paul@nessus.UUCP (05/21/87)
I have a question about compress -- running version 4.0 r1.1 86/10/21
Help,
I'm trying to get compress to run on my AT under SVR2. The problem is
that I am stuck with the same problem Xenix people have -- the damned
compiler won't create array's > 65536 bytes (thank's intel ..grrr).
I tried compiling with the following flags:
BITS=16
M_XENIX (breaks array's into little pieces)
Doing a compress/decompress with bits <= 12 seems to work ok. Uncompressing
a file (bits=16) that was compressed on another machine barfs. I ran
compress on my vax (under VMS) and it works beatifully.
Q1: Does it >really< work under Xenix? If so, why would SysV break it?
Q2: If it doesn't work under Xenix (as is), has someone already fixed
it?
Q3: Simply: Does anyone have a working compress that runs under
Microport's SystemV/AT port?
Thanks very much.
--
{ucbvax|cepu|sdcsvax|pyramid}!ucsbcsl!nessus!paul Paul Traina
paul%nessus%sbphy.ucsb@lbl.arpa 375 Ellwood Beach Dr. #18
paul%nessus@sbitp.bitnet Goleta, California 93117
pst@ai.ai.mit.edu +1 805 968-1658tweten@ames-prandtl.arpa (Dave Tweten) (05/23/87)
From: nessus!paul@ames-nas.ARPA I'm trying to get compress to run on my AT under SVR2. I tried compiling with the following flags: BITS=16 M_XENIX (breaks array's into little pieces) Doing a compress/decompress with bits <= 12 seems to work ok. Uncompressing a file (bits=16) that was compressed on another machine barfs. Q2: If it doesn't work under Xenix (as is), has someone already fixed it? When I ported Compress 4.0 to run under MS-DOS with 16-bit codes, I found several casting problems which should have been a problem for Xenix, as well as for Microsoft C, version 3.0 (or 4.0). They are all fixed in the version of Compress which is available by anonymous FTP from the Info-IBMPC software library at c.isi.edu. I did the MS-DOS specific parts of my work within ifdefs, and confirmed that compress still compiled and worked under Berkeley and System V large machine compilers. If my port doesn't cure your problem, it may put you a few steps along toward a solution.
stuart@bms-at.UUCP (Stuart D. Gathman) (05/25/87)
Compress requires the 'huge' model to run on am 80286. Actually, it runs much faster using a 'mixed' model which requires inserting 'huge' keywords at appropriate points in the code. Anyone wanting a mixed model version for Xenix can get it from me. (It runs twice as fast as a 68010 version!) Although the original poster mentioned Xenix, he seemed to imply that he was using Microport. I am not sure whether Microport supports mixed model, or even huge model for that matter. -- Stuart D. Gathman <..!seismo!dgis!bms-at!stuart>
neighorn@qiclab.UUCP (05/27/87)
In article <68@nessus.UUCP> paul@nessus.UUCP writes:
:I have a question about compress -- running version 4.0 r1.1 86/10/21
:Q3: Simply: Does anyone have a working compress that runs under
: Microport's SystemV/AT port?
:Thanks very much.
:--
:{ucbvax|cepu|sdcsvax|pyramid}!ucsbcsl!nessus!paul Paul Traina
Yes, compress works under System V/AT. Give me the word, and I'll whisk a
copy of the sources and the necessary compiler commands needed to make it
work.
--
Steven C. Neighorn tektronix!{psu-cs,reed}!qiclab!neighorn
Portland Public Schools "Where we train young Star Fighters to defend the
(503) 249-2000 ext 337 frontier against Xur and the Ko-dan Armada"mike@cimcor.UUCP (05/28/87)
I have a version of compress that runs under System V/AT for those
that want a copy. Someone was nice enough to send it to me and I'll forward
it to those you send me a line.
-Mike
ihnp4!meccts!cimcor!mike