From 1c9afd69051a64cf085c6fb3674a444ff9a43857 Mon Sep 17 00:00:00 2001 From: Rich Felker Date: Thu, 6 Feb 2020 11:34:54 -0500 Subject: fix excess precision in return value of i386 atan[2][f] for functions implemented in C, this is a requirement of C11 (F.6); strictly speaking that text does not apply to standard library functions, but it seems to be intended to apply to them, and C2x is expected to make it a requirement. failure to drop excess precision is particularly bad for inverse trig functions, where a value with excess precision can be outside the range of the function (entire range, or range for a particular subdomain), breaking reasonable invariants a caller may expect. --- src/math/i386/atan.s | 2 ++ 1 file changed, 2 insertions(+) (limited to 'src/math/i386/atan.s') diff --git a/src/math/i386/atan.s b/src/math/i386/atan.s index a26feae1..2c57f6b3 100644 --- a/src/math/i386/atan.s +++ b/src/math/i386/atan.s @@ -8,6 +8,8 @@ atan: jb 1f fld1 fpatan + fstpl 4(%esp) + fldl 4(%esp) ret # subnormal x, return x with underflow 1: fsts 4(%esp) -- cgit v1.2.1