i2c: designware: fix poll-after-enable regression

commit 06cb616b1bca7080824acfedb3d4c898e7a64836 upstream.

Not all revisions of DW I2C controller implement the enable status register.
On platforms where that's the case (e.g. BG2CD and SPEAr ARM SoCs), waiting
for enable will time out as reading the unimplemented register yields zero.

It was observed that reading the IC_ENABLE_STATUS register once suffices to
avoid getting it stuck on Bay Trail hardware, so replace polling with one
dummy read of the register.

Fixes: fba4adbbf670 ("i2c: designware: must wait for enable")
Signed-off-by: Alexander Monakov <amonakov@ispras.ru>
Tested-by: Ben Gardner <gardner.ben@gmail.com>
Acked-by: Jarkko Nikula <jarkko.nikula@linux.intel.com>
Signed-off-by: Wolfram Sang <wsa@the-dreams.de>
Cc: stable@kernel.org
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

diff --git a/drivers/i2c/busses/i2c-designware-core.c b/drivers/i2c/busses/i2c-designware-core.c
index 340e037b3224ffb87e4e9004794c35c32fa759d0..884c1ec61ac9e9b53c48354aa0397c7568dc5310 100644 (file)
--- a/drivers/i2c/busses/i2c-designware-core.c
+++ b/drivers/i2c/busses/i2c-designware-core.c
@@ -507,7 +507,10 @@ static void i2c_dw_xfer_init(struct dw_i2c_dev *dev)
        i2c_dw_disable_int(dev);
 
        /* Enable the adapter */
-       __i2c_dw_enable_and_wait(dev, true);
+       __i2c_dw_enable(dev, true);
+
+       /* Dummy read to avoid the register getting stuck on Bay Trail */
+       dw_readl(dev, DW_IC_ENABLE_STATUS);
 
        /* Clear and enable interrupts */
        dw_readl(dev, DW_IC_CLR_INTR);