From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:38535) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1VyqeM-0006zW-As for qemu-devel@nongnu.org; Thu, 02 Jan 2014 17:22:10 -0500 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1VyqeD-0004ij-LY for qemu-devel@nongnu.org; Thu, 02 Jan 2014 17:21:58 -0500 From: Tom Musta Date: Thu, 2 Jan 2014 16:21:18 -0600 Message-Id: <1388701295-29855-6-git-send-email-tommusta@gmail.com> In-Reply-To: <1388701295-29855-1-git-send-email-tommusta@gmail.com> References: <1388701295-29855-1-git-send-email-tommusta@gmail.com> Subject: [Qemu-devel] [V5 PATCH 05/22] softfloat: Fix float64_to_uint32_round_to_zero List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: qemu-devel@nongnu.org Cc: Tom Musta , qemu-ppc@nongnu.org The float64_to_uint32_round_to_zero routine is incorrect. For example, the following test pattern: 425F81378DC0CD1F / 0x1.f81378dc0cd1fp+38 will erroneously set the inexact flag. This patch re-implements the routine to use the float64_to_uint64_round_to_zero routine. If saturation occurs the invalid exception flag is raised (but not the inexact flag). This contribution can be licensed under either the softfloat-2a or -2b license. Signed-off-by: Tom Musta Reviewed-by: Peter Maydell --- V4: Correct commit commentary. Corrected code to properly handle the stickiness of the inexact flag. fpu/softfloat.c | 15 +++++++-------- 1 files changed, 7 insertions(+), 8 deletions(-) diff --git a/fpu/softfloat.c b/fpu/softfloat.c index 640cd71..2c598ca 100644 --- a/fpu/softfloat.c +++ b/fpu/softfloat.c @@ -6596,19 +6596,18 @@ uint32 float64_to_uint32( float64 a STATUS_PARAM ) uint32 float64_to_uint32_round_to_zero( float64 a STATUS_PARAM ) { - int64_t v; + uint64_t v; uint32 res; + int old_exc_flags = get_float_exception_flags(status); - v = float64_to_int64_round_to_zero(a STATUS_VAR); - if (v < 0) { - res = 0; - float_raise( float_flag_invalid STATUS_VAR); - } else if (v > 0xffffffff) { + v = float64_to_uint64_round_to_zero(a STATUS_VAR); + if (v > 0xffffffff) { res = 0xffffffff; - float_raise( float_flag_invalid STATUS_VAR); } else { - res = v; + return v; } + set_float_exception_flags(old_exc_flags, status); + float_raise(float_flag_invalid STATUS_VAR); return res; } -- 1.7.1