ldq takes a pointer to the first byte to load the 64-bit word in;
ldo takes a pointer to the first byte of the ZMMReg. Make them
consistent, which will be useful in the new SSE decoder's
load/writeback routines.
Signed-off-by: Paolo Bonzini <pbonz...@redhat.com>
---
target/i386/tcg/translate.c | 44 +++++++++++++++++++------------------
1 file changed, 23 insertions(+), 21 deletions(-)
diff --git a/target/i386/tcg/translate.c b/target/i386/tcg/translate.c
index 001af76663..9a85010dcd 100644
--- a/target/i386/tcg/translate.c
+++ b/target/i386/tcg/translate.c
@@ -2761,28 +2761,29 @@ static inline void gen_ldo_env_A0(DisasContext *s, int
offset)
{
int mem_index = s->mem_index;
tcg_gen_qemu_ld_i64(s->tmp1_i64, s->A0, mem_index, MO_LEUQ);
- tcg_gen_st_i64(s->tmp1_i64, cpu_env, offset + offsetof(ZMMReg, ZMM_Q(0)));
+ tcg_gen_st_i64(s->tmp1_i64, cpu_env, offset + offsetof(XMMReg, XMM_Q(0)));
tcg_gen_addi_tl(s->tmp0, s->A0, 8);
tcg_gen_qemu_ld_i64(s->tmp1_i64, s->tmp0, mem_index, MO_LEUQ);
- tcg_gen_st_i64(s->tmp1_i64, cpu_env, offset + offsetof(ZMMReg, ZMM_Q(1)));
+ tcg_gen_st_i64(s->tmp1_i64, cpu_env, offset + offsetof(XMMReg, XMM_Q(1)));
}
static inline void gen_sto_env_A0(DisasContext *s, int offset)
{
int mem_index = s->mem_index;
- tcg_gen_ld_i64(s->tmp1_i64, cpu_env, offset + offsetof(ZMMReg, ZMM_Q(0)));
+ offset -= offsetof(ZMMReg, ZMM_Q(0));
+ tcg_gen_ld_i64(s->tmp1_i64, cpu_env, offset + offsetof(XMMReg, XMM_Q(0)));
tcg_gen_qemu_st_i64(s->tmp1_i64, s->A0, mem_index, MO_LEUQ);
tcg_gen_addi_tl(s->tmp0, s->A0, 8);
- tcg_gen_ld_i64(s->tmp1_i64, cpu_env, offset + offsetof(ZMMReg, ZMM_Q(1)));
+ tcg_gen_ld_i64(s->tmp1_i64, cpu_env, offset + offsetof(XMMReg, XMM_Q(1)));
tcg_gen_qemu_st_i64(s->tmp1_i64, s->tmp0, mem_index, MO_LEUQ);
}
static inline void gen_op_movo(DisasContext *s, int d_offset, int s_offset)
{
- tcg_gen_ld_i64(s->tmp1_i64, cpu_env, s_offset + offsetof(ZMMReg,
ZMM_Q(0)));
- tcg_gen_st_i64(s->tmp1_i64, cpu_env, d_offset + offsetof(ZMMReg,
ZMM_Q(0)));
- tcg_gen_ld_i64(s->tmp1_i64, cpu_env, s_offset + offsetof(ZMMReg,
ZMM_Q(1)));
- tcg_gen_st_i64(s->tmp1_i64, cpu_env, d_offset + offsetof(ZMMReg,
ZMM_Q(1)));
+ tcg_gen_ld_i64(s->tmp1_i64, cpu_env, s_offset + offsetof(XMMReg,
XMM_Q(0)));
+ tcg_gen_st_i64(s->tmp1_i64, cpu_env, d_offset + offsetof(XMMReg,
XMM_Q(0)));
+ tcg_gen_ld_i64(s->tmp1_i64, cpu_env, s_offset + offsetof(XMMReg,
XMM_Q(1)));
+ tcg_gen_st_i64(s->tmp1_i64, cpu_env, d_offset + offsetof(XMMReg,
XMM_Q(1)));
}
static inline void gen_op_movq(DisasContext *s, int d_offset, int s_offset)
@@ -2804,6 +2805,7 @@ static inline void gen_op_movq_env_0(DisasContext *s, int
d_offset)
}
#define ZMM_OFFSET(reg) offsetof(CPUX86State, xmm_regs[reg])
+#define XMM_OFFSET(reg) offsetof(CPUX86State, xmm_regs[reg].ZMM_X(0))
typedef void (*SSEFunc_i_ep)(TCGv_i32 val, TCGv_ptr env, TCGv_ptr reg);
typedef void (*SSEFunc_l_ep)(TCGv_i64 val, TCGv_ptr env, TCGv_ptr reg);
@@ -3317,13 +3319,13 @@ static void gen_sse(CPUX86State *env, DisasContext *s,
int b,
if (mod == 3)
goto illegal_op;
gen_lea_modrm(env, s, modrm);
- gen_sto_env_A0(s, ZMM_OFFSET(reg));
+ gen_sto_env_A0(s, XMM_OFFSET(reg));
break;
case 0x3f0: /* lddqu */
if (mod == 3)
goto illegal_op;
gen_lea_modrm(env, s, modrm);
- gen_ldo_env_A0(s, ZMM_OFFSET(reg));
+ gen_ldo_env_A0(s, XMM_OFFSET(reg));
break;
case 0x22b: /* movntss */
case 0x32b: /* movntsd */
@@ -3392,10 +3394,10 @@ static void gen_sse(CPUX86State *env, DisasContext *s,
int b,
case 0x26f: /* movdqu xmm, ea */
if (mod != 3) {
gen_lea_modrm(env, s, modrm);
- gen_ldo_env_A0(s, ZMM_OFFSET(reg));
+ gen_ldo_env_A0(s, XMM_OFFSET(reg));
} else {
rm = (modrm & 7) | REX_B(s);
- gen_op_movo(s, ZMM_OFFSET(reg), ZMM_OFFSET(rm));
+ gen_op_movo(s, XMM_OFFSET(reg), XMM_OFFSET(rm));
}
break;
case 0x210: /* movss xmm, ea */
@@ -3451,7 +3453,7 @@ static void gen_sse(CPUX86State *env, DisasContext *s,
int b,
case 0x212: /* movsldup */
if (mod != 3) {
gen_lea_modrm(env, s, modrm);
- gen_ldo_env_A0(s, ZMM_OFFSET(reg));
+ gen_ldo_env_A0(s, XMM_OFFSET(reg));
} else {
rm = (modrm & 7) | REX_B(s);
gen_op_movl(s, offsetof(CPUX86State, xmm_regs[reg].ZMM_L(0)),
@@ -3493,7 +3495,7 @@ static void gen_sse(CPUX86State *env, DisasContext *s,
int b,
case 0x216: /* movshdup */
if (mod != 3) {
gen_lea_modrm(env, s, modrm);
- gen_ldo_env_A0(s, ZMM_OFFSET(reg));
+ gen_ldo_env_A0(s, XMM_OFFSET(reg));
} else {
rm = (modrm & 7) | REX_B(s);
gen_op_movl(s, offsetof(CPUX86State, xmm_regs[reg].ZMM_L(1)),
@@ -3587,10 +3589,10 @@ static void gen_sse(CPUX86State *env, DisasContext *s,
int b,
case 0x27f: /* movdqu ea, xmm */
if (mod != 3) {
gen_lea_modrm(env, s, modrm);
- gen_sto_env_A0(s, ZMM_OFFSET(reg));
+ gen_sto_env_A0(s, XMM_OFFSET(reg));
} else {
rm = (modrm & 7) | REX_B(s);
- gen_op_movo(s, ZMM_OFFSET(rm), ZMM_OFFSET(reg));
+ gen_op_movo(s, XMM_OFFSET(rm), XMM_OFFSET(reg));
}
break;
case 0x211: /* movss ea, xmm */
@@ -3742,7 +3744,7 @@ static void gen_sse(CPUX86State *env, DisasContext *s,
int b,
gen_helper_enter_mmx(cpu_env);
if (mod != 3) {
gen_lea_modrm(env, s, modrm);
- op2_offset = offsetof(CPUX86State,xmm_t0);
+ op2_offset = offsetof(CPUX86State,xmm_t0.ZMM_X(0));
gen_ldo_env_A0(s, op2_offset);
} else {
rm = (modrm & 7) | REX_B(s);
@@ -3906,9 +3908,9 @@ static void gen_sse(CPUX86State *env, DisasContext *s,
int b,
}
if (b1) {
- op1_offset = ZMM_OFFSET(reg);
+ op1_offset = XMM_OFFSET(reg);
if (mod == 3) {
- op2_offset = ZMM_OFFSET(rm | REX_B(s));
+ op2_offset = XMM_OFFSET(rm | REX_B(s));
} else {
op2_offset = offsetof(CPUX86State,xmm_t0);
gen_lea_modrm(env, s, modrm);
@@ -4516,7 +4518,7 @@ static void gen_sse(CPUX86State *env, DisasContext *s,
int b,
if (mod == 3) {
op2_offset = ZMM_OFFSET(rm | REX_B(s));
} else {
- op2_offset = offsetof(CPUX86State, xmm_t0);
+ op2_offset = offsetof(CPUX86State, xmm_t0.ZMM_X(0));
gen_lea_modrm(env, s, modrm);
gen_ldo_env_A0(s, op2_offset);
}
@@ -4625,7 +4627,7 @@ static void gen_sse(CPUX86State *env, DisasContext *s,
int b,
break;
default:
/* 128 bit access */
- gen_ldo_env_A0(s, op2_offset);
+ gen_ldo_env_A0(s, offsetof(CPUX86State, xmm_t0.ZMM_X(0)));
break;
}
} else {
--
2.37.2
