2021-04-08 13:55:44

by liulongfang

[permalink] [raw]
Subject: [PATCH v2 0/2] USB:ehci:fix the no SRBN register problem

(1) Add a whitelist for EHCI devices without SBRN registers.
(2) Add Kunpeng920's EHCI device to the whitelist.

Changes in v2:
- Fix some code style issues.
- Update function name.

Longfang Liu (2):
USB:ehci:Add a whitelist for EHCI controllers
USB:ehci:fix Kunpeng920 ehci hardware problem

drivers/usb/host/ehci-pci.c | 30 ++++++++++++++++++++++++++----
1 file changed, 26 insertions(+), 4 deletions(-)

--
2.8.1


2021-04-08 14:55:01

by Alan Stern

[permalink] [raw]
Subject: Re: [PATCH v2 0/2] USB:ehci:fix the no SRBN register problem

On Thu, Apr 08, 2021 at 09:49:18PM +0800, Longfang Liu wrote:
> (1) Add a whitelist for EHCI devices without SBRN registers.
> (2) Add Kunpeng920's EHCI device to the whitelist.
>
> Changes in v2:
> - Fix some code style issues.
> - Update function name.
>
> Longfang Liu (2):
> USB:ehci:Add a whitelist for EHCI controllers
> USB:ehci:fix Kunpeng920 ehci hardware problem
>
> drivers/usb/host/ehci-pci.c | 30 ++++++++++++++++++++++++++----
> 1 file changed, 26 insertions(+), 4 deletions(-)

I don't think we need a whole list, along with an associated lookup
routine, when there are only two entries. The total amount of code will
be smaller if you just add a check for the Kunpeng920 controller to
the existing check for the STMICRO controller.

Alan Stern

2021-04-09 02:44:12

by liulongfang

[permalink] [raw]
Subject: Re: [PATCH v2 0/2] USB:ehci:fix the no SRBN register problem

On 2021/4/8 22:53, Alan Stern wrote:
> On Thu, Apr 08, 2021 at 09:49:18PM +0800, Longfang Liu wrote:
>> (1) Add a whitelist for EHCI devices without SBRN registers.
>> (2) Add Kunpeng920's EHCI device to the whitelist.
>>
>> Changes in v2:
>> - Fix some code style issues.
>> - Update function name.
>>
>> Longfang Liu (2):
>> USB:ehci:Add a whitelist for EHCI controllers
>> USB:ehci:fix Kunpeng920 ehci hardware problem
>>
>> drivers/usb/host/ehci-pci.c | 30 ++++++++++++++++++++++++++----
>> 1 file changed, 26 insertions(+), 4 deletions(-)
>
> I don't think we need a whole list, along with an associated lookup
> routine, when there are only two entries. The total amount of code will
> be smaller if you just add a check for the Kunpeng920 controller to
> the existing check for the STMICRO controller.
>
> Alan Stern
> .
>
Now there are two EHCI controllers that do not have SBRN registers,
and there may be more in the future. This list is added for subsequent
compatibility if there are such controllers, instead of a series of if-else.
Thanks.
Longfang.

2021-04-09 06:48:36

by Greg Kroah-Hartman

[permalink] [raw]
Subject: Re: [PATCH v2 0/2] USB:ehci:fix the no SRBN register problem

On Fri, Apr 09, 2021 at 10:42:35AM +0800, liulongfang wrote:
> On 2021/4/8 22:53, Alan Stern wrote:
> > On Thu, Apr 08, 2021 at 09:49:18PM +0800, Longfang Liu wrote:
> >> (1) Add a whitelist for EHCI devices without SBRN registers.
> >> (2) Add Kunpeng920's EHCI device to the whitelist.
> >>
> >> Changes in v2:
> >> - Fix some code style issues.
> >> - Update function name.
> >>
> >> Longfang Liu (2):
> >> USB:ehci:Add a whitelist for EHCI controllers
> >> USB:ehci:fix Kunpeng920 ehci hardware problem
> >>
> >> drivers/usb/host/ehci-pci.c | 30 ++++++++++++++++++++++++++----
> >> 1 file changed, 26 insertions(+), 4 deletions(-)
> >
> > I don't think we need a whole list, along with an associated lookup
> > routine, when there are only two entries. The total amount of code will
> > be smaller if you just add a check for the Kunpeng920 controller to
> > the existing check for the STMICRO controller.
> >
> > Alan Stern
> > .
> >
> Now there are two EHCI controllers that do not have SBRN registers,
> and there may be more in the future. This list is added for subsequent
> compatibility if there are such controllers, instead of a series of if-else.

Why would more people create new EHCI controllers these days with the
cheapness of USB 3 cores? Anyway, let us worry about that if this list
gets "too long", a simple if statement is fine for now.

thanks,

greg k-h