The parsefloat routines -- which let us pass NaNs and infinities on to
strto(f|d|ld) -- come from NetBSD.
Also fix LP64's strtold to return a NaN, and fix all the architectures
to return quiet NaNs.
Also fix wcstof/wcstod/wcstold to use parsefloat so they support hex
floats.
Lots of new tests.
Bug: http://b/31101647
Change-Id: Id7d46ac2d8acb8770b5e8c445e87cfabfde6f111
This patch adds more tests for math functions to address coverage
issue of math functions discussed in:
https://android-review.googlesource.com/#/c/49653/https://android-review.googlesource.com/#/c/94780/
These are data sets used in regression tests for the Intel the math library (libm). They were collected over a long period of testing various libm implementations.
The data sets contain function specific data (special and corner cases such as +/-0, maximum/minimum normalized numbers, +/-infinity, QNaN/SNaN, maximum/minimum denormal numbers, arguments that would produce close to overflow/underflow results, known hard-to-round cases, etc), implementation specific data (arguments close to table look-up values for different polynomial approximations, worst cases for range reduction algorithms) and other data with interesting bit patterns.
The reference values are computed with Maple and were converted into hexadecimal format.
Change-Id: I7177c282937369eae98f25d02134e4fc3beadde8
Signed-off-by: Jingwei Zhang <jingwei.zhang@intel.com>
Signed-off-by: Mingwei Shi <mingwei.shi@intel.com>
At the moment our libm is only good enough for a 1 ulp bound on these tests,
but that's better than the 4 ulp bound you get from gtest by default.
I'm not really happy with the multiple structures and corresponding functions,
but at least they mean there's no duplication in the tests themselves, and it
should be easy enough for us to make further improvements in future.
Change-Id: I004e12970332e1d9531721361d6c34f908cfcecc
