On Wed, Feb 04, 2009 at 04:13:33PM -0200, Werner Almesberger wrote: > Hi Balaji, > > in commit 9d1fd9e7b19d66a35ee4e760b30a3b9a168c20a2, you introduced > a major change in s3cmci.c: you make it explicitly use the SD_3V3 > regulator and give this precedence over the platform set_power > callback. > > Unfortunately, there are a number of problems with this: > > - s3cmci.c shouldn't have to know about a platform's power > arrangements. That's why we have set_power. >
In mach-gta02.c, when we define SD_3V3, we also have to specify which device is the consumer - which means the supply namespace is not global and it depends on the device and not the platform. So, I think it's reasonable to assume that any regulator that supplies it will be named SD_3V3 ? > - In GTA02, SD_3V3 is actually used for the MMC interface of the > Glamo, so s3cmci.c is (unsuccessfully) trying to grab somebody > else's regulator. > > - If the whole thing worked, it would take precedence over the > callback to gta02_s3c_mmc_set_power, thus disabling the WLAN > reset logic. > No, this cannot happen. SD_3V3 is owned by glamo_mci and s3c_sdi_dev cannot claim it. Thanks, Balaji
