i2c: Spelling fixes

[JD: One more fix in i2c-dev.]

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Jean Delvare <khali@linux-fr.org>
diff --git a/drivers/i2c/busses/i2c-at91.c b/drivers/i2c/busses/i2c-at91.c
index 9c8b6d5..c09b036 100644
--- a/drivers/i2c/busses/i2c-at91.c
+++ b/drivers/i2c/busses/i2c-at91.c
@@ -135,7 +135,7 @@
  * Generic i2c master transfer entrypoint.
  *
  * Note: We do not use Atmel's feature of storing the "internal device address".
- * Instead the "internal device address" has to be written using a seperate
+ * Instead the "internal device address" has to be written using a separate
  * i2c message.
  * http://lists.arm.linux.org.uk/pipermail/linux-arm-kernel/2004-September/024411.html
  */
diff --git a/drivers/i2c/busses/i2c-powermac.c b/drivers/i2c/busses/i2c-powermac.c
index 0ab4f26..7813127 100644
--- a/drivers/i2c/busses/i2c-powermac.c
+++ b/drivers/i2c/busses/i2c-powermac.c
@@ -94,7 +94,7 @@
 	    	break;
 
 	/* Note that these are broken vs. the expected smbus API where
-	 * on reads, the lenght is actually returned from the function,
+	 * on reads, the length is actually returned from the function,
 	 * but I think the current API makes no sense and I don't want
 	 * any driver that I haven't verified for correctness to go
 	 * anywhere near a pmac i2c bus anyway ...