[v2] usb/xhci: fix compilation warning when !CONFIG_PCI && !CONFIG_PM

Message ID 20140218184741.GI28506@saruman.home
State New
Headers show

Commit Message

Felipe Balbi Feb. 18, 2014, 6:47 p.m.
On Tue, Feb 18, 2014 at 10:00:30AM -0800, David Cohen wrote:
> Hi Sarah,
> 
> On Mon, Jan 06, 2014 at 07:02:19PM -0800, David Cohen wrote:
> > When CONFIG_PCI and CONFIG_PM are not selected, xhci.c gets this
> > warning:
> > drivers/usb/host/xhci.c:409:13: warning: ‘xhci_msix_sync_irqs’ defined
> > but not used [-Wunused-function]
> > 
> > It happens due to lack of __maybe_unused flag on xhci_msix_sync_irqs()
> > function in case of !CONFIG_PCI.
> > 
> > Signed-off-by: David Cohen <david.a.cohen@linux.intel.com>
> > ---
> 
> Ping :)
> Any comments here?
> 
> Br, David
> 
> > 
> > Change v1 -> v2:
> >  - xhci_msix_sync_irqs() already uses __maybe_unused flag when CONFIG_PCI is
> >    set. Proper solution is to add same flag when !CONFIG_PCI instead of define
> >    function as inline.
> > 
> >  drivers/usb/host/xhci.c | 2 +-
> >  1 file changed, 1 insertion(+), 1 deletion(-)
> > 
> > diff --git a/drivers/usb/host/xhci.c b/drivers/usb/host/xhci.c
> > index 4265b48856f6..ed6b717b8ee1 100644
> > --- a/drivers/usb/host/xhci.c
> > +++ b/drivers/usb/host/xhci.c
> > @@ -406,7 +406,7 @@ static void xhci_cleanup_msix(struct xhci_hcd *xhci)
> >  {
> >  }
> >  
> > -static void xhci_msix_sync_irqs(struct xhci_hcd *xhci)
> > +static void __maybe_unused xhci_msix_sync_irqs(struct xhci_hcd *xhci)

bellow is likely a better fix. Usually stubs are marked inline, rather
than getting an unused attribute:

Comments

David Cohen March 21, 2014, 9:41 p.m. | #1
Hi Mathias,

On Tue, Feb 18, 2014 at 11:04:12AM -0800, David Cohen wrote:
> On Tue, Feb 18, 2014 at 12:47:41PM -0600, Felipe Balbi wrote:
> > On Tue, Feb 18, 2014 at 10:00:30AM -0800, David Cohen wrote:
> > > Hi Sarah,
> > > 
> > > On Mon, Jan 06, 2014 at 07:02:19PM -0800, David Cohen wrote:
> > > > When CONFIG_PCI and CONFIG_PM are not selected, xhci.c gets this
> > > > warning:
> > > > drivers/usb/host/xhci.c:409:13: warning: ‘xhci_msix_sync_irqs’ defined
> > > > but not used [-Wunused-function]
> > > > 
> > > > It happens due to lack of __maybe_unused flag on xhci_msix_sync_irqs()
> > > > function in case of !CONFIG_PCI.
> > > > 
> > > > Signed-off-by: David Cohen <david.a.cohen@linux.intel.com>
> > > > ---
> > > 
> > > Ping :)
> > > Any comments here?
> > > 
> > > Br, David
> > > 
> > > > 
> > > > Change v1 -> v2:
> > > >  - xhci_msix_sync_irqs() already uses __maybe_unused flag when CONFIG_PCI is
> > > >    set. Proper solution is to add same flag when !CONFIG_PCI instead of define
> > > >    function as inline.
> > > > 
> > > >  drivers/usb/host/xhci.c | 2 +-
> > > >  1 file changed, 1 insertion(+), 1 deletion(-)
> > > > 
> > > > diff --git a/drivers/usb/host/xhci.c b/drivers/usb/host/xhci.c
> > > > index 4265b48856f6..ed6b717b8ee1 100644
> > > > --- a/drivers/usb/host/xhci.c
> > > > +++ b/drivers/usb/host/xhci.c
> > > > @@ -406,7 +406,7 @@ static void xhci_cleanup_msix(struct xhci_hcd *xhci)
> > > >  {
> > > >  }
> > > >  
> > > > -static void xhci_msix_sync_irqs(struct xhci_hcd *xhci)
> > > > +static void __maybe_unused xhci_msix_sync_irqs(struct xhci_hcd *xhci)
> > 
> > bellow is likely a better fix. Usually stubs are marked inline, rather
> > than getting an unused attribute:
> 
> Thanks for commenting. That would be actually the v1 of my patch :)
> I changed after I see the proper function has __maybe_unused flag.
> 
> But I'm fine with Sarah picking any of the patch's versions.

Guess you're handling the review of this trivial patch now :)
Comments are very welcome.

Br, David

> 
> Br, David
> 
> > 
> > diff --git a/drivers/usb/host/xhci.c b/drivers/usb/host/xhci.c
> > index 3712359..8f1a6d5 100644
> > --- a/drivers/usb/host/xhci.c
> > +++ b/drivers/usb/host/xhci.c
> > @@ -404,16 +404,16 @@ static int xhci_try_enable_msi(struct usb_hcd *hcd)
> >  
> >  #else
> >  
> > -static int xhci_try_enable_msi(struct usb_hcd *hcd)
> > +static inline int xhci_try_enable_msi(struct usb_hcd *hcd)
> >  {
> >  	return 0;
> >  }
> >  
> > -static void xhci_cleanup_msix(struct xhci_hcd *xhci)
> > +static inline void xhci_cleanup_msix(struct xhci_hcd *xhci)
> >  {
> >  }
> >  
> > -static void xhci_msix_sync_irqs(struct xhci_hcd *xhci)
> > +static inline void xhci_msix_sync_irqs(struct xhci_hcd *xhci)
> >  {
> >  }
> >  
> > 
> > -- 
> > balbi
> 
> 
> 
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/
Mathias Nyman March 24, 2014, 11:44 a.m. | #2
On 03/21/2014 11:41 PM, David Cohen wrote:
> Hi Mathias,
>
>>
>> Thanks for commenting. That would be actually the v1 of my patch :)
>> I changed after I see the proper function has __maybe_unused flag.
>>
>> But I'm fine with Sarah picking any of the patch's versions.
>
> Guess you're handling the review of this trivial patch now :)
> Comments are very welcome.
>
> Br, David
>

Hi

Patch version 1 inlining the stubs looks good to me. I'll queue it for 
Greg’s usb-linus once 3.15-rc1 is out.

-Mathias

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

Patch

diff --git a/drivers/usb/host/xhci.c b/drivers/usb/host/xhci.c
index 3712359..8f1a6d5 100644
--- a/drivers/usb/host/xhci.c
+++ b/drivers/usb/host/xhci.c
@@ -404,16 +404,16 @@  static int xhci_try_enable_msi(struct usb_hcd *hcd)
 
 #else
 
-static int xhci_try_enable_msi(struct usb_hcd *hcd)
+static inline int xhci_try_enable_msi(struct usb_hcd *hcd)
 {
 	return 0;
 }
 
-static void xhci_cleanup_msix(struct xhci_hcd *xhci)
+static inline void xhci_cleanup_msix(struct xhci_hcd *xhci)
 {
 }
 
-static void xhci_msix_sync_irqs(struct xhci_hcd *xhci)
+static inline void xhci_msix_sync_irqs(struct xhci_hcd *xhci)
 {
 }