yuzu/unicorn
1
0
Fork 0
mirror of https://github.com/yuzu-emu/unicorn.git synced 2025-01-11 05:46:12 +00:00
Code Issues Packages Projects Releases Wiki Activity

target/arm: Convert Neon 3-reg-same VMUL, VMLA, VMLS, VSHL to decodetree

Browse source 
Convert the Neon VMUL, VMLA, VMLS and VSHL insns in the
3-reg-same grouping to decodetree.

Backports commit 0de34fd48ad4e44bf5caa2330657ebefa93cea7d from qemu
This commit is contained in:
Peter Maydell 2020-05-07 09:49:57 -04:00 committed by Lioncash
parent 17bd8930fc
commit 652165d671
3 changed files with 56 additions and 25 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

9
qemu/target/arm/neon-dp.decode
View file

@ -65,6 +65,9 @@ VCGT_U_3s 1111 001 1 0 . .. .... .... 0011 . . . 0 .... @3same
VCGE_S_3s 1111 001 0 0 . .. .... .... 0011 . . . 1 .... @3same
VCGE_U_3s 1111 001 1 0 . .. .... .... 0011 . . . 1 .... @3same
VSHL_S_3s 1111 001 0 0 . .. .... .... 0100 . . . 0 .... @3same
VSHL_U_3s 1111 001 1 0 . .. .... .... 0100 . . . 0 .... @3same
VMAX_S_3s 1111 001 0 0 . .. .... .... 0110 . . . 0 .... @3same
VMAX_U_3s 1111 001 1 0 . .. .... .... 0110 . . . 0 .... @3same
VMIN_S_3s 1111 001 0 0 . .. .... .... 0110 . . . 1 .... @3same
@ -75,3 +78,9 @@ VSUB_3s 1111 001 1 0 . .. .... .... 1000 . . . 0 .... @3same
VTST_3s 1111 001 0 0 . .. .... .... 1000 . . . 1 .... @3same
VCEQ_3s 1111 001 1 0 . .. .... .... 1000 . . . 1 .... @3same
VMLA_3s 1111 001 0 0 . .. .... .... 1001 . . . 0 .... @3same
VMLS_3s 1111 001 1 0 . .. .... .... 1001 . . . 0 .... @3same
VMUL_3s 1111 001 0 0 . .. .... .... 1001 . . . 1 .... @3same
VMUL_p_3s 1111 001 1 0 . .. .... .... 1001 . . . 1 .... @3same

44
qemu/target/arm/translate-neon.inc.c
View file

@ -644,6 +644,7 @@ DO_3SAME_NO_SZ_3(VMAX_S, tcg_gen_gvec_smax)
DO_3SAME_NO_SZ_3(VMAX_U, tcg_gen_gvec_umax)
DO_3SAME_NO_SZ_3(VMIN_S, tcg_gen_gvec_smin)
DO_3SAME_NO_SZ_3(VMIN_U, tcg_gen_gvec_umin)
DO_3SAME_NO_SZ_3(VMUL, tcg_gen_gvec_mul)
#define DO_3SAME_CMP(INSN, COND) \
static void gen_##INSN##_3s(TCGContext *s, unsigned vece, uint32_t rd_ofs, \
@ -681,3 +682,46 @@ DO_3SAME_GVEC4(VQADD_S, sqadd_op)
DO_3SAME_GVEC4(VQADD_U, uqadd_op)
DO_3SAME_GVEC4(VQSUB_S, sqsub_op)
DO_3SAME_GVEC4(VQSUB_U, uqsub_op)
static void gen_VMUL_p_3s(TCGContext *s, unsigned vece, uint32_t rd_ofs, uint32_t rn_ofs,
uint32_t rm_ofs, uint32_t oprsz, uint32_t maxsz)
{
tcg_gen_gvec_3_ool(s, rd_ofs, rn_ofs, rm_ofs, oprsz, maxsz,
0, gen_helper_gvec_pmul_b);
}
static bool trans_VMUL_p_3s(DisasContext *s, arg_3same *a)
{
if (a->size != 0) {
return false;
}
return do_3same(s, a, gen_VMUL_p_3s);
}
#define DO_3SAME_GVEC3_NO_SZ_3(INSN, OPARRAY) \
static void gen_##INSN##_3s(TCGContext *s, unsigned vece, uint32_t rd_ofs, \
uint32_t rn_ofs, uint32_t rm_ofs, \
uint32_t oprsz, uint32_t maxsz) \
{ \
tcg_gen_gvec_3(s, rd_ofs, rn_ofs, rm_ofs, \
oprsz, maxsz, &OPARRAY[vece]); \
} \
DO_3SAME_NO_SZ_3(INSN, gen_##INSN##_3s)
DO_3SAME_GVEC3_NO_SZ_3(VMLA, mla_op)
DO_3SAME_GVEC3_NO_SZ_3(VMLS, mls_op)
#define DO_3SAME_GVEC3_SHIFT(INSN, OPARRAY) \
static void gen_##INSN##_3s(TCGContext *s, unsigned vece, uint32_t rd_ofs, \
uint32_t rn_ofs, uint32_t rm_ofs, \
uint32_t oprsz, uint32_t maxsz) \
{ \
/* Note the operation is vshl vd,vm,vn */ \
tcg_gen_gvec_3(s, rd_ofs, rm_ofs, rn_ofs, \
oprsz, maxsz, &OPARRAY[vece]); \
} \
DO_3SAME(INSN, gen_##INSN##_3s)
DO_3SAME_GVEC3_SHIFT(VSHL_S, sshl_op)
DO_3SAME_GVEC3_SHIFT(VSHL_U, ushl_op)

28
qemu/target/arm/translate.c
View file

@ -4979,31 +4979,6 @@ static int disas_neon_data_insn(DisasContext *s, uint32_t insn)
}
return 1;
case NEON_3R_VMUL: /* VMUL */
if (u) {
/* Polynomial case allows only P8. */
if (size != 0) {
return 1;
}
tcg_gen_gvec_3_ool(tcg_ctx, rd_ofs, rn_ofs, rm_ofs, vec_size, vec_size,
0, gen_helper_gvec_pmul_b);
} else {
tcg_gen_gvec_mul(tcg_ctx, size, rd_ofs, rn_ofs, rm_ofs,
vec_size, vec_size);
}
return 0;
case NEON_3R_VML: /* VMLA, VMLS */
tcg_gen_gvec_3(tcg_ctx, rd_ofs, rn_ofs, rm_ofs, vec_size, vec_size,
u ? &mls_op[size] : &mla_op[size]);
return 0;
case NEON_3R_VSHL:
/* Note the operation is vshl vd,vm,vn */
tcg_gen_gvec_3(tcg_ctx, rd_ofs, rm_ofs, rn_ofs, vec_size, vec_size,
u ? &ushl_op[size] : &sshl_op[size]);
return 0;
case NEON_3R_VADD_VSUB:
case NEON_3R_LOGIC:
case NEON_3R_VMAX:
@ -5013,6 +4988,9 @@ static int disas_neon_data_insn(DisasContext *s, uint32_t insn)
case NEON_3R_VCGE:
case NEON_3R_VQADD:
case NEON_3R_VQSUB:
case NEON_3R_VMUL:
case NEON_3R_VML:
case NEON_3R_VSHL:
/* Already handled by decodetree */
return 1;
}