On Thu, Feb 9, 2017 at 7:11 AM, Peter Chen [off-list ref] wrote:

On Wed, Feb 08, 2017 at 12:43:00PM -0800, Jack Pham wrote:

quoted

Hi Peter, Sriram, Arnd,

On Mon, Feb 06, 2017 at 05:13:38PM +0800, Peter Chen wrote:

quoted

From: Arnd Bergmann <arnd@arndb.de>

For xhci-hcd platform device, all the DMA parameters are not
configured properly, notably dma ops for dwc3 devices. So, set
the dma for xhci from sysdev. sysdev is pointing to device that
is known to the system firmware or hardware.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Sriram Dash <redacted>
Tested-by: Baolin Wang <redacted>
Tested-by: Vivek Gautam <redacted>
Tested-by: Alexander Sverdlin <redacted>
Signed-off-by: Mathias Nyman <redacted>
---

diff --git a/drivers/usb/host/xhci-plat.c b/drivers/usb/host/xhci-plat.c
index 6d33b42..7a9c860 100644
--- a/drivers/usb/host/xhci-plat.c
+++ b/drivers/usb/host/xhci-plat.c

quoted

-   hcd = usb_create_hcd(driver, &pdev->dev, dev_name(&pdev->dev));
+   hcd = __usb_create_hcd(driver, sysdev, &pdev->dev,
+                          dev_name(&pdev->dev), NULL);

As mentioned already in [1], usb_create_shared_hcd() is called to create
the second bus, however it also needs to be converted.

Not exactly as Roger's suggestion but this worked for me:

-       xhci->shared_hcd = usb_create_shared_hcd(driver, &pdev->dev,
+       xhci->shared_hcd = __usb_create_hcd(driver, sysdev, &pdev->dev,
                        dev_name(&pdev->dev), hcd);
        if (!xhci->shared_hcd) {
                ret = -ENOMEM;

Without this, SuperSpeed devices fail to enumerate:

 usb 2-1: new SuperSpeed USB device number 2 using xhci-hcd
 usb 2-1: device descriptor read/8, error -11
 usb 2-1: new SuperSpeed USB device number 2 using xhci-hcd
 usb 2-1: device descriptor read/8, error -11
 usb 2-1: new SuperSpeed USB device number 3 using xhci-hcd
 usb 2-1: device descriptor read/8, error -11
 usb 2-1: new SuperSpeed USB device number 3 using xhci-hcd
 usb 2-1: device descriptor read/8, error -11
 <and continues...>

Thanks,
Jack

Do you work at same platform with Vivek Gautam? I see his tested-by
tag in this patch.

The patches [1] had worked for me on 4.9-rc4.
But with
(a) usb-next merged on 4.10-rc7
(b) this patch and [PATCH 3/9] in this series:
      [PATCH v12 3/9] usb: separate out sysdev pointer from usb_bus
and,
(c) phy patches for msm8996 [2] along with necessary device tree patches
I get a hang when SuperSpeed device is being enumerated on db820c.

The change above suggested by Jack works for me too, and i hope that's
the right way to do things.

I can give a Tested-by again with the changes for shared_hcd in the
next version.

[1] https://lkml.org/lkml/2016/11/17/339
[2] https://lkml.org/lkml/2017/1/10/225

I have no dwc3 platform to test, I only got patches from [1], and rebase to
Greg's usb-next, the rebase has a conflict, and I made a
change which is similar to revert below patch, but it is definitely
not your problem.

commit 488dc164914ff5ce5e913abd32048d28fc0d06b8
Author: Mathias Nyman [off-list ref]
Date:   Fri Jan 20 15:38:24 2017 +0200

    xhci: remove WARN_ON if dma mask is not set for platform devices


[1] http://www.spinics.net/lists/linux-usb/msg152375.html

--

Best Regards,
Peter Chen
--
To unsubscribe from this list: send the line "unsubscribe linux-usb" in
the body of a message to majordomo at vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

-- 
Qualcomm Innovation Center, Inc. is a member of Code Aurora Forum,
a Linux Foundation Collaborative Project