Updated bug report: gmp4.2.1 install on itanium2 running unix

Torbjorn Granlund tg-this-will-bounce-but-I-am-subscribed-to-the-list-honest at swox.com
Wed Nov 29 14:29:27 CET 2006


Paul Nanninga <paul.nanninga at anu.edu.au> writes:

  icc -c -DHAVE_CONFIG_H -I. -I. -I.. -D__GMP_WITHIN_GMP
  ...
  icc -c -DHAVE_CONFIG_H -I. -I. -I.. -D__GMP_WITHIN_GMP
  ...
  Output from gcc -v:

You're using icc, as it seems, not gcc.  Therefore, the gcc version is
not important.

     * If the bug is related to `configure', then the contents of config.log.
  
  config.log attached
  
Well, "appended" is the right term.

     * If the bug is related to an asm file not assembling, then the
  contents of config.m4 and the offending line or lines from the
  temporary mpn/tmp-<file>.s.
  
  config.m4 and tmp-add_n.s attached.
  
Please resend tmp-add_n.s as an *attachement* so that I can relate the
error messages to the line numbers in the file.

The file looks fine, BTW.

  I hope this is enough information for you.
  
No, but now I have a guess.  I suspect the Intel assembler used by icc
has a bug.  I've seen this before with that assembler.

I suppose we should have a feature test for this, and reject this
compiler.

But to come further, I need your exact tmp-add_n.s.

It is curious why gcc is not used, since you seem to have that in your
path, and since that is preferred (unless you explicityly tell
configure not to prefer it).

-- 
Torbjörn


More information about the gmp-bugs mailing list