On 03-02-23, 16:18, Richard Fitzgerald wrote:
On 03/02/2023 14:35, Pierre-Louis Bossart wrote:
On 2/2/23 09:42, Richard Fitzgerald wrote:
Define the ch_mask argument of sdw_ch_mask_to_ch() as an unsigned so that the shift right is guaranteed to eventually make the value of ch_mask==0.
Previously ch_mask was defined as a signed int, but a right shift of a signed value preserves the sign bit. So if the sign bit was 1, ch_mask would never become 0 and the for loop would be infinite. Signed-off-by: Richard Fitzgerald rf@opensource.cirrus.com
drivers/soundwire/bus.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/soundwire/bus.h b/drivers/soundwire/bus.h index 7631ef5e71fb..28bedc919b78 100644 --- a/drivers/soundwire/bus.h +++ b/drivers/soundwire/bus.h @@ -160,7 +160,7 @@ int sdw_fill_msg(struct sdw_msg *msg, struct sdw_slave *slave, u32 addr, size_t count, u16 dev_num, u8 flags, u8 *buf); /* Retrieve and return channel count from channel mask */ -static inline int sdw_ch_mask_to_ch(int ch_mask) +static inline int sdw_ch_mask_to_ch(unsigned int ch_mask) { int c = 0;
This patch1 is fine, but you remove this function in patch2, so is this patch needed at all?
-/* Retrieve and return channel count from channel mask */ -static inline int sdw_ch_mask_to_ch(unsigned int ch_mask) -{
- int c = 0;
- for (c = 0; ch_mask; ch_mask >>= 1)
c += ch_mask & 1;- return c;
-}
I'm happy to squash them, I did it in two steps so it didn't get overlooked that there's a bugfix happening.
I think this case is fine to squash and send as a single patch while documenting that we are fixing the bug in the log
Thanks