This is the mail archive of the libc-alpha@sources.redhat.com mailing list for the glibc project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Gcc 3.4 and glibc problem


When I used CFLAGS="-march=i686 -DNDEBUG=1 -finline-limit=2000 -g -O3"
to build glibc for i686-linux with gcc 3.4, I got

# cat /export/build/gnu/glibc-3.4/build-i686-linux/math/test-ifloat.out
testing float (inline functions)
Failure: Test: jn (3, 10.0) == 0.0583793793051868123429354784103409563
Result:
 is:          5.83793856203556060791e-02   0x1.de3e7200000000000000p-5
 should be:   5.83793781697750091553e-02   0x1.de3e6e00000000000000p-5
 difference:  7.45058059692382812500e-09   0x1.00000000000000000000p-27
 ulp       :  2.0000
 max.ulp   :  1.0000

Test suite completed:
  2462 test cases plus 2251 tests for exception flags executed.
  1 errors occurred.
# cat /export/build/gnu/glibc-3.4/build-i686-linux/math/test-float.out
testing float (without inline functions)
Failure: Test: jn (3, 10.0) == 0.0583793793051868123429354784103409563
Result:
 is:          5.83793856203556060791e-02   0x1.de3e7200000000000000p-5
 should be:   5.83793781697750091553e-02   0x1.de3e6e00000000000000p-5
 difference:  7.45058059692382812500e-09   0x1.00000000000000000000p-27
 ulp       :  2.0000
 max.ulp   :  1.0000

Test suite completed:
  2524 test cases plus 2313 tests for exception flags executed.
  1 errors occurred.

If I use -O2, it is OK. Is that a gcc bug?


H.J.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]