || dev->gadget.speed== USB_SPEED_UNKNOWN) {
570
571debug("%s: bogus device state\n", __func__);
572return -ESHUTDOWN;
573}
574
575ep->stopped = 0;
576ep->desc= desc;
577ep->pio_irqs= 0;
578ep->ep.maxpacket= le16_to_cpu(get_unaligned(&desc->wMaxPacketSize));
The error message is because:
line 561 wMaxPacketSize= 64, and
line 562 ep_maxpacket(ep) = 512
I am questioning this logic, because at line 578, the ep.maxpacket is being changed to the wMaxPacketSize anyway!
I am uncertain how to proceed - please advise!
Thanks in advance, Steve
On Wednesday, January 27, 2016 at 01:35:55 AM, Steve Rae wrote:
Hi Marek & Lukasz,
I have an issue running fastboot on my board (which is USB_SPEED_HIGH):
=> fastboot0 failed to enable in ep failed to enable in ep failed to enable in epBecause of this, fastboot does not even initialize properly, and I need to "^C" to break out…
I have traced this the following code in "drivers/usb/gadget/dwc2_udc_otg.c":
I will have to delegate this onto Lukasz, since he's the s3c gadget expert.
Take a look at: [PATCH 1/2] usb: gadget: s3c_udc_otg: fixed max packet size check for ep_in in high speed
Best regards, Marek Vasut
On 16-01-26 05:35 PM, Marek Vasut wrote:
On Wednesday, January 27, 2016 at 01:35:55 AM, Steve Rae wrote:
Hi Marek & Lukasz,
I have an issue running fastboot on my board (which is USB_SPEED_HIGH):
=> fastboot0 failed to enable in ep failed to enable in ep failed to enable in epBecause of this, fastboot does not even initialize properly, and I need to "^C" to break out…
I have traced this the following code in "drivers/usb/gadget/dwc2_udc_otg.c":
I will have to delegate this onto Lukasz, since he's the s3c gadget expert.
Take a look at: [PATCH 1/2] usb: gadget: s3c_udc_otg: fixed max packet size check for ep_in in high speed
Best regards, Marek Vasut
yes - I think that 'patchwork.ozlabs.org/patch/560029' would resolve my issue Thanks, Steve
On Wednesday, January 27, 2016 at 02:58:38 AM, Steve Rae wrote:
On 16-01-26 05:35 PM, Marek Vasut wrote:
On Wednesday, January 27, 2016 at 01:35:55 AM, Steve Rae wrote:
Hi Marek & Lukasz,
I have an issue running fastboot on my board (which is USB_SPEED_HIGH): => fastboot0
failed to enable in ep failed to enable in ep failed to enable in epBecause of this, fastboot does not even initialize properly, and I need to "^C" to break out…
I have traced this the following code in
"drivers/usb/gadget/dwc2_udc_otg.c":
I will have to delegate this onto Lukasz, since he's the s3c gadget expert.
Take a look at: [PATCH 1/2] usb: gadget: s3c_udc_otg: fixed max packet size check for ep_in in high speed
Best regards, Marek Vasut
yes - I think that 'patchwork.ozlabs.org/patch/560029' would resolve my issue
Can you please resubmit that patch if it helps so it finally gets in ? Thanks
Best regards, Marek Vasut
Hi Steve,
On 16-01-26 05:35 PM, Marek Vasut wrote:
On Wednesday, January 27, 2016 at 01:35:55 AM, Steve Rae wrote:
Hi Marek & Lukasz,
I have an issue running fastboot on my board (which is USB_SPEED_HIGH):
=> fastboot0 failed to enable in ep failed to enable in ep failed to enable in epBecause of this, fastboot does not even initialize properly, and I need to "^C" to break out…
I have traced this the following code in "drivers/usb/gadget/dwc2_udc_otg.c":
I will have to delegate this onto Lukasz, since he's the s3c gadget expert.
Take a look at: [PATCH 1/2] usb: gadget: s3c_udc_otg: fixed max packet size check for ep_in in high speed
Best regards, Marek Vasut
yes - I think that 'patchwork.ozlabs.org/patch/560029' would resolve my issue Thanks, Steve
Ok, I see the point. The above patch should not bring any regression, since if MPS < maxpacket(ep), then we will set the descriptor's wMaxPacketSize as ep->ep.maxpacket.
I'm just wondering if you could share the wMaxPacketSize of your platform (lsusb -v on HOST).
In my case it is: wMaxPacketSize 0x0200 1x 512 bytes
Hi Lukasz,
On 16-01-29 08:40 AM, Lukasz Majewski wrote:
Hi Steve,
On 16-01-26 05:35 PM, Marek Vasut wrote:
On Wednesday, January 27, 2016 at 01:35:55 AM, Steve Rae wrote:
Hi Marek & Lukasz,
I have an issue running fastboot on my board (which is USB_SPEED_HIGH):
=> fastboot0 failed to enable in ep failed to enable in ep failed to enable in epBecause of this, fastboot does not even initialize properly, and I need to "^C" to break out…
I have traced this the following code in "drivers/usb/gadget/dwc2_udc_otg.c":
I will have to delegate this onto Lukasz, since he's the s3c gadget expert.
Take a look at: [PATCH 1/2] usb: gadget: s3c_udc_otg: fixed max packet size check for ep_in in high speed
Best regards, Marek Vasut
yes - I think that 'patchwork.ozlabs.org/patch/560029' would resolve my issue Thanks, Steve
Ok, I see the point. The above patch should not bring any regression, since if MPS < maxpacket(ep), then we will set the descriptor's wMaxPacketSize as ep->ep.maxpacket.
I'm just wondering if you could share the wMaxPacketSize of your platform (lsusb -v on HOST).
In my case it is: wMaxPacketSize 0x0200 1x 512 bytes
on host: lsusb -v [... snip ...] Endpoint Descriptor: bLength 7 bDescriptorType 5 bEndpointAddress 0x81 EP 1 IN bmAttributes 2 Transfer Type Bulk Synch Type None Usage Type Data wMaxPacketSize 0x0040 1x 64 bytes bInterval 0 Endpoint Descriptor: bLength 7 bDescriptorType 5 bEndpointAddress 0x02 EP 2 OUT bmAttributes 2 Transfer Type Bulk Synch Type None Usage Type Data wMaxPacketSize 0x0200 1x 512 bytes bInterval 0 [... snip ...]
Thanks, Steve
participants (3)
-
Lukasz Majewski -
Marek Vasut -
Steve Rae