Message ID | alpine.DEB.2.20.1706021707100.29125@digraph.polyomino.org.uk |
---|---|
State | New |
Headers | show |
On Fri, Jun 2, 2017 at 1:07 PM, Joseph Myers <joseph@codesourcery.com> wrote: > [BZ #21543] > * signal/signal.h: Only include <bits/types/sigevent_t.h> and > <bits/sigevent-consts.h> if [__USE_POSIX199309], not if > [__USE_XOPEN_EXTENDED]. LGTM. Do you think you'll be able to fix signal.h namespace-wise all the way for 2.26? zw
On Fri, 2 Jun 2017, Zack Weinberg wrote: > On Fri, Jun 2, 2017 at 1:07 PM, Joseph Myers <joseph@codesourcery.com> wrote: > > [BZ #21543] > > * signal/signal.h: Only include <bits/types/sigevent_t.h> and > > <bits/sigevent-consts.h> if [__USE_POSIX199309], not if > > [__USE_XOPEN_EXTENDED]. > > LGTM. > > Do you think you'll be able to fix signal.h namespace-wise all the way > for 2.26? I don't know. There are two patches pending review <https://sourceware.org/ml/libc-alpha/2017-05/msg00744.html> <https://sourceware.org/ml/libc-alpha/2017-06/msg00042.html>, but the big part (for XPG42 and later) is the ucontext issues.
diff --git a/signal/signal.h b/signal/signal.h index 947873e..21d628d 100644 --- a/signal/signal.h +++ b/signal/signal.h @@ -55,8 +55,11 @@ typedef __uid_t uid_t; #if defined __USE_POSIX199309 || defined __USE_XOPEN_EXTENDED # include <bits/types/siginfo_t.h> -# include <bits/types/sigevent_t.h> # include <bits/siginfo-consts.h> +#endif + +#ifdef __USE_POSIX199309 +# include <bits/types/sigevent_t.h> # include <bits/sigevent-consts.h> #endif