MSYS build - stage 2 problem

[Simon Marlow]

> >   $ ../../ghc/utils/hsc2hs/hsc2hs-inplace -Iinclude -Icbits/regex
> > System/CPUTime.hsc -v
> 
> $ ../../ghc/utils/hsc2hs/hsc2hs-inplace -Iinclude 
> -Icbits/regex System/CPUTime.
> hsc -v
> Executing: c:/MinGW/bin/gcc -c -mno-cygwin 
> -D__GLASGOW_HASKELL__=602 -Ic:/ghc/6.2.1/ghc/includes 
> -Iinclude -Icbits/regex System\CPUTime_hsc_make.c -o 
> System\CPUTime_hsc_make.o
> Executing: c:/MinGW/bin/gcc System\CPUTime_hsc_make.o -o 
> System\CPUTime_hsc_make.exe
> 
> Fail: System\CPUTime_hsc_make.o: removeFile: does not exist 
> (No such file or directory)

Ok, so it looks like when hsc2hs runs these commands they don't work,
but when run by hand they do work.  So that indicates that there is a
problem with System.Cmd.system (which hsc2hs uses) on Win98.

We can't switch hsch2s over to rawSystem, easily at least, because it
uses redirection in one place which requires a shell.  Furthermore older
versions of GHC didn't have a working rawSystem on Win98.

I guess we need to find out what's wrong with System.Cmd.system on Win98
and work around it.  Debugging this remotely is going to take ages - it
would be a great help if you could play around with System.Cmd.system a
bit and try to find out what the problem is.  A start would be to make a
small program that tries to invoke c:/mingw/bin/gcc using
System.Cmd.system, compile and run it and see what happens.

Cheers,
	Simon