S.D.Mechveliani writes:
> I am trying to build  ghc-2.04  on  linux, i486 ...
> 
> using  ghc-0.29  installed earlier.  The questions arise.
> 

Hi, sorry about the delay..


> -------------------------------------------------------------------
> I downloaded   ghc-2.04-src-pl2.tar.gz,   unzipped-untarred it.
> 
> Then no *patches* have to be applied?

That's right, no patches beyond this have been made available.

> -------------------------------------------------------------------
> config.mk.in   contains   GhcProjectPatchLevel =0
> 
> is this right?

Not really, but it is not used anywhere, AFAIK.

> -------------------------------------------------------------------
> config.mk.in   contains    mkdependHS_1_2 = mkdependHS-1.2 
> 
> But where is defined  mkdependHS-1.2  ?
> -------------------------------------------------------------------

It's the 0.29 one (as Alex answered).

> Many lines in  config.mk.in   contain   @..@  values.  Say,
> 

The config.mk.in is a template from which the configure script will
substitute appropriate values between the @..@s. See the installation
guide for more details of the new Makefile setup.

> 
> -----------------------------------------------------------------
> I made a link copy tree    ..../myfptools-linux   (== $(TOP))
> Then commanded
>                 cd $(TOP);  ./configure;  make boot
> 
> `make boot'     reports
> 
>   info.c:69: sgtty.h: No such file or directory
>   In file included from /usr/include/pwd.h:30,
>                    from makeinfo.c:79:
> 
> among its messages. Will this do?

Not very nice, but I would ignore failures in literate/info-utils with
2,04, these are peripheral tools at best. (meanwhile, we'll dig up an
updated version of info.c and include it in the next release)

> 
> -----------------------------------------------------------
> Then commanded   make all
> 
> It works for several hours, reports many good messages, and also
> this one (for many modules):
>   Warning: InitTime not found in stats file
>   Warning: InitElapsed not found in stats file
> 

Pls ignore these.

> Then it comes
> 
> ghc/PrelBase.lhs:10:9: parse error on input: Segmentation fault 
>                                        caught, address = 75640072
> make[2]: *** [ghc/PrelBase.o] Error 1
> make[1]: *** [all] Error 2
> 

Ouch..could you send the output from compiling PrelBase.o, i.e., what
does the following report

 make ghc/PrelBase.o EXTRA_HC_OPTS='-v -dshow-rn-trace'

Thanks,

--Sigbjorn

Reply via email to