Please apply commit 057d476fff778f1 ("xhci: fix USB3 device initiated resume race with roothub autosuspend") to v4.4.y and v4.14.y

Please apply commit 057d476fff778f1 ("xhci: fix USB3 device initiated resume race with roothub autosuspend") to v4.4.y and v4.14.y

[Lee, Chiasheng]

Hi,

Commit 057d476fff778f1 ("xhci: fix USB3 device initiated resume race with roothub autosuspend")
fixes race conditions when we're dealing with a USB3 modem using v4.4 and USB3 hubs using 4.14.
Kindly apply the patch to v4.4.y and v4.14.y.


Thanks,
Chiasheng

Re: Please apply commit 057d476fff778f1 ("xhci: fix USB3 device initiated resume race with roothub autosuspend") to v4.4.y and v4.14.y

[Greg Kroah-Hartman]

On Thu, Dec 19, 2019 at 08:20:12AM +0000, Lee, Chiasheng wrote:
> Hi,
> 
> Commit 057d476fff778f1 ("xhci: fix USB3 device initiated resume race with roothub autosuspend")
> fixes race conditions when we're dealing with a USB3 modem using v4.4 and USB3 hubs using 4.14.
> Kindly apply the patch to v4.4.y and v4.14.y.

Why not 4.9.y and 4.19.y?  You can not just skip stable kernel trees,
otherwise people upgrading will have a regression.

Anyway, the reason I did not backport the patch to older kernels is that
it does not apply at all.  If you have already done the backport (and I
am guessing you have as otherwise how would you have tested this),
please provide it to us so that we can apply it to all trees.

thanks,

greg k-h

Re: Please apply commit 057d476fff778f1 ("xhci: fix USB3 device initiated resume race with roothub autosuspend") to v4.4.y and v4.14.y

[Mathias Nyman]

On 19.12.2019 10.29, Greg Kroah-Hartman wrote:
> On Thu, Dec 19, 2019 at 08:20:12AM +0000, Lee, Chiasheng wrote:
>> Hi,
>>
>> Commit 057d476fff778f1 ("xhci: fix USB3 device initiated resume race with roothub autosuspend")
>> fixes race conditions when we're dealing with a USB3 modem using v4.4 and USB3 hubs using 4.14.
>> Kindly apply the patch to v4.4.y and v4.14.y.
> 
> Why not 4.9.y and 4.19.y?  You can not just skip stable kernel trees,
> otherwise people upgrading will have a regression.
> 
> Anyway, the reason I did not backport the patch to older kernels is that
> it does not apply at all.  If you have already done the backport (and I
> am guessing you have as otherwise how would you have tested this),
> please provide it to us so that we can apply it to all trees.
> 

Let me fix up and submit the backport, give me a few minutes.
I'll check it applies to the other stable trees as well.

-Mathias

RE: Please apply commit 057d476fff778f1 ("xhci: fix USB3 device initiated resume race with roothub autosuspend") to v4.4.y and v4.14.y

[Lee, Chiasheng]

Dear Greg,
Thanks for the promptly reply.

> -----Original Message-----
> From: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
> Sent: Thursday, December 19, 2019 4:29 PM
> To: Lee, Chiasheng <chiasheng.lee@intel.com>
> Cc: stable@vger.kernel.org; Mathias Nyman <mathias.nyman@linux.intel.com>;
> Guenter Roeck <groeck@chromium.org>; Lee, Hou-hsun <hou-
> hsun.lee@intel.com>; Pan, Harry <harry.pan@intel.com>
> Subject: Re: Please apply commit 057d476fff778f1 ("xhci: fix USB3 device
> initiated resume race with roothub autosuspend") to v4.4.y and v4.14.y
> 
> On Thu, Dec 19, 2019 at 08:20:12AM +0000, Lee, Chiasheng wrote:
> > Hi,
> >
> > Commit 057d476fff778f1 ("xhci: fix USB3 device initiated resume race
> > with roothub autosuspend") fixes race conditions when we're dealing with a
> USB3 modem using v4.4 and USB3 hubs using 4.14.
> > Kindly apply the patch to v4.4.y and v4.14.y.
> 
> Why not 4.9.y and 4.19.y?  You can not just skip stable kernel trees, otherwise
> people upgrading will have a regression.
I (mistakenly?) thought it would be merged automagically to all the stable branches since stable@vger.kernel.org is in the CC list.
And we encountered, root-caused, and fixed/verified the issue only with v4.4 and v4.14 Chromium OS kernel.
> 
> Anyway, the reason I did not backport the patch to older kernels is that it does
> not apply at all.  If you have already done the backport (and I am guessing you
> have as otherwise how would you have tested this), please provide it to us so
> that we can apply it to all trees.
We did the backport to Chromium OS kernel v4.4 and v4.14 for test indeed. Will check with our colleagues and get back to you soon.

> 
> thanks,
> 
> greg k-h

Thanks,
Chiasheng