qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Richard Henderson <richard.henderson@linaro.org>
To: qemu-devel@nongnu.org
Cc: peter.maydell@linaro.org, "Alex Bennée" <alex.bennee@linaro.org>,
	"David Hildenbrand" <david@redhat.com>
Subject: [PULL 10/46] softfloat: Add float_cmask and constants
Date: Sun, 16 May 2021 07:33:55 -0500	[thread overview]
Message-ID: <20210516123431.718318-11-richard.henderson@linaro.org> (raw)
In-Reply-To: <20210516123431.718318-1-richard.henderson@linaro.org>

Testing more than one class at a time is better done with masks.
This reduces the static branch count.

Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
Reviewed-by: David Hildenbrand <david@redhat.com>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
---
 fpu/softfloat.c | 30 +++++++++++++++++++++++-------
 1 file changed, 23 insertions(+), 7 deletions(-)

diff --git a/fpu/softfloat.c b/fpu/softfloat.c
index ee4b5073b6..64edb23793 100644
--- a/fpu/softfloat.c
+++ b/fpu/softfloat.c
@@ -469,6 +469,20 @@ typedef enum __attribute__ ((__packed__)) {
     float_class_snan,
 } FloatClass;
 
+#define float_cmask(bit)  (1u << (bit))
+
+enum {
+    float_cmask_zero    = float_cmask(float_class_zero),
+    float_cmask_normal  = float_cmask(float_class_normal),
+    float_cmask_inf     = float_cmask(float_class_inf),
+    float_cmask_qnan    = float_cmask(float_class_qnan),
+    float_cmask_snan    = float_cmask(float_class_snan),
+
+    float_cmask_infzero = float_cmask_zero | float_cmask_inf,
+    float_cmask_anynan  = float_cmask_qnan | float_cmask_snan,
+};
+
+
 /* Simple helpers for checking if, or what kind of, NaN we have */
 static inline __attribute__((unused)) bool is_nan(FloatClass c)
 {
@@ -1338,26 +1352,28 @@ bfloat16 QEMU_FLATTEN bfloat16_mul(bfloat16 a, bfloat16 b, float_status *status)
 static FloatParts muladd_floats(FloatParts a, FloatParts b, FloatParts c,
                                 int flags, float_status *s)
 {
-    bool inf_zero = ((1 << a.cls) | (1 << b.cls)) ==
-                    ((1 << float_class_inf) | (1 << float_class_zero));
-    bool p_sign;
+    bool inf_zero, p_sign;
     bool sign_flip = flags & float_muladd_negate_result;
     FloatClass p_class;
     uint64_t hi, lo;
     int p_exp;
+    int ab_mask, abc_mask;
+
+    ab_mask = float_cmask(a.cls) | float_cmask(b.cls);
+    abc_mask = float_cmask(c.cls) | ab_mask;
+    inf_zero = ab_mask == float_cmask_infzero;
 
     /* It is implementation-defined whether the cases of (0,inf,qnan)
      * and (inf,0,qnan) raise InvalidOperation or not (and what QNaN
      * they return if they do), so we have to hand this information
      * off to the target-specific pick-a-NaN routine.
      */
-    if (is_nan(a.cls) || is_nan(b.cls) || is_nan(c.cls)) {
+    if (unlikely(abc_mask & float_cmask_anynan)) {
         return pick_nan_muladd(a, b, c, inf_zero, s);
     }
 
     if (inf_zero) {
         float_raise(float_flag_invalid, s);
-        s->float_exception_flags |= float_flag_invalid;
         return parts_default_nan(s);
     }
 
@@ -1371,9 +1387,9 @@ static FloatParts muladd_floats(FloatParts a, FloatParts b, FloatParts c,
         p_sign ^= 1;
     }
 
-    if (a.cls == float_class_inf || b.cls == float_class_inf) {
+    if (ab_mask & float_cmask_inf) {
         p_class = float_class_inf;
-    } else if (a.cls == float_class_zero || b.cls == float_class_zero) {
+    } else if (ab_mask & float_cmask_zero) {
         p_class = float_class_zero;
     } else {
         p_class = float_class_normal;
-- 
2.25.1



  parent reply	other threads:[~2021-05-16 12:45 UTC|newest]

Thread overview: 52+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-05-16 12:33 [PULL 00/46] softfloat: Convert float128 to FloatParts (part 1) Richard Henderson
2021-05-16 12:33 ` [PULL 01/46] qemu/host-utils: Use __builtin_bitreverseN Richard Henderson
2021-05-16 12:33 ` [PULL 02/46] qemu/host-utils: Add wrappers for overflow builtins Richard Henderson
2021-05-16 12:33 ` [PULL 03/46] qemu/host-utils: Add wrappers for carry builtins Richard Henderson
2021-05-16 12:33 ` [PULL 04/46] accel/tcg: Use add/sub overflow routines in tcg-runtime-gvec.c Richard Henderson
2021-05-16 12:33 ` [PULL 05/46] tests/fp: add quad support to the benchmark utility Richard Henderson
2021-05-16 12:33 ` [PULL 06/46] softfloat: Move the binary point to the msb Richard Henderson
2021-05-16 12:33 ` [PULL 07/46] softfloat: Inline float_raise Richard Henderson
2021-05-16 12:33 ` [PULL 08/46] softfloat: Use float_raise in more places Richard Henderson
2021-05-16 12:33 ` [PULL 09/46] softfloat: Tidy a * b + inf return Richard Henderson
2021-05-16 12:33 ` Richard Henderson [this message]
2021-05-16 12:33 ` [PULL 11/46] softfloat: Use return_nan in float_to_float Richard Henderson
2021-05-16 12:33 ` [PULL 12/46] softfloat: fix return_nan vs default_nan_mode Richard Henderson
2021-05-16 12:33 ` [PULL 13/46] target/mips: Set set_default_nan_mode with set_snan_bit_is_one Richard Henderson
2021-05-16 12:33 ` [PULL 14/46] softfloat: Do not produce a default_nan from parts_silence_nan Richard Henderson
2021-05-16 12:34 ` [PULL 15/46] softfloat: Rename FloatParts to FloatParts64 Richard Henderson
2021-05-16 12:34 ` [PULL 16/46] softfloat: Move type-specific pack/unpack routines Richard Henderson
2021-05-16 12:34 ` [PULL 17/46] softfloat: Use pointers with parts_default_nan Richard Henderson
2021-05-16 12:34 ` [PULL 18/46] softfloat: Use pointers with unpack_raw Richard Henderson
2021-05-16 12:34 ` [PULL 19/46] softfloat: Use pointers with ftype_unpack_raw Richard Henderson
2021-05-16 12:34 ` [PULL 20/46] softfloat: Use pointers with pack_raw Richard Henderson
2021-05-16 12:34 ` [PULL 21/46] softfloat: Use pointers with ftype_pack_raw Richard Henderson
2021-05-16 12:34 ` [PULL 22/46] softfloat: Use pointers with ftype_unpack_canonical Richard Henderson
2021-05-16 12:34 ` [PULL 23/46] softfloat: Use pointers with ftype_round_pack_canonical Richard Henderson
2021-05-16 12:34 ` [PULL 24/46] softfloat: Use pointers with parts_silence_nan Richard Henderson
2021-05-16 12:34 ` [PULL 25/46] softfloat: Rearrange FloatParts64 Richard Henderson
2021-05-16 12:34 ` [PULL 26/46] softfloat: Convert float128_silence_nan to parts Richard Henderson
2021-05-16 12:34 ` [PULL 27/46] softfloat: Convert float128_default_nan " Richard Henderson
2021-05-16 12:34 ` [PULL 28/46] softfloat: Move return_nan to softfloat-parts.c.inc Richard Henderson
2021-05-16 12:34 ` [PULL 29/46] softfloat: Move pick_nan " Richard Henderson
2021-05-16 12:34 ` [PULL 30/46] softfloat: Move pick_nan_muladd " Richard Henderson
2021-05-16 12:34 ` [PULL 31/46] softfloat: Move sf_canonicalize " Richard Henderson
2021-05-16 12:34 ` [PULL 32/46] softfloat: Move round_canonical " Richard Henderson
2021-05-16 12:34 ` [PULL 33/46] softfloat: Use uadd64_carry, usub64_borrow in softfloat-macros.h Richard Henderson
2021-05-16 12:34 ` [PULL 34/46] softfloat: Move addsub_floats to softfloat-parts.c.inc Richard Henderson
2021-05-16 12:34 ` [PULL 35/46] softfloat: Implement float128_add/sub via parts Richard Henderson
2021-05-16 12:34 ` [PULL 36/46] softfloat: Move mul_floats to softfloat-parts.c.inc Richard Henderson
2021-05-16 12:34 ` [PULL 37/46] softfloat: Move muladd_floats " Richard Henderson
2021-05-16 12:34 ` [PULL 38/46] softfloat: Use mulu64 for mul64To128 Richard Henderson
2021-05-16 12:34 ` [PULL 39/46] softfloat: Use add192 in mul128To256 Richard Henderson
2021-05-16 12:34 ` [PULL 40/46] softfloat: Tidy mul128By64To192 Richard Henderson
2021-05-16 12:34 ` [PULL 41/46] softfloat: Introduce sh[lr]_double primitives Richard Henderson
2021-05-16 12:34 ` [PULL 42/46] softfloat: Move div_floats to softfloat-parts.c.inc Richard Henderson
2021-05-20 13:40   ` Peter Maydell
2021-05-20 16:15     ` Richard Henderson
2021-05-20 17:04       ` Peter Maydell
2021-05-16 12:34 ` [PULL 43/46] softfloat: Split float_to_float Richard Henderson
2021-05-16 12:34 ` [PULL 44/46] softfloat: Convert float-to-float conversions with float128 Richard Henderson
2021-05-16 12:34 ` [PULL 45/46] softfloat: Move round_to_int to softfloat-parts.c.inc Richard Henderson
2021-05-16 12:34 ` [PULL 46/46] softfloat: Move round_to_int_and_pack " Richard Henderson
2021-05-16 13:19 ` [PULL 00/46] softfloat: Convert float128 to FloatParts (part 1) no-reply
2021-05-18 10:11 ` Peter Maydell

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=20210516123431.718318-11-richard.henderson@linaro.org \
    --to=richard.henderson@linaro.org \
    --cc=alex.bennee@linaro.org \
    --cc=david@redhat.com \
    --cc=peter.maydell@linaro.org \
    --cc=qemu-devel@nongnu.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).