On Fri, Sep 08, 2017 at 01:53:43PM +0300, Dan Carpenter wrote: > The background of this code is that we can either use the default > tables or load our own table with sysfs. The default tables are three > element arrays of struct tsl2x7x_lux. If we load the table with sysfs > then we can have as many as nine elements. Which ever way we do it, the > last element is always zeroed out. > > The most interesting part of this patch is in the > in_illuminance0_lux_table_show() function. We were using the wrong > limit, "TSL2X7X_MAX_LUX_TABLE_SIZE * 3", when it should have been just > "TSL2X7X_MAX_LUX_TABLE_SIZE". This creates a static checker warning > that we are going of of bounds. However, since the last element is > always zeroed out, that means we hit the break statement and the code > works correctly despite the wrong limit check. > > I made several related readability changes. The most notable that I > changed the MAX_DEFAULT_TABLE_BYTES define which was: > > #define MAX_DEFAULT_TABLE_BYTES (sizeof(int) * TSL2X7X_MAX_LUX_TABLE_SIZE) > > I renamed the define to TSL2X7X_DEFAULT_TABLE_BYTES because it's not the > max size, it's the only size. Also the size should really be expressed > as sizeof(struct tsl2x7x_lux) * 3. In other words, 12 * 3 instead of > 4 * 9. It's 36 bytes either way, so this doesn't change the behavior. > > Finally, I created the TSL2X7X_DEF_LUX_TABLE_SZ define instead of using > the magic number 3. I declared the default tables using that define > to hopefully signal to future programmers that if they want to use a > different size they have to update all the related code. > > Signed-off-by: Dan Carpenter <dan.carpenter@xxxxxxxxxx> Acked-by: Brian Masney <masneyb@xxxxxxxxxxxxx> Brian -- To unsubscribe from this list: send the line "unsubscribe kernel-janitors" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
