Re: [mmc:mmc-next 17/28] sdhci-pltfm.c:(.text+0x80): multiple definition of `sdhci_runtime_pm_get'

[Fengguang Wu <fengguang.wu@intel.com>]

On Wed, Nov 07, 2012 at 03:43:05PM -0500, Chris Ball wrote:
> Hi,
> 
> On Wed, Nov 07 2012, kbuild test robot wrote:
> > tree:   git://git.kernel.org/pub/scm/linux/kernel/git/cjb/mmc.git mmc-next
> > head:   a6a567a4c25762e36228d2d0c8c1859085132e15
> > commit: 90ba86ed414e8ab4d84fe3342f10948c607efa7a [17/28] mmc:
> > sdhci-s3c: ensure non-transaction of bus before clk_disable
> > config: i386-randconfig-b646 (attached as .config)
> >
> > All error/warnings:
> >
> > drivers/mmc/host/sdhci-pltfm.o: In function `sdhci_runtime_pm_get':
> > sdhci-pltfm.c:(.text+0x80): multiple definition of `sdhci_runtime_pm_get'
> > drivers/mmc/host/sdhci.o:sdhci.c:(.text+0x7550): first defined here
> > drivers/mmc/host/sdhci-pltfm.o: In function `sdhci_runtime_pm_put':
> > sdhci-pltfm.c:(.text+0xa0): multiple definition of `sdhci_runtime_pm_put'
> > drivers/mmc/host/sdhci.o:sdhci.c:(.text+0x7570): first defined here
> 
> Yay for buildbots.  Seungwon, I've dropped this patch from mmc-next now,
> please investigate and fix.
> 
> (Wu, are you sure that this is the correct config?  It's i386 without
> MMC_SDHCI_S3C=y, and MMC_SDHCI_S3C depends on PLAT_SAMSUNG/ARM.)

Sorry I don't understand that. MMC_SDHCI_S3C seems to not matter at all?

Looking at the patch, these inline functions in .h should be prefixed
with "static":

+inline int sdhci_runtime_pm_get(struct sdhci_host *host)
+{
+       return 0;
+}
+inline int sdhci_runtime_pm_put(struct sdhci_host *host)
+{
+       return 0;
+}

Thanks,
Fengguang