Path: utzoo!yunexus!telly!attcan!utgpu!jarvis.csri.toronto.edu!mailrus!tut.cis.ohio-state.edu!SAS1.DREA.DND.CA!Bruce.Skinner From: Bruce.Skinner@SAS1.DREA.DND.CA Newsgroups: gnu.emacs.vms Subject: RE: VMSFNS does not compile under VAXC030 Message-ID: Date: 19 Jul 89 17:11:36 GMT Article-I.D.: AENEAS.8907191512.AA25024 Sender: daemon@tut.cis.ohio-state.edu Reply-To: Bruce.Skinner@sas1.drea.dnd.ca Distribution: gnu Organization: GNUs Not Usenet Lines: 76 Anil Khullar writes -- =========================================================================== > > From: (ANIL KHULLAR) > Subject: VMSFNS does not compile under VAXC030 > To: vms-gnu-emacs@prep.ai.mit.edu > X-Original-To: vms-gnu-emacs@prep.ai.mit.edu > > Today I tried to compile GNU Emacs 18.52 under VMS 5.0-2 using the > new VAX C 3.0 and found that VMSFNS did not compile. It seems all > references to PRV$V_ are not declared. > > I would be obliged if someone in the net has any ideas what needs to > be changed. > > Thanks in advance > > Anil > =========================================================================== I ran into this problem a couple of days ago. Release 3 of VAXC changed some of the .H files. You will find the following on page 9 of sys$help:vaxc030.release_notes A few of the above files changed to a degree that programs that use those files may need to be modified: . . . o PRVDEF.H -- bit mask macros became bit field members of structs . . . As the above list shows, the most common change was that bit mask macros became bit field members of structs. Previously, bit field members of data structures were specified using macros that defined masks for accessing the bits. The programmer was responsible for using the masks in conjunction with logical AND, OR, NOT, and shift operations to manipulate the target bits. The header files have been changed to use bitfield members of structs to accomplish the same task. This allows simple assignment statements to be used to set or clear bits. For example: #include prvdef union prvdef privs; set() { privs.prv$vcmkrnl = 1; } I'm not a C programmer, but I think you can probably get away with using the old prvdef.H that you had with VAXC V2.4. Any comments. By the way where is Mukesh Prasad? He's the author and I'm sure can answer this one. regards++ Bruce S. P.S. I think the VMS port of GNU Emacs is really well done. My heartfelt thanks to those who toiled ... +-----------------------------------------+--------------------------------+ | Bruce S. Skinner | Bruce.Skinner@sas1.drea.dnd.ca | +-----------------------------------------+ | | Defence Research Establishment Atlantic | (902) 426-3100 (168) | | Dartmouth, Nova Scotia, CANADA, B2Y 3Z7 | (902) 228-2001 (home) | +-----------------------------------------+--------------------------------+