Path: csiph.com!x330-a1.tempe.blueboxinc.net!newsfeed.hal-mli.net!feeder1.hal-mli.net!border3.nntp.dca.giganews.com!border1.nntp.dca.giganews.com!nntp.giganews.com!nx02.iad01.newshosting.com!newshosting.com!news.glorb.com!dotsrc.org!filter.dotsrc.org!news.dotsrc.org!not-for-mail
From: koehler@eisner.nospam.encompasserve.org (Bob Koehler)
Newsgroups: alt.sys.pdp11,comp.sys.dec,vmsnet.pdp-11
Subject: Re: Y3K for PDP-11 Operating Systems
Message-ID: <8muWrc1fNvZf@eisner.encompasserve.org>
Date: 26 May 2011 08:01:49 -0500
References: <4d99c3b6$0$23756$14726298@news.sunsite.dk>   <93vbcaF3luU1@mid.individual.net> <ireog6$ksn$1@Iltempo.Update.UU.SE> <940bm5Fcc3U1@mid.individual.net> <ireupe$mqj$1@Iltempo.Update.UU.SE> <941ob1Fre1U1@mid.individual.net> <irgfp7$382$1@Iltempo.Update.UU.SE> <9425gcFhnpU1@mid.individual.net> <irgsnm$7ov$1@Iltempo.Update.UU.SE> <942hghFhukU1@mid.individual.net> <irilvv$p6t$1@Iltempo.Update.UU.SE> <944mhrF19qU1@mid.individual.net> <mddk4de2s77.fsf@panix5.panix.com> <irka34$sqr$1@dont-email.me>
Organization: Encompasserve
Lines: 33
NNTP-Posting-Host: 67.53.90.116
X-Trace: news.sunsite.dk DXC=M<A=FI_P]LW07<M:5AA`SVYSB=nbEKnk[j;0@n>kLbiZT1og<J9`okR\J]h`co[Te[?b]2B[Je5dSP_U0^8jMJZQCRW2dGYa4<Pm?l@NAI6W@\
X-Complaints-To: staff@sunsite.dk
Xref: x330-a1.tempe.blueboxinc.net comp.sys.dec:503

In article <irka34$sqr$1@dont-email.me>, glen herrmannsfeldt <gah@ugcs.caltech.edu> writes:
> 
> It has been a while since I thought about this.  I did used to read
> EBCDIC tapes from OS/360 on TOPS-10, with my own conversion program
> written in Fortran.  I don't remember how the blocking came out now,
> but the favorite format (RECFM) at the time was fixed length 80
> byte records, blocked at something like 3200 bytes.
> DCB=(RECFM=FB,LRECL=80,BLKSIZE=3200)

   Once upon a time 80 byte EBCDIC records (one per block) on an 800
   BPI tape was the all-purpose media for data exchange.  A VAX 11/780
   running VMS 1.x was the first system I ran into that didn't ship with
   an EBCDIC tape reading progream.  The next 11/780 I ran into went
   into a shop with 3 IBM 360, so the local DEC office gave us a
   program that would read and write EBCDIC tapes.
   
> 
> I also remember doing direct access files in Fortran, I believe
> with records that were multiples of the disk block size.

   I've done a lot of direct access in Fortran, and never had to worry
   about the disk block size.  I was suprized to discover that the
   IBM Fortran H Extended Enhanced I/O subsystem would do direct access
   on a veriable-length record file, but not a fixed length record file.
   Perhaps IBM figured they only had to do the harder part.

   We had an IBM Macro routine that would do direct access on a fixed
   length record file.  It had a variable calling sequence.  When we
   ported the code to our VAX, I wrote a Fortran subroutine to do the
   direct access file I/O and a Macro-32 routine to accept the variable
   length argument list and fill in the arguments to the Fortran
   subroutine.