target/riscv: vector single-width integer reduction instructions

Backports fe5c9ab1fc185e96bf7e034954127429ca74d386
This commit is contained in:
LIU Zhiwei 2021-03-07 12:07:49 -05:00 committed by Lioncash
parent e925927e23
commit 4b1e548fd0
7 changed files with 229 additions and 0 deletions

View file

@ -7201,6 +7201,38 @@ riscv_symbols = (
'helper_vfncvt_f_x_v_w',
'helper_vfncvt_f_f_v_h',
'helper_vfncvt_f_f_v_w',
'helper_vredsum_vs_b',
'helper_vredsum_vs_h',
'helper_vredsum_vs_w',
'helper_vredsum_vs_d',
'helper_vredmaxu_vs_b',
'helper_vredmaxu_vs_h',
'helper_vredmaxu_vs_w',
'helper_vredmaxu_vs_d',
'helper_vredmax_vs_b',
'helper_vredmax_vs_h',
'helper_vredmax_vs_w',
'helper_vredmax_vs_d',
'helper_vredminu_vs_b',
'helper_vredminu_vs_h',
'helper_vredminu_vs_w',
'helper_vredminu_vs_d',
'helper_vredmin_vs_b',
'helper_vredmin_vs_h',
'helper_vredmin_vs_w',
'helper_vredmin_vs_d',
'helper_vredand_vs_b',
'helper_vredand_vs_h',
'helper_vredand_vs_w',
'helper_vredand_vs_d',
'helper_vredor_vs_b',
'helper_vredor_vs_h',
'helper_vredor_vs_w',
'helper_vredor_vs_d',
'helper_vredxor_vs_b',
'helper_vredxor_vs_h',
'helper_vredxor_vs_w',
'helper_vredxor_vs_d',
'pmp_hart_has_privs',
'pmpaddr_csr_read',
'pmpaddr_csr_write',

View file

@ -4637,6 +4637,38 @@
#define helper_vfncvt_f_x_v_w helper_vfncvt_f_x_v_w_riscv32
#define helper_vfncvt_f_f_v_h helper_vfncvt_f_f_v_h_riscv32
#define helper_vfncvt_f_f_v_w helper_vfncvt_f_f_v_w_riscv32
#define helper_vredsum_vs_b helper_vredsum_vs_b_riscv32
#define helper_vredsum_vs_h helper_vredsum_vs_h_riscv32
#define helper_vredsum_vs_w helper_vredsum_vs_w_riscv32
#define helper_vredsum_vs_d helper_vredsum_vs_d_riscv32
#define helper_vredmaxu_vs_b helper_vredmaxu_vs_b_riscv32
#define helper_vredmaxu_vs_h helper_vredmaxu_vs_h_riscv32
#define helper_vredmaxu_vs_w helper_vredmaxu_vs_w_riscv32
#define helper_vredmaxu_vs_d helper_vredmaxu_vs_d_riscv32
#define helper_vredmax_vs_b helper_vredmax_vs_b_riscv32
#define helper_vredmax_vs_h helper_vredmax_vs_h_riscv32
#define helper_vredmax_vs_w helper_vredmax_vs_w_riscv32
#define helper_vredmax_vs_d helper_vredmax_vs_d_riscv32
#define helper_vredminu_vs_b helper_vredminu_vs_b_riscv32
#define helper_vredminu_vs_h helper_vredminu_vs_h_riscv32
#define helper_vredminu_vs_w helper_vredminu_vs_w_riscv32
#define helper_vredminu_vs_d helper_vredminu_vs_d_riscv32
#define helper_vredmin_vs_b helper_vredmin_vs_b_riscv32
#define helper_vredmin_vs_h helper_vredmin_vs_h_riscv32
#define helper_vredmin_vs_w helper_vredmin_vs_w_riscv32
#define helper_vredmin_vs_d helper_vredmin_vs_d_riscv32
#define helper_vredand_vs_b helper_vredand_vs_b_riscv32
#define helper_vredand_vs_h helper_vredand_vs_h_riscv32
#define helper_vredand_vs_w helper_vredand_vs_w_riscv32
#define helper_vredand_vs_d helper_vredand_vs_d_riscv32
#define helper_vredor_vs_b helper_vredor_vs_b_riscv32
#define helper_vredor_vs_h helper_vredor_vs_h_riscv32
#define helper_vredor_vs_w helper_vredor_vs_w_riscv32
#define helper_vredor_vs_d helper_vredor_vs_d_riscv32
#define helper_vredxor_vs_b helper_vredxor_vs_b_riscv32
#define helper_vredxor_vs_h helper_vredxor_vs_h_riscv32
#define helper_vredxor_vs_w helper_vredxor_vs_w_riscv32
#define helper_vredxor_vs_d helper_vredxor_vs_d_riscv32
#define pmp_hart_has_privs pmp_hart_has_privs_riscv32
#define pmpaddr_csr_read pmpaddr_csr_read_riscv32
#define pmpaddr_csr_write pmpaddr_csr_write_riscv32

View file

@ -4637,6 +4637,38 @@
#define helper_vfncvt_f_x_v_w helper_vfncvt_f_x_v_w_riscv64
#define helper_vfncvt_f_f_v_h helper_vfncvt_f_f_v_h_riscv64
#define helper_vfncvt_f_f_v_w helper_vfncvt_f_f_v_w_riscv64
#define helper_vredsum_vs_b helper_vredsum_vs_b_riscv64
#define helper_vredsum_vs_h helper_vredsum_vs_h_riscv64
#define helper_vredsum_vs_w helper_vredsum_vs_w_riscv64
#define helper_vredsum_vs_d helper_vredsum_vs_d_riscv64
#define helper_vredmaxu_vs_b helper_vredmaxu_vs_b_riscv64
#define helper_vredmaxu_vs_h helper_vredmaxu_vs_h_riscv64
#define helper_vredmaxu_vs_w helper_vredmaxu_vs_w_riscv64
#define helper_vredmaxu_vs_d helper_vredmaxu_vs_d_riscv64
#define helper_vredmax_vs_b helper_vredmax_vs_b_riscv64
#define helper_vredmax_vs_h helper_vredmax_vs_h_riscv64
#define helper_vredmax_vs_w helper_vredmax_vs_w_riscv64
#define helper_vredmax_vs_d helper_vredmax_vs_d_riscv64
#define helper_vredminu_vs_b helper_vredminu_vs_b_riscv64
#define helper_vredminu_vs_h helper_vredminu_vs_h_riscv64
#define helper_vredminu_vs_w helper_vredminu_vs_w_riscv64
#define helper_vredminu_vs_d helper_vredminu_vs_d_riscv64
#define helper_vredmin_vs_b helper_vredmin_vs_b_riscv64
#define helper_vredmin_vs_h helper_vredmin_vs_h_riscv64
#define helper_vredmin_vs_w helper_vredmin_vs_w_riscv64
#define helper_vredmin_vs_d helper_vredmin_vs_d_riscv64
#define helper_vredand_vs_b helper_vredand_vs_b_riscv64
#define helper_vredand_vs_h helper_vredand_vs_h_riscv64
#define helper_vredand_vs_w helper_vredand_vs_w_riscv64
#define helper_vredand_vs_d helper_vredand_vs_d_riscv64
#define helper_vredor_vs_b helper_vredor_vs_b_riscv64
#define helper_vredor_vs_h helper_vredor_vs_h_riscv64
#define helper_vredor_vs_w helper_vredor_vs_w_riscv64
#define helper_vredor_vs_d helper_vredor_vs_d_riscv64
#define helper_vredxor_vs_b helper_vredxor_vs_b_riscv64
#define helper_vredxor_vs_h helper_vredxor_vs_h_riscv64
#define helper_vredxor_vs_w helper_vredxor_vs_w_riscv64
#define helper_vredxor_vs_d helper_vredxor_vs_d_riscv64
#define pmp_hart_has_privs pmp_hart_has_privs_riscv64
#define pmpaddr_csr_read pmpaddr_csr_read_riscv64
#define pmpaddr_csr_write pmpaddr_csr_write_riscv64

View file

@ -1042,3 +1042,36 @@ DEF_HELPER_5(vfncvt_f_x_v_h, void, ptr, ptr, ptr, env, i32)
DEF_HELPER_5(vfncvt_f_x_v_w, void, ptr, ptr, ptr, env, i32)
DEF_HELPER_5(vfncvt_f_f_v_h, void, ptr, ptr, ptr, env, i32)
DEF_HELPER_5(vfncvt_f_f_v_w, void, ptr, ptr, ptr, env, i32)
DEF_HELPER_6(vredsum_vs_b, void, ptr, ptr, ptr, ptr, env, i32)
DEF_HELPER_6(vredsum_vs_h, void, ptr, ptr, ptr, ptr, env, i32)
DEF_HELPER_6(vredsum_vs_w, void, ptr, ptr, ptr, ptr, env, i32)
DEF_HELPER_6(vredsum_vs_d, void, ptr, ptr, ptr, ptr, env, i32)
DEF_HELPER_6(vredmaxu_vs_b, void, ptr, ptr, ptr, ptr, env, i32)
DEF_HELPER_6(vredmaxu_vs_h, void, ptr, ptr, ptr, ptr, env, i32)
DEF_HELPER_6(vredmaxu_vs_w, void, ptr, ptr, ptr, ptr, env, i32)
DEF_HELPER_6(vredmaxu_vs_d, void, ptr, ptr, ptr, ptr, env, i32)
DEF_HELPER_6(vredmax_vs_b, void, ptr, ptr, ptr, ptr, env, i32)
DEF_HELPER_6(vredmax_vs_h, void, ptr, ptr, ptr, ptr, env, i32)
DEF_HELPER_6(vredmax_vs_w, void, ptr, ptr, ptr, ptr, env, i32)
DEF_HELPER_6(vredmax_vs_d, void, ptr, ptr, ptr, ptr, env, i32)
DEF_HELPER_6(vredminu_vs_b, void, ptr, ptr, ptr, ptr, env, i32)
DEF_HELPER_6(vredminu_vs_h, void, ptr, ptr, ptr, ptr, env, i32)
DEF_HELPER_6(vredminu_vs_w, void, ptr, ptr, ptr, ptr, env, i32)
DEF_HELPER_6(vredminu_vs_d, void, ptr, ptr, ptr, ptr, env, i32)
DEF_HELPER_6(vredmin_vs_b, void, ptr, ptr, ptr, ptr, env, i32)
DEF_HELPER_6(vredmin_vs_h, void, ptr, ptr, ptr, ptr, env, i32)
DEF_HELPER_6(vredmin_vs_w, void, ptr, ptr, ptr, ptr, env, i32)
DEF_HELPER_6(vredmin_vs_d, void, ptr, ptr, ptr, ptr, env, i32)
DEF_HELPER_6(vredand_vs_b, void, ptr, ptr, ptr, ptr, env, i32)
DEF_HELPER_6(vredand_vs_h, void, ptr, ptr, ptr, ptr, env, i32)
DEF_HELPER_6(vredand_vs_w, void, ptr, ptr, ptr, ptr, env, i32)
DEF_HELPER_6(vredand_vs_d, void, ptr, ptr, ptr, ptr, env, i32)
DEF_HELPER_6(vredor_vs_b, void, ptr, ptr, ptr, ptr, env, i32)
DEF_HELPER_6(vredor_vs_h, void, ptr, ptr, ptr, ptr, env, i32)
DEF_HELPER_6(vredor_vs_w, void, ptr, ptr, ptr, ptr, env, i32)
DEF_HELPER_6(vredor_vs_d, void, ptr, ptr, ptr, ptr, env, i32)
DEF_HELPER_6(vredxor_vs_b, void, ptr, ptr, ptr, ptr, env, i32)
DEF_HELPER_6(vredxor_vs_h, void, ptr, ptr, ptr, ptr, env, i32)
DEF_HELPER_6(vredxor_vs_w, void, ptr, ptr, ptr, ptr, env, i32)
DEF_HELPER_6(vredxor_vs_d, void, ptr, ptr, ptr, ptr, env, i32)

View file

@ -531,6 +531,14 @@ vfncvt_x_f_v 100010 . ..... 10001 001 ..... 1010111 @r2_vm
vfncvt_f_xu_v 100010 . ..... 10010 001 ..... 1010111 @r2_vm
vfncvt_f_x_v 100010 . ..... 10011 001 ..... 1010111 @r2_vm
vfncvt_f_f_v 100010 . ..... 10100 001 ..... 1010111 @r2_vm
vredsum_vs 000000 . ..... ..... 010 ..... 1010111 @r_vm
vredand_vs 000001 . ..... ..... 010 ..... 1010111 @r_vm
vredor_vs 000010 . ..... ..... 010 ..... 1010111 @r_vm
vredxor_vs 000011 . ..... ..... 010 ..... 1010111 @r_vm
vredminu_vs 000100 . ..... ..... 010 ..... 1010111 @r_vm
vredmin_vs 000101 . ..... ..... 010 ..... 1010111 @r_vm
vredmaxu_vs 000110 . ..... ..... 010 ..... 1010111 @r_vm
vredmax_vs 000111 . ..... ..... 010 ..... 1010111 @r_vm
vsetvli 0 ........... ..... 111 ..... 1010111 @r2_zimm
vsetvl 1000000 ..... ..... 111 ..... 1010111 @r

View file

@ -2361,3 +2361,21 @@ GEN_OPFV_NARROW_TRANS(vfncvt_x_f_v)
GEN_OPFV_NARROW_TRANS(vfncvt_f_xu_v)
GEN_OPFV_NARROW_TRANS(vfncvt_f_x_v)
GEN_OPFV_NARROW_TRANS(vfncvt_f_f_v)
/*
*** Vector Reduction Operations
*/
/* Vector Single-Width Integer Reduction Instructions */
static bool reduction_check(DisasContext *s, arg_rmrr *a)
{
return vext_check_isa_ill(s) && vext_check_reg(s, a->rs2, false);
}
GEN_OPIVV_TRANS(vredsum_vs, reduction_check)
GEN_OPIVV_TRANS(vredmaxu_vs, reduction_check)
GEN_OPIVV_TRANS(vredmax_vs, reduction_check)
GEN_OPIVV_TRANS(vredminu_vs, reduction_check)
GEN_OPIVV_TRANS(vredmin_vs, reduction_check)
GEN_OPIVV_TRANS(vredand_vs, reduction_check)
GEN_OPIVV_TRANS(vredor_vs, reduction_check)
GEN_OPIVV_TRANS(vredxor_vs, reduction_check)

View file

@ -4308,3 +4308,77 @@ RVVCALL(OPFVV1, vfncvt_f_f_v_h, NOP_UU_H, H2, H4, vfncvtffv16)
RVVCALL(OPFVV1, vfncvt_f_f_v_w, NOP_UU_W, H4, H8, float64_to_float32)
GEN_VEXT_V_ENV(vfncvt_f_f_v_h, 2, 2, clearh)
GEN_VEXT_V_ENV(vfncvt_f_f_v_w, 4, 4, clearl)
/*
*** Vector Reduction Operations
*/
/* Vector Single-Width Integer Reduction Instructions */
#define GEN_VEXT_RED(NAME, TD, TS2, HD, HS2, OP, CLEAR_FN)\
void HELPER(NAME)(void *vd, void *v0, void *vs1, \
void *vs2, CPURISCVState *env, uint32_t desc) \
{ \
uint32_t mlen = vext_mlen(desc); \
uint32_t vm = vext_vm(desc); \
uint32_t vl = env->vl; \
uint32_t i; \
uint32_t tot = env_archcpu(env)->cfg.vlen / 8; \
TD s1 = *((TD *)vs1 + HD(0)); \
\
for (i = 0; i < vl; i++) { \
TS2 s2 = *((TS2 *)vs2 + HS2(i)); \
if (!vm && !vext_elem_mask(v0, mlen, i)) { \
continue; \
} \
s1 = OP(s1, (TD)s2); \
} \
*((TD *)vd + HD(0)) = s1; \
CLEAR_FN(vd, 1, sizeof(TD), tot); \
}
/* vd[0] = sum(vs1[0], vs2[*]) */
GEN_VEXT_RED(vredsum_vs_b, int8_t, int8_t, H1, H1, DO_ADD, clearb)
GEN_VEXT_RED(vredsum_vs_h, int16_t, int16_t, H2, H2, DO_ADD, clearh)
GEN_VEXT_RED(vredsum_vs_w, int32_t, int32_t, H4, H4, DO_ADD, clearl)
GEN_VEXT_RED(vredsum_vs_d, int64_t, int64_t, H8, H8, DO_ADD, clearq)
/* vd[0] = maxu(vs1[0], vs2[*]) */
GEN_VEXT_RED(vredmaxu_vs_b, uint8_t, uint8_t, H1, H1, DO_MAX, clearb)
GEN_VEXT_RED(vredmaxu_vs_h, uint16_t, uint16_t, H2, H2, DO_MAX, clearh)
GEN_VEXT_RED(vredmaxu_vs_w, uint32_t, uint32_t, H4, H4, DO_MAX, clearl)
GEN_VEXT_RED(vredmaxu_vs_d, uint64_t, uint64_t, H8, H8, DO_MAX, clearq)
/* vd[0] = max(vs1[0], vs2[*]) */
GEN_VEXT_RED(vredmax_vs_b, int8_t, int8_t, H1, H1, DO_MAX, clearb)
GEN_VEXT_RED(vredmax_vs_h, int16_t, int16_t, H2, H2, DO_MAX, clearh)
GEN_VEXT_RED(vredmax_vs_w, int32_t, int32_t, H4, H4, DO_MAX, clearl)
GEN_VEXT_RED(vredmax_vs_d, int64_t, int64_t, H8, H8, DO_MAX, clearq)
/* vd[0] = minu(vs1[0], vs2[*]) */
GEN_VEXT_RED(vredminu_vs_b, uint8_t, uint8_t, H1, H1, DO_MIN, clearb)
GEN_VEXT_RED(vredminu_vs_h, uint16_t, uint16_t, H2, H2, DO_MIN, clearh)
GEN_VEXT_RED(vredminu_vs_w, uint32_t, uint32_t, H4, H4, DO_MIN, clearl)
GEN_VEXT_RED(vredminu_vs_d, uint64_t, uint64_t, H8, H8, DO_MIN, clearq)
/* vd[0] = min(vs1[0], vs2[*]) */
GEN_VEXT_RED(vredmin_vs_b, int8_t, int8_t, H1, H1, DO_MIN, clearb)
GEN_VEXT_RED(vredmin_vs_h, int16_t, int16_t, H2, H2, DO_MIN, clearh)
GEN_VEXT_RED(vredmin_vs_w, int32_t, int32_t, H4, H4, DO_MIN, clearl)
GEN_VEXT_RED(vredmin_vs_d, int64_t, int64_t, H8, H8, DO_MIN, clearq)
/* vd[0] = and(vs1[0], vs2[*]) */
GEN_VEXT_RED(vredand_vs_b, int8_t, int8_t, H1, H1, DO_AND, clearb)
GEN_VEXT_RED(vredand_vs_h, int16_t, int16_t, H2, H2, DO_AND, clearh)
GEN_VEXT_RED(vredand_vs_w, int32_t, int32_t, H4, H4, DO_AND, clearl)
GEN_VEXT_RED(vredand_vs_d, int64_t, int64_t, H8, H8, DO_AND, clearq)
/* vd[0] = or(vs1[0], vs2[*]) */
GEN_VEXT_RED(vredor_vs_b, int8_t, int8_t, H1, H1, DO_OR, clearb)
GEN_VEXT_RED(vredor_vs_h, int16_t, int16_t, H2, H2, DO_OR, clearh)
GEN_VEXT_RED(vredor_vs_w, int32_t, int32_t, H4, H4, DO_OR, clearl)
GEN_VEXT_RED(vredor_vs_d, int64_t, int64_t, H8, H8, DO_OR, clearq)
/* vd[0] = xor(vs1[0], vs2[*]) */
GEN_VEXT_RED(vredxor_vs_b, int8_t, int8_t, H1, H1, DO_XOR, clearb)
GEN_VEXT_RED(vredxor_vs_h, int16_t, int16_t, H2, H2, DO_XOR, clearh)
GEN_VEXT_RED(vredxor_vs_w, int32_t, int32_t, H4, H4, DO_XOR, clearl)
GEN_VEXT_RED(vredxor_vs_d, int64_t, int64_t, H8, H8, DO_XOR, clearq)