Hi,
On 4/4/2017 11:03 PM, John Stultz wrote:
> On Tue, Apr 4, 2017 at 12:38 AM, Felipe Balbi
> <felipe.balbi@xxxxxxxxxxxxxxx> wrote:
>>
>> Hi,
>>
>> Minas Harutyunyan <Minas.Harutyunyan@xxxxxxxxxxxx> writes:
>>>>>>>> We've noticed that when using usb ethernet adapters on HiKey, we
>>>>>>>> occasionally see errors like:
>>>>>>>>
>>>>>>>> dwc2 f72c0000.usb: dwc2_hc_chhltd_intr_dma: Channel 0 - ChHltd set,
>>>>>>>> but reason is unknown
>>>>>>>> dwc2 f72c0000.usb: hcint 0x00000002, intsts 0x06200029
>>>>>>>>
>>>>>>>> dwc2 f72c0000.usb: dwc2_hc_chhltd_intr_dma: Channel 11 - ChHltd set,
>>>>>>>> but reason is unknown
>>>>>>>> dwc2 f72c0000.usb: hcint 0x00000002, intsts 0x04200029
>>>>>>>>
>>>>>>>> Sometimes followed up by a usb error in the driver, something like:
>>>>>>>> asix 1-1.2:1.0 eth0: asix_rx_fixup() Bad Header Length 0x36000807, offset 68
>>>>>>>>
>>>>>>>> Curious if you've seen any reports like this?
>>>>>
>>>>> The core version is less than 2.71a, am I right?
>>>>
>>>> So it looks like its reporting 0x4f54300a for hsotg->regs + GSNPSID
>>>> which looks like DWC2_CORE_REV_3_00a
>>>>
>>>>> Please send full debug log to do more investigation.
>>>>
>>>> Full dmesg, or is there special debugging you want me to enable?
>>>
>>> Full dmesg around issue.
>>>>
>>>>> Also send us regdump after connecting ethernet adapter.
>>>>
>>>> Sorry, can you clarify how to generate this?
>>>
>>> cat regdump. To locate dwc2 regdump file: cd /; find -name regdump
>>
>> this won't work if his distro doesn't mount debugfs. Please give
>> complete instructions ;-)
>>
>> # mkdir -p /d
>> # mount -t debugfs none /d
>> # cd /d
>> # find . -name regdump
>>
>> The directory name is the same name as the dwc2 device name, AFAICT. So,
>> check your DTS for the name of the device.
>
> Thanks for the extra details! I didn't have DEBUG_FS built in, so this
> helped clue me in.
>
> Attached are dmesg including the issue and the regdump.
>
> I did notice when cating the regdump file, I saw:
> dwc2 f72c0000.usb: Mode Mismatch Interrupt: currently in Host mode
> twice. (You'll see it 4 times in the dmesg around 1077 as I cat'ed
> regdump again to verify it wasn't just chance).
>
> Let me know if there is anything else you need!
>
> thanks
> -john
>
Could you please apply attached patch and try again.
Thanks,
Minas
From 412bf7e3dad8a20af937f3682a08116b654deb02 Mon Sep 17 00:00:00 2001
Message-Id: <412bf7e3dad8a20af937f3682a08116b654deb02.1491410224.git.hminas@xxxxxxxxxxxx>
From: Minas Harutyunyan <hminas@xxxxxxxxxxxx>
Date: Mon, 11 Jul 2016 03:31:17 -0700
Subject: [PATCH] usb: dwc2: hcd: Fix host channel halt flow
According databook in Buffer and External DMA mode
non-split periodic channels can't be halted.
Change-Id: I7c7c4e16309dda7a3b5af34020d46366c357ed49
Signed-off-by: Minas Harutyunyan <hminas@xxxxxxxxxxxx>
---
drivers/usb/dwc2/hcd.c | 19 +++++++++++++++++++
1 file changed, 19 insertions(+)
diff --git a/drivers/usb/dwc2/hcd.c b/drivers/usb/dwc2/hcd.c
index f4ef159b538e..4e6ee398efb7 100644
--- a/drivers/usb/dwc2/hcd.c
+++ b/drivers/usb/dwc2/hcd.c
@@ -979,6 +979,25 @@ void dwc2_hc_halt(struct dwc2_hsotg *hsotg, struct dwc2_host_chan *chan,
if (dbg_hc(chan))
dev_vdbg(hsotg->dev, "%s()\n", __func__);
+
+ /*
+ * In buffer DMA or external DMA mode channel can't be halted
+ * for non-split periodic channels. At the end of the next
+ * uframe/frame (in the worst case), the core generates a channel
+ * halted and disables the channel automatically.
+ */
+ if ((hsotg->core_params->dma_enable > 0 &&
+ hsotg->core_params->dma_desc_enable <= 0) ||
+ (hsotg->hw_params.arch == GHWCFG2_EXT_DMA_ARCH)) {
+ if (!chan->do_split &&
+ (chan->ep_type == USB_ENDPOINT_XFER_ISOC ||
+ chan->ep_type == USB_ENDPOINT_XFER_INT)) {
+ dev_err(hsotg->dev, "%s() Channel can't be halted\n",
+ __func__);
+ return;
+ }
+ }
+
if (halt_status == DWC2_HC_XFER_NO_HALT_STATUS)
dev_err(hsotg->dev, "!!! halt_status = %d !!!\n", halt_status);
--
2.11.0
