From owner-svn-src-all@FreeBSD.ORG Sat Mar 20 22:32:14 2010 Return-Path: Delivered-To: svn-src-all@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 50322106566C; Sat, 20 Mar 2010 22:32:14 +0000 (UTC) (envelope-from marius@FreeBSD.org) Received: from svn.freebsd.org (svn.freebsd.org [IPv6:2001:4f8:fff6::2c]) by mx1.freebsd.org (Postfix) with ESMTP id 33C598FC0C; Sat, 20 Mar 2010 22:32:14 +0000 (UTC) Received: from svn.freebsd.org (localhost [127.0.0.1]) by svn.freebsd.org (8.14.3/8.14.3) with ESMTP id o2KMWD7I007312; Sat, 20 Mar 2010 22:32:13 GMT (envelope-from marius@svn.freebsd.org) Received: (from marius@localhost) by svn.freebsd.org (8.14.3/8.14.3/Submit) id o2KMWDeE007310; Sat, 20 Mar 2010 22:32:13 GMT (envelope-from marius@svn.freebsd.org) Message-Id: <201003202232.o2KMWDeE007310@svn.freebsd.org> From: Marius Strobl Date: Sat, 20 Mar 2010 22:32:13 +0000 (UTC) To: src-committers@freebsd.org, svn-src-all@freebsd.org, svn-src-head@freebsd.org X-SVN-Group: head MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Cc: Subject: svn commit: r205397 - head/lib/libc/sparc64/fpu X-BeenThere: svn-src-all@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: "SVN commit messages for the entire src tree \(except for " user" and " projects" \)" List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sat, 20 Mar 2010 22:32:14 -0000 Author: marius Date: Sat Mar 20 22:32:13 2010 New Revision: 205397 URL: http://svn.freebsd.org/changeset/base/205397 Log: - While SPARC V9 allows tininess to be detected either before or after rounding (impl. dep. #55), the SPARC JPS1 responsible for SPARC64 and UltraSPARC processors defines that in all cases tinyness is detected before rounding, therefore rounding up to the smallest normalised number should set the underflow flag. - If an infinite result is rounded down, the result should have an exponent 1 less than the value for infinity. PR: 144900 Submitted by: Peter Jeremy MFC after: 3 days Modified: head/lib/libc/sparc64/fpu/fpu_implode.c Modified: head/lib/libc/sparc64/fpu/fpu_implode.c ============================================================================== --- head/lib/libc/sparc64/fpu/fpu_implode.c Sat Mar 20 22:12:15 2010 (r205396) +++ head/lib/libc/sparc64/fpu/fpu_implode.c Sat Mar 20 22:32:13 2010 (r205397) @@ -329,8 +329,9 @@ __fpu_ftos(fe, fp) * right to introduce leading zeroes. Rounding then acts * differently for normals and subnormals: the largest subnormal * may round to the smallest normal (1.0 x 2^minexp), or may - * remain subnormal. In the latter case, signal an underflow - * if the result was inexact or if underflow traps are enabled. + * remain subnormal. A number that is subnormal before rounding + * will signal an underflow if the result is inexact or if underflow + * traps are enabled. * * Rounding a normal, on the other hand, always produces another * normal (although either way the result might be too big for @@ -345,8 +346,10 @@ __fpu_ftos(fe, fp) if ((exp = fp->fp_exp + SNG_EXP_BIAS) <= 0) { /* subnormal */ /* -NG for g,r; -SNG_FRACBITS-exp for fraction */ (void) __fpu_shr(fp, FP_NMANT - FP_NG - SNG_FRACBITS - exp); - if (fpround(fe, fp) && fp->fp_mant[3] == SNG_EXP(1)) + if (fpround(fe, fp) && fp->fp_mant[3] == SNG_EXP(1)) { + fe->fe_cx |= FSR_UF; return (sign | SNG_EXP(1) | 0); + } if ((fe->fe_cx & FSR_NX) || (fe->fe_fsr & (FSR_UF << FSR_TEM_SHIFT))) fe->fe_cx |= FSR_UF; @@ -407,6 +410,7 @@ zero: res[1] = 0; if ((exp = fp->fp_exp + DBL_EXP_BIAS) <= 0) { (void) __fpu_shr(fp, FP_NMANT - FP_NG - DBL_FRACBITS - exp); if (fpround(fe, fp) && fp->fp_mant[2] == DBL_EXP(1)) { + fe->fe_cx |= FSR_UF; res[1] = 0; return (sign | DBL_EXP(1) | 0); } @@ -426,7 +430,7 @@ zero: res[1] = 0; return (sign | DBL_EXP(DBL_EXP_INFNAN) | 0); } res[1] = ~0; - return (sign | DBL_EXP(DBL_EXP_INFNAN) | DBL_MASK); + return (sign | DBL_EXP(DBL_EXP_INFNAN - 1) | DBL_MASK); } done: res[1] = fp->fp_mant[3]; @@ -468,6 +472,7 @@ zero: res[1] = res[2] = res[3] = 0; if ((exp = fp->fp_exp + EXT_EXP_BIAS) <= 0) { (void) __fpu_shr(fp, FP_NMANT - FP_NG - EXT_FRACBITS - exp); if (fpround(fe, fp) && fp->fp_mant[0] == EXT_EXP(1)) { + fe->fe_cx |= FSR_UF; res[1] = res[2] = res[3] = 0; return (sign | EXT_EXP(1) | 0); } @@ -487,7 +492,7 @@ zero: res[1] = res[2] = res[3] = 0; return (sign | EXT_EXP(EXT_EXP_INFNAN) | 0); } res[1] = res[2] = res[3] = ~0; - return (sign | EXT_EXP(EXT_EXP_INFNAN) | EXT_MASK); + return (sign | EXT_EXP(EXT_EXP_INFNAN - 1) | EXT_MASK); } done: res[1] = fp->fp_mant[1];