Before falling back to softfloat FMA, we do not restore the original
values of inputs A and C. Fix it.
This bug was caught by running gcc's testsuite on RISC-V qemu.
Note that this change gives a small perf increase for fp-bench:
Host: Intel(R) Core(TM) i7-4790K CPU @ 4.00GHz
Command: perf stat -r 3 taskset -c 0 ./fp-bench -o mulAdd -p $prec
- $prec = single:
- before:
101.71 MFlops
102.18 MFlops
100.96 MFlops
- after:
103.63 MFlops
103.05 MFlops
102.96 MFlops
- $prec = double:
- before:
173.10 MFlops
173.93 MFlops
172.11 MFlops
- after:
178.49 MFlops
178.88 MFlops
178.66 MFlops
Signed-off-by: Kito Cheng <kito.cheng@gmail.com>
Signed-off-by: Emilio G. Cota <cota@braap.org>
Message-Id: <
20190322204320.17777-1-cota@braap.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
}
ur.h = up.h + uc.h;
} else {
+ union_float32 ua_orig = ua;
+ union_float32 uc_orig = uc;
+
if (flags & float_muladd_negate_product) {
ua.h = -ua.h;
}
if (unlikely(f32_is_inf(ur))) {
s->float_exception_flags |= float_flag_overflow;
} else if (unlikely(fabsf(ur.h) <= FLT_MIN)) {
+ ua = ua_orig;
+ uc = uc_orig;
goto soft;
}
}
}
ur.h = up.h + uc.h;
} else {
+ union_float64 ua_orig = ua;
+ union_float64 uc_orig = uc;
+
if (flags & float_muladd_negate_product) {
ua.h = -ua.h;
}
if (unlikely(f64_is_inf(ur))) {
s->float_exception_flags |= float_flag_overflow;
} else if (unlikely(fabs(ur.h) <= FLT_MIN)) {
+ ua = ua_orig;
+ uc = uc_orig;
goto soft;
}
}