Hi Sergei,
On Tue, 2015-09-08 at 00:24 +0300, Sergei Shtylyov wrote:
> On 09/07/2015 11:50 PM, Alexey Brodkin wrote:
>
> > Current implementation via IS_ERR(phydev) may make no sense because
> > of_phy_attach() returns NULL on failure instead of error value.
>
> Not of_phy_connect()?
I alre
On 09/07/2015 11:50 PM, Alexey Brodkin wrote:
Current implementation via IS_ERR(phydev) may make no sense because
of_phy_attach() returns NULL on failure instead of error value.
Not of_phy_connect()?
Still for checking result of phy_connect() IS_ERR() is useful.
To address both situation
Hi Sergei,
On Mon, 2015-09-07 at 23:53 +0300, Sergei Shtylyov wrote:
> On 09/07/2015 11:50 PM, Alexey Brodkin wrote:
>
> > Current implementation via IS_ERR(phydev) may make no sense because
> > of_phy_attach() returns NULL on failure instead of error value.
> >
> > Still for checking result of
On 09/07/2015 11:50 PM, Alexey Brodkin wrote:
Current implementation via IS_ERR(phydev) may make no sense because
of_phy_attach() returns NULL on failure instead of error value.
Still for checking result of phy_connect() IS_ERR() is useful.
To address both situations we use combined IS_ERR_OR_
Current implementation via IS_ERR(phydev) may make no sense because
of_phy_attach() returns NULL on failure instead of error value.
Still for checking result of phy_connect() IS_ERR() is useful.
To address both situations we use combined IS_ERR_OR_NULL() check.
Cc: Giuseppe Cavallaro
Cc: linux-
