Anyone? I'll commit this soon if I don't hear back, I don't think this is contentious.
On Fri, Feb 07, 2020 at 03:59:50PM +0100, Florian Obser wrote: > Moving from misc to tech. > > This is effectively Ulrich's diff from github with a bit of whitespace > shuffling. > > OK? > > diff --git def.h def.h > index d4f00e84e59..0db023973e0 100644 > --- def.h > +++ def.h > @@ -337,7 +337,7 @@ void ttnowindow(void); > void ttcolor(int); > void ttresize(void); > > -volatile sig_atomic_t winch_flag; > +extern volatile sig_atomic_t winch_flag; > > /* ttyio.c */ > void ttopen(void); > @@ -752,11 +752,7 @@ extern char cinfo[]; > extern char *keystrings[]; > extern char pat[NPAT]; > extern char prompt[]; > - > -/* > - * Globals. > - */ > -int tceeol; > -int tcinsl; > -int tcdell; > -int rptcount; /* successive invocation count */ > +extern int tceeol; > +extern int tcinsl; > +extern int tcdell; > +extern int rptcount; /* successive invocation count */ > diff --git kbd.c kbd.c > index 06d6c9fcf48..5f9b0a9efa6 100644 > --- kbd.c > +++ kbd.c > @@ -26,13 +26,13 @@ char prompt[PROMPTL] = "", *promptp = prompt; > > static int mgwrap(PF, int, int); > > -static int use_metakey = TRUE; > -static int pushed = FALSE; > -static int pushedc; > +static int use_metakey = TRUE; > +static int pushed = FALSE; > +static int pushedc; > > struct map_element *ele; > - > -struct key key; > +struct key key; > +int rptcount; > > /* > * Toggle the value of use_metakey > diff --git tty.c tty.c > index 0b64c4b5453..c378cb240dd 100644 > --- tty.c > +++ tty.c > @@ -45,6 +45,11 @@ static const char *scroll_fwd; /* How to scroll > forward. */ > > static void winchhandler(int); > > +volatile sig_atomic_t winch_flag; > +int tceeol; > +int tcinsl; > +int tcdell; > + > /* ARGSUSED */ > static void > winchhandler(int sig) > > > On Tue, Feb 04, 2020 at 12:51:46AM +0100, Han Boetes wrote: > > The latest version of gcc is more picky about global variables resulting in > > this bug report for my portable version of mg: > > https://github.com/hboetes/mg/issues/12 > > > > To which Ulrich Müller created a pull request which fixed the problem: > > https://github.com/hboetes/mg/pull/13/files > > > > Is this worth applying to the upstream branch? > > -- > I'm not entirely sure you are real. > -- I'm not entirely sure you are real.
