This is the mail archive of the
gcc-bugs@gcc.gnu.org
mailing list for the GCC project.
Re: Compiler Error that cannot be reproduced.
- To: Alexandre Oliva <oliva at lsd dot ic dot unicamp dot br>, jazir <entropic at mpx dot com dot au>
- Subject: Re: Compiler Error that cannot be reproduced.
- From: Andris Pavenis <andris at stargate dot astr dot lu dot lv>
- Date: Wed, 12 Jan 2000 13:06:51 +0200
- Cc: gcc-bugs at gcc dot gnu dot org
- References: <387C40EA.34C912F3@mpx.com.au> <orr9fnmx26.fsf@garnize.lsd.ic.unicamp.br>
On Wed, 12 Jan 2000, Alexandre Oliva wrote:
> On Jan 12, 2000, jazir <entropic@mpx.com.au> wrote:
>
> > The next time I tried make the error didn't occur, so I'm not sure what to
> > do about it.
>
> This usually indicates a hardware problem, but since you're running on
> MS-Windows95, I wouldn't be surprised if it were just a random bug in
> the... ermh.... operating system?
>
gcc-2.95.2 built for DJGPP does not behave so badly under Win95. I have used
it rather much.
However there is one thing that may happen for big projects:
Wincrap9X does not reuse descriptors for DPMI clients unless one quits
all DPMI executables in current MS-DOS session. As result after make
runs about 2000 processes such resource leak does in works: we're
getting some crash (DJGPP build of make is also 32 bit DPMI executable).
Usually windows says something like "Your program executed illegal
operation", or I'm getting message about inability to allocate memory
in 1st Mb, but I guess sometimes one may get also internal compiler
error. If one reruns make or compiles that source from command line
such problem will not repeat.
Andris