The current definition of hive_int8 is a naked char, without any sign specifier. This is incorrect on platforms such as arm, where char is unsigned. Fortunately nothing in the kernel actually uses a hive_int8 type, but in case it gets used later rather than removed, this makes it explicitly signed. Cc: Mauro Carvalho Chehab <mchehab@xxxxxxxxxx> Cc: Sakari Ailus <sakari.ailus@xxxxxxxxxxxxxxx> Cc: Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx> Signed-off-by: Jason A. Donenfeld <Jason@xxxxxxxxx> --- Greg - if you're going to take this for 6.1, that's fine with me. Otherwise, if it's for 6.2, I'll take this in my unsigned-char tree to keep all of these fixups together. -Jason drivers/staging/media/atomisp/pci/hive_types.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/staging/media/atomisp/pci/hive_types.h b/drivers/staging/media/atomisp/pci/hive_types.h index 4b8a679fb672..55d36931f079 100644 --- a/drivers/staging/media/atomisp/pci/hive_types.h +++ b/drivers/staging/media/atomisp/pci/hive_types.h @@ -42,7 +42,7 @@ typedef unsigned int hive_bool; #define hive_false 0 #define hive_true 1 -typedef char hive_int8; +typedef signed char hive_int8; typedef short hive_int16; typedef int hive_int32; typedef long long hive_int64; -- 2.38.1
