Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

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

4 views
Skip to first unread message

David Cohen

unread,
Jan 6, 2014, 8:40:02 PM1/6/14
to
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]

Instead of creating nested #ifdefs, this patch fixes it by defining the
xHCI PCI stubs as inline.

Signed-off-by: David Cohen <david....@linux.intel.com>
---
drivers/usb/host/xhci.c | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/usb/host/xhci.c b/drivers/usb/host/xhci.c
index 4265b48856f6..672df9840014 100644
--- a/drivers/usb/host/xhci.c
+++ b/drivers/usb/host/xhci.c
@@ -397,16 +397,16 @@ 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)
{
}

--
1.8.4.2

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

David Cohen

unread,
Jan 6, 2014, 10:00:01 PM1/6/14
to
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....@linux.intel.com>
---

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)

David Cohen

unread,
Feb 18, 2014, 1:00:01 PM2/18/14
to
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....@linux.intel.com>
> ---

Ping :)
Any comments here?

Br, David

Felipe Balbi

unread,
Feb 18, 2014, 1:50:01 PM2/18/14
to
bellow is likely a better fix. Usually stubs are marked inline, rather
than getting an unused attribute:

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
signature.asc

David Cohen

unread,
Feb 18, 2014, 2:00:02 PM2/18/14
to
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.

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


David Cohen

unread,
Mar 21, 2014, 5:50:02 PM3/21/14
to
Hi Mathias,
Guess you're handling the review of this trivial patch now :)
Comments are very welcome.

Br, David

Mathias Nyman

unread,
Mar 24, 2014, 7:40:01 AM3/24/14
to
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
0 new messages