cvs commit: src/lib/libc_r/uthread uthread_write.c

[Nate Lawson]

On Wed, 17 Dec 2003, Daniel Eischen wrote:
> On Wed, 17 Dec 2003, Nate Lawson wrote:
>
> > On Wed, 17 Dec 2003, Daniel Eischen wrote:
> > >   Modified files:
> > >     lib/libc_r/uthread   uthread_write.c
> > >   Log:
> > >   Return to the caller if write() returns 0.
> > >
> > >   PR:     59291
> > >
> > >   Revision  Changes    Path
> > >   1.23      +2 -0      src/lib/libc_r/uthread/uthread_write.c
> > >
> > > --- src/lib/libc_r/uthread/uthread_write.c:1.22	Mon Sep 29 06:41:26 2003
> > > +++ src/lib/libc_r/uthread/uthread_write.c	Wed Dec 17 08:44:03 2003
> > >  #include <sys/types.h>
> > > @@ -138,6 +138,8 @@
> > >  					ret = num;
> > >  				else
> > >  					ret = n;
> > > +				if (n == 0)
> > > +					break;
> > >
> > >  			/* Check if the write has completed: */
> > >  			} else if (num >= nbytes)
> >
> > Thanks!  Oh what a difference one retval makes.
>
> Does this fix something other than bacula?

Anyone else who expected this semantics.  I have no specific examples.

-Nate