On Tue, May 18, 2021 at 10:23:30AM +0200, Takashi Iwai wrote: > On Tue, 18 May 2021 10:13:34 +0200, > Takashi Sakamoto wrote: > > > > On Tue, May 18, 2021 at 09:02:54AM +0200, Takashi Iwai wrote: > > > On Tue, 18 May 2021 04:43:17 +0200, > > > Takashi Sakamoto wrote: > > > > > > > > It brings some inconvenience in practice to use enumerated type for > > > > variable to which bitwise OR with enumerator constant is assigned. > > > > > > > > This commit replaces declarations of enumerated type with int type. > > > > > > Better to use unsigned int for bit flags. Otherwise the highest bit > > > becomes harder to use. > > > > I can't imagine such situation that the signed value causes issue. Would > > I request actual example with such issue? At least, the highest bit is > > still available as bit even if the value is negative by assigning > > 0x80000000... > > It's available in signed int, but this is inconvenient, e.g. if you > shift the bit. Maybe I forgot something else, too. > > You may still use signed int if you are sure that you'll never reach > to the highest number, but other than that, using unsigned for bit > flags is a *VERY* common practice in the kernel programming, so there > is no reason to ignore it. Ok. I just follow to the convention under the Linux kernel development. Thanks Takashi Sakamoto
