gnu mp 4.2.4 uncovered gcc bug???

[Stoddard, Isaac A]

In preparation for building GCC 4, I am upgrading my local copies of gmp
and mpfr. Gcc 3.2.3 is all I have to use at the moment, as the system
default compiler version. Given that "make check" reports one issue, is
it still safe to ocntinue with the GCC build, using the old gcc (3.2.3)
or should an interim version with patches be installed to transition
across this bug?

"make check" reveals one failure:
mpf_cmp_si wrong on data[6]
  a=-0.1e1 (-1)
  b=1 (1)
  got=0
  want=-1
FAIL: t-cmp_si

    * The GMP version number, and if pre-packaged or patched then say
so.
4.2.4

    * A test program that makes it possible for us to reproduce the bug.
Include instructions on how to run the program.
tests/mpf/t-cmp_si.c in the distribution

    * A description of what is wrong. If the results are incorrect, in
what way. If you get a crash, say so.
1 of 26 tests failed
Please report to gmp-bugs@...

    * The configuration options you used when building GMP, if any.
./configure --prefix=$HOME --enable-assert --enable-cxx

    * The name of the compiler and its version. For gcc, get the version
with `gcc -v', otherwise perhaps `what `which cc`', or similar.
gcc -v
Reading specs from /usr/lib/gcc-lib/x86_64-redhat-linux/3.2.3/specs
Configured with: ../configure --prefix=/usr --mandir=/usr/share/man
--infodir=/usr/share/info --enable-shared --enable-threads=posix
--disable-checking --with-system-zlib --enable-__cxa_atexit
--host=x86_64-redhat-linux
Thread model: posix
gcc version 3.2.3 20030502 (Red Hat Linux 3.2.3-52)

    * The output from running `uname -a'.
uname -a
Linux ewslc1 2.4.21-32.ELsmp #1 SMP Fri Apr 15 21:03:28 EDT 2005 x86_64
GNU/Linux

    * The output from running `./config.guess', and from running
`./configfsf.guess' (might be the same).
./config.guess
athlon64-unknown-linux-gnu




Isaac Stoddard
Guidance, Navigation & Control System Integration,
Boeing International Space Station Program,
Tower 2, Mail Stop HM5-40, cube 5146,
13100 Space Center Blvd., Houston TX 77059
_______________________________________________
gmp-bugs mailing list
gmp-bugs@...
http://gmplib.org/mailman/listinfo/gmp-bugs

[Torbjorn Granlund-2]

"Stoddard, Isaac A" <isaac.a.stoddard@...> writes:

  In preparation for building GCC 4, I am upgrading my local copies of gmp
  and mpfr. Gcc 3.2.3 is all I have to use at the moment, as the system
  default compiler version. Given that "make check" reports one issue, is
  it still safe to ocntinue with the GCC build, using the old gcc (3.2.3)
  or should an interim version with patches be installed to transition
  across this bug?
 
I wouldn't take the risk, if I were you.

Perhaps disabling optimization when building gmp could avoid tripping
the compiler bug?  Try ".../configure CFLAGS="" && make", but you should
add any non-optimizing compiler options to CFLAGS that configure
originally chose.

--
Torbjörn
_______________________________________________
gmp-bugs mailing list
gmp-bugs@...
http://gmplib.org/mailman/listinfo/gmp-bugs