On Thu, Feb 02, 2012 at 04:55:23PM +0800, Shawn Guo wrote: > On Thu, Feb 02, 2012 at 10:12:05AM +0800, Richard Zhao wrote: > > Signed-off-by: Richard Zhao <richard.z...@linaro.org> > > --- > > arch/arm/mach-imx/mach-mx31ads.c | 10 ++++++++++ > > arch/arm/plat-mxc/include/mach/mx31.h | 1 + > > 2 files changed, 11 insertions(+), 0 deletions(-) > > > Hmm, let's see who are actually using mxc_audmux_v2_configure_port(). > > $ git grep -n mxc_audmux_v2_configure_port arch/arm/ > arch/arm/mach-imx/eukrea_mbimxsd25-baseboard.c:246: > mxc_audmux_v2_configure_port(0, > arch/arm/mach-imx/eukrea_mbimxsd25-baseboard.c:254: > mxc_audmux_v2_configure_port(4, > arch/arm/mach-imx/eukrea_mbimxsd35-baseboard.c:257: > mxc_audmux_v2_configure_port(0, > arch/arm/mach-imx/eukrea_mbimxsd35-baseboard.c:265: > mxc_audmux_v2_configure_port(3, > arch/arm/mach-imx/mach-pcm043.c:365: mxc_audmux_v2_configure_port(3, > arch/arm/mach-imx/mach-pcm043.c:371: mxc_audmux_v2_configure_port(0, > As we are moving audmux into sound/soc/imx, it makes less sense to still keep these calls in board files. Instead, I prefer to call it from machine driver like what wm1133-ev1 does below. Or we can simply make the it a audmux-self call with 3 parameters it needs retrieved from platform_data or device tree, so that machine driver does not even bother with the call. Makes sense?
Regards, Shawn > $ git grep -n mxc_audmux_v2_configure_port sound/soc/imx/ > sound/soc/imx/wm1133-ev1.c:277: > mxc_audmux_v2_configure_port(MX31_AUDMUX_PORT1_SSI0, ptcr, pdcr); > sound/soc/imx/wm1133-ev1.c:281: > mxc_audmux_v2_configure_port(MX31_AUDMUX_PORT5_SSI_PINS_5, ptcr, pdcr); > > I guess audmux device needs to be added for all these users. And for > sake of bisect, it should be added as part of patch #5. > _______________________________________________ devicetree-discuss mailing list devicetree-discuss@lists.ozlabs.org https://lists.ozlabs.org/listinfo/devicetree-discuss