From bcefc90208f8a1d6f619d61c2647281d92277015 Mon Sep 17 00:00:00 2001 From: Richard Henderson Date: Wed, 17 Apr 2019 13:53:02 -1000 Subject: [PATCH] tcg: Add support for vector absolute value MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Reviewed-by: Alex Bennée Signed-off-by: Richard Henderson --- accel/tcg/tcg-runtime-gvec.c | 48 +++++++++++++++++++++++++++ accel/tcg/tcg-runtime.h | 5 +++ tcg/README | 4 +++ tcg/aarch64/tcg-target.h | 1 + tcg/i386/tcg-target.h | 1 + tcg/tcg-op-gvec.c | 63 ++++++++++++++++++++++++++++++++++++ tcg/tcg-op-gvec.h | 2 ++ tcg/tcg-op-vec.c | 39 ++++++++++++++++++++++ tcg/tcg-opc.h | 1 + tcg/tcg.c | 2 ++ tcg/tcg.h | 1 + 11 files changed, 167 insertions(+) diff --git a/accel/tcg/tcg-runtime-gvec.c b/accel/tcg/tcg-runtime-gvec.c index 2152fb6903..0f09e0ef38 100644 --- a/accel/tcg/tcg-runtime-gvec.c +++ b/accel/tcg/tcg-runtime-gvec.c @@ -398,6 +398,54 @@ void HELPER(gvec_neg64)(void *d, void *a, uint32_t desc) clear_high(d, oprsz, desc); } +void HELPER(gvec_abs8)(void *d, void *a, uint32_t desc) +{ + intptr_t oprsz = simd_oprsz(desc); + intptr_t i; + + for (i = 0; i < oprsz; i += sizeof(int8_t)) { + int8_t aa = *(int8_t *)(a + i); + *(int8_t *)(d + i) = aa < 0 ? -aa : aa; + } + clear_high(d, oprsz, desc); +} + +void HELPER(gvec_abs16)(void *d, void *a, uint32_t desc) +{ + intptr_t oprsz = simd_oprsz(desc); + intptr_t i; + + for (i = 0; i < oprsz; i += sizeof(int16_t)) { + int16_t aa = *(int16_t *)(a + i); + *(int16_t *)(d + i) = aa < 0 ? -aa : aa; + } + clear_high(d, oprsz, desc); +} + +void HELPER(gvec_abs32)(void *d, void *a, uint32_t desc) +{ + intptr_t oprsz = simd_oprsz(desc); + intptr_t i; + + for (i = 0; i < oprsz; i += sizeof(int32_t)) { + int32_t aa = *(int32_t *)(a + i); + *(int32_t *)(d + i) = aa < 0 ? -aa : aa; + } + clear_high(d, oprsz, desc); +} + +void HELPER(gvec_abs64)(void *d, void *a, uint32_t desc) +{ + intptr_t oprsz = simd_oprsz(desc); + intptr_t i; + + for (i = 0; i < oprsz; i += sizeof(int64_t)) { + int64_t aa = *(int64_t *)(a + i); + *(int64_t *)(d + i) = aa < 0 ? -aa : aa; + } + clear_high(d, oprsz, desc); +} + void HELPER(gvec_mov)(void *d, void *a, uint32_t desc) { intptr_t oprsz = simd_oprsz(desc); diff --git a/accel/tcg/tcg-runtime.h b/accel/tcg/tcg-runtime.h index ed3ce5fd91..6d73dc2d65 100644 --- a/accel/tcg/tcg-runtime.h +++ b/accel/tcg/tcg-runtime.h @@ -225,6 +225,11 @@ DEF_HELPER_FLAGS_3(gvec_neg16, TCG_CALL_NO_RWG, void, ptr, ptr, i32) DEF_HELPER_FLAGS_3(gvec_neg32, TCG_CALL_NO_RWG, void, ptr, ptr, i32) DEF_HELPER_FLAGS_3(gvec_neg64, TCG_CALL_NO_RWG, void, ptr, ptr, i32) +DEF_HELPER_FLAGS_3(gvec_abs8, TCG_CALL_NO_RWG, void, ptr, ptr, i32) +DEF_HELPER_FLAGS_3(gvec_abs16, TCG_CALL_NO_RWG, void, ptr, ptr, i32) +DEF_HELPER_FLAGS_3(gvec_abs32, TCG_CALL_NO_RWG, void, ptr, ptr, i32) +DEF_HELPER_FLAGS_3(gvec_abs64, TCG_CALL_NO_RWG, void, ptr, ptr, i32) + DEF_HELPER_FLAGS_3(gvec_not, TCG_CALL_NO_RWG, void, ptr, ptr, i32) DEF_HELPER_FLAGS_4(gvec_and, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i32) DEF_HELPER_FLAGS_4(gvec_or, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i32) diff --git a/tcg/README b/tcg/README index c30e5418a6..cbdfd3b6bc 100644 --- a/tcg/README +++ b/tcg/README @@ -561,6 +561,10 @@ E.g. VECL=1 -> 64 << 1 -> v128, and VECE=2 -> 1 << 2 -> i32. Similarly, v0 = -v1. +* abs_vec v0, v1 + + Similarly, v0 = v1 < 0 ? -v1 : v1, in elements across the vector. + * smin_vec: * umin_vec: diff --git a/tcg/aarch64/tcg-target.h b/tcg/aarch64/tcg-target.h index f5640a229b..21d06d928c 100644 --- a/tcg/aarch64/tcg-target.h +++ b/tcg/aarch64/tcg-target.h @@ -132,6 +132,7 @@ typedef enum { #define TCG_TARGET_HAS_orc_vec 1 #define TCG_TARGET_HAS_not_vec 1 #define TCG_TARGET_HAS_neg_vec 1 +#define TCG_TARGET_HAS_abs_vec 0 #define TCG_TARGET_HAS_shi_vec 1 #define TCG_TARGET_HAS_shs_vec 0 #define TCG_TARGET_HAS_shv_vec 1 diff --git a/tcg/i386/tcg-target.h b/tcg/i386/tcg-target.h index 618aa520d2..7445f05885 100644 --- a/tcg/i386/tcg-target.h +++ b/tcg/i386/tcg-target.h @@ -182,6 +182,7 @@ extern bool have_avx2; #define TCG_TARGET_HAS_orc_vec 0 #define TCG_TARGET_HAS_not_vec 0 #define TCG_TARGET_HAS_neg_vec 0 +#define TCG_TARGET_HAS_abs_vec 0 #define TCG_TARGET_HAS_shi_vec 1 #define TCG_TARGET_HAS_shs_vec 1 #define TCG_TARGET_HAS_shv_vec have_avx2 diff --git a/tcg/tcg-op-gvec.c b/tcg/tcg-op-gvec.c index c69c7960b8..338ddd9d9e 100644 --- a/tcg/tcg-op-gvec.c +++ b/tcg/tcg-op-gvec.c @@ -2177,6 +2177,69 @@ void tcg_gen_gvec_neg(unsigned vece, uint32_t dofs, uint32_t aofs, tcg_gen_gvec_2(dofs, aofs, oprsz, maxsz, &g[vece]); } +static void gen_absv_mask(TCGv_i64 d, TCGv_i64 b, unsigned vece) +{ + TCGv_i64 t = tcg_temp_new_i64(); + int nbit = 8 << vece; + + /* Create -1 for each negative element. */ + tcg_gen_shri_i64(t, b, nbit - 1); + tcg_gen_andi_i64(t, t, dup_const(vece, 1)); + tcg_gen_muli_i64(t, t, (1 << nbit) - 1); + + /* + * Invert (via xor -1) and add one (via sub -1). + * Because of the ordering the msb is cleared, + * so we never have carry into the next element. + */ + tcg_gen_xor_i64(d, b, t); + tcg_gen_sub_i64(d, d, t); + + tcg_temp_free_i64(t); +} + +static void tcg_gen_vec_abs8_i64(TCGv_i64 d, TCGv_i64 b) +{ + gen_absv_mask(d, b, MO_8); +} + +static void tcg_gen_vec_abs16_i64(TCGv_i64 d, TCGv_i64 b) +{ + gen_absv_mask(d, b, MO_16); +} + +void tcg_gen_gvec_abs(unsigned vece, uint32_t dofs, uint32_t aofs, + uint32_t oprsz, uint32_t maxsz) +{ + static const TCGOpcode vecop_list[] = { INDEX_op_abs_vec, 0 }; + static const GVecGen2 g[4] = { + { .fni8 = tcg_gen_vec_abs8_i64, + .fniv = tcg_gen_abs_vec, + .fno = gen_helper_gvec_abs8, + .opt_opc = vecop_list, + .vece = MO_8 }, + { .fni8 = tcg_gen_vec_abs16_i64, + .fniv = tcg_gen_abs_vec, + .fno = gen_helper_gvec_abs16, + .opt_opc = vecop_list, + .vece = MO_16 }, + { .fni4 = tcg_gen_abs_i32, + .fniv = tcg_gen_abs_vec, + .fno = gen_helper_gvec_abs32, + .opt_opc = vecop_list, + .vece = MO_32 }, + { .fni8 = tcg_gen_abs_i64, + .fniv = tcg_gen_abs_vec, + .fno = gen_helper_gvec_abs64, + .opt_opc = vecop_list, + .prefer_i64 = TCG_TARGET_REG_BITS == 64, + .vece = MO_64 }, + }; + + tcg_debug_assert(vece <= MO_64); + tcg_gen_gvec_2(dofs, aofs, oprsz, maxsz, &g[vece]); +} + void tcg_gen_gvec_and(unsigned vece, uint32_t dofs, uint32_t aofs, uint32_t bofs, uint32_t oprsz, uint32_t maxsz) { diff --git a/tcg/tcg-op-gvec.h b/tcg/tcg-op-gvec.h index 6ee98f3378..52a398c190 100644 --- a/tcg/tcg-op-gvec.h +++ b/tcg/tcg-op-gvec.h @@ -228,6 +228,8 @@ void tcg_gen_gvec_not(unsigned vece, uint32_t dofs, uint32_t aofs, uint32_t oprsz, uint32_t maxsz); void tcg_gen_gvec_neg(unsigned vece, uint32_t dofs, uint32_t aofs, uint32_t oprsz, uint32_t maxsz); +void tcg_gen_gvec_abs(unsigned vece, uint32_t dofs, uint32_t aofs, + uint32_t oprsz, uint32_t maxsz); void tcg_gen_gvec_add(unsigned vece, uint32_t dofs, uint32_t aofs, uint32_t bofs, uint32_t oprsz, uint32_t maxsz); diff --git a/tcg/tcg-op-vec.c b/tcg/tcg-op-vec.c index 16062f5995..543508d545 100644 --- a/tcg/tcg-op-vec.c +++ b/tcg/tcg-op-vec.c @@ -110,6 +110,14 @@ bool tcg_can_emit_vecop_list(const TCGOpcode *list, continue; } break; + case INDEX_op_abs_vec: + if (tcg_can_emit_vec_op(INDEX_op_sub_vec, type, vece) + && (tcg_can_emit_vec_op(INDEX_op_smax_vec, type, vece) > 0 + || tcg_can_emit_vec_op(INDEX_op_sari_vec, type, vece) > 0 + || tcg_can_emit_vec_op(INDEX_op_cmp_vec, type, vece))) { + continue; + } + break; default: break; } @@ -429,6 +437,37 @@ void tcg_gen_neg_vec(unsigned vece, TCGv_vec r, TCGv_vec a) tcg_swap_vecop_list(hold_list); } +void tcg_gen_abs_vec(unsigned vece, TCGv_vec r, TCGv_vec a) +{ + const TCGOpcode *hold_list; + + tcg_assert_listed_vecop(INDEX_op_abs_vec); + hold_list = tcg_swap_vecop_list(NULL); + + if (!do_op2(vece, r, a, INDEX_op_abs_vec)) { + TCGType type = tcgv_vec_temp(r)->base_type; + TCGv_vec t = tcg_temp_new_vec(type); + + tcg_debug_assert(tcg_can_emit_vec_op(INDEX_op_sub_vec, type, vece)); + if (tcg_can_emit_vec_op(INDEX_op_smax_vec, type, vece) > 0) { + tcg_gen_neg_vec(vece, t, a); + tcg_gen_smax_vec(vece, r, a, t); + } else { + if (tcg_can_emit_vec_op(INDEX_op_sari_vec, type, vece) > 0) { + tcg_gen_sari_vec(vece, t, a, (8 << vece) - 1); + } else { + do_dupi_vec(t, MO_REG, 0); + tcg_gen_cmp_vec(TCG_COND_LT, vece, t, a, t); + } + tcg_gen_xor_vec(vece, r, a, t); + tcg_gen_sub_vec(vece, r, r, t); + } + + tcg_temp_free_vec(t); + } + tcg_swap_vecop_list(hold_list); +} + static void do_shifti(TCGOpcode opc, unsigned vece, TCGv_vec r, TCGv_vec a, int64_t i) { diff --git a/tcg/tcg-opc.h b/tcg/tcg-opc.h index 4bf71f261f..4a2dd116eb 100644 --- a/tcg/tcg-opc.h +++ b/tcg/tcg-opc.h @@ -225,6 +225,7 @@ DEF(add_vec, 1, 2, 0, IMPLVEC) DEF(sub_vec, 1, 2, 0, IMPLVEC) DEF(mul_vec, 1, 2, 0, IMPLVEC | IMPL(TCG_TARGET_HAS_mul_vec)) DEF(neg_vec, 1, 1, 0, IMPLVEC | IMPL(TCG_TARGET_HAS_neg_vec)) +DEF(abs_vec, 1, 1, 0, IMPLVEC | IMPL(TCG_TARGET_HAS_abs_vec)) DEF(ssadd_vec, 1, 2, 0, IMPLVEC | IMPL(TCG_TARGET_HAS_sat_vec)) DEF(usadd_vec, 1, 2, 0, IMPLVEC | IMPL(TCG_TARGET_HAS_sat_vec)) DEF(sssub_vec, 1, 2, 0, IMPLVEC | IMPL(TCG_TARGET_HAS_sat_vec)) diff --git a/tcg/tcg.c b/tcg/tcg.c index 3b80feb344..24083b8c00 100644 --- a/tcg/tcg.c +++ b/tcg/tcg.c @@ -1616,6 +1616,8 @@ bool tcg_op_supported(TCGOpcode op) return have_vec && TCG_TARGET_HAS_not_vec; case INDEX_op_neg_vec: return have_vec && TCG_TARGET_HAS_neg_vec; + case INDEX_op_abs_vec: + return have_vec && TCG_TARGET_HAS_abs_vec; case INDEX_op_andc_vec: return have_vec && TCG_TARGET_HAS_andc_vec; case INDEX_op_orc_vec: diff --git a/tcg/tcg.h b/tcg/tcg.h index 2c7315da25..0e01a70d66 100644 --- a/tcg/tcg.h +++ b/tcg/tcg.h @@ -176,6 +176,7 @@ typedef uint64_t TCGRegSet; && !defined(TCG_TARGET_HAS_v128) \ && !defined(TCG_TARGET_HAS_v256) #define TCG_TARGET_MAYBE_vec 0 +#define TCG_TARGET_HAS_abs_vec 0 #define TCG_TARGET_HAS_neg_vec 0 #define TCG_TARGET_HAS_not_vec 0 #define TCG_TARGET_HAS_andc_vec 0