Large errors found in the GNU libc using mpcheck

The mpcheck program is available here.

Note: for each function, only the largest known errors (in ulps) are given. The error is positive when the GNU libc result is larger than the correct result, and negative otherwise.

The details the goals for functions expected to be within a few ulps are in the GNU libc manual, section Errors in Math Functions.

Single precision

Known issues: gamma 2542, 2543, j0 14469, j1 14470, log1p 6792, sincos 15467, y0 14471, y1 14472, yn 2555, 2577

See also: 14473 (cpowf)

glibc versionfunctionargumentsrounding modeerror in ulpstest programbugzilla
2.18.90erf0x1.c5bf94p-127RNDZspurious underflowtest4.c16516
2.18.90exp2-0x1.f8002ep+6RNDUmissing underflowtest8.c16521
2.18.90sinh0x1.4p-146RNDNmissing underflowtest6.c16519
2.18.90tan0x1p-149RNDZmissing underflowtest5.c16517
2.18.90tanh0x1p-149RNDNmissing underflowtest7.c16520
2.18.90yn42, 0x1.0c2d26p+9RNDZ-62183test1.c

Double precision

Known issues: exp 16284, gamma 2558, log1p 6792, pow 16315, sincos 15467, yn 2567

glibc versionfunctionargumentrounding modeerror in ulpstest programbugzilla
2.18.90sincos0x1p-1074RNDNmissing underflow16526

Long Double precision

Known issues: atanh 16357, exp 16361, log1p 6792, sincos 15467

glibc versionfunctionargumentrounding modeerror in ulpstest programbugzilla
2.18.90sin0x1p-16445RNDZmissing underflow16538
2.18.90yn17, 0xd.563c0f9f9f83a9cp+6RNDU745252115test3.c16507

Note: Anything 10 ulps or more, in any rounding mode, should be reported to Bugzilla. See the meta-bug 14759.