From: Tulio Magno Quites Machado Filho Date: Fri, 12 Jan 2018 22:39:42 +0000 (-0200) Subject: powerpc: Fix llround spurious inexact on 32-bit POWER4 [BZ #22697] X-Git-Tag: upstream/2.30~1816 X-Git-Url: http://review.tizen.org/git/?a=commitdiff_plain;h=af1e6be4ba60067907ea9e19e0cd6897d194e9cf;p=platform%2Fupstream%2Fglibc.git powerpc: Fix llround spurious inexact on 32-bit POWER4 [BZ #22697] This issue is similar to BZ #19235, where spurious exceptions are created from adding 0.5 then converting to an integer. The solution is based on Joseph's fix for BZ #19235. [BZ #22697] * sysdeps/powerpc/powerpc32/power4/fpu/s_llround.S (__llround): Do not add 0.5 to integer or out-of-range arguments. --- diff --git a/ChangeLog b/ChangeLog index e9035fe..f33212f 100644 --- a/ChangeLog +++ b/ChangeLog @@ -1,3 +1,9 @@ +2018-01-12 Tulio Magno Quites Machado Filho + + [BZ #22697] + * sysdeps/powerpc/powerpc32/power4/fpu/s_llround.S (__llround): + Do not add 0.5 to integer or out-of-range arguments. + 2018-01-12 Dmitry V. Levin * po/bg.po: Update translations. diff --git a/sysdeps/powerpc/powerpc32/power4/fpu/s_llround.S b/sysdeps/powerpc/powerpc32/power4/fpu/s_llround.S index d0b00ac..5ad89a1 100644 --- a/sysdeps/powerpc/powerpc32/power4/fpu/s_llround.S +++ b/sysdeps/powerpc/powerpc32/power4/fpu/s_llround.S @@ -71,6 +71,11 @@ ENTRY (__llround) fcmpu cr1,fp1,fp12 /* x is negative? x < 0.0 */ blt- cr6,.Lretzero /* 0.5 > x < -0.5 so just return 0. */ bge- cr7,.Lnobias /* 2^52 > x < -2^52 just convert with no bias. */ + /* Test whether an integer to avoid spurious "inexact". */ + fadd fp3,fp2,fp9 + fsub fp3,fp3,fp9 + fcmpu cr5,fp2,fp3 + beq cr5,.Lnobias fadd fp3,fp2,fp10 /* |x|+=0.5 bias to prepare to round. */ bge cr1,.Lconvert /* x is positive so don't negate x. */ fnabs fp3,fp3 /* -(|x|+=0.5) */