On Sat, Feb 22, 2025 at 11:24:26AM +0100, Willy Tarreau wrote: > On Mon, Feb 03, 2025 at 10:05:14AM +0100, Thomas Weißschuh wrote: > > nolibc does not provide this header, instead its definitions are > > available unconditionally. > > Please think about reminding which one you're talking about so that a > simple "git log" shows what header you're talking about (limits.h) > without requiring to also see the patch itself. Ack. > BTW, I think that limits.h is common enough that we could probably > provide it as well with nolibc to ease porting (and the current patch > is a good example of this). Maybe it could simply start by including > stdint.h to provide the various limits we rely on. I remember that in > the early days of nolibc-test we had to exclude it as well for nolibc. > > What do you think? The less we need to patch programs to insert #ifndef > NOLIBC, the better. Sounds good, I'll do that for v2. Given that any nolibc header always also includes the global nolibc.h, I think limits.h can directly include nolibc.h. > Cheers, > Willy
