qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Peter Maydell <peter.maydell@linaro.org>
To: qemu-devel@nongnu.org
Cc: "Tom Musta" <tommusta@gmail.com>,
	"Peter Crosthwaite" <peter.crosthwaite@xilinx.com>,
	patches@linaro.org, "Michael Matz" <matz@suse.de>,
	"Alexander Graf" <agraf@suse.de>,
	"Claudio Fontana" <claudio.fontana@linaro.org>,
	"Dirk Mueller" <dmueller@suse.de>,
	"Will Newton" <will.newton@linaro.org>,
	"Laurent Desnogues" <laurent.desnogues@gmail.com>,
	"Alex Bennée" <alex.bennee@linaro.org>,
	kvmarm@lists.cs.columbia.edu,
	"Christoffer Dall" <christoffer.dall@linaro.org>,
	"Richard Henderson" <rth@twiddle.net>
Subject: [Qemu-devel] [PATCH v2 06/24] softfloat: Only raise Invalid when conversions to int are out of range
Date: Mon,  6 Jan 2014 13:11:03 +0000	[thread overview]
Message-ID: <1389013881-15726-7-git-send-email-peter.maydell@linaro.org> (raw)
In-Reply-To: <1389013881-15726-1-git-send-email-peter.maydell@linaro.org>

We implement a number of float-to-integer conversions using conversion
to an integer type with a wider range and then a check against the
narrower range we are actually converting to. If we find the result to
be out of range we correctly raise the Invalid exception, but we must
also suppress other exceptions which might have been raised by the
conversion function we called.

This won't throw away exceptions we should have preserved, because for
the 'core' exception flags the IEEE spec mandates that the only valid
combinations of exception that can be raised by a single operation are
Inexact + Overflow and Inexact + Underflow. For the non-IEEE softfloat
flag for input denormals, we can guarantee that that flag won't have
been set for out of range float-to-int conversions because a squashed
denormal by definition goes to plus or minus zero, which is always in
range after conversion to integer zero.

This bug has been fixed for some of the float-to-int conversion routines
by previous patches; fix it for the remaining functions as well, so
that they all restore the pre-conversion status flags prior to raising
Invalid.

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Aurelien Jarno <aurelien@aurel32.net>
---
 fpu/softfloat.c | 28 ++++++++++++++++------------
 1 file changed, 16 insertions(+), 12 deletions(-)

diff --git a/fpu/softfloat.c b/fpu/softfloat.c
index 2364513..6312e0c 100644
--- a/fpu/softfloat.c
+++ b/fpu/softfloat.c
@@ -6509,17 +6509,18 @@ uint32 float32_to_uint32( float32 a STATUS_PARAM )
 {
     int64_t v;
     uint32 res;
+    int old_exc_flags = get_float_exception_flags(status);
 
     v = float32_to_int64(a STATUS_VAR);
     if (v < 0) {
         res = 0;
-        float_raise( float_flag_invalid STATUS_VAR);
     } else 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;
 }
 
@@ -6527,17 +6528,18 @@ uint32 float32_to_uint32_round_to_zero( float32 a STATUS_PARAM )
 {
     int64_t v;
     uint32 res;
+    int old_exc_flags = get_float_exception_flags(status);
 
     v = float32_to_int64_round_to_zero(a STATUS_VAR);
     if (v < 0) {
         res = 0;
-        float_raise( float_flag_invalid STATUS_VAR);
     } else 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;
 }
 
@@ -6585,17 +6587,18 @@ uint_fast16_t float32_to_uint16_round_to_zero(float32 a STATUS_PARAM)
 {
     int64_t v;
     uint_fast16_t res;
+    int old_exc_flags = get_float_exception_flags(status);
 
     v = float32_to_int64_round_to_zero(a STATUS_VAR);
     if (v < 0) {
         res = 0;
-        float_raise( float_flag_invalid STATUS_VAR);
     } else if (v > 0xffff) {
         res = 0xffff;
-        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;
 }
 
@@ -6679,17 +6682,18 @@ uint_fast16_t float64_to_uint16_round_to_zero(float64 a STATUS_PARAM)
 {
     int64_t v;
     uint_fast16_t 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 > 0xffff) {
         res = 0xffff;
-        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;
 }
 
-- 
1.8.5

  parent reply	other threads:[~2014-01-06 13:11 UTC|newest]

Thread overview: 40+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-01-06 13:10 [Qemu-devel] [PATCH v2 00/24] A64 decoder patchset 6: rest of floating point Peter Maydell
2014-01-06 13:10 ` [Qemu-devel] [PATCH v2 01/24] softfloat: Fix exception flag handling for float32_to_float16() Peter Maydell
2014-01-06 18:18   ` Richard Henderson
2014-01-06 13:10 ` [Qemu-devel] [PATCH v2 02/24] softfloat: Add float to 16bit integer conversions Peter Maydell
2014-01-06 13:11 ` [Qemu-devel] [PATCH v2 03/24] softfloat: Add 16 bit integer to float conversions Peter Maydell
2014-01-06 18:19   ` Richard Henderson
2014-01-06 13:11 ` [Qemu-devel] [PATCH v2 04/24] softfloat: Make the int-to-float functions take exact-width types Peter Maydell
2014-01-06 18:20   ` Richard Henderson
2014-01-06 13:11 ` [Qemu-devel] [PATCH v2 05/24] softfloat: Fix float64_to_uint64 Peter Maydell
2014-01-06 18:23   ` Richard Henderson
2014-01-06 13:11 ` Peter Maydell [this message]
2014-01-06 18:24   ` [Qemu-devel] [PATCH v2 06/24] softfloat: Only raise Invalid when conversions to int are out of range Richard Henderson
2014-01-06 13:11 ` [Qemu-devel] [PATCH v2 07/24] softfloat: Fix factor 2 error for scalbn on denormal inputs Peter Maydell
2014-01-06 13:11 ` [Qemu-devel] [PATCH v2 08/24] softfloat: Add float32_to_uint64() Peter Maydell
2014-01-06 18:25   ` Richard Henderson
2014-01-06 13:11 ` [Qemu-devel] [PATCH v2 09/24] softfloat: Fix float64_to_uint64_round_to_zero Peter Maydell
2014-01-06 13:11 ` [Qemu-devel] [PATCH v2 10/24] softfloat: Fix float64_to_uint32 Peter Maydell
2014-01-06 13:11 ` [Qemu-devel] [PATCH v2 11/24] softfloat: Fix float64_to_uint32_round_to_zero Peter Maydell
2014-01-06 13:11 ` [Qemu-devel] [PATCH v2 12/24] softfloat: Provide complete set of accessors for fp state Peter Maydell
2014-01-06 13:11 ` [Qemu-devel] [PATCH v2 13/24] softfloat: Factor out RoundAndPackFloat16 and NormalizeFloat16Subnormal Peter Maydell
2014-01-06 18:28   ` Richard Henderson
2014-01-06 13:11 ` [Qemu-devel] [PATCH v2 14/24] softfloat: Add float16 <=> float64 conversion functions Peter Maydell
2014-01-06 18:30   ` Richard Henderson
2014-01-06 13:11 ` [Qemu-devel] [PATCH v2 15/24] softfloat: Refactor code handling various rounding modes Peter Maydell
2014-01-06 18:34   ` Richard Henderson
2014-01-06 13:11 ` [Qemu-devel] [PATCH v2 16/24] softfloat: Add support for ties-away rounding Peter Maydell
2014-01-06 18:36   ` Richard Henderson
2014-01-06 13:11 ` [Qemu-devel] [PATCH v2 17/24] target-arm: Prepare VFP_CONV_FIX helpers for A64 uses Peter Maydell
2014-01-06 18:37   ` Richard Henderson
2014-01-06 13:11 ` [Qemu-devel] [PATCH v2 18/24] target-arm: Rename A32 VFP conversion helpers Peter Maydell
2014-01-06 13:11 ` [Qemu-devel] [PATCH v2 19/24] target-arm: Ignore most exceptions from scalbn when doing fixpoint conversion Peter Maydell
2014-01-06 13:11 ` [Qemu-devel] [PATCH v2 20/24] target-arm: A64: Add extra VFP fixed point conversion helpers Peter Maydell
2014-01-06 13:11 ` [Qemu-devel] [PATCH v2 21/24] target-arm: A64: Add "Floating-point<->fixed-point" instructions Peter Maydell
2014-01-06 18:44   ` Richard Henderson
2014-01-06 13:11 ` [Qemu-devel] [PATCH v2 22/24] target-arm: A64: Add floating-point<->integer conversion instructions Peter Maydell
2014-01-06 18:47   ` Richard Henderson
2014-01-06 13:11 ` [Qemu-devel] [PATCH v2 23/24] target-arm: A64: Add 1-source 32-to-32 and 64-to-64 FP instructions Peter Maydell
2014-01-06 18:52   ` Richard Henderson
2014-01-06 13:11 ` [Qemu-devel] [PATCH v2 24/24] target-arm: A64: Add support for FCVT between half, single and double Peter Maydell
2014-01-06 18:55   ` Richard Henderson

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=1389013881-15726-7-git-send-email-peter.maydell@linaro.org \
    --to=peter.maydell@linaro.org \
    --cc=agraf@suse.de \
    --cc=alex.bennee@linaro.org \
    --cc=christoffer.dall@linaro.org \
    --cc=claudio.fontana@linaro.org \
    --cc=dmueller@suse.de \
    --cc=kvmarm@lists.cs.columbia.edu \
    --cc=laurent.desnogues@gmail.com \
    --cc=matz@suse.de \
    --cc=patches@linaro.org \
    --cc=peter.crosthwaite@xilinx.com \
    --cc=qemu-devel@nongnu.org \
    --cc=rth@twiddle.net \
    --cc=tommusta@gmail.com \
    --cc=will.newton@linaro.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).