Re: [Qemu-devel] [PATCH v2 43/67] target/arm: Implement SVE Floating Point Arithmetic - Unpredicated Group

2018-02-23 Thread Richard Henderson 
On 02/23/2018 09:25 AM, Peter Maydell wrote:
> On 17 February 2018 at 18:22, Richard Henderson
>  wrote:
>> Signed-off-by: Richard Henderson 
>> ---
>>  target/arm/helper-sve.h| 14 +++
>>  target/arm/helper.h| 19 ++
>>  target/arm/translate-sve.c | 41 
>>  target/arm/vec_helper.c| 94 
>> ++
>>  target/arm/Makefile.objs   |  2 +-
>>  target/arm/sve.decode  | 10 +
>>  6 files changed, 179 insertions(+), 1 deletion(-)
>>  create mode 100644 target/arm/vec_helper.c
>>
> 
>> +/* Floating-point trigonometric starting value.
>> + * See the ARM ARM pseudocode function FPTrigSMul.
>> + */
>> +static float16 float16_ftsmul(float16 op1, uint16_t op2, float_status *stat)
>> +{
>> +float16 result = float16_mul(op1, op1, stat);
>> +if (!float16_is_any_nan(result)) {
>> +result = float16_set_sign(result, op2 & 1);
>> +}
>> +return result;
>> +}
>> +
>> +static float32 float32_ftsmul(float32 op1, uint32_t op2, float_status *stat)
>> +{
>> +float32 result = float32_mul(op1, op1, stat);
>> +if (!float32_is_any_nan(result)) {
>> +result = float32_set_sign(result, op2 & 1);
>> +}
>> +return result;
>> +}
>> +
>> +static float64 float64_ftsmul(float64 op1, uint64_t op2, float_status *stat)
>> +{
>> +float64 result = float64_mul(op1, op1, stat);
>> +if (!float64_is_any_nan(result)) {
>> +result = float64_set_sign(result, op2 & 1);
>> +}
>> +return result;
>> +}
>> +
>> +#define DO_3OP(NAME, FUNC, TYPE) \
>> +void HELPER(NAME)(void *vd, void *vn, void *vm, void *stat, uint32_t desc) \
>> +{  \
>> +intptr_t i, oprsz = simd_oprsz(desc);  \
>> +TYPE *d = vd, *n = vn, *m = vm;\
>> +for (i = 0; i < oprsz / sizeof(TYPE); i++) {   \
>> +d[i] = FUNC(n[i], m[i], stat); \
>> +}  \
>> +}
>> +
>> +DO_3OP(gvec_fadd_h, float16_add, float16)
>> +DO_3OP(gvec_fadd_s, float32_add, float32)
>> +DO_3OP(gvec_fadd_d, float64_add, float64)
>> +
>> +DO_3OP(gvec_fsub_h, float16_sub, float16)
>> +DO_3OP(gvec_fsub_s, float32_sub, float32)
>> +DO_3OP(gvec_fsub_d, float64_sub, float64)
>> +
>> +DO_3OP(gvec_fmul_h, float16_mul, float16)
>> +DO_3OP(gvec_fmul_s, float32_mul, float32)
>> +DO_3OP(gvec_fmul_d, float64_mul, float64)
>> +
>> +DO_3OP(gvec_ftsmul_h, float16_ftsmul, float16)
>> +DO_3OP(gvec_ftsmul_s, float32_ftsmul, float32)
>> +DO_3OP(gvec_ftsmul_d, float64_ftsmul, float64)
>> +
>> +#ifdef TARGET_AARCH64
> 
> This seems a bit odd given SVE is AArch64-only anyway...

Ah right.

The thing to notice here is that the helpers have been placed such that the
helpers can be shared with AA32 and AA64 AdvSIMD.  One call to one of these
would replace the 2-8 calls that we currently generate for such an operation.

I thought it better to plan ahead for that cleanup as opposed to moving them 
later.

Here you see where AA64 differs from AA32 (and in particular where the scalar
operation is also conditionalized).


r~

Re: [Qemu-devel] [PATCH v2 43/67] target/arm: Implement SVE Floating Point Arithmetic - Unpredicated Group

2018-02-23 Thread Peter Maydell 
On 17 February 2018 at 18:22, Richard Henderson
 wrote:
> Signed-off-by: Richard Henderson 
> ---
>  target/arm/helper-sve.h| 14 +++
>  target/arm/helper.h| 19 ++
>  target/arm/translate-sve.c | 41 
>  target/arm/vec_helper.c| 94 
> ++
>  target/arm/Makefile.objs   |  2 +-
>  target/arm/sve.decode  | 10 +
>  6 files changed, 179 insertions(+), 1 deletion(-)
>  create mode 100644 target/arm/vec_helper.c
>

> +/* Floating-point trigonometric starting value.
> + * See the ARM ARM pseudocode function FPTrigSMul.
> + */
> +static float16 float16_ftsmul(float16 op1, uint16_t op2, float_status *stat)
> +{
> +float16 result = float16_mul(op1, op1, stat);
> +if (!float16_is_any_nan(result)) {
> +result = float16_set_sign(result, op2 & 1);
> +}
> +return result;
> +}
> +
> +static float32 float32_ftsmul(float32 op1, uint32_t op2, float_status *stat)
> +{
> +float32 result = float32_mul(op1, op1, stat);
> +if (!float32_is_any_nan(result)) {
> +result = float32_set_sign(result, op2 & 1);
> +}
> +return result;
> +}
> +
> +static float64 float64_ftsmul(float64 op1, uint64_t op2, float_status *stat)
> +{
> +float64 result = float64_mul(op1, op1, stat);
> +if (!float64_is_any_nan(result)) {
> +result = float64_set_sign(result, op2 & 1);
> +}
> +return result;
> +}
> +
> +#define DO_3OP(NAME, FUNC, TYPE) \
> +void HELPER(NAME)(void *vd, void *vn, void *vm, void *stat, uint32_t desc) \
> +{  \
> +intptr_t i, oprsz = simd_oprsz(desc);  \
> +TYPE *d = vd, *n = vn, *m = vm;\
> +for (i = 0; i < oprsz / sizeof(TYPE); i++) {   \
> +d[i] = FUNC(n[i], m[i], stat); \
> +}  \
> +}
> +
> +DO_3OP(gvec_fadd_h, float16_add, float16)
> +DO_3OP(gvec_fadd_s, float32_add, float32)
> +DO_3OP(gvec_fadd_d, float64_add, float64)
> +
> +DO_3OP(gvec_fsub_h, float16_sub, float16)
> +DO_3OP(gvec_fsub_s, float32_sub, float32)
> +DO_3OP(gvec_fsub_d, float64_sub, float64)
> +
> +DO_3OP(gvec_fmul_h, float16_mul, float16)
> +DO_3OP(gvec_fmul_s, float32_mul, float32)
> +DO_3OP(gvec_fmul_d, float64_mul, float64)
> +
> +DO_3OP(gvec_ftsmul_h, float16_ftsmul, float16)
> +DO_3OP(gvec_ftsmul_s, float32_ftsmul, float32)
> +DO_3OP(gvec_ftsmul_d, float64_ftsmul, float64)
> +
> +#ifdef TARGET_AARCH64

This seems a bit odd given SVE is AArch64-only anyway...

> +
> +DO_3OP(gvec_recps_h, helper_recpsf_f16, float16)
> +DO_3OP(gvec_recps_s, helper_recpsf_f32, float32)
> +DO_3OP(gvec_recps_d, helper_recpsf_f64, float64)
> +
> +DO_3OP(gvec_rsqrts_h, helper_rsqrtsf_f16, float16)
> +DO_3OP(gvec_rsqrts_s, helper_rsqrtsf_f32, float32)
> +DO_3OP(gvec_rsqrts_d, helper_rsqrtsf_f64, float64)
> +
> +#endif
> +#undef DO_3OP

> +### SVE Floating Point Arithmetic - Unpredicated Group
> +
> +# SVE floating-point arithmetic (unpredicated)
> +FADD_zzz   01100101 .. 0 . 000 000 . . @rd_rn_rm
> +FSUB_zzz   01100101 .. 0 . 000 001 . . @rd_rn_rm
> +FMUL_zzz   01100101 .. 0 . 000 010 . . @rd_rn_rm
> +FTSMUL 01100101 .. 0 . 000 011 . . @rd_rn_rm
> +FRECPS 01100101 .. 0 . 000 110 . . @rd_rn_rm
> +FRSQRTS01100101 .. 0 . 000 111 . . 
> @rd_rn_rm

Another misaligned line.

Reviewed-by: Peter Maydell 

thanks
-- PMM

[Qemu-devel] [PATCH v2 43/67] target/arm: Implement SVE Floating Point Arithmetic - Unpredicated Group

2018-02-17 Thread Richard Henderson 
Signed-off-by: Richard Henderson 
---
 target/arm/helper-sve.h| 14 +++
 target/arm/helper.h| 19 ++
 target/arm/translate-sve.c | 41 
 target/arm/vec_helper.c| 94 ++
 target/arm/Makefile.objs   |  2 +-
 target/arm/sve.decode  | 10 +
 6 files changed, 179 insertions(+), 1 deletion(-)
 create mode 100644 target/arm/vec_helper.c

diff --git a/target/arm/helper-sve.h b/target/arm/helper-sve.h
index 97bfe0f47b..2e76084992 100644
--- a/target/arm/helper-sve.h
+++ b/target/arm/helper-sve.h
@@ -705,3 +705,17 @@ DEF_HELPER_FLAGS_4(sve_umini_b, TCG_CALL_NO_RWG, void, 
ptr, ptr, i64, i32)
 DEF_HELPER_FLAGS_4(sve_umini_h, TCG_CALL_NO_RWG, void, ptr, ptr, i64, i32)
 DEF_HELPER_FLAGS_4(sve_umini_s, TCG_CALL_NO_RWG, void, ptr, ptr, i64, i32)
 DEF_HELPER_FLAGS_4(sve_umini_d, TCG_CALL_NO_RWG, void, ptr, ptr, i64, i32)
+
+DEF_HELPER_FLAGS_5(gvec_recps_h, TCG_CALL_NO_RWG,
+   void, ptr, ptr, ptr, ptr, i32)
+DEF_HELPER_FLAGS_5(gvec_recps_s, TCG_CALL_NO_RWG,
+   void, ptr, ptr, ptr, ptr, i32)
+DEF_HELPER_FLAGS_5(gvec_recps_d, TCG_CALL_NO_RWG,
+   void, ptr, ptr, ptr, ptr, i32)
+
+DEF_HELPER_FLAGS_5(gvec_rsqrts_h, TCG_CALL_NO_RWG,
+   void, ptr, ptr, ptr, ptr, i32)
+DEF_HELPER_FLAGS_5(gvec_rsqrts_s, TCG_CALL_NO_RWG,
+   void, ptr, ptr, ptr, ptr, i32)
+DEF_HELPER_FLAGS_5(gvec_rsqrts_d, TCG_CALL_NO_RWG,
+   void, ptr, ptr, ptr, ptr, i32)
diff --git a/target/arm/helper.h b/target/arm/helper.h
index be3c2fcdc0..f3ce58e276 100644
--- a/target/arm/helper.h
+++ b/target/arm/helper.h
@@ -565,6 +565,25 @@ DEF_HELPER_2(dc_zva, void, env, i64)
 DEF_HELPER_FLAGS_2(neon_pmull_64_lo, TCG_CALL_NO_RWG_SE, i64, i64, i64)
 DEF_HELPER_FLAGS_2(neon_pmull_64_hi, TCG_CALL_NO_RWG_SE, i64, i64, i64)
 
+DEF_HELPER_FLAGS_5(gvec_fadd_h, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, ptr, i32)
+DEF_HELPER_FLAGS_5(gvec_fadd_s, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, ptr, i32)
+DEF_HELPER_FLAGS_5(gvec_fadd_d, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, ptr, i32)
+
+DEF_HELPER_FLAGS_5(gvec_fsub_h, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, ptr, i32)
+DEF_HELPER_FLAGS_5(gvec_fsub_s, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, ptr, i32)
+DEF_HELPER_FLAGS_5(gvec_fsub_d, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, ptr, i32)
+
+DEF_HELPER_FLAGS_5(gvec_fmul_h, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, ptr, i32)
+DEF_HELPER_FLAGS_5(gvec_fmul_s, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, ptr, i32)
+DEF_HELPER_FLAGS_5(gvec_fmul_d, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, ptr, i32)
+
+DEF_HELPER_FLAGS_5(gvec_ftsmul_h, TCG_CALL_NO_RWG,
+   void, ptr, ptr, ptr, ptr, i32)
+DEF_HELPER_FLAGS_5(gvec_ftsmul_s, TCG_CALL_NO_RWG,
+   void, ptr, ptr, ptr, ptr, i32)
+DEF_HELPER_FLAGS_5(gvec_ftsmul_d, TCG_CALL_NO_RWG,
+   void, ptr, ptr, ptr, ptr, i32)
+
 #ifdef TARGET_AARCH64
 #include "helper-a64.h"
 #include "helper-sve.h"
diff --git a/target/arm/translate-sve.c b/target/arm/translate-sve.c
index 72abcb543a..f9a3ad1434 100644
--- a/target/arm/translate-sve.c
+++ b/target/arm/translate-sve.c
@@ -3109,6 +3109,47 @@ DO_ZZI(UMIN, umin)
 
 #undef DO_ZZI
 
+/*
+ *** SVE Floating Point Arithmetic - Unpredicated Group
+ */
+
+static void do_zzz_fp(DisasContext *s, arg_rrr_esz *a,
+  gen_helper_gvec_3_ptr *fn)
+{
+unsigned vsz = vec_full_reg_size(s);
+TCGv_ptr status;
+
+if (fn == NULL) {
+unallocated_encoding(s);
+return;
+}
+status = get_fpstatus_ptr(a->esz == MO_16);
+tcg_gen_gvec_3_ptr(vec_full_reg_offset(s, a->rd),
+   vec_full_reg_offset(s, a->rn),
+   vec_full_reg_offset(s, a->rm),
+   status, vsz, vsz, 0, fn);
+}
+
+
+#define DO_FP3(NAME, name) \
+static void trans_##NAME(DisasContext *s, arg_rrr_esz *a, uint32_t insn) \
+{   \
+static gen_helper_gvec_3_ptr * const fns[4] = { \
+NULL, gen_helper_gvec_##name##_h,   \
+gen_helper_gvec_##name##_s, gen_helper_gvec_##name##_d  \
+};  \
+do_zzz_fp(s, a, fns[a->esz]);   \
+}
+
+DO_FP3(FADD_zzz, fadd)
+DO_FP3(FSUB_zzz, fsub)
+DO_FP3(FMUL_zzz, fmul)
+DO_FP3(FTSMUL, ftsmul)
+DO_FP3(FRECPS, recps)
+DO_FP3(FRSQRTS, rsqrts)
+
+#undef DO_FP3
+
 /*
  *** SVE Memory - 32-bit Gather and Unsized Contiguous Group
  */
diff --git a/target/arm/vec_helper.c b/target/arm/vec_helper.c
new file mode 100644
index 00..ad5c29cdd5
--- /dev/null
+++ b/target/arm/vec_helper.c
@@ -0,0 +1,94 @@
+/*
+ * ARM Shared AdvSIMD / SVE Operations
+ *
+ * Copyright (c) 2018 Linaro
+ *
+ * This library is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Lesser