On Wed, Jun 25, 2014 at 10:55:31AM -0700, Doug Anderson wrote:

Sachin,

On Wed, Jun 25, 2014 at 3:02 AM, Sachin Kamat [off-list ref] wrote:

quoted

All Exynos5 platforms have HSI2C controllers and are needed by
various IPs connected to the boards based on these SoCs. Thus
select this by default for Exynos5 platforms.

Signed-off-by: Sachin Kamat <redacted>
Cc: Doug Anderson <redacted>
---
 drivers/i2c/busses/Kconfig |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/i2c/busses/Kconfig b/drivers/i2c/busses/Kconfig
index 9f7d5859cf65..c7918cffe790 100644
--- a/drivers/i2c/busses/Kconfig
+++ b/drivers/i2c/busses/Kconfig

@@ -465,9 +465,9 @@ config I2C_EG20T
 config I2C_EXYNOS5
        tristate "Exynos5 high-speed I2C driver"
        depends on ARCH_EXYNOS5 && OF
+       default y
        help
-         Say Y here to include support for high-speed I2C controller in the
-         Exynos5 based Samsung SoCs.
+         High-speed I2C controller on Exynos5 based Samsung SoCs.

 config I2C_GPIO
        tristate "GPIO-based bitbanging I2C"
--

1.7.9.5

This seems reasonable to me and I will also take the blame for
suggesting this.  It's hard to imagine running a real exynos5 system
without I2C.  One could argue that on an exynos5250 the high speed I2C
controller is not mandatory (since all the ports can be muxed to use
the old controller) but on newer exynos5 products you're expected to
have the main PMIC on one of the i2c ports.  An exynos system without
access to its PMIC will just sorta limp by.

...but I've added Arnd to this thread as he expressed some
reservations about this type of thing, so we'll see what he says.

Ping. Arnd? I have no strong opinion and would take this patch...

Reviewed-by: Doug Anderson <redacted>