[2415] in Release_Engineering
rkinit for 7.1
daemon@ATHENA.MIT.EDU (qjb@ATHENA.MIT.EDU)
Mon Jul 16 14:46:43 1990
From: qjb@ATHENA.MIT.EDU
Date: Mon, 16 Jul 90 14:46:24 -0400
To: rel-eng@ATHENA.MIT.EDU
Cc: jik@ATHENA.MIT.EDU, qjb@ATHENA.MIT.EDU
In response to some mail that I received from jik concerning my
rkinit patches, I have ansi-ified all of rkinit. Since every
source file was touched, I will not send patches but instead
suggest that you copy all of /mit/qjb/source/rkinit. I have
tested make clean, depend, all, and install on the RT, the VAX,
and the PMAX and tested the clients and servers on all three
machine types as well as on the various combinations (i.e.,
rkinit from RT to pmax, etc...) All programs work; all builds
are clean (no warnings).
Summary of changes:
These sources include the two patches that I have previously
sent to bugs:
1. rkinitd does a setuid before calling k_userok
2. rkinitd no longer loops on a client error (i.e., the client
sends the proper shutdown request to the server)
In addition, all functions are declared both ansi and old style
(conditionalized on __STDC__), rkinit.h includes ansi
prototypes, all #else and #endif have comments around lables if
they are present (i.e. #endif /* __STDC__ */ instead of #endif
__STDC__), all $Header lines have been changed to $Id lines, and
all rcsid variables are now #ifdefed SABER, lint, LOCORE, and
RCS_HDRS.
Please let me know what you decide and when you have recopied
the sources. I appologize for exceeding code cut by almost six
hours, but I wasn't asked for these changes until Sunday and I
was away for the weekend.
Jay
