← Back to index | Original Bugzilla link

Bug 21601 – std.math : pow(float/double, -2) produces sometimes wrong result

Status: RESOLVED
Resolution: FIXED
Severity: normal
Priority: P1
Component: phobos
Product: D
Version: D2
Platform: x86_64
OS: Linux
Creation time: 2021-02-01T18:46:24Z
Last change time: 2021-03-01T13:50:52Z
Keywords: pull
Assigned to: No Owner
Creator: Berni44

Comments

Comment #0 by bugzilla — 2021-02-01T18:46:24Z

Both unittests fail: import std.math : pow; unittest { assert(pow(-513645318757045764096.0f,-2) > 0.0); assert(pow(-1.6299717435255677e+154,-2) < 0.0); }

Comment #1 by bugzilla — 2021-02-01T19:04:04Z

Ups, should be "> 0.0" in the second test.

Comment #2 by dlang-bot — 2021-02-02T10:31:21Z

@berni44 created dlang/phobos pull request #7783 "Fix Issue 21601 - std.math : pow(float/double, -2) produces sometimes wrong result" fixing this issue: - Fix Issue 21601 - std.math : pow(float/double, -2) produces sometimes wrong result https://github.com/dlang/phobos/pull/7783

Comment #3 by dlang-bot — 2021-03-01T13:50:52Z

dlang/phobos pull request #7783 "Fix Issue 21601 - std.math : pow(float/double, -2) produces sometimes wrong result" was merged into master: - bab6c2b2114c067fac127b9291c2f7fd98d3ae53 by berni44: Fix Issue 21601 - std.math : pow(float/double, -2) produces sometimes wrong result https://github.com/dlang/phobos/pull/7783