summaryrefslogtreecommitdiff
path: root/src/locale/newlocale.c
diff options
context:
space:
mode:
authorSzabolcs Nagy <nsz@port70.net>2020-06-12 17:34:28 +0000
committerRich Felker <dalias@aerifal.cx>2020-08-05 23:05:36 -0400
commitb1756ec8848623b5ec5ca8f6705832323176e0cb (patch)
treefd2f2cc8b3511b17a9559a148d3fa17bbdd02b89 /src/locale/newlocale.c
parent97e9b73d59b65d445f2ba0b6294605eac1d72ecb (diff)
downloadmusl-b1756ec8848623b5ec5ca8f6705832323176e0cb.tar.gz
musl-b1756ec8848623b5ec5ca8f6705832323176e0cb.tar.bz2
musl-b1756ec8848623b5ec5ca8f6705832323176e0cb.tar.xz
musl-b1756ec8848623b5ec5ca8f6705832323176e0cb.zip
math: new software sqrtf
same method as in sqrt, this was tested on all inputs against an sqrtf instruction. (the only difference found was that x86 sqrtf does not signal the x86 specific input-denormal exception on negative subnormal inputs while the software sqrtf does, this is fine as it was designed for ieee754 exceptions only.) there is known faster method: "Computing Floating-Point Square Roots via Bivariate Polynomial Evaluation" that computes sqrtf directly via pipelined polynomial evaluation which allows more parallelism, but the design does not generalize easily to higher precisions.
Diffstat (limited to 'src/locale/newlocale.c')
0 files changed, 0 insertions, 0 deletions