On Wed, 2006-08-02 at 21:15 +0200, Michael Hanselmann wrote: > +config SENSORS_AMS > + tristate "Apple Motion Sensor driver" > + depends on HWMON && PPC_PMAC && INPUT && EXPERIMENTAL This causes the build failure, because now SENSORS_AMS can be 'y' while > +config SENSORS_AMS_I2C > + bool "I2C variant" > + depends on SENSORS_AMS && I2C is satisfied by I2C=m. I think. SENSORS_AMS should probably have depends on HWMON && PPC_PMAC && INPUT && (ADB_PMU || I2C) && EXPERIMENTAL Is it possible to check further constraints? This still doesn't solve all problems since I2C=m and ADB_PMU=y would satisfy SENSORS_AMS=y, but that'd still lead to a build failure with SENSORS_AMS_I2C=y. I'm not sure if that's possible to solve at all. But since ADB_PMU is practically required even on those machines where we have the I2C version of the sensor (they'll boot IIRC, but most things won't work without it), how about just depending on that anyway? Hence making the dependency depends on HWMON && PPC_PMAC && INPUT && ADB_PMU && I2C && EXPERIMENTAL Also, "default y" should go away, and be put into the ppc32_defconfig. We don't ever need this on ppc64. The dependency on ADB_PMU would of course stop people from building it on ppc64 :) johannes -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 828 bytes Desc: This is a digitally signed message part Url : http://lists.lm-sensors.org/pipermail/lm-sensors/attachments/20060806/29aaade5/attachment.bin
