Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

If you're using DJ G++, could you read this?

3 views
Skip to first unread message

Ajay Shah

unread,
Sep 3, 1991, 9:15:26 AM9/3/91
to
I was able to get djg++ to work fine. The only price paid was
that I had to strip my config.sys bare of everything which made
my machine fast! Saying "gcc hello.c" now takes 45 seconds.
Do you have suggestions on how to do this better? While
retaining some of my old functionality?

The system is a noname 386 clone with 4Meg of RAM, 40Meg of disk,
Hercules card, running DOS 3.3.

My old config.sys was:

device=\tools\386.max\386max.sys
device=\tools\386.max\386load.sys PROG=\bin\sys\dmdrvr.bin
device=\tools\386.max\386load.sys PROG=\bin\sys\smartdrv.sys 1500 /a
device=\tools\386.max\386load.sys PROG=\bin\sys\ramdrive.sys 512 /a
device=\tools\386.max\386load.sys PROG=\bin\sys\mouse.sys /p3
files=30
BUFFERS=3
break=on
shell=c:\bin\sh.exe -L -0 c:/bin/sh.exe -R 0

where smartdrv.sys and ramdrive.sys are the only useful things I
found inside Windoze 3. dmdrvr.bin is because I'm still on DOS
3.3. The last line is because I boot into MKS kornshell.

The config.sys needed to allow g++ to work is:

device=\bin\sys\dmdrvr.bin
device=\bin\sys\mouse.sys /p3
files=30
BUFFERS=3
break=on
shell=c:\bin\sh.exe -L -0 c:/bin/sh.exe -R 0

where I've lost disk cacheing, the ramdisk used to speed up
MKS Toolkit, and all expanded memory. (I don't mind losing the
expanded memory.) How do I make gcc run faster?

--
_______________________________________________________________________________
Ajay Shah, (213)734-3930, ajay...@usc.edu
The more things change, the more they stay insane.
_______________________________________________________________________________

DJ Delorie

unread,
Sep 3, 1991, 10:04:01 AM9/3/91
to
In article <35...@usc.edu>, ajay...@alhena.usc.edu (Ajay Shah) writes:
> I was able to get djg++ to work fine. The only price paid was
> that I had to strip my config.sys bare of everything which made
> my machine fast! Saying "gcc hello.c" now takes 45 seconds.
> Do you have suggestions on how to do this better? While
> retaining some of my old functionality?

> device=\tools\386.max\386max.sys

> device=\bin\sys\dmdrvr.bin

djgpp goes fastest under the following conditions:

1. disk cache available, or run off RAM drive.

2. Sufficient EXTENDED memory available to hold all of the program
you're running (about 1Mb for gcc, 1.5Mb for gcc -O)

Patch level #4 does NOT run under 386MAX, qemm, or any other program
that puts the CPU into V86 mode (that's in the README). PL#5 will
support VCPI, so this won't be a problem any more.

If you don't have enough physical memory available, then go32 pages
your program in and out of disk. This slows down the program quite a
bit. Make sure any disk cache or ramdrive drivers leave enough
extended memory for your application.

My system is a 4Mb system, with 2Mb of cache and 1Mb left for gcc,
running with extended memory, and I get good times with gcc.

DJ
d...@ctron.com

Joao M.S. Lourenco

unread,
Sep 3, 1991, 12:21:02 PM9/3/91
to
In article <35...@usc.edu> ajay...@alhena.usc.edu (Ajay Shah) writes:
> I was able to get djg++ to work fine. The only price paid was
> that I had to strip my config.sys bare of everything which made
> my machine fast! Saying "gcc hello.c" now takes 45 seconds.
> Do you have suggestions on how to do this better? While
> retaining some of my old functionality?
> [description of old & new config.sys file]

Why do you have BUFFERS=3 ????
Put it like BUFFERS=20 and it will be faster...

--
---
Joao Lourenco BITNET/Internet: j...@fct.unl.pt
+---------------------------------+ UUCP: j...@unl.uucp
| Departamento de Informatica +----------------------------------+
| Universidade Nova de Lisboa 2825 Monte Caparica, PORTUGAL |
+--------------------------------------------------------------------+

Kevin D. Quitt

unread,
Sep 3, 1991, 8:30:43 PM9/3/91
to
In article <1991Sep3....@fct.unl.pt> j...@fct.unl.pt (Joao M.S. Lourenco) writes:
>Why do you have BUFFERS=3 ????
>Put it like BUFFERS=20 and it will be faster...

Not true if you've got a good disk cache utility running. In that case,
adding buffer increases the overhead, since there is a duplication of effort
in DOS and the disk cache.


--
_
Kevin D. Quitt srhqla!venus!kdq kdq%ve...@sr.com
3D systems, inc. 26081 Avenue Hall Valencia, CA 91355
VOICE (805) 295-5600 x430 FAX (805) 257-1200

96.37% of all statistics are made up.

Thomas Neumann

unread,
Sep 4, 1991, 1:51:06 AM9/4/91
to
ajay...@alhena.usc.edu (Ajay Shah) writes:

>My old config.sys was:

>device=\tools\386.max\386max.sys
>device=\tools\386.max\386load.sys PROG=\bin\sys\dmdrvr.bin
>device=\tools\386.max\386load.sys PROG=\bin\sys\smartdrv.sys 1500 /a
>device=\tools\386.max\386load.sys PROG=\bin\sys\ramdrive.sys 512 /a
>device=\tools\386.max\386load.sys PROG=\bin\sys\mouse.sys /p3
>files=30
>BUFFERS=3
>break=on
>shell=c:\bin\sh.exe -L -0 c:/bin/sh.exe -R 0

Hmm...the only culprit is 386max. DJ's DOS extender is not VCPI
conformant, and it expects the machine _not_ to be in V86 mode.
However, the same problems occure with other EMS managers such as
Qemm386.

>The config.sys needed to allow g++ to work is:

>device=\bin\sys\dmdrvr.bin
>device=\bin\sys\mouse.sys /p3
>files=30
>BUFFERS=3
>break=on
>shell=c:\bin\sh.exe -L -0 c:/bin/sh.exe -R 0

>where I've lost disk cacheing, the ramdisk used to speed up
>MKS Toolkit, and all expanded memory. (I don't mind losing the
>expanded memory.) How do I make gcc run faster?

First, BUFFERS=3 may have been a good idea as long as some
disk caching was included in your config.sys . Without a
disk cache 3 buffers much too few. I suggest you increase
your buffers (maybe something like 30 or so) or re-install
a disk cache. Since you have to disable your memory manager
386max you can't use EMS anymore, so you might want to install
a cache program that works with plain extended memory as well,
such as hyperdisk.
--
Thomas Neumann EMAIL: t...@smart.ruhr.de
Luchweg 54 FIDO: 2:241/5800.4
4630 Bochum, FRG VOICE: +49-234-283497

0 new messages