* This library is free software; you can redistribute it and/or
* modify it under the terms of the GNU Lesser General Public
* License as published by the Free Software Foundation; either
- * version 2 of the License, or (at your option) any later version.
+ * version 2.1 of the License, or (at your option) any later version.
*
* This library is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
*/
#include "qemu/osdep.h"
-#include "qemu-common.h"
#include "tcg.h"
#include "tcg-op.h"
#include "tcg-op-gvec.h"
+#include "qemu/main-loop.h"
#include "tcg-gvec-desc.h"
#define MAX_UNROLL 4
+#ifdef CONFIG_DEBUG_TCG
+static const TCGOpcode vecop_list_empty[1] = { 0 };
+#else
+#define vecop_list_empty NULL
+#endif
+
+
/* Verify vector size and alignment rules. OFS should be the OR of all
of the operand offsets so that we can check them all at once. */
static void check_size_align(uint32_t oprsz, uint32_t maxsz, uint32_t ofs)
tcg_temp_free_i32(desc);
}
+/* Generate a call to a gvec-style helper with two vector operands
+ and one scalar operand. */
+void tcg_gen_gvec_2i_ool(uint32_t dofs, uint32_t aofs, TCGv_i64 c,
+ uint32_t oprsz, uint32_t maxsz, int32_t data,
+ gen_helper_gvec_2i *fn)
+{
+ TCGv_ptr a0, a1;
+ TCGv_i32 desc = tcg_const_i32(simd_desc(oprsz, maxsz, data));
+
+ a0 = tcg_temp_new_ptr();
+ a1 = tcg_temp_new_ptr();
+
+ tcg_gen_addi_ptr(a0, cpu_env, dofs);
+ tcg_gen_addi_ptr(a1, cpu_env, aofs);
+
+ fn(a0, a1, c, desc);
+
+ tcg_temp_free_ptr(a0);
+ tcg_temp_free_ptr(a1);
+ tcg_temp_free_i32(desc);
+}
+
/* Generate a call to a gvec-style helper with three vector operands. */
void tcg_gen_gvec_3_ool(uint32_t dofs, uint32_t aofs, uint32_t bofs,
uint32_t oprsz, uint32_t maxsz, int32_t data,
in units of LNSZ. This limits the expansion of inline code. */
static inline bool check_size_impl(uint32_t oprsz, uint32_t lnsz)
{
- uint32_t lnct = oprsz / lnsz;
- return lnct >= 1 && lnct <= MAX_UNROLL;
+ if (oprsz % lnsz == 0) {
+ uint32_t lnct = oprsz / lnsz;
+ return lnct >= 1 && lnct <= MAX_UNROLL;
+ }
+ return false;
}
static void expand_clr(uint32_t dofs, uint32_t maxsz);
}
}
+/* Select a supported vector type for implementing an operation on SIZE
+ * bytes. If OP is 0, assume that the real operation to be performed is
+ * required by all backends. Otherwise, make sure than OP can be performed
+ * on elements of size VECE in the selected type. Do not select V64 if
+ * PREFER_I64 is true. Return 0 if no vector type is selected.
+ */
+static TCGType choose_vector_type(const TCGOpcode *list, unsigned vece,
+ uint32_t size, bool prefer_i64)
+{
+ if (TCG_TARGET_HAS_v256 && check_size_impl(size, 32)) {
+ /*
+ * Recall that ARM SVE allows vector sizes that are not a
+ * power of 2, but always a multiple of 16. The intent is
+ * that e.g. size == 80 would be expanded with 2x32 + 1x16.
+ * It is hard to imagine a case in which v256 is supported
+ * but v128 is not, but check anyway.
+ */
+ if (tcg_can_emit_vecop_list(list, TCG_TYPE_V256, vece)
+ && (size % 32 == 0
+ || tcg_can_emit_vecop_list(list, TCG_TYPE_V128, vece))) {
+ return TCG_TYPE_V256;
+ }
+ }
+ if (TCG_TARGET_HAS_v128 && check_size_impl(size, 16)
+ && tcg_can_emit_vecop_list(list, TCG_TYPE_V128, vece)) {
+ return TCG_TYPE_V128;
+ }
+ if (TCG_TARGET_HAS_v64 && !prefer_i64 && check_size_impl(size, 8)
+ && tcg_can_emit_vecop_list(list, TCG_TYPE_V64, vece)) {
+ return TCG_TYPE_V64;
+ }
+ return 0;
+}
+
+static void do_dup_store(TCGType type, uint32_t dofs, uint32_t oprsz,
+ uint32_t maxsz, TCGv_vec t_vec)
+{
+ uint32_t i = 0;
+
+ switch (type) {
+ case TCG_TYPE_V256:
+ /*
+ * Recall that ARM SVE allows vector sizes that are not a
+ * power of 2, but always a multiple of 16. The intent is
+ * that e.g. size == 80 would be expanded with 2x32 + 1x16.
+ */
+ for (; i + 32 <= oprsz; i += 32) {
+ tcg_gen_stl_vec(t_vec, cpu_env, dofs + i, TCG_TYPE_V256);
+ }
+ /* fallthru */
+ case TCG_TYPE_V128:
+ for (; i + 16 <= oprsz; i += 16) {
+ tcg_gen_stl_vec(t_vec, cpu_env, dofs + i, TCG_TYPE_V128);
+ }
+ break;
+ case TCG_TYPE_V64:
+ for (; i < oprsz; i += 8) {
+ tcg_gen_stl_vec(t_vec, cpu_env, dofs + i, TCG_TYPE_V64);
+ }
+ break;
+ default:
+ g_assert_not_reached();
+ }
+
+ if (oprsz < maxsz) {
+ expand_clr(dofs + oprsz, maxsz - oprsz);
+ }
+}
+
/* Set OPRSZ bytes at DOFS to replications of IN_32, IN_64 or IN_C.
* Only one of IN_32 or IN_64 may be set;
* IN_C is used if IN_32 and IN_64 are unset.
}
}
- type = 0;
- if (TCG_TARGET_HAS_v256 && check_size_impl(oprsz, 32)) {
- type = TCG_TYPE_V256;
- } else if (TCG_TARGET_HAS_v128 && check_size_impl(oprsz, 16)) {
- type = TCG_TYPE_V128;
- } else if (TCG_TARGET_HAS_v64 && check_size_impl(oprsz, 8)
- /* Prefer integer when 64-bit host and no variable dup. */
- && !(TCG_TARGET_REG_BITS == 64 && in_32 == NULL
- && (in_64 == NULL || vece == MO_64))) {
- type = TCG_TYPE_V64;
- }
-
- /* Implement inline with a vector type, if possible. */
+ /* Implement inline with a vector type, if possible.
+ * Prefer integer when 64-bit host and no variable dup.
+ */
+ type = choose_vector_type(NULL, vece, oprsz,
+ (TCG_TARGET_REG_BITS == 64 && in_32 == NULL
+ && (in_64 == NULL || vece == MO_64)));
if (type != 0) {
TCGv_vec t_vec = tcg_temp_new_vec(type);
} else if (in_64) {
tcg_gen_dup_i64_vec(vece, t_vec, in_64);
} else {
- switch (vece) {
- case MO_8:
- tcg_gen_dup8i_vec(t_vec, in_c);
- break;
- case MO_16:
- tcg_gen_dup16i_vec(t_vec, in_c);
- break;
- case MO_32:
- tcg_gen_dup32i_vec(t_vec, in_c);
- break;
- default:
- tcg_gen_dup64i_vec(t_vec, in_c);
- break;
- }
- }
-
- i = 0;
- if (TCG_TARGET_HAS_v256) {
- for (; i + 32 <= oprsz; i += 32) {
- tcg_gen_stl_vec(t_vec, cpu_env, dofs + i, TCG_TYPE_V256);
- }
- }
- if (TCG_TARGET_HAS_v128) {
- for (; i + 16 <= oprsz; i += 16) {
- tcg_gen_stl_vec(t_vec, cpu_env, dofs + i, TCG_TYPE_V128);
- }
- }
- if (TCG_TARGET_HAS_v64) {
- for (; i < oprsz; i += 8) {
- tcg_gen_stl_vec(t_vec, cpu_env, dofs + i, TCG_TYPE_V64);
- }
+ tcg_gen_dupi_vec(vece, t_vec, in_c);
}
+ do_dup_store(type, dofs, oprsz, maxsz, t_vec);
tcg_temp_free_vec(t_vec);
- goto done;
+ return;
}
/* Otherwise, inline with an integer type, unless "large". */
}
tcg_temp_free_i64(t_64);
goto done;
- }
+ }
}
/* Otherwise implement out of line. */
tcg_temp_free_i32(t1);
}
+static void expand_2s_i32(uint32_t dofs, uint32_t aofs, uint32_t oprsz,
+ TCGv_i32 c, bool scalar_first,
+ void (*fni)(TCGv_i32, TCGv_i32, TCGv_i32))
+{
+ TCGv_i32 t0 = tcg_temp_new_i32();
+ TCGv_i32 t1 = tcg_temp_new_i32();
+ uint32_t i;
+
+ for (i = 0; i < oprsz; i += 4) {
+ tcg_gen_ld_i32(t0, cpu_env, aofs + i);
+ if (scalar_first) {
+ fni(t1, c, t0);
+ } else {
+ fni(t1, t0, c);
+ }
+ tcg_gen_st_i32(t1, cpu_env, dofs + i);
+ }
+ tcg_temp_free_i32(t0);
+ tcg_temp_free_i32(t1);
+}
+
/* Expand OPSZ bytes worth of three-operand operations using i32 elements. */
static void expand_3_i32(uint32_t dofs, uint32_t aofs,
uint32_t bofs, uint32_t oprsz, bool load_dest,
tcg_temp_free_i32(t0);
}
+static void expand_3i_i32(uint32_t dofs, uint32_t aofs, uint32_t bofs,
+ uint32_t oprsz, int32_t c, bool load_dest,
+ void (*fni)(TCGv_i32, TCGv_i32, TCGv_i32, int32_t))
+{
+ TCGv_i32 t0 = tcg_temp_new_i32();
+ TCGv_i32 t1 = tcg_temp_new_i32();
+ TCGv_i32 t2 = tcg_temp_new_i32();
+ uint32_t i;
+
+ for (i = 0; i < oprsz; i += 4) {
+ tcg_gen_ld_i32(t0, cpu_env, aofs + i);
+ tcg_gen_ld_i32(t1, cpu_env, bofs + i);
+ if (load_dest) {
+ tcg_gen_ld_i32(t2, cpu_env, dofs + i);
+ }
+ fni(t2, t0, t1, c);
+ tcg_gen_st_i32(t2, cpu_env, dofs + i);
+ }
+ tcg_temp_free_i32(t0);
+ tcg_temp_free_i32(t1);
+ tcg_temp_free_i32(t2);
+}
+
/* Expand OPSZ bytes worth of three-operand operations using i32 elements. */
static void expand_4_i32(uint32_t dofs, uint32_t aofs, uint32_t bofs,
- uint32_t cofs, uint32_t oprsz,
+ uint32_t cofs, uint32_t oprsz, bool write_aofs,
void (*fni)(TCGv_i32, TCGv_i32, TCGv_i32, TCGv_i32))
{
TCGv_i32 t0 = tcg_temp_new_i32();
tcg_gen_ld_i32(t3, cpu_env, cofs + i);
fni(t0, t1, t2, t3);
tcg_gen_st_i32(t0, cpu_env, dofs + i);
+ if (write_aofs) {
+ tcg_gen_st_i32(t1, cpu_env, aofs + i);
+ }
}
tcg_temp_free_i32(t3);
tcg_temp_free_i32(t2);
tcg_temp_free_i64(t1);
}
+static void expand_2s_i64(uint32_t dofs, uint32_t aofs, uint32_t oprsz,
+ TCGv_i64 c, bool scalar_first,
+ void (*fni)(TCGv_i64, TCGv_i64, TCGv_i64))
+{
+ TCGv_i64 t0 = tcg_temp_new_i64();
+ TCGv_i64 t1 = tcg_temp_new_i64();
+ uint32_t i;
+
+ for (i = 0; i < oprsz; i += 8) {
+ tcg_gen_ld_i64(t0, cpu_env, aofs + i);
+ if (scalar_first) {
+ fni(t1, c, t0);
+ } else {
+ fni(t1, t0, c);
+ }
+ tcg_gen_st_i64(t1, cpu_env, dofs + i);
+ }
+ tcg_temp_free_i64(t0);
+ tcg_temp_free_i64(t1);
+}
+
/* Expand OPSZ bytes worth of three-operand operations using i64 elements. */
static void expand_3_i64(uint32_t dofs, uint32_t aofs,
uint32_t bofs, uint32_t oprsz, bool load_dest,
tcg_temp_free_i64(t0);
}
+static void expand_3i_i64(uint32_t dofs, uint32_t aofs, uint32_t bofs,
+ uint32_t oprsz, int64_t c, bool load_dest,
+ void (*fni)(TCGv_i64, TCGv_i64, TCGv_i64, int64_t))
+{
+ TCGv_i64 t0 = tcg_temp_new_i64();
+ TCGv_i64 t1 = tcg_temp_new_i64();
+ TCGv_i64 t2 = tcg_temp_new_i64();
+ uint32_t i;
+
+ for (i = 0; i < oprsz; i += 8) {
+ tcg_gen_ld_i64(t0, cpu_env, aofs + i);
+ tcg_gen_ld_i64(t1, cpu_env, bofs + i);
+ if (load_dest) {
+ tcg_gen_ld_i64(t2, cpu_env, dofs + i);
+ }
+ fni(t2, t0, t1, c);
+ tcg_gen_st_i64(t2, cpu_env, dofs + i);
+ }
+ tcg_temp_free_i64(t0);
+ tcg_temp_free_i64(t1);
+ tcg_temp_free_i64(t2);
+}
+
/* Expand OPSZ bytes worth of three-operand operations using i64 elements. */
static void expand_4_i64(uint32_t dofs, uint32_t aofs, uint32_t bofs,
- uint32_t cofs, uint32_t oprsz,
+ uint32_t cofs, uint32_t oprsz, bool write_aofs,
void (*fni)(TCGv_i64, TCGv_i64, TCGv_i64, TCGv_i64))
{
TCGv_i64 t0 = tcg_temp_new_i64();
tcg_gen_ld_i64(t3, cpu_env, cofs + i);
fni(t0, t1, t2, t3);
tcg_gen_st_i64(t0, cpu_env, dofs + i);
+ if (write_aofs) {
+ tcg_gen_st_i64(t1, cpu_env, aofs + i);
+ }
}
tcg_temp_free_i64(t3);
tcg_temp_free_i64(t2);
tcg_temp_free_vec(t1);
}
+static void expand_2s_vec(unsigned vece, uint32_t dofs, uint32_t aofs,
+ uint32_t oprsz, uint32_t tysz, TCGType type,
+ TCGv_vec c, bool scalar_first,
+ void (*fni)(unsigned, TCGv_vec, TCGv_vec, TCGv_vec))
+{
+ TCGv_vec t0 = tcg_temp_new_vec(type);
+ TCGv_vec t1 = tcg_temp_new_vec(type);
+ uint32_t i;
+
+ for (i = 0; i < oprsz; i += tysz) {
+ tcg_gen_ld_vec(t0, cpu_env, aofs + i);
+ if (scalar_first) {
+ fni(vece, t1, c, t0);
+ } else {
+ fni(vece, t1, t0, c);
+ }
+ tcg_gen_st_vec(t1, cpu_env, dofs + i);
+ }
+ tcg_temp_free_vec(t0);
+ tcg_temp_free_vec(t1);
+}
+
/* Expand OPSZ bytes worth of three-operand operations using host vectors. */
static void expand_3_vec(unsigned vece, uint32_t dofs, uint32_t aofs,
uint32_t bofs, uint32_t oprsz,
tcg_temp_free_vec(t0);
}
+/*
+ * Expand OPSZ bytes worth of three-vector operands and an immediate operand
+ * using host vectors.
+ */
+static void expand_3i_vec(unsigned vece, uint32_t dofs, uint32_t aofs,
+ uint32_t bofs, uint32_t oprsz, uint32_t tysz,
+ TCGType type, int64_t c, bool load_dest,
+ void (*fni)(unsigned, TCGv_vec, TCGv_vec, TCGv_vec,
+ int64_t))
+{
+ TCGv_vec t0 = tcg_temp_new_vec(type);
+ TCGv_vec t1 = tcg_temp_new_vec(type);
+ TCGv_vec t2 = tcg_temp_new_vec(type);
+ uint32_t i;
+
+ for (i = 0; i < oprsz; i += tysz) {
+ tcg_gen_ld_vec(t0, cpu_env, aofs + i);
+ tcg_gen_ld_vec(t1, cpu_env, bofs + i);
+ if (load_dest) {
+ tcg_gen_ld_vec(t2, cpu_env, dofs + i);
+ }
+ fni(vece, t2, t0, t1, c);
+ tcg_gen_st_vec(t2, cpu_env, dofs + i);
+ }
+ tcg_temp_free_vec(t0);
+ tcg_temp_free_vec(t1);
+ tcg_temp_free_vec(t2);
+}
+
/* Expand OPSZ bytes worth of four-operand operations using host vectors. */
static void expand_4_vec(unsigned vece, uint32_t dofs, uint32_t aofs,
uint32_t bofs, uint32_t cofs, uint32_t oprsz,
- uint32_t tysz, TCGType type,
+ uint32_t tysz, TCGType type, bool write_aofs,
void (*fni)(unsigned, TCGv_vec, TCGv_vec,
TCGv_vec, TCGv_vec))
{
tcg_gen_ld_vec(t3, cpu_env, cofs + i);
fni(vece, t0, t1, t2, t3);
tcg_gen_st_vec(t0, cpu_env, dofs + i);
+ if (write_aofs) {
+ tcg_gen_st_vec(t1, cpu_env, aofs + i);
+ }
}
tcg_temp_free_vec(t3);
tcg_temp_free_vec(t2);
void tcg_gen_gvec_2(uint32_t dofs, uint32_t aofs,
uint32_t oprsz, uint32_t maxsz, const GVecGen2 *g)
{
+ const TCGOpcode *this_list = g->opt_opc ? : vecop_list_empty;
+ const TCGOpcode *hold_list = tcg_swap_vecop_list(this_list);
+ TCGType type;
+ uint32_t some;
+
check_size_align(oprsz, maxsz, dofs | aofs);
check_overlap_2(dofs, aofs, maxsz);
- /* Recall that ARM SVE allows vector sizes that are not a power of 2.
- Expand with successively smaller host vector sizes. The intent is
- that e.g. oprsz == 80 would be expanded with 2x32 + 1x16. */
- /* ??? For maxsz > oprsz, the host may be able to use an opr-sized
- operation, zeroing the balance of the register. We can then
- use a max-sized store to implement the clearing without an extra
- store operation. This is true for aarch64 and x86_64 hosts. */
-
- if (TCG_TARGET_HAS_v256 && g->fniv && check_size_impl(oprsz, 32)
- && (!g->opc || tcg_can_emit_vec_op(g->opc, TCG_TYPE_V256, g->vece))) {
- uint32_t some = QEMU_ALIGN_DOWN(oprsz, 32);
+ type = 0;
+ if (g->fniv) {
+ type = choose_vector_type(g->opt_opc, g->vece, oprsz, g->prefer_i64);
+ }
+ switch (type) {
+ case TCG_TYPE_V256:
+ /* Recall that ARM SVE allows vector sizes that are not a
+ * power of 2, but always a multiple of 16. The intent is
+ * that e.g. size == 80 would be expanded with 2x32 + 1x16.
+ */
+ some = QEMU_ALIGN_DOWN(oprsz, 32);
expand_2_vec(g->vece, dofs, aofs, some, 32, TCG_TYPE_V256, g->fniv);
if (some == oprsz) {
- goto done;
+ break;
}
dofs += some;
aofs += some;
oprsz -= some;
maxsz -= some;
- }
-
- if (TCG_TARGET_HAS_v128 && g->fniv && check_size_impl(oprsz, 16)
- && (!g->opc || tcg_can_emit_vec_op(g->opc, TCG_TYPE_V128, g->vece))) {
+ /* fallthru */
+ case TCG_TYPE_V128:
expand_2_vec(g->vece, dofs, aofs, oprsz, 16, TCG_TYPE_V128, g->fniv);
- } else if (TCG_TARGET_HAS_v64 && !g->prefer_i64
- && g->fniv && check_size_impl(oprsz, 8)
- && (!g->opc
- || tcg_can_emit_vec_op(g->opc, TCG_TYPE_V64, g->vece))) {
+ break;
+ case TCG_TYPE_V64:
expand_2_vec(g->vece, dofs, aofs, oprsz, 8, TCG_TYPE_V64, g->fniv);
- } else if (g->fni8 && check_size_impl(oprsz, 8)) {
- expand_2_i64(dofs, aofs, oprsz, g->fni8);
- } else if (g->fni4 && check_size_impl(oprsz, 4)) {
- expand_2_i32(dofs, aofs, oprsz, g->fni4);
- } else {
- assert(g->fno != NULL);
- tcg_gen_gvec_2_ool(dofs, aofs, oprsz, maxsz, g->data, g->fno);
- return;
+ break;
+
+ case 0:
+ if (g->fni8 && check_size_impl(oprsz, 8)) {
+ expand_2_i64(dofs, aofs, oprsz, g->fni8);
+ } else if (g->fni4 && check_size_impl(oprsz, 4)) {
+ expand_2_i32(dofs, aofs, oprsz, g->fni4);
+ } else {
+ assert(g->fno != NULL);
+ tcg_gen_gvec_2_ool(dofs, aofs, oprsz, maxsz, g->data, g->fno);
+ oprsz = maxsz;
+ }
+ break;
+
+ default:
+ g_assert_not_reached();
}
+ tcg_swap_vecop_list(hold_list);
- done:
if (oprsz < maxsz) {
expand_clr(dofs + oprsz, maxsz - oprsz);
}
}
+/* Expand a vector operation with two vectors and an immediate. */
void tcg_gen_gvec_2i(uint32_t dofs, uint32_t aofs, uint32_t oprsz,
uint32_t maxsz, int64_t c, const GVecGen2i *g)
{
+ const TCGOpcode *this_list = g->opt_opc ? : vecop_list_empty;
+ const TCGOpcode *hold_list = tcg_swap_vecop_list(this_list);
+ TCGType type;
+ uint32_t some;
+
check_size_align(oprsz, maxsz, dofs | aofs);
check_overlap_2(dofs, aofs, maxsz);
- /* Recall that ARM SVE allows vector sizes that are not a power of 2.
- Expand with successively smaller host vector sizes. The intent is
- that e.g. oprsz == 80 would be expanded with 2x32 + 1x16. */
-
- if (TCG_TARGET_HAS_v256 && g->fniv && check_size_impl(oprsz, 32)
- && (!g->opc || tcg_can_emit_vec_op(g->opc, TCG_TYPE_V256, g->vece))) {
- uint32_t some = QEMU_ALIGN_DOWN(oprsz, 32);
+ type = 0;
+ if (g->fniv) {
+ type = choose_vector_type(g->opt_opc, g->vece, oprsz, g->prefer_i64);
+ }
+ switch (type) {
+ case TCG_TYPE_V256:
+ /* Recall that ARM SVE allows vector sizes that are not a
+ * power of 2, but always a multiple of 16. The intent is
+ * that e.g. size == 80 would be expanded with 2x32 + 1x16.
+ */
+ some = QEMU_ALIGN_DOWN(oprsz, 32);
expand_2i_vec(g->vece, dofs, aofs, some, 32, TCG_TYPE_V256,
c, g->load_dest, g->fniv);
if (some == oprsz) {
- goto done;
+ break;
}
dofs += some;
aofs += some;
oprsz -= some;
maxsz -= some;
- }
-
- if (TCG_TARGET_HAS_v128 && g->fniv && check_size_impl(oprsz, 16)
- && (!g->opc || tcg_can_emit_vec_op(g->opc, TCG_TYPE_V128, g->vece))) {
+ /* fallthru */
+ case TCG_TYPE_V128:
expand_2i_vec(g->vece, dofs, aofs, oprsz, 16, TCG_TYPE_V128,
c, g->load_dest, g->fniv);
- } else if (TCG_TARGET_HAS_v64 && !g->prefer_i64
- && g->fniv && check_size_impl(oprsz, 8)
- && (!g->opc
- || tcg_can_emit_vec_op(g->opc, TCG_TYPE_V64, g->vece))) {
+ break;
+ case TCG_TYPE_V64:
expand_2i_vec(g->vece, dofs, aofs, oprsz, 8, TCG_TYPE_V64,
c, g->load_dest, g->fniv);
+ break;
+
+ case 0:
+ if (g->fni8 && check_size_impl(oprsz, 8)) {
+ expand_2i_i64(dofs, aofs, oprsz, c, g->load_dest, g->fni8);
+ } else if (g->fni4 && check_size_impl(oprsz, 4)) {
+ expand_2i_i32(dofs, aofs, oprsz, c, g->load_dest, g->fni4);
+ } else {
+ if (g->fno) {
+ tcg_gen_gvec_2_ool(dofs, aofs, oprsz, maxsz, c, g->fno);
+ } else {
+ TCGv_i64 tcg_c = tcg_const_i64(c);
+ tcg_gen_gvec_2i_ool(dofs, aofs, tcg_c, oprsz,
+ maxsz, c, g->fnoi);
+ tcg_temp_free_i64(tcg_c);
+ }
+ oprsz = maxsz;
+ }
+ break;
+
+ default:
+ g_assert_not_reached();
+ }
+ tcg_swap_vecop_list(hold_list);
+
+ if (oprsz < maxsz) {
+ expand_clr(dofs + oprsz, maxsz - oprsz);
+ }
+}
+
+/* Expand a vector operation with two vectors and a scalar. */
+void tcg_gen_gvec_2s(uint32_t dofs, uint32_t aofs, uint32_t oprsz,
+ uint32_t maxsz, TCGv_i64 c, const GVecGen2s *g)
+{
+ TCGType type;
+
+ check_size_align(oprsz, maxsz, dofs | aofs);
+ check_overlap_2(dofs, aofs, maxsz);
+
+ type = 0;
+ if (g->fniv) {
+ type = choose_vector_type(g->opt_opc, g->vece, oprsz, g->prefer_i64);
+ }
+ if (type != 0) {
+ const TCGOpcode *this_list = g->opt_opc ? : vecop_list_empty;
+ const TCGOpcode *hold_list = tcg_swap_vecop_list(this_list);
+ TCGv_vec t_vec = tcg_temp_new_vec(type);
+ uint32_t some;
+
+ tcg_gen_dup_i64_vec(g->vece, t_vec, c);
+
+ switch (type) {
+ case TCG_TYPE_V256:
+ /* Recall that ARM SVE allows vector sizes that are not a
+ * power of 2, but always a multiple of 16. The intent is
+ * that e.g. size == 80 would be expanded with 2x32 + 1x16.
+ */
+ some = QEMU_ALIGN_DOWN(oprsz, 32);
+ expand_2s_vec(g->vece, dofs, aofs, some, 32, TCG_TYPE_V256,
+ t_vec, g->scalar_first, g->fniv);
+ if (some == oprsz) {
+ break;
+ }
+ dofs += some;
+ aofs += some;
+ oprsz -= some;
+ maxsz -= some;
+ /* fallthru */
+
+ case TCG_TYPE_V128:
+ expand_2s_vec(g->vece, dofs, aofs, oprsz, 16, TCG_TYPE_V128,
+ t_vec, g->scalar_first, g->fniv);
+ break;
+
+ case TCG_TYPE_V64:
+ expand_2s_vec(g->vece, dofs, aofs, oprsz, 8, TCG_TYPE_V64,
+ t_vec, g->scalar_first, g->fniv);
+ break;
+
+ default:
+ g_assert_not_reached();
+ }
+ tcg_temp_free_vec(t_vec);
+ tcg_swap_vecop_list(hold_list);
} else if (g->fni8 && check_size_impl(oprsz, 8)) {
- expand_2i_i64(dofs, aofs, oprsz, c, g->load_dest, g->fni8);
+ TCGv_i64 t64 = tcg_temp_new_i64();
+
+ gen_dup_i64(g->vece, t64, c);
+ expand_2s_i64(dofs, aofs, oprsz, t64, g->scalar_first, g->fni8);
+ tcg_temp_free_i64(t64);
} else if (g->fni4 && check_size_impl(oprsz, 4)) {
- expand_2i_i32(dofs, aofs, oprsz, c, g->load_dest, g->fni4);
+ TCGv_i32 t32 = tcg_temp_new_i32();
+
+ tcg_gen_extrl_i64_i32(t32, c);
+ gen_dup_i32(g->vece, t32, t32);
+ expand_2s_i32(dofs, aofs, oprsz, t32, g->scalar_first, g->fni4);
+ tcg_temp_free_i32(t32);
} else {
- tcg_gen_gvec_2_ool(dofs, aofs, oprsz, maxsz, c, g->fno);
+ tcg_gen_gvec_2i_ool(dofs, aofs, c, oprsz, maxsz, 0, g->fno);
return;
}
- done:
if (oprsz < maxsz) {
expand_clr(dofs + oprsz, maxsz - oprsz);
}
void tcg_gen_gvec_3(uint32_t dofs, uint32_t aofs, uint32_t bofs,
uint32_t oprsz, uint32_t maxsz, const GVecGen3 *g)
{
+ const TCGOpcode *this_list = g->opt_opc ? : vecop_list_empty;
+ const TCGOpcode *hold_list = tcg_swap_vecop_list(this_list);
+ TCGType type;
+ uint32_t some;
+
check_size_align(oprsz, maxsz, dofs | aofs | bofs);
check_overlap_3(dofs, aofs, bofs, maxsz);
- /* Recall that ARM SVE allows vector sizes that are not a power of 2.
- Expand with successively smaller host vector sizes. The intent is
- that e.g. oprsz == 80 would be expanded with 2x32 + 1x16. */
-
- if (TCG_TARGET_HAS_v256 && g->fniv && check_size_impl(oprsz, 32)
- && (!g->opc || tcg_can_emit_vec_op(g->opc, TCG_TYPE_V256, g->vece))) {
- uint32_t some = QEMU_ALIGN_DOWN(oprsz, 32);
+ type = 0;
+ if (g->fniv) {
+ type = choose_vector_type(g->opt_opc, g->vece, oprsz, g->prefer_i64);
+ }
+ switch (type) {
+ case TCG_TYPE_V256:
+ /* Recall that ARM SVE allows vector sizes that are not a
+ * power of 2, but always a multiple of 16. The intent is
+ * that e.g. size == 80 would be expanded with 2x32 + 1x16.
+ */
+ some = QEMU_ALIGN_DOWN(oprsz, 32);
expand_3_vec(g->vece, dofs, aofs, bofs, some, 32, TCG_TYPE_V256,
g->load_dest, g->fniv);
if (some == oprsz) {
- goto done;
+ break;
}
dofs += some;
aofs += some;
bofs += some;
oprsz -= some;
maxsz -= some;
- }
-
- if (TCG_TARGET_HAS_v128 && g->fniv && check_size_impl(oprsz, 16)
- && (!g->opc || tcg_can_emit_vec_op(g->opc, TCG_TYPE_V128, g->vece))) {
+ /* fallthru */
+ case TCG_TYPE_V128:
expand_3_vec(g->vece, dofs, aofs, bofs, oprsz, 16, TCG_TYPE_V128,
g->load_dest, g->fniv);
- } else if (TCG_TARGET_HAS_v64 && !g->prefer_i64
- && g->fniv && check_size_impl(oprsz, 8)
- && (!g->opc
- || tcg_can_emit_vec_op(g->opc, TCG_TYPE_V64, g->vece))) {
+ break;
+ case TCG_TYPE_V64:
expand_3_vec(g->vece, dofs, aofs, bofs, oprsz, 8, TCG_TYPE_V64,
g->load_dest, g->fniv);
- } else if (g->fni8 && check_size_impl(oprsz, 8)) {
- expand_3_i64(dofs, aofs, bofs, oprsz, g->load_dest, g->fni8);
- } else if (g->fni4 && check_size_impl(oprsz, 4)) {
- expand_3_i32(dofs, aofs, bofs, oprsz, g->load_dest, g->fni4);
- } else {
- assert(g->fno != NULL);
- tcg_gen_gvec_3_ool(dofs, aofs, bofs, oprsz, maxsz, g->data, g->fno);
- }
+ break;
+
+ case 0:
+ if (g->fni8 && check_size_impl(oprsz, 8)) {
+ expand_3_i64(dofs, aofs, bofs, oprsz, g->load_dest, g->fni8);
+ } else if (g->fni4 && check_size_impl(oprsz, 4)) {
+ expand_3_i32(dofs, aofs, bofs, oprsz, g->load_dest, g->fni4);
+ } else {
+ assert(g->fno != NULL);
+ tcg_gen_gvec_3_ool(dofs, aofs, bofs, oprsz,
+ maxsz, g->data, g->fno);
+ oprsz = maxsz;
+ }
+ break;
+
+ default:
+ g_assert_not_reached();
+ }
+ tcg_swap_vecop_list(hold_list);
+
+ if (oprsz < maxsz) {
+ expand_clr(dofs + oprsz, maxsz - oprsz);
+ }
+}
+
+/* Expand a vector operation with three vectors and an immediate. */
+void tcg_gen_gvec_3i(uint32_t dofs, uint32_t aofs, uint32_t bofs,
+ uint32_t oprsz, uint32_t maxsz, int64_t c,
+ const GVecGen3i *g)
+{
+ const TCGOpcode *this_list = g->opt_opc ? : vecop_list_empty;
+ const TCGOpcode *hold_list = tcg_swap_vecop_list(this_list);
+ TCGType type;
+ uint32_t some;
+
+ check_size_align(oprsz, maxsz, dofs | aofs | bofs);
+ check_overlap_3(dofs, aofs, bofs, maxsz);
+
+ type = 0;
+ if (g->fniv) {
+ type = choose_vector_type(g->opt_opc, g->vece, oprsz, g->prefer_i64);
+ }
+ switch (type) {
+ case TCG_TYPE_V256:
+ /*
+ * Recall that ARM SVE allows vector sizes that are not a
+ * power of 2, but always a multiple of 16. The intent is
+ * that e.g. size == 80 would be expanded with 2x32 + 1x16.
+ */
+ some = QEMU_ALIGN_DOWN(oprsz, 32);
+ expand_3i_vec(g->vece, dofs, aofs, bofs, some, 32, TCG_TYPE_V256,
+ c, g->load_dest, g->fniv);
+ if (some == oprsz) {
+ break;
+ }
+ dofs += some;
+ aofs += some;
+ bofs += some;
+ oprsz -= some;
+ maxsz -= some;
+ /* fallthru */
+ case TCG_TYPE_V128:
+ expand_3i_vec(g->vece, dofs, aofs, bofs, oprsz, 16, TCG_TYPE_V128,
+ c, g->load_dest, g->fniv);
+ break;
+ case TCG_TYPE_V64:
+ expand_3i_vec(g->vece, dofs, aofs, bofs, oprsz, 8, TCG_TYPE_V64,
+ c, g->load_dest, g->fniv);
+ break;
+
+ case 0:
+ if (g->fni8 && check_size_impl(oprsz, 8)) {
+ expand_3i_i64(dofs, aofs, bofs, oprsz, c, g->load_dest, g->fni8);
+ } else if (g->fni4 && check_size_impl(oprsz, 4)) {
+ expand_3i_i32(dofs, aofs, bofs, oprsz, c, g->load_dest, g->fni4);
+ } else {
+ assert(g->fno != NULL);
+ tcg_gen_gvec_3_ool(dofs, aofs, bofs, oprsz, maxsz, c, g->fno);
+ oprsz = maxsz;
+ }
+ break;
+
+ default:
+ g_assert_not_reached();
+ }
+ tcg_swap_vecop_list(hold_list);
- done:
if (oprsz < maxsz) {
expand_clr(dofs + oprsz, maxsz - oprsz);
}
void tcg_gen_gvec_4(uint32_t dofs, uint32_t aofs, uint32_t bofs, uint32_t cofs,
uint32_t oprsz, uint32_t maxsz, const GVecGen4 *g)
{
+ const TCGOpcode *this_list = g->opt_opc ? : vecop_list_empty;
+ const TCGOpcode *hold_list = tcg_swap_vecop_list(this_list);
+ TCGType type;
+ uint32_t some;
+
check_size_align(oprsz, maxsz, dofs | aofs | bofs | cofs);
check_overlap_4(dofs, aofs, bofs, cofs, maxsz);
- /* Recall that ARM SVE allows vector sizes that are not a power of 2.
- Expand with successively smaller host vector sizes. The intent is
- that e.g. oprsz == 80 would be expanded with 2x32 + 1x16. */
-
- if (TCG_TARGET_HAS_v256 && g->fniv && check_size_impl(oprsz, 32)
- && (!g->opc || tcg_can_emit_vec_op(g->opc, TCG_TYPE_V256, g->vece))) {
- uint32_t some = QEMU_ALIGN_DOWN(oprsz, 32);
+ type = 0;
+ if (g->fniv) {
+ type = choose_vector_type(g->opt_opc, g->vece, oprsz, g->prefer_i64);
+ }
+ switch (type) {
+ case TCG_TYPE_V256:
+ /* Recall that ARM SVE allows vector sizes that are not a
+ * power of 2, but always a multiple of 16. The intent is
+ * that e.g. size == 80 would be expanded with 2x32 + 1x16.
+ */
+ some = QEMU_ALIGN_DOWN(oprsz, 32);
expand_4_vec(g->vece, dofs, aofs, bofs, cofs, some,
- 32, TCG_TYPE_V256, g->fniv);
+ 32, TCG_TYPE_V256, g->write_aofs, g->fniv);
if (some == oprsz) {
- goto done;
+ break;
}
dofs += some;
aofs += some;
cofs += some;
oprsz -= some;
maxsz -= some;
- }
-
- if (TCG_TARGET_HAS_v128 && g->fniv && check_size_impl(oprsz, 16)
- && (!g->opc || tcg_can_emit_vec_op(g->opc, TCG_TYPE_V128, g->vece))) {
+ /* fallthru */
+ case TCG_TYPE_V128:
expand_4_vec(g->vece, dofs, aofs, bofs, cofs, oprsz,
- 16, TCG_TYPE_V128, g->fniv);
- } else if (TCG_TARGET_HAS_v64 && !g->prefer_i64
- && g->fniv && check_size_impl(oprsz, 8)
- && (!g->opc
- || tcg_can_emit_vec_op(g->opc, TCG_TYPE_V64, g->vece))) {
+ 16, TCG_TYPE_V128, g->write_aofs, g->fniv);
+ break;
+ case TCG_TYPE_V64:
expand_4_vec(g->vece, dofs, aofs, bofs, cofs, oprsz,
- 8, TCG_TYPE_V64, g->fniv);
- } else if (g->fni8 && check_size_impl(oprsz, 8)) {
- expand_4_i64(dofs, aofs, bofs, cofs, oprsz, g->fni8);
- } else if (g->fni4 && check_size_impl(oprsz, 4)) {
- expand_4_i32(dofs, aofs, bofs, cofs, oprsz, g->fni4);
- } else {
- assert(g->fno != NULL);
- tcg_gen_gvec_4_ool(dofs, aofs, bofs, cofs,
- oprsz, maxsz, g->data, g->fno);
- return;
+ 8, TCG_TYPE_V64, g->write_aofs, g->fniv);
+ break;
+
+ case 0:
+ if (g->fni8 && check_size_impl(oprsz, 8)) {
+ expand_4_i64(dofs, aofs, bofs, cofs, oprsz,
+ g->write_aofs, g->fni8);
+ } else if (g->fni4 && check_size_impl(oprsz, 4)) {
+ expand_4_i32(dofs, aofs, bofs, cofs, oprsz,
+ g->write_aofs, g->fni4);
+ } else {
+ assert(g->fno != NULL);
+ tcg_gen_gvec_4_ool(dofs, aofs, bofs, cofs,
+ oprsz, maxsz, g->data, g->fno);
+ oprsz = maxsz;
+ }
+ break;
+
+ default:
+ g_assert_not_reached();
}
+ tcg_swap_vecop_list(hold_list);
- done:
if (oprsz < maxsz) {
expand_clr(dofs + oprsz, maxsz - oprsz);
}
void tcg_gen_gvec_dup_mem(unsigned vece, uint32_t dofs, uint32_t aofs,
uint32_t oprsz, uint32_t maxsz)
{
- if (vece <= MO_32) {
- TCGv_i32 in = tcg_temp_new_i32();
- switch (vece) {
- case MO_8:
- tcg_gen_ld8u_i32(in, cpu_env, aofs);
- break;
- case MO_16:
- tcg_gen_ld16u_i32(in, cpu_env, aofs);
- break;
- case MO_32:
- tcg_gen_ld_i32(in, cpu_env, aofs);
- break;
+ check_size_align(oprsz, maxsz, dofs);
+ if (vece <= MO_64) {
+ TCGType type = choose_vector_type(NULL, vece, oprsz, 0);
+ if (type != 0) {
+ TCGv_vec t_vec = tcg_temp_new_vec(type);
+ tcg_gen_dup_mem_vec(vece, t_vec, cpu_env, aofs);
+ do_dup_store(type, dofs, oprsz, maxsz, t_vec);
+ tcg_temp_free_vec(t_vec);
+ } else if (vece <= MO_32) {
+ TCGv_i32 in = tcg_temp_new_i32();
+ switch (vece) {
+ case MO_8:
+ tcg_gen_ld8u_i32(in, cpu_env, aofs);
+ break;
+ case MO_16:
+ tcg_gen_ld16u_i32(in, cpu_env, aofs);
+ break;
+ default:
+ tcg_gen_ld_i32(in, cpu_env, aofs);
+ break;
+ }
+ do_dup(vece, dofs, oprsz, maxsz, in, NULL, 0);
+ tcg_temp_free_i32(in);
+ } else {
+ TCGv_i64 in = tcg_temp_new_i64();
+ tcg_gen_ld_i64(in, cpu_env, aofs);
+ do_dup(vece, dofs, oprsz, maxsz, NULL, in, 0);
+ tcg_temp_free_i64(in);
}
- tcg_gen_gvec_dup_i32(vece, dofs, oprsz, maxsz, in);
- tcg_temp_free_i32(in);
- } else if (vece == MO_64) {
- TCGv_i64 in = tcg_temp_new_i64();
- tcg_gen_ld_i64(in, cpu_env, aofs);
- tcg_gen_gvec_dup_i64(MO_64, dofs, oprsz, maxsz, in);
- tcg_temp_free_i64(in);
} else {
/* 128-bit duplicate. */
/* ??? Dup to 256-bit vector. */
tcg_temp_free_i64(in0);
tcg_temp_free_i64(in1);
}
+ if (oprsz < maxsz) {
+ expand_clr(dofs + oprsz, maxsz - oprsz);
+ }
}
}
tcg_temp_free_i64(t2);
}
+static const TCGOpcode vecop_list_add[] = { INDEX_op_add_vec, 0 };
+
void tcg_gen_gvec_add(unsigned vece, uint32_t dofs, uint32_t aofs,
uint32_t bofs, uint32_t oprsz, uint32_t maxsz)
{
{ .fni8 = tcg_gen_vec_add8_i64,
.fniv = tcg_gen_add_vec,
.fno = gen_helper_gvec_add8,
- .opc = INDEX_op_add_vec,
+ .opt_opc = vecop_list_add,
.vece = MO_8 },
{ .fni8 = tcg_gen_vec_add16_i64,
.fniv = tcg_gen_add_vec,
.fno = gen_helper_gvec_add16,
- .opc = INDEX_op_add_vec,
+ .opt_opc = vecop_list_add,
.vece = MO_16 },
{ .fni4 = tcg_gen_add_i32,
.fniv = tcg_gen_add_vec,
.fno = gen_helper_gvec_add32,
- .opc = INDEX_op_add_vec,
+ .opt_opc = vecop_list_add,
.vece = MO_32 },
{ .fni8 = tcg_gen_add_i64,
.fniv = tcg_gen_add_vec,
.fno = gen_helper_gvec_add64,
- .opc = INDEX_op_add_vec,
+ .opt_opc = vecop_list_add,
.prefer_i64 = TCG_TARGET_REG_BITS == 64,
.vece = MO_64 },
};
tcg_gen_gvec_3(dofs, aofs, bofs, oprsz, maxsz, &g[vece]);
}
+void tcg_gen_gvec_adds(unsigned vece, uint32_t dofs, uint32_t aofs,
+ TCGv_i64 c, uint32_t oprsz, uint32_t maxsz)
+{
+ static const GVecGen2s g[4] = {
+ { .fni8 = tcg_gen_vec_add8_i64,
+ .fniv = tcg_gen_add_vec,
+ .fno = gen_helper_gvec_adds8,
+ .opt_opc = vecop_list_add,
+ .vece = MO_8 },
+ { .fni8 = tcg_gen_vec_add16_i64,
+ .fniv = tcg_gen_add_vec,
+ .fno = gen_helper_gvec_adds16,
+ .opt_opc = vecop_list_add,
+ .vece = MO_16 },
+ { .fni4 = tcg_gen_add_i32,
+ .fniv = tcg_gen_add_vec,
+ .fno = gen_helper_gvec_adds32,
+ .opt_opc = vecop_list_add,
+ .vece = MO_32 },
+ { .fni8 = tcg_gen_add_i64,
+ .fniv = tcg_gen_add_vec,
+ .fno = gen_helper_gvec_adds64,
+ .opt_opc = vecop_list_add,
+ .prefer_i64 = TCG_TARGET_REG_BITS == 64,
+ .vece = MO_64 },
+ };
+
+ tcg_debug_assert(vece <= MO_64);
+ tcg_gen_gvec_2s(dofs, aofs, oprsz, maxsz, c, &g[vece]);
+}
+
+void tcg_gen_gvec_addi(unsigned vece, uint32_t dofs, uint32_t aofs,
+ int64_t c, uint32_t oprsz, uint32_t maxsz)
+{
+ TCGv_i64 tmp = tcg_const_i64(c);
+ tcg_gen_gvec_adds(vece, dofs, aofs, tmp, oprsz, maxsz);
+ tcg_temp_free_i64(tmp);
+}
+
+static const TCGOpcode vecop_list_sub[] = { INDEX_op_sub_vec, 0 };
+
+void tcg_gen_gvec_subs(unsigned vece, uint32_t dofs, uint32_t aofs,
+ TCGv_i64 c, uint32_t oprsz, uint32_t maxsz)
+{
+ static const GVecGen2s g[4] = {
+ { .fni8 = tcg_gen_vec_sub8_i64,
+ .fniv = tcg_gen_sub_vec,
+ .fno = gen_helper_gvec_subs8,
+ .opt_opc = vecop_list_sub,
+ .vece = MO_8 },
+ { .fni8 = tcg_gen_vec_sub16_i64,
+ .fniv = tcg_gen_sub_vec,
+ .fno = gen_helper_gvec_subs16,
+ .opt_opc = vecop_list_sub,
+ .vece = MO_16 },
+ { .fni4 = tcg_gen_sub_i32,
+ .fniv = tcg_gen_sub_vec,
+ .fno = gen_helper_gvec_subs32,
+ .opt_opc = vecop_list_sub,
+ .vece = MO_32 },
+ { .fni8 = tcg_gen_sub_i64,
+ .fniv = tcg_gen_sub_vec,
+ .fno = gen_helper_gvec_subs64,
+ .opt_opc = vecop_list_sub,
+ .prefer_i64 = TCG_TARGET_REG_BITS == 64,
+ .vece = MO_64 },
+ };
+
+ tcg_debug_assert(vece <= MO_64);
+ tcg_gen_gvec_2s(dofs, aofs, oprsz, maxsz, c, &g[vece]);
+}
+
/* Perform a vector subtraction using normal subtraction and a mask.
Compare gen_addv_mask above. */
static void gen_subv_mask(TCGv_i64 d, TCGv_i64 a, TCGv_i64 b, TCGv_i64 m)
{ .fni8 = tcg_gen_vec_sub8_i64,
.fniv = tcg_gen_sub_vec,
.fno = gen_helper_gvec_sub8,
- .opc = INDEX_op_sub_vec,
+ .opt_opc = vecop_list_sub,
.vece = MO_8 },
{ .fni8 = tcg_gen_vec_sub16_i64,
.fniv = tcg_gen_sub_vec,
.fno = gen_helper_gvec_sub16,
- .opc = INDEX_op_sub_vec,
+ .opt_opc = vecop_list_sub,
.vece = MO_16 },
{ .fni4 = tcg_gen_sub_i32,
.fniv = tcg_gen_sub_vec,
.fno = gen_helper_gvec_sub32,
- .opc = INDEX_op_sub_vec,
+ .opt_opc = vecop_list_sub,
.vece = MO_32 },
{ .fni8 = tcg_gen_sub_i64,
.fniv = tcg_gen_sub_vec,
.fno = gen_helper_gvec_sub64,
- .opc = INDEX_op_sub_vec,
+ .opt_opc = vecop_list_sub,
.prefer_i64 = TCG_TARGET_REG_BITS == 64,
.vece = MO_64 },
};
tcg_gen_gvec_3(dofs, aofs, bofs, oprsz, maxsz, &g[vece]);
}
+static const TCGOpcode vecop_list_mul[] = { INDEX_op_mul_vec, 0 };
+
void tcg_gen_gvec_mul(unsigned vece, uint32_t dofs, uint32_t aofs,
uint32_t bofs, uint32_t oprsz, uint32_t maxsz)
{
static const GVecGen3 g[4] = {
{ .fniv = tcg_gen_mul_vec,
.fno = gen_helper_gvec_mul8,
- .opc = INDEX_op_mul_vec,
+ .opt_opc = vecop_list_mul,
.vece = MO_8 },
{ .fniv = tcg_gen_mul_vec,
.fno = gen_helper_gvec_mul16,
- .opc = INDEX_op_mul_vec,
+ .opt_opc = vecop_list_mul,
.vece = MO_16 },
{ .fni4 = tcg_gen_mul_i32,
.fniv = tcg_gen_mul_vec,
.fno = gen_helper_gvec_mul32,
- .opc = INDEX_op_mul_vec,
+ .opt_opc = vecop_list_mul,
.vece = MO_32 },
{ .fni8 = tcg_gen_mul_i64,
.fniv = tcg_gen_mul_vec,
.fno = gen_helper_gvec_mul64,
- .opc = INDEX_op_mul_vec,
+ .opt_opc = vecop_list_mul,
.prefer_i64 = TCG_TARGET_REG_BITS == 64,
.vece = MO_64 },
};
tcg_gen_gvec_3(dofs, aofs, bofs, oprsz, maxsz, &g[vece]);
}
+void tcg_gen_gvec_muls(unsigned vece, uint32_t dofs, uint32_t aofs,
+ TCGv_i64 c, uint32_t oprsz, uint32_t maxsz)
+{
+ static const GVecGen2s g[4] = {
+ { .fniv = tcg_gen_mul_vec,
+ .fno = gen_helper_gvec_muls8,
+ .opt_opc = vecop_list_mul,
+ .vece = MO_8 },
+ { .fniv = tcg_gen_mul_vec,
+ .fno = gen_helper_gvec_muls16,
+ .opt_opc = vecop_list_mul,
+ .vece = MO_16 },
+ { .fni4 = tcg_gen_mul_i32,
+ .fniv = tcg_gen_mul_vec,
+ .fno = gen_helper_gvec_muls32,
+ .opt_opc = vecop_list_mul,
+ .vece = MO_32 },
+ { .fni8 = tcg_gen_mul_i64,
+ .fniv = tcg_gen_mul_vec,
+ .fno = gen_helper_gvec_muls64,
+ .opt_opc = vecop_list_mul,
+ .prefer_i64 = TCG_TARGET_REG_BITS == 64,
+ .vece = MO_64 },
+ };
+
+ tcg_debug_assert(vece <= MO_64);
+ tcg_gen_gvec_2s(dofs, aofs, oprsz, maxsz, c, &g[vece]);
+}
+
+void tcg_gen_gvec_muli(unsigned vece, uint32_t dofs, uint32_t aofs,
+ int64_t c, uint32_t oprsz, uint32_t maxsz)
+{
+ TCGv_i64 tmp = tcg_const_i64(c);
+ tcg_gen_gvec_muls(vece, dofs, aofs, tmp, oprsz, maxsz);
+ tcg_temp_free_i64(tmp);
+}
+
void tcg_gen_gvec_ssadd(unsigned vece, uint32_t dofs, uint32_t aofs,
uint32_t bofs, uint32_t oprsz, uint32_t maxsz)
{
+ static const TCGOpcode vecop_list[] = { INDEX_op_ssadd_vec, 0 };
static const GVecGen3 g[4] = {
- { .fno = gen_helper_gvec_ssadd8, .vece = MO_8 },
- { .fno = gen_helper_gvec_ssadd16, .vece = MO_16 },
- { .fno = gen_helper_gvec_ssadd32, .vece = MO_32 },
- { .fno = gen_helper_gvec_ssadd64, .vece = MO_64 }
+ { .fniv = tcg_gen_ssadd_vec,
+ .fno = gen_helper_gvec_ssadd8,
+ .opt_opc = vecop_list,
+ .vece = MO_8 },
+ { .fniv = tcg_gen_ssadd_vec,
+ .fno = gen_helper_gvec_ssadd16,
+ .opt_opc = vecop_list,
+ .vece = MO_16 },
+ { .fniv = tcg_gen_ssadd_vec,
+ .fno = gen_helper_gvec_ssadd32,
+ .opt_opc = vecop_list,
+ .vece = MO_32 },
+ { .fniv = tcg_gen_ssadd_vec,
+ .fno = gen_helper_gvec_ssadd64,
+ .opt_opc = vecop_list,
+ .vece = MO_64 },
};
tcg_debug_assert(vece <= MO_64);
tcg_gen_gvec_3(dofs, aofs, bofs, oprsz, maxsz, &g[vece]);
void tcg_gen_gvec_sssub(unsigned vece, uint32_t dofs, uint32_t aofs,
uint32_t bofs, uint32_t oprsz, uint32_t maxsz)
{
+ static const TCGOpcode vecop_list[] = { INDEX_op_sssub_vec, 0 };
static const GVecGen3 g[4] = {
- { .fno = gen_helper_gvec_sssub8, .vece = MO_8 },
- { .fno = gen_helper_gvec_sssub16, .vece = MO_16 },
- { .fno = gen_helper_gvec_sssub32, .vece = MO_32 },
- { .fno = gen_helper_gvec_sssub64, .vece = MO_64 }
+ { .fniv = tcg_gen_sssub_vec,
+ .fno = gen_helper_gvec_sssub8,
+ .opt_opc = vecop_list,
+ .vece = MO_8 },
+ { .fniv = tcg_gen_sssub_vec,
+ .fno = gen_helper_gvec_sssub16,
+ .opt_opc = vecop_list,
+ .vece = MO_16 },
+ { .fniv = tcg_gen_sssub_vec,
+ .fno = gen_helper_gvec_sssub32,
+ .opt_opc = vecop_list,
+ .vece = MO_32 },
+ { .fniv = tcg_gen_sssub_vec,
+ .fno = gen_helper_gvec_sssub64,
+ .opt_opc = vecop_list,
+ .vece = MO_64 },
};
tcg_debug_assert(vece <= MO_64);
tcg_gen_gvec_3(dofs, aofs, bofs, oprsz, maxsz, &g[vece]);
}
-static void tcg_gen_vec_usadd32_i32(TCGv_i32 d, TCGv_i32 a, TCGv_i32 b)
+static void tcg_gen_usadd_i32(TCGv_i32 d, TCGv_i32 a, TCGv_i32 b)
{
TCGv_i32 max = tcg_const_i32(-1);
tcg_gen_add_i32(d, a, b);
tcg_temp_free_i32(max);
}
-static void tcg_gen_vec_usadd32_i64(TCGv_i64 d, TCGv_i64 a, TCGv_i64 b)
+static void tcg_gen_usadd_i64(TCGv_i64 d, TCGv_i64 a, TCGv_i64 b)
{
TCGv_i64 max = tcg_const_i64(-1);
tcg_gen_add_i64(d, a, b);
void tcg_gen_gvec_usadd(unsigned vece, uint32_t dofs, uint32_t aofs,
uint32_t bofs, uint32_t oprsz, uint32_t maxsz)
{
+ static const TCGOpcode vecop_list[] = { INDEX_op_usadd_vec, 0 };
static const GVecGen3 g[4] = {
- { .fno = gen_helper_gvec_usadd8, .vece = MO_8 },
- { .fno = gen_helper_gvec_usadd16, .vece = MO_16 },
- { .fni4 = tcg_gen_vec_usadd32_i32,
+ { .fniv = tcg_gen_usadd_vec,
+ .fno = gen_helper_gvec_usadd8,
+ .opt_opc = vecop_list,
+ .vece = MO_8 },
+ { .fniv = tcg_gen_usadd_vec,
+ .fno = gen_helper_gvec_usadd16,
+ .opt_opc = vecop_list,
+ .vece = MO_16 },
+ { .fni4 = tcg_gen_usadd_i32,
+ .fniv = tcg_gen_usadd_vec,
.fno = gen_helper_gvec_usadd32,
+ .opt_opc = vecop_list,
.vece = MO_32 },
- { .fni8 = tcg_gen_vec_usadd32_i64,
+ { .fni8 = tcg_gen_usadd_i64,
+ .fniv = tcg_gen_usadd_vec,
.fno = gen_helper_gvec_usadd64,
+ .opt_opc = vecop_list,
.vece = MO_64 }
};
tcg_debug_assert(vece <= MO_64);
tcg_gen_gvec_3(dofs, aofs, bofs, oprsz, maxsz, &g[vece]);
}
-static void tcg_gen_vec_ussub32_i32(TCGv_i32 d, TCGv_i32 a, TCGv_i32 b)
+static void tcg_gen_ussub_i32(TCGv_i32 d, TCGv_i32 a, TCGv_i32 b)
{
TCGv_i32 min = tcg_const_i32(0);
tcg_gen_sub_i32(d, a, b);
tcg_temp_free_i32(min);
}
-static void tcg_gen_vec_ussub32_i64(TCGv_i64 d, TCGv_i64 a, TCGv_i64 b)
+static void tcg_gen_ussub_i64(TCGv_i64 d, TCGv_i64 a, TCGv_i64 b)
{
TCGv_i64 min = tcg_const_i64(0);
tcg_gen_sub_i64(d, a, b);
void tcg_gen_gvec_ussub(unsigned vece, uint32_t dofs, uint32_t aofs,
uint32_t bofs, uint32_t oprsz, uint32_t maxsz)
{
+ static const TCGOpcode vecop_list[] = { INDEX_op_ussub_vec, 0 };
static const GVecGen3 g[4] = {
- { .fno = gen_helper_gvec_ussub8, .vece = MO_8 },
- { .fno = gen_helper_gvec_ussub16, .vece = MO_16 },
- { .fni4 = tcg_gen_vec_ussub32_i32,
+ { .fniv = tcg_gen_ussub_vec,
+ .fno = gen_helper_gvec_ussub8,
+ .opt_opc = vecop_list,
+ .vece = MO_8 },
+ { .fniv = tcg_gen_ussub_vec,
+ .fno = gen_helper_gvec_ussub16,
+ .opt_opc = vecop_list,
+ .vece = MO_16 },
+ { .fni4 = tcg_gen_ussub_i32,
+ .fniv = tcg_gen_ussub_vec,
.fno = gen_helper_gvec_ussub32,
+ .opt_opc = vecop_list,
.vece = MO_32 },
- { .fni8 = tcg_gen_vec_ussub32_i64,
+ { .fni8 = tcg_gen_ussub_i64,
+ .fniv = tcg_gen_ussub_vec,
.fno = gen_helper_gvec_ussub64,
+ .opt_opc = vecop_list,
+ .vece = MO_64 }
+ };
+ tcg_debug_assert(vece <= MO_64);
+ tcg_gen_gvec_3(dofs, aofs, bofs, oprsz, maxsz, &g[vece]);
+}
+
+void tcg_gen_gvec_smin(unsigned vece, uint32_t dofs, uint32_t aofs,
+ uint32_t bofs, uint32_t oprsz, uint32_t maxsz)
+{
+ static const TCGOpcode vecop_list[] = { INDEX_op_smin_vec, 0 };
+ static const GVecGen3 g[4] = {
+ { .fniv = tcg_gen_smin_vec,
+ .fno = gen_helper_gvec_smin8,
+ .opt_opc = vecop_list,
+ .vece = MO_8 },
+ { .fniv = tcg_gen_smin_vec,
+ .fno = gen_helper_gvec_smin16,
+ .opt_opc = vecop_list,
+ .vece = MO_16 },
+ { .fni4 = tcg_gen_smin_i32,
+ .fniv = tcg_gen_smin_vec,
+ .fno = gen_helper_gvec_smin32,
+ .opt_opc = vecop_list,
+ .vece = MO_32 },
+ { .fni8 = tcg_gen_smin_i64,
+ .fniv = tcg_gen_smin_vec,
+ .fno = gen_helper_gvec_smin64,
+ .opt_opc = vecop_list,
+ .vece = MO_64 }
+ };
+ tcg_debug_assert(vece <= MO_64);
+ tcg_gen_gvec_3(dofs, aofs, bofs, oprsz, maxsz, &g[vece]);
+}
+
+void tcg_gen_gvec_umin(unsigned vece, uint32_t dofs, uint32_t aofs,
+ uint32_t bofs, uint32_t oprsz, uint32_t maxsz)
+{
+ static const TCGOpcode vecop_list[] = { INDEX_op_umin_vec, 0 };
+ static const GVecGen3 g[4] = {
+ { .fniv = tcg_gen_umin_vec,
+ .fno = gen_helper_gvec_umin8,
+ .opt_opc = vecop_list,
+ .vece = MO_8 },
+ { .fniv = tcg_gen_umin_vec,
+ .fno = gen_helper_gvec_umin16,
+ .opt_opc = vecop_list,
+ .vece = MO_16 },
+ { .fni4 = tcg_gen_umin_i32,
+ .fniv = tcg_gen_umin_vec,
+ .fno = gen_helper_gvec_umin32,
+ .opt_opc = vecop_list,
+ .vece = MO_32 },
+ { .fni8 = tcg_gen_umin_i64,
+ .fniv = tcg_gen_umin_vec,
+ .fno = gen_helper_gvec_umin64,
+ .opt_opc = vecop_list,
+ .vece = MO_64 }
+ };
+ tcg_debug_assert(vece <= MO_64);
+ tcg_gen_gvec_3(dofs, aofs, bofs, oprsz, maxsz, &g[vece]);
+}
+
+void tcg_gen_gvec_smax(unsigned vece, uint32_t dofs, uint32_t aofs,
+ uint32_t bofs, uint32_t oprsz, uint32_t maxsz)
+{
+ static const TCGOpcode vecop_list[] = { INDEX_op_smax_vec, 0 };
+ static const GVecGen3 g[4] = {
+ { .fniv = tcg_gen_smax_vec,
+ .fno = gen_helper_gvec_smax8,
+ .opt_opc = vecop_list,
+ .vece = MO_8 },
+ { .fniv = tcg_gen_smax_vec,
+ .fno = gen_helper_gvec_smax16,
+ .opt_opc = vecop_list,
+ .vece = MO_16 },
+ { .fni4 = tcg_gen_smax_i32,
+ .fniv = tcg_gen_smax_vec,
+ .fno = gen_helper_gvec_smax32,
+ .opt_opc = vecop_list,
+ .vece = MO_32 },
+ { .fni8 = tcg_gen_smax_i64,
+ .fniv = tcg_gen_smax_vec,
+ .fno = gen_helper_gvec_smax64,
+ .opt_opc = vecop_list,
+ .vece = MO_64 }
+ };
+ tcg_debug_assert(vece <= MO_64);
+ tcg_gen_gvec_3(dofs, aofs, bofs, oprsz, maxsz, &g[vece]);
+}
+
+void tcg_gen_gvec_umax(unsigned vece, uint32_t dofs, uint32_t aofs,
+ uint32_t bofs, uint32_t oprsz, uint32_t maxsz)
+{
+ static const TCGOpcode vecop_list[] = { INDEX_op_umax_vec, 0 };
+ static const GVecGen3 g[4] = {
+ { .fniv = tcg_gen_umax_vec,
+ .fno = gen_helper_gvec_umax8,
+ .opt_opc = vecop_list,
+ .vece = MO_8 },
+ { .fniv = tcg_gen_umax_vec,
+ .fno = gen_helper_gvec_umax16,
+ .opt_opc = vecop_list,
+ .vece = MO_16 },
+ { .fni4 = tcg_gen_umax_i32,
+ .fniv = tcg_gen_umax_vec,
+ .fno = gen_helper_gvec_umax32,
+ .opt_opc = vecop_list,
+ .vece = MO_32 },
+ { .fni8 = tcg_gen_umax_i64,
+ .fniv = tcg_gen_umax_vec,
+ .fno = gen_helper_gvec_umax64,
+ .opt_opc = vecop_list,
.vece = MO_64 }
};
tcg_debug_assert(vece <= MO_64);
void tcg_gen_gvec_neg(unsigned vece, uint32_t dofs, uint32_t aofs,
uint32_t oprsz, uint32_t maxsz)
{
+ static const TCGOpcode vecop_list[] = { INDEX_op_neg_vec, 0 };
static const GVecGen2 g[4] = {
{ .fni8 = tcg_gen_vec_neg8_i64,
.fniv = tcg_gen_neg_vec,
.fno = gen_helper_gvec_neg8,
- .opc = INDEX_op_neg_vec,
+ .opt_opc = vecop_list,
.vece = MO_8 },
{ .fni8 = tcg_gen_vec_neg16_i64,
.fniv = tcg_gen_neg_vec,
.fno = gen_helper_gvec_neg16,
- .opc = INDEX_op_neg_vec,
+ .opt_opc = vecop_list,
.vece = MO_16 },
{ .fni4 = tcg_gen_neg_i32,
.fniv = tcg_gen_neg_vec,
.fno = gen_helper_gvec_neg32,
- .opc = INDEX_op_neg_vec,
+ .opt_opc = vecop_list,
.vece = MO_32 },
{ .fni8 = tcg_gen_neg_i64,
.fniv = tcg_gen_neg_vec,
.fno = gen_helper_gvec_neg64,
- .opc = INDEX_op_neg_vec,
+ .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]);
+}
+
+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 },
};
.fni8 = tcg_gen_and_i64,
.fniv = tcg_gen_and_vec,
.fno = gen_helper_gvec_and,
- .opc = INDEX_op_and_vec,
.prefer_i64 = TCG_TARGET_REG_BITS == 64,
};
- tcg_gen_gvec_3(dofs, aofs, bofs, oprsz, maxsz, &g);
+
+ if (aofs == bofs) {
+ tcg_gen_gvec_mov(vece, dofs, aofs, oprsz, maxsz);
+ } else {
+ tcg_gen_gvec_3(dofs, aofs, bofs, oprsz, maxsz, &g);
+ }
}
void tcg_gen_gvec_or(unsigned vece, uint32_t dofs, uint32_t aofs,
.fni8 = tcg_gen_or_i64,
.fniv = tcg_gen_or_vec,
.fno = gen_helper_gvec_or,
- .opc = INDEX_op_or_vec,
.prefer_i64 = TCG_TARGET_REG_BITS == 64,
};
- tcg_gen_gvec_3(dofs, aofs, bofs, oprsz, maxsz, &g);
+
+ if (aofs == bofs) {
+ tcg_gen_gvec_mov(vece, dofs, aofs, oprsz, maxsz);
+ } else {
+ tcg_gen_gvec_3(dofs, aofs, bofs, oprsz, maxsz, &g);
+ }
}
void tcg_gen_gvec_xor(unsigned vece, uint32_t dofs, uint32_t aofs,
.fni8 = tcg_gen_xor_i64,
.fniv = tcg_gen_xor_vec,
.fno = gen_helper_gvec_xor,
- .opc = INDEX_op_xor_vec,
.prefer_i64 = TCG_TARGET_REG_BITS == 64,
};
- tcg_gen_gvec_3(dofs, aofs, bofs, oprsz, maxsz, &g);
+
+ if (aofs == bofs) {
+ tcg_gen_gvec_dup8i(dofs, oprsz, maxsz, 0);
+ } else {
+ tcg_gen_gvec_3(dofs, aofs, bofs, oprsz, maxsz, &g);
+ }
}
void tcg_gen_gvec_andc(unsigned vece, uint32_t dofs, uint32_t aofs,
.fni8 = tcg_gen_andc_i64,
.fniv = tcg_gen_andc_vec,
.fno = gen_helper_gvec_andc,
- .opc = INDEX_op_andc_vec,
.prefer_i64 = TCG_TARGET_REG_BITS == 64,
};
- tcg_gen_gvec_3(dofs, aofs, bofs, oprsz, maxsz, &g);
+
+ if (aofs == bofs) {
+ tcg_gen_gvec_dup8i(dofs, oprsz, maxsz, 0);
+ } else {
+ tcg_gen_gvec_3(dofs, aofs, bofs, oprsz, maxsz, &g);
+ }
}
void tcg_gen_gvec_orc(unsigned vece, uint32_t dofs, uint32_t aofs,
.fni8 = tcg_gen_orc_i64,
.fniv = tcg_gen_orc_vec,
.fno = gen_helper_gvec_orc,
- .opc = INDEX_op_orc_vec,
.prefer_i64 = TCG_TARGET_REG_BITS == 64,
};
- tcg_gen_gvec_3(dofs, aofs, bofs, oprsz, maxsz, &g);
+
+ if (aofs == bofs) {
+ tcg_gen_gvec_dup8i(dofs, oprsz, maxsz, -1);
+ } else {
+ tcg_gen_gvec_3(dofs, aofs, bofs, oprsz, maxsz, &g);
+ }
+}
+
+void tcg_gen_gvec_nand(unsigned vece, uint32_t dofs, uint32_t aofs,
+ uint32_t bofs, uint32_t oprsz, uint32_t maxsz)
+{
+ static const GVecGen3 g = {
+ .fni8 = tcg_gen_nand_i64,
+ .fniv = tcg_gen_nand_vec,
+ .fno = gen_helper_gvec_nand,
+ .prefer_i64 = TCG_TARGET_REG_BITS == 64,
+ };
+
+ if (aofs == bofs) {
+ tcg_gen_gvec_not(vece, dofs, aofs, oprsz, maxsz);
+ } else {
+ tcg_gen_gvec_3(dofs, aofs, bofs, oprsz, maxsz, &g);
+ }
+}
+
+void tcg_gen_gvec_nor(unsigned vece, uint32_t dofs, uint32_t aofs,
+ uint32_t bofs, uint32_t oprsz, uint32_t maxsz)
+{
+ static const GVecGen3 g = {
+ .fni8 = tcg_gen_nor_i64,
+ .fniv = tcg_gen_nor_vec,
+ .fno = gen_helper_gvec_nor,
+ .prefer_i64 = TCG_TARGET_REG_BITS == 64,
+ };
+
+ if (aofs == bofs) {
+ tcg_gen_gvec_not(vece, dofs, aofs, oprsz, maxsz);
+ } else {
+ tcg_gen_gvec_3(dofs, aofs, bofs, oprsz, maxsz, &g);
+ }
+}
+
+void tcg_gen_gvec_eqv(unsigned vece, uint32_t dofs, uint32_t aofs,
+ uint32_t bofs, uint32_t oprsz, uint32_t maxsz)
+{
+ static const GVecGen3 g = {
+ .fni8 = tcg_gen_eqv_i64,
+ .fniv = tcg_gen_eqv_vec,
+ .fno = gen_helper_gvec_eqv,
+ .prefer_i64 = TCG_TARGET_REG_BITS == 64,
+ };
+
+ if (aofs == bofs) {
+ tcg_gen_gvec_dup8i(dofs, oprsz, maxsz, -1);
+ } else {
+ tcg_gen_gvec_3(dofs, aofs, bofs, oprsz, maxsz, &g);
+ }
+}
+
+static const GVecGen2s gop_ands = {
+ .fni8 = tcg_gen_and_i64,
+ .fniv = tcg_gen_and_vec,
+ .fno = gen_helper_gvec_ands,
+ .prefer_i64 = TCG_TARGET_REG_BITS == 64,
+ .vece = MO_64
+};
+
+void tcg_gen_gvec_ands(unsigned vece, uint32_t dofs, uint32_t aofs,
+ TCGv_i64 c, uint32_t oprsz, uint32_t maxsz)
+{
+ TCGv_i64 tmp = tcg_temp_new_i64();
+ gen_dup_i64(vece, tmp, c);
+ tcg_gen_gvec_2s(dofs, aofs, oprsz, maxsz, tmp, &gop_ands);
+ tcg_temp_free_i64(tmp);
+}
+
+void tcg_gen_gvec_andi(unsigned vece, uint32_t dofs, uint32_t aofs,
+ int64_t c, uint32_t oprsz, uint32_t maxsz)
+{
+ TCGv_i64 tmp = tcg_const_i64(dup_const(vece, c));
+ tcg_gen_gvec_2s(dofs, aofs, oprsz, maxsz, tmp, &gop_ands);
+ tcg_temp_free_i64(tmp);
+}
+
+static const GVecGen2s gop_xors = {
+ .fni8 = tcg_gen_xor_i64,
+ .fniv = tcg_gen_xor_vec,
+ .fno = gen_helper_gvec_xors,
+ .prefer_i64 = TCG_TARGET_REG_BITS == 64,
+ .vece = MO_64
+};
+
+void tcg_gen_gvec_xors(unsigned vece, uint32_t dofs, uint32_t aofs,
+ TCGv_i64 c, uint32_t oprsz, uint32_t maxsz)
+{
+ TCGv_i64 tmp = tcg_temp_new_i64();
+ gen_dup_i64(vece, tmp, c);
+ tcg_gen_gvec_2s(dofs, aofs, oprsz, maxsz, tmp, &gop_xors);
+ tcg_temp_free_i64(tmp);
+}
+
+void tcg_gen_gvec_xori(unsigned vece, uint32_t dofs, uint32_t aofs,
+ int64_t c, uint32_t oprsz, uint32_t maxsz)
+{
+ TCGv_i64 tmp = tcg_const_i64(dup_const(vece, c));
+ tcg_gen_gvec_2s(dofs, aofs, oprsz, maxsz, tmp, &gop_xors);
+ tcg_temp_free_i64(tmp);
+}
+
+static const GVecGen2s gop_ors = {
+ .fni8 = tcg_gen_or_i64,
+ .fniv = tcg_gen_or_vec,
+ .fno = gen_helper_gvec_ors,
+ .prefer_i64 = TCG_TARGET_REG_BITS == 64,
+ .vece = MO_64
+};
+
+void tcg_gen_gvec_ors(unsigned vece, uint32_t dofs, uint32_t aofs,
+ TCGv_i64 c, uint32_t oprsz, uint32_t maxsz)
+{
+ TCGv_i64 tmp = tcg_temp_new_i64();
+ gen_dup_i64(vece, tmp, c);
+ tcg_gen_gvec_2s(dofs, aofs, oprsz, maxsz, tmp, &gop_ors);
+ tcg_temp_free_i64(tmp);
+}
+
+void tcg_gen_gvec_ori(unsigned vece, uint32_t dofs, uint32_t aofs,
+ int64_t c, uint32_t oprsz, uint32_t maxsz)
+{
+ TCGv_i64 tmp = tcg_const_i64(dup_const(vece, c));
+ tcg_gen_gvec_2s(dofs, aofs, oprsz, maxsz, tmp, &gop_ors);
+ tcg_temp_free_i64(tmp);
}
void tcg_gen_vec_shl8i_i64(TCGv_i64 d, TCGv_i64 a, int64_t c)
void tcg_gen_gvec_shli(unsigned vece, uint32_t dofs, uint32_t aofs,
int64_t shift, uint32_t oprsz, uint32_t maxsz)
{
+ static const TCGOpcode vecop_list[] = { INDEX_op_shli_vec, 0 };
static const GVecGen2i g[4] = {
{ .fni8 = tcg_gen_vec_shl8i_i64,
.fniv = tcg_gen_shli_vec,
.fno = gen_helper_gvec_shl8i,
- .opc = INDEX_op_shli_vec,
+ .opt_opc = vecop_list,
.vece = MO_8 },
{ .fni8 = tcg_gen_vec_shl16i_i64,
.fniv = tcg_gen_shli_vec,
.fno = gen_helper_gvec_shl16i,
- .opc = INDEX_op_shli_vec,
+ .opt_opc = vecop_list,
.vece = MO_16 },
{ .fni4 = tcg_gen_shli_i32,
.fniv = tcg_gen_shli_vec,
.fno = gen_helper_gvec_shl32i,
- .opc = INDEX_op_shli_vec,
+ .opt_opc = vecop_list,
.vece = MO_32 },
{ .fni8 = tcg_gen_shli_i64,
.fniv = tcg_gen_shli_vec,
.fno = gen_helper_gvec_shl64i,
- .opc = INDEX_op_shli_vec,
+ .opt_opc = vecop_list,
.prefer_i64 = TCG_TARGET_REG_BITS == 64,
.vece = MO_64 },
};
void tcg_gen_gvec_shri(unsigned vece, uint32_t dofs, uint32_t aofs,
int64_t shift, uint32_t oprsz, uint32_t maxsz)
{
+ static const TCGOpcode vecop_list[] = { INDEX_op_shri_vec, 0 };
static const GVecGen2i g[4] = {
{ .fni8 = tcg_gen_vec_shr8i_i64,
.fniv = tcg_gen_shri_vec,
.fno = gen_helper_gvec_shr8i,
- .opc = INDEX_op_shri_vec,
+ .opt_opc = vecop_list,
.vece = MO_8 },
{ .fni8 = tcg_gen_vec_shr16i_i64,
.fniv = tcg_gen_shri_vec,
.fno = gen_helper_gvec_shr16i,
- .opc = INDEX_op_shri_vec,
+ .opt_opc = vecop_list,
.vece = MO_16 },
{ .fni4 = tcg_gen_shri_i32,
.fniv = tcg_gen_shri_vec,
.fno = gen_helper_gvec_shr32i,
- .opc = INDEX_op_shri_vec,
+ .opt_opc = vecop_list,
.vece = MO_32 },
{ .fni8 = tcg_gen_shri_i64,
.fniv = tcg_gen_shri_vec,
.fno = gen_helper_gvec_shr64i,
- .opc = INDEX_op_shri_vec,
+ .opt_opc = vecop_list,
.prefer_i64 = TCG_TARGET_REG_BITS == 64,
.vece = MO_64 },
};
void tcg_gen_gvec_sari(unsigned vece, uint32_t dofs, uint32_t aofs,
int64_t shift, uint32_t oprsz, uint32_t maxsz)
{
+ static const TCGOpcode vecop_list[] = { INDEX_op_sari_vec, 0 };
static const GVecGen2i g[4] = {
{ .fni8 = tcg_gen_vec_sar8i_i64,
.fniv = tcg_gen_sari_vec,
.fno = gen_helper_gvec_sar8i,
- .opc = INDEX_op_sari_vec,
+ .opt_opc = vecop_list,
.vece = MO_8 },
{ .fni8 = tcg_gen_vec_sar16i_i64,
.fniv = tcg_gen_sari_vec,
.fno = gen_helper_gvec_sar16i,
- .opc = INDEX_op_sari_vec,
+ .opt_opc = vecop_list,
.vece = MO_16 },
{ .fni4 = tcg_gen_sari_i32,
.fniv = tcg_gen_sari_vec,
.fno = gen_helper_gvec_sar32i,
- .opc = INDEX_op_sari_vec,
+ .opt_opc = vecop_list,
.vece = MO_32 },
{ .fni8 = tcg_gen_sari_i64,
.fniv = tcg_gen_sari_vec,
.fno = gen_helper_gvec_sar64i,
- .opc = INDEX_op_sari_vec,
+ .opt_opc = vecop_list,
.prefer_i64 = TCG_TARGET_REG_BITS == 64,
.vece = MO_64 },
};
}
}
+/*
+ * Specialized generation vector shifts by a non-constant scalar.
+ */
+
+typedef struct {
+ void (*fni4)(TCGv_i32, TCGv_i32, TCGv_i32);
+ void (*fni8)(TCGv_i64, TCGv_i64, TCGv_i64);
+ void (*fniv_s)(unsigned, TCGv_vec, TCGv_vec, TCGv_i32);
+ void (*fniv_v)(unsigned, TCGv_vec, TCGv_vec, TCGv_vec);
+ gen_helper_gvec_2 *fno[4];
+ TCGOpcode s_list[2];
+ TCGOpcode v_list[2];
+} GVecGen2sh;
+
+static void expand_2sh_vec(unsigned vece, uint32_t dofs, uint32_t aofs,
+ uint32_t oprsz, uint32_t tysz, TCGType type,
+ TCGv_i32 shift,
+ void (*fni)(unsigned, TCGv_vec, TCGv_vec, TCGv_i32))
+{
+ TCGv_vec t0 = tcg_temp_new_vec(type);
+ uint32_t i;
+
+ for (i = 0; i < oprsz; i += tysz) {
+ tcg_gen_ld_vec(t0, cpu_env, aofs + i);
+ fni(vece, t0, t0, shift);
+ tcg_gen_st_vec(t0, cpu_env, dofs + i);
+ }
+ tcg_temp_free_vec(t0);
+}
+
+static void
+do_gvec_shifts(unsigned vece, uint32_t dofs, uint32_t aofs, TCGv_i32 shift,
+ uint32_t oprsz, uint32_t maxsz, const GVecGen2sh *g)
+{
+ TCGType type;
+ uint32_t some;
+
+ check_size_align(oprsz, maxsz, dofs | aofs);
+ check_overlap_2(dofs, aofs, maxsz);
+
+ /* If the backend has a scalar expansion, great. */
+ type = choose_vector_type(g->s_list, vece, oprsz, vece == MO_64);
+ if (type) {
+ const TCGOpcode *hold_list = tcg_swap_vecop_list(NULL);
+ switch (type) {
+ case TCG_TYPE_V256:
+ some = QEMU_ALIGN_DOWN(oprsz, 32);
+ expand_2sh_vec(vece, dofs, aofs, some, 32,
+ TCG_TYPE_V256, shift, g->fniv_s);
+ if (some == oprsz) {
+ break;
+ }
+ dofs += some;
+ aofs += some;
+ oprsz -= some;
+ maxsz -= some;
+ /* fallthru */
+ case TCG_TYPE_V128:
+ expand_2sh_vec(vece, dofs, aofs, oprsz, 16,
+ TCG_TYPE_V128, shift, g->fniv_s);
+ break;
+ case TCG_TYPE_V64:
+ expand_2sh_vec(vece, dofs, aofs, oprsz, 8,
+ TCG_TYPE_V64, shift, g->fniv_s);
+ break;
+ default:
+ g_assert_not_reached();
+ }
+ tcg_swap_vecop_list(hold_list);
+ goto clear_tail;
+ }
+
+ /* If the backend supports variable vector shifts, also cool. */
+ type = choose_vector_type(g->v_list, vece, oprsz, vece == MO_64);
+ if (type) {
+ const TCGOpcode *hold_list = tcg_swap_vecop_list(NULL);
+ TCGv_vec v_shift = tcg_temp_new_vec(type);
+
+ if (vece == MO_64) {
+ TCGv_i64 sh64 = tcg_temp_new_i64();
+ tcg_gen_extu_i32_i64(sh64, shift);
+ tcg_gen_dup_i64_vec(MO_64, v_shift, sh64);
+ tcg_temp_free_i64(sh64);
+ } else {
+ tcg_gen_dup_i32_vec(vece, v_shift, shift);
+ }
+
+ switch (type) {
+ case TCG_TYPE_V256:
+ some = QEMU_ALIGN_DOWN(oprsz, 32);
+ expand_2s_vec(vece, dofs, aofs, some, 32, TCG_TYPE_V256,
+ v_shift, false, g->fniv_v);
+ if (some == oprsz) {
+ break;
+ }
+ dofs += some;
+ aofs += some;
+ oprsz -= some;
+ maxsz -= some;
+ /* fallthru */
+ case TCG_TYPE_V128:
+ expand_2s_vec(vece, dofs, aofs, oprsz, 16, TCG_TYPE_V128,
+ v_shift, false, g->fniv_v);
+ break;
+ case TCG_TYPE_V64:
+ expand_2s_vec(vece, dofs, aofs, oprsz, 8, TCG_TYPE_V64,
+ v_shift, false, g->fniv_v);
+ break;
+ default:
+ g_assert_not_reached();
+ }
+ tcg_temp_free_vec(v_shift);
+ tcg_swap_vecop_list(hold_list);
+ goto clear_tail;
+ }
+
+ /* Otherwise fall back to integral... */
+ if (vece == MO_32 && check_size_impl(oprsz, 4)) {
+ expand_2s_i32(dofs, aofs, oprsz, shift, false, g->fni4);
+ } else if (vece == MO_64 && check_size_impl(oprsz, 8)) {
+ TCGv_i64 sh64 = tcg_temp_new_i64();
+ tcg_gen_extu_i32_i64(sh64, shift);
+ expand_2s_i64(dofs, aofs, oprsz, sh64, false, g->fni8);
+ tcg_temp_free_i64(sh64);
+ } else {
+ TCGv_ptr a0 = tcg_temp_new_ptr();
+ TCGv_ptr a1 = tcg_temp_new_ptr();
+ TCGv_i32 desc = tcg_temp_new_i32();
+
+ tcg_gen_shli_i32(desc, shift, SIMD_DATA_SHIFT);
+ tcg_gen_ori_i32(desc, desc, simd_desc(oprsz, maxsz, 0));
+ tcg_gen_addi_ptr(a0, cpu_env, dofs);
+ tcg_gen_addi_ptr(a1, cpu_env, aofs);
+
+ g->fno[vece](a0, a1, desc);
+
+ tcg_temp_free_ptr(a0);
+ tcg_temp_free_ptr(a1);
+ tcg_temp_free_i32(desc);
+ return;
+ }
+
+ clear_tail:
+ if (oprsz < maxsz) {
+ expand_clr(dofs + oprsz, maxsz - oprsz);
+ }
+}
+
+void tcg_gen_gvec_shls(unsigned vece, uint32_t dofs, uint32_t aofs,
+ TCGv_i32 shift, uint32_t oprsz, uint32_t maxsz)
+{
+ static const GVecGen2sh g = {
+ .fni4 = tcg_gen_shl_i32,
+ .fni8 = tcg_gen_shl_i64,
+ .fniv_s = tcg_gen_shls_vec,
+ .fniv_v = tcg_gen_shlv_vec,
+ .fno = {
+ gen_helper_gvec_shl8i,
+ gen_helper_gvec_shl16i,
+ gen_helper_gvec_shl32i,
+ gen_helper_gvec_shl64i,
+ },
+ .s_list = { INDEX_op_shls_vec, 0 },
+ .v_list = { INDEX_op_shlv_vec, 0 },
+ };
+
+ tcg_debug_assert(vece <= MO_64);
+ do_gvec_shifts(vece, dofs, aofs, shift, oprsz, maxsz, &g);
+}
+
+void tcg_gen_gvec_shrs(unsigned vece, uint32_t dofs, uint32_t aofs,
+ TCGv_i32 shift, uint32_t oprsz, uint32_t maxsz)
+{
+ static const GVecGen2sh g = {
+ .fni4 = tcg_gen_shr_i32,
+ .fni8 = tcg_gen_shr_i64,
+ .fniv_s = tcg_gen_shrs_vec,
+ .fniv_v = tcg_gen_shrv_vec,
+ .fno = {
+ gen_helper_gvec_shr8i,
+ gen_helper_gvec_shr16i,
+ gen_helper_gvec_shr32i,
+ gen_helper_gvec_shr64i,
+ },
+ .s_list = { INDEX_op_shrs_vec, 0 },
+ .v_list = { INDEX_op_shrv_vec, 0 },
+ };
+
+ tcg_debug_assert(vece <= MO_64);
+ do_gvec_shifts(vece, dofs, aofs, shift, oprsz, maxsz, &g);
+}
+
+void tcg_gen_gvec_sars(unsigned vece, uint32_t dofs, uint32_t aofs,
+ TCGv_i32 shift, uint32_t oprsz, uint32_t maxsz)
+{
+ static const GVecGen2sh g = {
+ .fni4 = tcg_gen_sar_i32,
+ .fni8 = tcg_gen_sar_i64,
+ .fniv_s = tcg_gen_sars_vec,
+ .fniv_v = tcg_gen_sarv_vec,
+ .fno = {
+ gen_helper_gvec_sar8i,
+ gen_helper_gvec_sar16i,
+ gen_helper_gvec_sar32i,
+ gen_helper_gvec_sar64i,
+ },
+ .s_list = { INDEX_op_sars_vec, 0 },
+ .v_list = { INDEX_op_sarv_vec, 0 },
+ };
+
+ tcg_debug_assert(vece <= MO_64);
+ do_gvec_shifts(vece, dofs, aofs, shift, oprsz, maxsz, &g);
+}
+
+/*
+ * Expand D = A << (B % element bits)
+ *
+ * Unlike scalar shifts, where it is easy for the target front end
+ * to include the modulo as part of the expansion. If the target
+ * naturally includes the modulo as part of the operation, great!
+ * If the target has some other behaviour from out-of-range shifts,
+ * then it could not use this function anyway, and would need to
+ * do it's own expansion with custom functions.
+ */
+static void tcg_gen_shlv_mod_vec(unsigned vece, TCGv_vec d,
+ TCGv_vec a, TCGv_vec b)
+{
+ TCGv_vec t = tcg_temp_new_vec_matching(d);
+
+ tcg_gen_dupi_vec(vece, t, (8 << vece) - 1);
+ tcg_gen_and_vec(vece, t, t, b);
+ tcg_gen_shlv_vec(vece, d, a, t);
+ tcg_temp_free_vec(t);
+}
+
+static void tcg_gen_shl_mod_i32(TCGv_i32 d, TCGv_i32 a, TCGv_i32 b)
+{
+ TCGv_i32 t = tcg_temp_new_i32();
+
+ tcg_gen_andi_i32(t, b, 31);
+ tcg_gen_shl_i32(d, a, t);
+ tcg_temp_free_i32(t);
+}
+
+static void tcg_gen_shl_mod_i64(TCGv_i64 d, TCGv_i64 a, TCGv_i64 b)
+{
+ TCGv_i64 t = tcg_temp_new_i64();
+
+ tcg_gen_andi_i64(t, b, 63);
+ tcg_gen_shl_i64(d, a, t);
+ tcg_temp_free_i64(t);
+}
+
+void tcg_gen_gvec_shlv(unsigned vece, uint32_t dofs, uint32_t aofs,
+ uint32_t bofs, uint32_t oprsz, uint32_t maxsz)
+{
+ static const TCGOpcode vecop_list[] = { INDEX_op_shlv_vec, 0 };
+ static const GVecGen3 g[4] = {
+ { .fniv = tcg_gen_shlv_mod_vec,
+ .fno = gen_helper_gvec_shl8v,
+ .opt_opc = vecop_list,
+ .vece = MO_8 },
+ { .fniv = tcg_gen_shlv_mod_vec,
+ .fno = gen_helper_gvec_shl16v,
+ .opt_opc = vecop_list,
+ .vece = MO_16 },
+ { .fni4 = tcg_gen_shl_mod_i32,
+ .fniv = tcg_gen_shlv_mod_vec,
+ .fno = gen_helper_gvec_shl32v,
+ .opt_opc = vecop_list,
+ .vece = MO_32 },
+ { .fni8 = tcg_gen_shl_mod_i64,
+ .fniv = tcg_gen_shlv_mod_vec,
+ .fno = gen_helper_gvec_shl64v,
+ .opt_opc = vecop_list,
+ .prefer_i64 = TCG_TARGET_REG_BITS == 64,
+ .vece = MO_64 },
+ };
+
+ tcg_debug_assert(vece <= MO_64);
+ tcg_gen_gvec_3(dofs, aofs, bofs, oprsz, maxsz, &g[vece]);
+}
+
+/*
+ * Similarly for logical right shifts.
+ */
+
+static void tcg_gen_shrv_mod_vec(unsigned vece, TCGv_vec d,
+ TCGv_vec a, TCGv_vec b)
+{
+ TCGv_vec t = tcg_temp_new_vec_matching(d);
+
+ tcg_gen_dupi_vec(vece, t, (8 << vece) - 1);
+ tcg_gen_and_vec(vece, t, t, b);
+ tcg_gen_shrv_vec(vece, d, a, t);
+ tcg_temp_free_vec(t);
+}
+
+static void tcg_gen_shr_mod_i32(TCGv_i32 d, TCGv_i32 a, TCGv_i32 b)
+{
+ TCGv_i32 t = tcg_temp_new_i32();
+
+ tcg_gen_andi_i32(t, b, 31);
+ tcg_gen_shr_i32(d, a, t);
+ tcg_temp_free_i32(t);
+}
+
+static void tcg_gen_shr_mod_i64(TCGv_i64 d, TCGv_i64 a, TCGv_i64 b)
+{
+ TCGv_i64 t = tcg_temp_new_i64();
+
+ tcg_gen_andi_i64(t, b, 63);
+ tcg_gen_shr_i64(d, a, t);
+ tcg_temp_free_i64(t);
+}
+
+void tcg_gen_gvec_shrv(unsigned vece, uint32_t dofs, uint32_t aofs,
+ uint32_t bofs, uint32_t oprsz, uint32_t maxsz)
+{
+ static const TCGOpcode vecop_list[] = { INDEX_op_shrv_vec, 0 };
+ static const GVecGen3 g[4] = {
+ { .fniv = tcg_gen_shrv_mod_vec,
+ .fno = gen_helper_gvec_shr8v,
+ .opt_opc = vecop_list,
+ .vece = MO_8 },
+ { .fniv = tcg_gen_shrv_mod_vec,
+ .fno = gen_helper_gvec_shr16v,
+ .opt_opc = vecop_list,
+ .vece = MO_16 },
+ { .fni4 = tcg_gen_shr_mod_i32,
+ .fniv = tcg_gen_shrv_mod_vec,
+ .fno = gen_helper_gvec_shr32v,
+ .opt_opc = vecop_list,
+ .vece = MO_32 },
+ { .fni8 = tcg_gen_shr_mod_i64,
+ .fniv = tcg_gen_shrv_mod_vec,
+ .fno = gen_helper_gvec_shr64v,
+ .opt_opc = vecop_list,
+ .prefer_i64 = TCG_TARGET_REG_BITS == 64,
+ .vece = MO_64 },
+ };
+
+ tcg_debug_assert(vece <= MO_64);
+ tcg_gen_gvec_3(dofs, aofs, bofs, oprsz, maxsz, &g[vece]);
+}
+
+/*
+ * Similarly for arithmetic right shifts.
+ */
+
+static void tcg_gen_sarv_mod_vec(unsigned vece, TCGv_vec d,
+ TCGv_vec a, TCGv_vec b)
+{
+ TCGv_vec t = tcg_temp_new_vec_matching(d);
+
+ tcg_gen_dupi_vec(vece, t, (8 << vece) - 1);
+ tcg_gen_and_vec(vece, t, t, b);
+ tcg_gen_sarv_vec(vece, d, a, t);
+ tcg_temp_free_vec(t);
+}
+
+static void tcg_gen_sar_mod_i32(TCGv_i32 d, TCGv_i32 a, TCGv_i32 b)
+{
+ TCGv_i32 t = tcg_temp_new_i32();
+
+ tcg_gen_andi_i32(t, b, 31);
+ tcg_gen_sar_i32(d, a, t);
+ tcg_temp_free_i32(t);
+}
+
+static void tcg_gen_sar_mod_i64(TCGv_i64 d, TCGv_i64 a, TCGv_i64 b)
+{
+ TCGv_i64 t = tcg_temp_new_i64();
+
+ tcg_gen_andi_i64(t, b, 63);
+ tcg_gen_sar_i64(d, a, t);
+ tcg_temp_free_i64(t);
+}
+
+void tcg_gen_gvec_sarv(unsigned vece, uint32_t dofs, uint32_t aofs,
+ uint32_t bofs, uint32_t oprsz, uint32_t maxsz)
+{
+ static const TCGOpcode vecop_list[] = { INDEX_op_sarv_vec, 0 };
+ static const GVecGen3 g[4] = {
+ { .fniv = tcg_gen_sarv_mod_vec,
+ .fno = gen_helper_gvec_sar8v,
+ .opt_opc = vecop_list,
+ .vece = MO_8 },
+ { .fniv = tcg_gen_sarv_mod_vec,
+ .fno = gen_helper_gvec_sar16v,
+ .opt_opc = vecop_list,
+ .vece = MO_16 },
+ { .fni4 = tcg_gen_sar_mod_i32,
+ .fniv = tcg_gen_sarv_mod_vec,
+ .fno = gen_helper_gvec_sar32v,
+ .opt_opc = vecop_list,
+ .vece = MO_32 },
+ { .fni8 = tcg_gen_sar_mod_i64,
+ .fniv = tcg_gen_sarv_mod_vec,
+ .fno = gen_helper_gvec_sar64v,
+ .opt_opc = vecop_list,
+ .prefer_i64 = TCG_TARGET_REG_BITS == 64,
+ .vece = MO_64 },
+ };
+
+ tcg_debug_assert(vece <= MO_64);
+ tcg_gen_gvec_3(dofs, aofs, bofs, oprsz, maxsz, &g[vece]);
+}
+
/* Expand OPSZ bytes worth of three-operand operations using i32 elements. */
static void expand_cmp_i32(uint32_t dofs, uint32_t aofs, uint32_t bofs,
uint32_t oprsz, TCGCond cond)
uint32_t aofs, uint32_t bofs,
uint32_t oprsz, uint32_t maxsz)
{
+ static const TCGOpcode cmp_list[] = { INDEX_op_cmp_vec, 0 };
static gen_helper_gvec_3 * const eq_fn[4] = {
gen_helper_gvec_eq8, gen_helper_gvec_eq16,
gen_helper_gvec_eq32, gen_helper_gvec_eq64
[TCG_COND_LEU] = leu_fn,
};
+ const TCGOpcode *hold_list;
+ TCGType type;
+ uint32_t some;
+
check_size_align(oprsz, maxsz, dofs | aofs | bofs);
check_overlap_3(dofs, aofs, bofs, maxsz);
return;
}
- /* Recall that ARM SVE allows vector sizes that are not a power of 2.
- Expand with successively smaller host vector sizes. The intent is
- that e.g. oprsz == 80 would be expanded with 2x32 + 1x16. */
-
- if (TCG_TARGET_HAS_v256 && check_size_impl(oprsz, 32)
- && tcg_can_emit_vec_op(INDEX_op_cmp_vec, TCG_TYPE_V256, vece)) {
- uint32_t some = QEMU_ALIGN_DOWN(oprsz, 32);
+ /*
+ * Implement inline with a vector type, if possible.
+ * Prefer integer when 64-bit host and 64-bit comparison.
+ */
+ hold_list = tcg_swap_vecop_list(cmp_list);
+ type = choose_vector_type(cmp_list, vece, oprsz,
+ TCG_TARGET_REG_BITS == 64 && vece == MO_64);
+ switch (type) {
+ case TCG_TYPE_V256:
+ /* Recall that ARM SVE allows vector sizes that are not a
+ * power of 2, but always a multiple of 16. The intent is
+ * that e.g. size == 80 would be expanded with 2x32 + 1x16.
+ */
+ some = QEMU_ALIGN_DOWN(oprsz, 32);
expand_cmp_vec(vece, dofs, aofs, bofs, some, 32, TCG_TYPE_V256, cond);
if (some == oprsz) {
- goto done;
+ break;
}
dofs += some;
aofs += some;
bofs += some;
oprsz -= some;
maxsz -= some;
- }
-
- if (TCG_TARGET_HAS_v128 && check_size_impl(oprsz, 16)
- && tcg_can_emit_vec_op(INDEX_op_cmp_vec, TCG_TYPE_V128, vece)) {
+ /* fallthru */
+ case TCG_TYPE_V128:
expand_cmp_vec(vece, dofs, aofs, bofs, oprsz, 16, TCG_TYPE_V128, cond);
- } else if (TCG_TARGET_HAS_v64
- && check_size_impl(oprsz, 8)
- && (TCG_TARGET_REG_BITS == 32 || vece != MO_64)
- && tcg_can_emit_vec_op(INDEX_op_cmp_vec, TCG_TYPE_V64, vece)) {
+ break;
+ case TCG_TYPE_V64:
expand_cmp_vec(vece, dofs, aofs, bofs, oprsz, 8, TCG_TYPE_V64, cond);
- } else if (vece == MO_64 && check_size_impl(oprsz, 8)) {
- expand_cmp_i64(dofs, aofs, bofs, oprsz, cond);
- } else if (vece == MO_32 && check_size_impl(oprsz, 4)) {
- expand_cmp_i32(dofs, aofs, bofs, oprsz, cond);
- } else {
- gen_helper_gvec_3 * const *fn = fns[cond];
-
- if (fn == NULL) {
- uint32_t tmp;
- tmp = aofs, aofs = bofs, bofs = tmp;
- cond = tcg_swap_cond(cond);
- fn = fns[cond];
- assert(fn != NULL);
+ break;
+
+ case 0:
+ if (vece == MO_64 && check_size_impl(oprsz, 8)) {
+ expand_cmp_i64(dofs, aofs, bofs, oprsz, cond);
+ } else if (vece == MO_32 && check_size_impl(oprsz, 4)) {
+ expand_cmp_i32(dofs, aofs, bofs, oprsz, cond);
+ } else {
+ gen_helper_gvec_3 * const *fn = fns[cond];
+
+ if (fn == NULL) {
+ uint32_t tmp;
+ tmp = aofs, aofs = bofs, bofs = tmp;
+ cond = tcg_swap_cond(cond);
+ fn = fns[cond];
+ assert(fn != NULL);
+ }
+ tcg_gen_gvec_3_ool(dofs, aofs, bofs, oprsz, maxsz, 0, fn[vece]);
+ oprsz = maxsz;
}
- tcg_gen_gvec_3_ool(dofs, aofs, bofs, oprsz, maxsz, 0, fn[vece]);
- return;
+ break;
+
+ default:
+ g_assert_not_reached();
}
+ tcg_swap_vecop_list(hold_list);
- done:
if (oprsz < maxsz) {
expand_clr(dofs + oprsz, maxsz - oprsz);
}
}
+
+static void tcg_gen_bitsel_i64(TCGv_i64 d, TCGv_i64 a, TCGv_i64 b, TCGv_i64 c)
+{
+ TCGv_i64 t = tcg_temp_new_i64();
+
+ tcg_gen_and_i64(t, b, a);
+ tcg_gen_andc_i64(d, c, a);
+ tcg_gen_or_i64(d, d, t);
+ tcg_temp_free_i64(t);
+}
+
+void tcg_gen_gvec_bitsel(unsigned vece, uint32_t dofs, uint32_t aofs,
+ uint32_t bofs, uint32_t cofs,
+ uint32_t oprsz, uint32_t maxsz)
+{
+ static const GVecGen4 g = {
+ .fni8 = tcg_gen_bitsel_i64,
+ .fniv = tcg_gen_bitsel_vec,
+ .fno = gen_helper_gvec_bitsel,
+ };
+
+ tcg_gen_gvec_4(dofs, aofs, bofs, cofs, oprsz, maxsz, &g);
+}
projects / qemu.git / blobdiff