[v5] fpu: Export some internals for targets

[PATCH v5 28/30] fpu: Introduce float_status.e4m3_nan_is_snan

Richard Henderson posted 30 patches 2 months, 3 weeks ago

Diff against v1 v2 v3 v4
Download mbox

Maintainers: Aurelien Jarno <aurelien@aurel32.net>, Peter Maydell <peter.maydell@linaro.org>, "Alex Bennée" <alex.bennee@linaro.org>, Richard Henderson <richard.henderson@linaro.org>, Brian Cain <brian.cain@oss.qualcomm.com>, Pierrick Bouvier <pierrick.bouvier@oss.qualcomm.com>, Helge Deller <deller@gmx.de>, Paolo Bonzini <pbonzini@redhat.com>, Zhao Liu <zhao1.liu@intel.com>, Laurent Vivier <laurent@vivier.eu>, "Philippe Mathieu-Daudé" <philmd@linaro.org>, Jiaxun Yang <jiaxun.yang@flygoat.com>, Aleksandar Rikalo <arikalo@gmail.com>, Nicholas Piggin <npiggin@gmail.com>, Chinmay Rath <rathc@linux.ibm.com>, Glenn Miles <milesg@linux.ibm.com>, Harsh Prateek Bora <harshpb@linux.ibm.com>, Palmer Dabbelt <palmer@dabbelt.com>, Alistair Francis <alistair.francis@wdc.com>, Weiwei Li <liwei1518@gmail.com>, Daniel Henrique Barboza <daniel.barboza@oss.qualcomm.com>, Liu Zhiwei <zhiwei_liu@linux.alibaba.com>, Chao Liu <chao.liu.zevorn@gmail.com>, Ilya Leoshkevich <iii@linux.ibm.com>, David Hildenbrand <david@kernel.org>, Cornelia Huck <cohuck@redhat.com>, Eric Farman <farman@linux.ibm.com>, Matthew Rosato <mjrosato@linux.ibm.com>, Yoshinori Sato <yoshinori.sato@nifty.com>, Max Filippov <jcmvbkbc@gmail.com>

Expand all Fold all

[PATCH v5 28/30] fpu: Introduce float_status.e4m3_nan_is_snan

Posted by Richard Henderson 2 months, 3 weeks ago

Introduce a separate control from float_snan_rule
that applies only to the OCP E4M3 format.

Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
---
 include/fpu/softfloat-helpers.h | 5 +++++
 include/fpu/softfloat-types.h   | 5 +++++
 fpu/softfloat-parts.c.inc       | 2 +-
 fpu/softfloat-specialize.c.inc  | 5 +++++
 4 files changed, 16 insertions(+), 1 deletion(-)

diff --git a/include/fpu/softfloat-helpers.h b/include/fpu/softfloat-helpers.h
index 4791a7fc87..8df0273132 100644
--- a/include/fpu/softfloat-helpers.h
+++ b/include/fpu/softfloat-helpers.h
@@ -131,6 +131,11 @@ static inline void set_snan_rule(FloatSNaNRule val, float_status *status)
     status->float_snan_rule = val;
 }
 
+static inline void set_float_e4m3_nan_is_snan(bool val, float_status *status)
+{
+    status->e4m3_nan_is_snan = val;
+}
+
 static inline void set_float_rebias_overflow(bool val, float_status *status)
 {
     status->rebias_overflow = val;
diff --git a/include/fpu/softfloat-types.h b/include/fpu/softfloat-types.h
index c5a6684763..d7f5303a6b 100644
--- a/include/fpu/softfloat-types.h
+++ b/include/fpu/softfloat-types.h
@@ -431,6 +431,11 @@ typedef struct float_status {
     /* do we detect and flush denormal results before or after rounding? */
     bool ftz_before_rounding : 1;
     FloatSNaNRule float_snan_rule : 2;
+    /*
+     * Overriding float_snan_rule, is the single NaN representation for
+     * the OCP E4M3 format an SNaN or QNaN?
+     */
+    bool e4m3_nan_is_snan : 1;
     Float2NaNPropRule float_2nan_prop_rule : 3;
     Float3NaNPropRule float_3nan_prop_rule : 7;
     FloatInfZeroNaNRule float_infzeronan_rule: 3;
diff --git a/fpu/softfloat-parts.c.inc b/fpu/softfloat-parts.c.inc
index 184614ea47..559e40d196 100644
--- a/fpu/softfloat-parts.c.inc
+++ b/fpu/softfloat-parts.c.inc
@@ -246,7 +246,7 @@ static void partsN(canonicalize)(FloatPartsN *p, float_status *status,
         case float_expmax_e4m3:
             if (p->frac_hi == 0b111) {
                 fracN(shl)(p, fmt->frac_shift);
-                p->cls = (parts_is_snan_frac(p->frac_hi, status)
+                p->cls = (get_float_e4m3_nan_is_snan(status)
                           ? float_class_snan : float_class_qnan);
                 return;
             }
diff --git a/fpu/softfloat-specialize.c.inc b/fpu/softfloat-specialize.c.inc
index b97c971ae6..8b765ef623 100644
--- a/fpu/softfloat-specialize.c.inc
+++ b/fpu/softfloat-specialize.c.inc
@@ -104,6 +104,11 @@ static inline bool get_float_rebias_underflow(const float_status *status)
     return status->rebias_underflow;
 }
 
+static inline bool get_float_e4m3_nan_is_snan(const float_status *status)
+{
+    return status->e4m3_nan_is_snan;
+}
+
 /*----------------------------------------------------------------------------
 | For the deconstructed floating-point with fraction FRAC, return true
 | if the fraction represents a signalling NaN; otherwise false.
-- 
2.43.0

Re: [PATCH v5 28/30] fpu: Introduce float_status.e4m3_nan_is_snan

Posted by Peter Maydell 2 months, 3 weeks ago

On Wed, 20 May 2026 at 18:22, Richard Henderson
<richard.henderson@linaro.org> wrote:
>
> Introduce a separate control from float_snan_rule
> that applies only to the OCP E4M3 format.
>
> Signed-off-by: Richard Henderson <richard.henderson@linaro.org>

Reviewed-by: Peter Maydell <peter.maydell@linaro.org>

thanks
-- PMM