On 06/28/13 20:54, Seiji Aguchi wrote: > >>> diff --git a/include/qemu/time.h b/include/qemu/time.h >>> new file mode 100644 >>> index 0000000..f70739b >>> --- /dev/null >>> +++ b/include/qemu/time.h >>> @@ -0,0 +1,11 @@ >>> +#ifndef TIME_H >>> +#define TIME_H >> >> I wonder if TIME_H is maybe a bit too nondescript and could conflict >> with other guards. >> >> The guards currently used in "include/qemu/*.h" files are inconsistent >> -- some use a QEMU_ prefix, some a __QEMU_ one, and others use none. >> >> Don't respin just because of this, but maybe it's something to consider. > > This should be discussed in other thread...
Ah, right, apologies for the ambiguity. I just meant that either QEMU_TIME_H or __QEMU_TIME_H would have a lower chance to clash with another guard than plain TIME_H -- "some" prefix would be nice, but I couldn't point out a specific one, because the current practice varies. I didn't mean to suggest that you should unify the prefixes across all guards! Anyway, even my QEMU_TIME_H / __QEMU_TIME_H suggestion is tangential; feel free to ignore it. Thanks Laszlo
