glibc 2.2: math test failures

[Michael Deutschmann]

(using list since GNATS is still not available.  It might be -accepting-
reports, but I'm uncomfortable submitting blind.)

On i486-linux-pc-gnu,  I've got "make check" failures on test-ldouble.  
The .out file is appended.

This is using Linux-2.0.39 ('final' pre-patch), gcc-2.95.2 and 
binutils-2.10.1, attempting to move from glibc 2.1.3.

There was one irregularity in my build procedure -- I forgot to use a
seperate build directory this time.  But I don't see how that could cause
this.

The CPU is (just in case it's an FPU problem):
processor       : 0
cpu             : 486
model           : 486 DX/4
vendor_id       : GenuineIntel
stepping        : 3
fdiv_bug        : no
hlt_bug         : no
f00f_bug        : no
fpu             : yes
fpu_exception   : yes
cpuid           : yes
wp              : yes
flags           : fpu vme pse
bogomips        : 49.66

---- Michael Deutschmann <address@hidden>

testing long double (without inline functions)
Failure: Test: log (e) == 1
Result:
 is:          1.00000000000000000011e+00   0x8.00000000000000100000p-3
 should be:   1.00000000000000000000e+00   0x8.00000000000000000000p-3
 difference:  1.08420217248550443401e-19   0x8.00000000000000000000p-66
 ulp       :  1.0000
 max.ulp   :  0.0000
Failure: Test: log (1.0 / M_El) == -1
Result:
 is:         -1.00000000000000000011e+00  -0x8.00000000000000100000p-3
 should be:  -1.00000000000000000000e+00  -0x8.00000000000000000000p-3
 difference:  1.08420217248550443401e-19   0x8.00000000000000000000p-66
 ulp       :  1.0000
 max.ulp   :  0.0000
Failure: Test: log1p (M_El - 1.0) == 1
Result:
 is:          1.00000000000000000011e+00   0x8.00000000000000100000p-3
 should be:   1.00000000000000000000e+00   0x8.00000000000000000000p-3
 difference:  1.08420217248550443401e-19   0x8.00000000000000000000p-66
 ulp       :  1.0000
 max.ulp   :  0.0000
Failure: Test: Imaginary part of: cpow (e + 0 i, 0 + 2 * M_PIl i) == 1.0 + 0.0 i
Result:
 is:          9.75781955236953990607e-19   0x9.00000000000000000000p-63
 should be:   0.00000000000000000000e+00   0x0.00000000000000000000p+0
 difference:  9.75781955236953990607e-19   0x9.00000000000000000000p-63
 ulp       :  9.0000
 max.ulp   :  1.0000
Failure: Test: Real part of: cpow (2 + 3 i, 4 + 0 i) == -119.0 - 120.0 i
Result:
 is:         -1.18999999999999999986e+02  -0xe.dfffffffffffffe00000p+3
 should be:  -1.19000000000000000000e+02  -0xe.e0000000000000000000p+3
 difference:  1.38777878078144567553e-17   0x8.00000000000000000000p-59
 ulp       :  2.0000
 max.ulp   :  0.0000
Failure: Test: Imaginary part of: cpow (2 + 3 i, 4 + 0 i) == -119.0 - 120.0 i
Result:
 is:         -1.19999999999999999965e+02  -0xe.ffffffffffffffb00000p+3
 should be:  -1.20000000000000000000e+02  -0xf.00000000000000000000p+3
 difference:  3.46944695195361418882e-17   0xa.00000000000000000000p-58
 ulp       :  5.0000
 max.ulp   :  4.0000
Maximal error of real part of: cpow
 is      :  2.0000 ulp
 accepted:  0.0000 ulp
Maximal error of imaginary part of: cpow
 is      :  9.0000 ulp
 accepted:  4.0000 ulp

Test suite completed:
  2271 test cases plus 2084 tests for exception flags executed.
  7 errors occured.