egcs-980205 report (Linux/Alpha)

[Richard Henderson]

On Sun, Feb 15, 1998 at 01:20:22PM -0700, Michal Jaegermann wrote:
> Compilation was performed, eventually, with "./configure ; make"
> (prefix changed).  Initially this did not work due to a line in Makefile
> which says:
>         cd $(srcdir) && autoheader

This is simply a problem of timestamps on the auto-generated files.
Snapshots, particularly when patched up, are going to have this problem.

> It turned out that an existing /usr/bin/as, which is gas from
> binutils-2.8.1.0.1, is inadequate for compiling the current egcs.
> It will not accept emitted '.arch' assembler directive.

Yep.

> (A bunch of
> "macro requires $at register while noat in effect" could have been
> worked around by forcing an option '-m21164a' in an assembler).

Except that some folks were reporting "macro requires..." even
with -m21164a using binutils 2.7.0.2.  So to force people to
upgrade to unbroken binutils I turned on .arch generation.  It's
nicer for the assembly anyway.

> Resolved by replacing this assmbler by a version from ftp.cygnus.com.
> This assembler was configured with "--target=alpha-elf64-linux-gnu",
> which is a guess.  Seems to be lucky one.  My OS was not automatically
> recognized by 'configure' utility ...

Really?  I hadn't had that problem.  What _does_ config.guess report?

> gcc -D__KERNEL__ -I/usr/src/linux-2.0.33/include -Wall -Wstrict-prototypes \
> -g -O2 -pipe -mno-fp-regs  -c -o pci.o pci.c
> In file included from pci.c:13:
> gcc: Internal compiler error: program cc1 got fatal signal 6

Ouch.  Since this is 2.0.x, and those are typically heavily patched,
would you send me a preprocessed copy of pci.c?

> Closer scrutiny revealed that
> 'configure' decided to set 'integer' Fortran type to 'long int'
> (it was 'int' for  egcs-1.0.1).

I wonder if configure didn't get regenerated properly.  Cause that
should have been fixed by this patch

Mon Feb  2 11:08:49 1998  Richard Henderson  <rth@cygnus.com>

        * configure.in: Update F2C_INTEGER and F2C_LONGINT tests
        for the new placement in the hierarchy.

> Even after that 'configure' failed to detect a number of system
> utilities and libU77/u77-test.f program suffered from strange errors.

Odd.  Will you send me the config.log from the libf2c/libU77
directory?


r~