There's a meaningless check in isp1763_run. According to the
similar implement in isp1760_run, the proper check should remove
retval = 0;
Fixes: 60d789f3bfbb ("usb: isp1760: add support for isp1763")
Signed-off-by: Tong Tiangen <[email protected]>
---
drivers/usb/isp1760/isp1760-hcd.c | 1 -
1 file changed, 1 deletion(-)
diff --git a/drivers/usb/isp1760/isp1760-hcd.c b/drivers/usb/isp1760/isp1760-hcd.c
index 016a54ea76f4..27168b4a4ef2 100644
--- a/drivers/usb/isp1760/isp1760-hcd.c
+++ b/drivers/usb/isp1760/isp1760-hcd.c
@@ -1648,7 +1648,6 @@ static int isp1763_run(struct usb_hcd *hcd)
down_write(&ehci_cf_port_reset_rwsem);
retval = isp1760_hcd_set_and_wait(hcd, FLAG_CF, 250 * 1000);
up_write(&ehci_cf_port_reset_rwsem);
- retval = 0;
if (retval)
return retval;
--
2.18.0.huawei.25
On Tue, Jun 01, 2021 at 06:03:11PM +0800, Tong Tiangen wrote:
> There's a meaningless check in isp1763_run. According to the
> similar implement in isp1760_run, the proper check should remove
> retval = 0;
>
> Fixes: 60d789f3bfbb ("usb: isp1760: add support for isp1763")
> Signed-off-by: Tong Tiangen <[email protected]>
> ---
> drivers/usb/isp1760/isp1760-hcd.c | 1 -
> 1 file changed, 1 deletion(-)
>
> diff --git a/drivers/usb/isp1760/isp1760-hcd.c b/drivers/usb/isp1760/isp1760-hcd.c
> index 016a54ea76f4..27168b4a4ef2 100644
> --- a/drivers/usb/isp1760/isp1760-hcd.c
> +++ b/drivers/usb/isp1760/isp1760-hcd.c
> @@ -1648,7 +1648,6 @@ static int isp1763_run(struct usb_hcd *hcd)
> down_write(&ehci_cf_port_reset_rwsem);
> retval = isp1760_hcd_set_and_wait(hcd, FLAG_CF, 250 * 1000);
> up_write(&ehci_cf_port_reset_rwsem);
> - retval = 0;
> if (retval)
> return retval;
>
> --
> 2.18.0.huawei.25
>
Did you test this change to verify that the driver still works properly?
You are now checking something that never was checked before...
thanks,
greg k-h
On 2021/6/3 19:45, Greg Kroah-Hartman wrote:
> On Tue, Jun 01, 2021 at 06:03:11PM +0800, Tong Tiangen wrote:
>> There's a meaningless check in isp1763_run. According to the
>> similar implement in isp1760_run, the proper check should remove
>> retval = 0;
>>
>> Fixes: 60d789f3bfbb ("usb: isp1760: add support for isp1763")
>> Signed-off-by: Tong Tiangen <[email protected]>
>> ---
>> drivers/usb/isp1760/isp1760-hcd.c | 1 -
>> 1 file changed, 1 deletion(-)
>>
>> diff --git a/drivers/usb/isp1760/isp1760-hcd.c b/drivers/usb/isp1760/isp1760-hcd.c
>> index 016a54ea76f4..27168b4a4ef2 100644
>> --- a/drivers/usb/isp1760/isp1760-hcd.c
>> +++ b/drivers/usb/isp1760/isp1760-hcd.c
>> @@ -1648,7 +1648,6 @@ static int isp1763_run(struct usb_hcd *hcd)
>> down_write(&ehci_cf_port_reset_rwsem);
>> retval = isp1760_hcd_set_and_wait(hcd, FLAG_CF, 250 * 1000);
>> up_write(&ehci_cf_port_reset_rwsem);
>> - retval = 0;
>> if (retval)
>> return retval;
>>
>> --
>> 2.18.0.huawei.25
>>
> Did you test this change to verify that the driver still works properly?
> You are now checking something that never was checked before...
>
> thanks,
>
> greg k-h
> .
Sorry, this fix was not send to Rui.
From the point of view of code logic, there should be a problem here. I
don't have the actual hardware to verify whether it works properly. Rui
may know if the patch affects the original workflow.
thanks
.
>
Hi Tiangen,
On Fri Jun 4, 2021 at 3:09 AM WEST, tongtiangen wrote:
> On 2021/6/3 19:45, Greg Kroah-Hartman wrote:
> > On Tue, Jun 01, 2021 at 06:03:11PM +0800, Tong Tiangen wrote:
> >> There's a meaningless check in isp1763_run. According to the
> >> similar implement in isp1760_run, the proper check should remove
> >> retval = 0;
> >>
> >> Fixes: 60d789f3bfbb ("usb: isp1760: add support for isp1763")
> >> Signed-off-by: Tong Tiangen <[email protected]>
> >> ---
> >> drivers/usb/isp1760/isp1760-hcd.c | 1 -
> >> 1 file changed, 1 deletion(-)
> >>
> >> diff --git a/drivers/usb/isp1760/isp1760-hcd.c b/drivers/usb/isp1760/isp1760-hcd.c
> >> index 016a54ea76f4..27168b4a4ef2 100644
> >> --- a/drivers/usb/isp1760/isp1760-hcd.c
> >> +++ b/drivers/usb/isp1760/isp1760-hcd.c
> >> @@ -1648,7 +1648,6 @@ static int isp1763_run(struct usb_hcd *hcd)
> >> down_write(&ehci_cf_port_reset_rwsem);
> >> retval = isp1760_hcd_set_and_wait(hcd, FLAG_CF, 250 * 1000);
> >> up_write(&ehci_cf_port_reset_rwsem);
> >> - retval = 0;
> >> if (retval)
> >> return retval;
> >>
> >> --
> >> 2.18.0.huawei.25
> >>
> > Did you test this change to verify that the driver still works properly?
> > You are now checking something that never was checked before...
Thanks Greg for asking the right question here :)
> >
> > thanks,
> >
> > greg k-h
> > .
> Sorry, this fix was not send to Rui.
Yeah, I did not get the original, no. I will need to add entries in
MAINTAINERS for this driver to be easier for others.
> From the point of view of code logic, there should be a problem here.
It looks like. Only if we fail to setup the Configure Flag,
something that it's not really expected to fail.
> I don't have the actual hardware to verify whether it works
> properly. Rui may know if the patch affects the original workflow.
Thanks for the patch and I will test this and get back to you.
------
Cheers,
Rui
Hi,
Managed to test this and looks good. Everything is working as
expected.
On Fri Jun 4, 2021 at 3:09 AM WEST, tongtiangen wrote:
> On 2021/6/3 19:45, Greg Kroah-Hartman wrote:
> > On Tue, Jun 01, 2021 at 06:03:11PM +0800, Tong Tiangen wrote:
> >> There's a meaningless check in isp1763_run. According to the
> >> similar implement in isp1760_run, the proper check should remove
> >> retval = 0;
however I think a more descriptive changelog instead of comparing to
similar function would be better, maybe something around:
"Remove attribution to retval before check, which make it completely
meaningless, and does't check what it was supposed: the return
value of the timed function to set up configuration flag."
> >>
> >> Fixes: 60d789f3bfbb ("usb: isp1760: add support for isp1763")
> >> Signed-off-by: Tong Tiangen <[email protected]>
With changelog changed you can add:
Tested-by: Rui Miguel Silva <[email protected]>
Reviewed-by: Rui Miguel Silva <[email protected]>
------
Cheers,
Rui
> >> ---
> >> drivers/usb/isp1760/isp1760-hcd.c | 1 -
> >> 1 file changed, 1 deletion(-)
> >>
> >> diff --git a/drivers/usb/isp1760/isp1760-hcd.c b/drivers/usb/isp1760/isp1760-hcd.c
> >> index 016a54ea76f4..27168b4a4ef2 100644
> >> --- a/drivers/usb/isp1760/isp1760-hcd.c
> >> +++ b/drivers/usb/isp1760/isp1760-hcd.c
> >> @@ -1648,7 +1648,6 @@ static int isp1763_run(struct usb_hcd *hcd)
> >> down_write(&ehci_cf_port_reset_rwsem);
> >> retval = isp1760_hcd_set_and_wait(hcd, FLAG_CF, 250 * 1000);
> >> up_write(&ehci_cf_port_reset_rwsem);
> >> - retval = 0;
> >> if (retval)
> >> return retval;
> >>
> >> --
> >> 2.18.0.huawei.25
> >>
> > Did you test this change to verify that the driver still works properly?
> > You are now checking something that never was checked before...
> >
> > thanks,
> >
> > greg k-h
> > .
> Sorry, this fix was not send to Rui.
> From the point of view of code logic, there should be a problem here. I
> don't have the actual hardware to verify whether it works properly. Rui
> may know if the patch affects the original workflow.
>
> thanks
> .
> >