target/arm: Handle FPCR.AH in FRECPS and FRSQRTS vector insns

Handle the FPCR.AH "don't negate the sign of a NaN" semantics
in the vector versions of FRECPS and FRSQRTS, by implementing
new vector wrappers that call the _ah_ scalar helpers.

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
This commit is contained in:
Peter Maydell 2025-02-01 16:39:43 +00:00
parent 28048a3d18
commit fdf89638dc
4 changed files with 44 additions and 6 deletions

View File

@ -980,6 +980,20 @@ DEF_HELPER_FLAGS_5(gvec_rsqrts_s, TCG_CALL_NO_RWG,
DEF_HELPER_FLAGS_5(gvec_rsqrts_d, TCG_CALL_NO_RWG, DEF_HELPER_FLAGS_5(gvec_rsqrts_d, TCG_CALL_NO_RWG,
void, ptr, ptr, ptr, fpst, i32) void, ptr, ptr, ptr, fpst, i32)
DEF_HELPER_FLAGS_5(gvec_ah_recps_h, TCG_CALL_NO_RWG,
void, ptr, ptr, ptr, fpst, i32)
DEF_HELPER_FLAGS_5(gvec_ah_recps_s, TCG_CALL_NO_RWG,
void, ptr, ptr, ptr, fpst, i32)
DEF_HELPER_FLAGS_5(gvec_ah_recps_d, TCG_CALL_NO_RWG,
void, ptr, ptr, ptr, fpst, i32)
DEF_HELPER_FLAGS_5(gvec_ah_rsqrts_h, TCG_CALL_NO_RWG,
void, ptr, ptr, ptr, fpst, i32)
DEF_HELPER_FLAGS_5(gvec_ah_rsqrts_s, TCG_CALL_NO_RWG,
void, ptr, ptr, ptr, fpst, i32)
DEF_HELPER_FLAGS_5(gvec_ah_rsqrts_d, TCG_CALL_NO_RWG,
void, ptr, ptr, ptr, fpst, i32)
DEF_HELPER_FLAGS_5(gvec_ah_fmax_h, TCG_CALL_NO_RWG, DEF_HELPER_FLAGS_5(gvec_ah_fmax_h, TCG_CALL_NO_RWG,
void, ptr, ptr, ptr, fpst, i32) void, ptr, ptr, ptr, fpst, i32)
DEF_HELPER_FLAGS_5(gvec_ah_fmax_s, TCG_CALL_NO_RWG, DEF_HELPER_FLAGS_5(gvec_ah_fmax_s, TCG_CALL_NO_RWG,

View File

@ -5767,10 +5767,11 @@ static bool do_fp3_vector_2fn(DisasContext *s, arg_qrrr_e *a, int data,
return do_fp3_vector(s, a, data, s->fpcr_ah ? fah : fnormal); return do_fp3_vector(s, a, data, s->fpcr_ah ? fah : fnormal);
} }
static bool do_fp3_vector_ah(DisasContext *s, arg_qrrr_e *a, int data, static bool do_fp3_vector_ah_2fn(DisasContext *s, arg_qrrr_e *a, int data,
gen_helper_gvec_3_ptr * const f[3]) gen_helper_gvec_3_ptr * const fnormal[3],
gen_helper_gvec_3_ptr * const fah[3])
{ {
return do_fp3_vector_with_fpsttype(s, a, data, f, return do_fp3_vector_with_fpsttype(s, a, data, s->fpcr_ah ? fah : fnormal,
select_ah_fpst(s, a->esz)); select_ah_fpst(s, a->esz));
} }
@ -5913,14 +5914,24 @@ static gen_helper_gvec_3_ptr * const f_vector_frecps[3] = {
gen_helper_gvec_recps_s, gen_helper_gvec_recps_s,
gen_helper_gvec_recps_d, gen_helper_gvec_recps_d,
}; };
TRANS(FRECPS_v, do_fp3_vector_ah, a, 0, f_vector_frecps) static gen_helper_gvec_3_ptr * const f_vector_ah_frecps[3] = {
gen_helper_gvec_ah_recps_h,
gen_helper_gvec_ah_recps_s,
gen_helper_gvec_ah_recps_d,
};
TRANS(FRECPS_v, do_fp3_vector_ah_2fn, a, 0, f_vector_frecps, f_vector_ah_frecps)
static gen_helper_gvec_3_ptr * const f_vector_frsqrts[3] = { static gen_helper_gvec_3_ptr * const f_vector_frsqrts[3] = {
gen_helper_gvec_rsqrts_h, gen_helper_gvec_rsqrts_h,
gen_helper_gvec_rsqrts_s, gen_helper_gvec_rsqrts_s,
gen_helper_gvec_rsqrts_d, gen_helper_gvec_rsqrts_d,
}; };
TRANS(FRSQRTS_v, do_fp3_vector_ah, a, 0, f_vector_frsqrts) static gen_helper_gvec_3_ptr * const f_vector_ah_frsqrts[3] = {
gen_helper_gvec_ah_rsqrts_h,
gen_helper_gvec_ah_rsqrts_s,
gen_helper_gvec_ah_rsqrts_d,
};
TRANS(FRSQRTS_v, do_fp3_vector_ah_2fn, a, 0, f_vector_frsqrts, f_vector_ah_frsqrts)
static gen_helper_gvec_3_ptr * const f_vector_faddp[3] = { static gen_helper_gvec_3_ptr * const f_vector_faddp[3] = {
gen_helper_gvec_faddp_h, gen_helper_gvec_faddp_h,

View File

@ -3741,7 +3741,12 @@ static bool trans_FADDA(DisasContext *s, arg_rprr_esz *a)
NULL, gen_helper_gvec_##name##_h, \ NULL, gen_helper_gvec_##name##_h, \
gen_helper_gvec_##name##_s, gen_helper_gvec_##name##_d \ gen_helper_gvec_##name##_s, gen_helper_gvec_##name##_d \
}; \ }; \
TRANS_FEAT(NAME, aa64_sve, gen_gvec_fpst_ah_arg_zzz, name##_fns[a->esz], a, 0) static gen_helper_gvec_3_ptr * const name##_ah_fns[4] = { \
NULL, gen_helper_gvec_ah_##name##_h, \
gen_helper_gvec_ah_##name##_s, gen_helper_gvec_ah_##name##_d \
}; \
TRANS_FEAT(NAME, aa64_sve, gen_gvec_fpst_ah_arg_zzz, \
s->fpcr_ah ? name##_ah_fns[a->esz] : name##_fns[a->esz], a, 0)
DO_FP3(FADD_zzz, fadd) DO_FP3(FADD_zzz, fadd)
DO_FP3(FSUB_zzz, fsub) DO_FP3(FSUB_zzz, fsub)

View File

@ -1477,6 +1477,14 @@ DO_3OP(gvec_rsqrts_h, helper_rsqrtsf_f16, float16)
DO_3OP(gvec_rsqrts_s, helper_rsqrtsf_f32, float32) DO_3OP(gvec_rsqrts_s, helper_rsqrtsf_f32, float32)
DO_3OP(gvec_rsqrts_d, helper_rsqrtsf_f64, float64) DO_3OP(gvec_rsqrts_d, helper_rsqrtsf_f64, float64)
DO_3OP(gvec_ah_recps_h, helper_recpsf_ah_f16, float16)
DO_3OP(gvec_ah_recps_s, helper_recpsf_ah_f32, float32)
DO_3OP(gvec_ah_recps_d, helper_recpsf_ah_f64, float64)
DO_3OP(gvec_ah_rsqrts_h, helper_rsqrtsf_ah_f16, float16)
DO_3OP(gvec_ah_rsqrts_s, helper_rsqrtsf_ah_f32, float32)
DO_3OP(gvec_ah_rsqrts_d, helper_rsqrtsf_ah_f64, float64)
DO_3OP(gvec_ah_fmax_h, helper_vfp_ah_maxh, float16) DO_3OP(gvec_ah_fmax_h, helper_vfp_ah_maxh, float16)
DO_3OP(gvec_ah_fmax_s, helper_vfp_ah_maxs, float32) DO_3OP(gvec_ah_fmax_s, helper_vfp_ah_maxs, float32)
DO_3OP(gvec_ah_fmax_d, helper_vfp_ah_maxd, float64) DO_3OP(gvec_ah_fmax_d, helper_vfp_ah_maxd, float64)