Re: /cygdrive in CVS? Time for 1.5.6 soon.

[Nicholas Wourms <nwourms at netscape dot net>]

vinschen@redhat.com wrote:

> On Mon, Dec 01, 2003 at 01:23:47PM -0500, Nicholas Wourms wrote:
>
> > Actually, you might want check this against the original LTP source.  If 
> > you look at the original, I think it makes more sense (at least to me), 
> > since this logic was setting the l_type based on the size of a counter 
> > in a for loop.  Breifly, this is a portion of the difference as compared 
> > to our version 0f fcntl09:
> >
> > <<<diff -u fcntl09.c.ltp fcntl09.c.cygwin>>>
> >  for (lc=0; TESTLOOPING
> > -  int type;
> > -  for (type = 0, type < 2; type++){
> >
> >    Tst_count=0;
> >
> > -   flocks.l_type = type ? F_RDLCK : F_WRLCK;
> > +   flocks.l_type = F_RDLCK : F_WRLCK;
>
>
> Yes, that makes a lot more sense.  Interested to submit a patch?

Sure I'll cook one up shortly.  I'm also currently porting some of the 
LTP flock tests, so when I'm finished I'd like to submit them in one shot.

Forgive me for this, but...  while I was poking around in Cygwin, I 
wondered about something.  I know I'm gonna get slammed for asking this, 
but why not define the mkfifo() syscall as:

extern "C" int
mkfifo (const char *path, mode_t mode)
{
  return (mknod (path, (mode_t) (mode|S_IFIFO), 0));
}

It seems to work here on win2k, and certainly anything's better then 
being a stub that returns ENOSYS?  I'm sure it isn't close to being as 
robust as BSD's or Linux's mkfifo and it's a bit kludgy, but it's a 
start, isn't it?

Cheers,
Nicholas