Backwards?

[Conrad Scott]

Yesterday the following fragment of code from tty.cc confused me
because it used small_print rather than system_printf (speedily fixed
by Chris).  Now I'm confused because I don't understand the logic:

  if (wincap.has_security ()
       && cygserver_running == CYGSERVER_OK
       && (SetKernelObjectSecurity (hMainProc,
            ACL_SECURITY_INFORMATION,
            get_null_sd ()) == FALSE))
    small_printf ("Can't set process security, %E");

The call to SetKernelObjectSecurity was in the file before the
cygserver changes were added, i.e. the code used to be (before the
first cygwin_daemon merge):

  if ((iswinnt) &&
      (SetKernelObjectSecurity (hMainProc,
               DACL_SECURITY_INFORMATION,
			               get_null_sd ()) == FALSE))
    small_printf ("Can't set process security, %E");

On that basis, shouldn't the test for cygserver be reversed:

    if (... && cygserver != CYGSERVER_OK && ...)

i.e. if cygserver isn't running, act as before?

I don't understand quite this code is trying to achieve or why but,
assuming it's wrong, I've attached a patch to reverse the test. I've
checked this on the cygwin_daemon branch, both with and without
cygserver running, and can see no difference (this is with both
processes running as the same user tho').

If someone could confirm / deny / explain this or even just wave their
hands around a bit and waffle, it would make me happier :-)

// Conrad

-------------- next part --------------
A non-text attachment was scrubbed...
Name: tty.patch
Type: application/octet-stream
Size: 667 bytes
Desc: not available
URL: <http://cygwin.com/pipermail/cygwin-patches/attachments/20020703/8a37f287/attachment.obj>
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: ChangeLog.txt
URL: <http://cygwin.com/pipermail/cygwin-patches/attachments/20020703/8a37f287/attachment.txt>