The difference being that INSN_* has been pre-shifted into place.
This should result in less runtime shifting of constants.
The patch could be split into smaller pieces for clarity...
Signed-off-by: Richard Henderson <r...@twiddle.net>
---
tcg/aarch64/tcg-target.c | 518 ++++++++++++++++++++++++-----------------------
1 file changed, 269 insertions(+), 249 deletions(-)
diff --git a/tcg/aarch64/tcg-target.c b/tcg/aarch64/tcg-target.c
index 76595b4..3640486 100644
--- a/tcg/aarch64/tcg-target.c
+++ b/tcg/aarch64/tcg-target.c
@@ -212,7 +212,76 @@ static inline int tcg_target_const_match(tcg_target_long
val,
return 0;
}
-enum aarch64_cond_code {
+typedef enum {
+ /* Logical immediate instructions */
+ INSN_ANDI = 0x12000000,
+ INSN_ORRI = 0x32000000,
+ INSN_EORI = 0x52000000,
+
+ /* Logical shifted register instructions */
+ INSN_AND = 0x0a000000,
+ INSN_BIC = 0x0a200000,
+ INSN_ORR = 0x2a000000,
+ INSN_ORN = 0x2a200000,
+ INSN_EOR = 0x4a000000,
+ INSN_EON = 0x4a200000,
+
+ /* Move wide immediate instructions */
+ INSN_MOVN = 0x12800000,
+ INSN_MOVZ = 0x52800000,
+ INSN_MOVK = 0x72800000,
+
+ /* Add/subtract immediate instructions */
+ INSN_ADDI = 0x11000000,
+ INSN_ADDSI = 0x31000000,
+ INSN_SUBI = 0x51000000,
+ INSN_SUBSI = 0x71000000,
+
+ /* Add/subtract shifted register instructions */
+ INSN_ADD = 0x0b000000,
+ INSN_SUB = 0x4b000000,
+ INSN_SUBS = 0x6b000000,
+
+ /* Data-processing (1 source) instructions */
+ INSN_REV16 = 0x5ac00400,
+ INSN_REVx = 0xdac00c00,
+ INSN_REVw = 0x5ac00800,
+
+ /* Data-processing (2 source) instructions */
+ INSN_LSLV = 0x1ac02000,
+ INSN_LSRV = 0x1ac02400,
+ INSN_ASRV = 0x1ac02800,
+ INSN_RORV = 0x1ac02c00,
+ INSN_MUL = 0x1b007c00, /* MADD alias with Ra = xzr */
+
+ /* Bitfield instructions */
+ INSN_SBFM = 0x13000000,
+ INSN_UBFM = 0x53000000,
+ INSN_EXTR = 0x13800000,
+
+ /* Conditional select instructions */
+ INSN_CSINC = 0x1a800400,
+
+ /* Branch instructions */
+ INSN_B = 0x14000000,
+ INSN_BL = 0x94000000,
+ INSN_BR = 0xd61f0000,
+ INSN_BLR = 0xd63f0000,
+ INSN_RET = 0xd65f0000,
+ INSN_B_C = 0x54000000,
+
+ /* System instructions */
+ INSN_NOP = 0xd503201f,
+} AArch64Insn;
+
+typedef enum {
+ E32 = 0,
+ E64 = 0x80000000u,
+} AArch64Ext;
+
+#define EXT(cond) (cond ? E64 : E32)
+
+typedef enum {
COND_EQ = 0x0,
COND_NE = 0x1,
COND_CS = 0x2, /* Unsigned greater or equal */
@@ -231,9 +300,9 @@ enum aarch64_cond_code {
COND_LE = 0xd,
COND_AL = 0xe,
COND_NV = 0xf, /* behaves like COND_AL here */
-};
+} AArch64Cond;
-static const enum aarch64_cond_code tcg_cond_to_aarch64[] = {
+static const AArch64Cond tcg_cond_to_aarch64[] = {
[TCG_COND_EQ] = COND_EQ,
[TCG_COND_NE] = COND_NE,
[TCG_COND_LT] = COND_LT,
@@ -261,31 +330,6 @@ enum aarch64_ldst_op_type { /* type of operation */
LDST_LD_S_W = 0xc, /* load and sign-extend into Wt */
};
-enum aarch64_arith_opc {
- ARITH_AND = 0x0a,
- ARITH_ADD = 0x0b,
- ARITH_ADDI = 0x11,
- ARITH_ANDI = 0x12,
- ARITH_OR = 0x2a,
- ARITH_ADDS = 0x2b,
- ARITH_ADDSI = 0x31,
- ARITH_ORI = 0x32,
- ARITH_XOR = 0x4a,
- ARITH_SUB = 0x4b,
- ARITH_SUBI = 0x51,
- ARITH_XORI = 0x52,
- ARITH_ANDS = 0x6a,
- ARITH_SUBS = 0x6b,
- ARITH_SUBSI = 0x71,
-};
-
-enum aarch64_srr_opc {
- SRR_SHL = 0x0,
- SRR_SHR = 0x4,
- SRR_SAR = 0x8,
- SRR_ROR = 0xc
-};
-
static inline enum aarch64_ldst_op_data
aarch64_ldst_get_data(TCGOpcode tcg_op)
{
@@ -395,46 +439,78 @@ static inline void tcg_out_ldst_12(TCGContext *s,
| op_type << 20 | scaled_uimm << 10 | rn << 5 | rd);
}
-static inline void tcg_out_movr(TCGContext *s, int ext, TCGReg rd, TCGReg src)
+/* Suitable for add/sub and logical shifted register instructions. */
+static inline void tcg_out_arith(TCGContext *s, AArch64Insn insn,
+ AArch64Ext ext, TCGReg rd, TCGReg rn,
+ TCGReg rm, int shift_imm)
{
- /* register to register move using MOV (shifted register with no shift) */
- /* using MOV 0x2a0003e0 | (shift).. */
- unsigned int base = ext ? 0xaa0003e0 : 0x2a0003e0;
- tcg_out32(s, base | src << 16 | rd);
+ unsigned int shift;
+ if (shift_imm == 0) {
+ shift = 0;
+ } else if (shift_imm > 0) {
+ shift = shift_imm << 10 | 1 << 22;
+ } else /* (shift_imm < 0) */ {
+ shift = (-shift_imm) << 10;
+ }
+ tcg_out32(s, insn | ext | shift | rm << 16 | rn << 5 | rd);
}
-static inline void tcg_out_movi_aux(TCGContext *s,
- TCGReg rd, uint64_t value)
+static inline void tcg_out_aimm(TCGContext *s, AArch64Insn insn,
+ AArch64Ext ext, TCGReg rd, TCGReg rn,
+ uint64_t aimm)
{
- uint32_t half, base, shift, movk = 0;
- /* construct halfwords of the immediate with MOVZ/MOVK with LSL */
- /* using MOVZ 0x52800000 | extended reg.. */
- base = (value > 0xffffffff) ? 0xd2800000 : 0x52800000;
- /* count trailing zeros in 16 bit steps, mapping 64 to 0. Emit the
- first MOVZ with the half-word immediate skipping the zeros, with a shift
- (LSL) equal to this number. Then morph all next instructions into MOVKs.
+ if (aimm > 0xfff) {
+ assert((aimm & 0xfff) == 0);
+ aimm >>= 12;
+ assert(aimm <= 0xfff);
+ aimm |= 1 << 12; /* apply LSL 12 */
+ }
+ tcg_out32(s, insn | ext | aimm << 10 | rn << 5 | rd);
+}
+
+static void tcg_out_limm(TCGContext *s, AArch64Insn insn, AArch64Ext ext,
+ TCGReg rd, TCGReg rn, tcg_target_long val)
+{
+ int index = (ext ? find_bitmask64(val) : find_bitmask32(val));
+ assert(index >= 0);
+ tcg_out32(s, insn | ext | bitmask_enc[index] << 10 | rn << 5 | rd);
+}
+
+/* Register-register move, assuming XSP not XZR. */
+static inline void tcg_out_movr(TCGContext *s, AArch64Ext ext,
+ TCGReg dest, TCGReg src)
+{
+ tcg_out_aimm(s, INSN_ADDI, ext, dest, src, 0);
+}
+
+static void tcg_out_movi(TCGContext *s, TCGType type, TCGReg rd,
+ tcg_target_long value)
+{
+ AArch64Insn insn = INSN_MOVZ;
+
+ if (type == TCG_TYPE_I32) {
+ value = (uint32_t)value;
+ }
+
+ /* Construct halfwords of the immediate with MOVZ/MOVK with LSL.
+ Count trailing zeros in 16 bit steps, mapping 64 to 0. Emit the
+ first MOVZ with the half-word immediate skipping the zeros, with
+ a shift (LSL) equal to this number. Then all other insns are MOVKs.
Zero the processed half-word in the value, continue until empty.
We build the final result 16bits at a time with up to 4 instructions,
but do not emit instructions for 16bit zero holes. */
do {
- shift = ctz64(value) & (63 & -16);
- half = (value >> shift) & 0xffff;
- tcg_out32(s, base | movk | shift << 17 | half << 5 | rd);
- movk = 0x20000000; /* morph next MOVZs into MOVKs */
+ unsigned shift = ctz64(value) & (63 & -16);
+ unsigned half = (value >> shift) & 0xffff;
+ AArch64Ext ext = EXT(shift >= 32);
+
+ tcg_out32(s, insn | ext | shift << 17 | half << 5 | rd);
+
+ insn = INSN_MOVK;
value &= ~(0xffffUL << shift);
} while (value);
}
-static inline void tcg_out_movi(TCGContext *s, TCGType type,
- TCGReg rd, tcg_target_long value)
-{
- if (type == TCG_TYPE_I64) {
- tcg_out_movi_aux(s, rd, value);
- } else {
- tcg_out_movi_aux(s, rd, value & 0xffffffff);
- }
-}
-
static inline void tcg_out_ldst_r(TCGContext *s,
enum aarch64_ldst_op_data op_data,
enum aarch64_ldst_op_type op_type,
@@ -475,19 +551,11 @@ static inline void tcg_out_ldst(TCGContext *s, enum
aarch64_ldst_op_data data,
tcg_out_ldst_r(s, data, type, rd, rn, TCG_REG_TMP);
}
-/* mov alias implemented with add immediate, useful to move to/from SP */
-static inline void tcg_out_movr_sp(TCGContext *s, int ext, TCGReg rd, TCGReg
rn)
-{
- /* using ADD 0x11000000 | (ext) | rn << 5 | rd */
- unsigned int base = ext ? 0x91000000 : 0x11000000;
- tcg_out32(s, base | rn << 5 | rd);
-}
-
static inline void tcg_out_mov(TCGContext *s,
TCGType type, TCGReg ret, TCGReg arg)
{
if (ret != arg) {
- tcg_out_movr(s, type == TCG_TYPE_I64, ret, arg);
+ tcg_out_movr(s, EXT(type == TCG_TYPE_I64), ret, arg);
}
}
@@ -505,90 +573,35 @@ static inline void tcg_out_st(TCGContext *s, TCGType
type, TCGReg arg,
arg, arg1, arg2);
}
-static inline void tcg_out_arith(TCGContext *s, enum aarch64_arith_opc opc,
- int ext, TCGReg rd, TCGReg rn, TCGReg rm,
- int shift_imm)
-{
- /* Using shifted register arithmetic operations */
- /* if extended register operation (64bit) just OR with 0x80 << 24 */
- unsigned int shift, base = ext ? (0x80 | opc) << 24 : opc << 24;
- if (shift_imm == 0) {
- shift = 0;
- } else if (shift_imm > 0) {
- shift = shift_imm << 10 | 1 << 22;
- } else /* (shift_imm < 0) */ {
- shift = (-shift_imm) << 10;
- }
- tcg_out32(s, base | rm << 16 | shift | rn << 5 | rd);
-}
-
-static inline void tcg_out_aimm(TCGContext *s, enum aarch64_arith_opc opc,
- int ext, TCGReg rd, TCGReg rn, uint64_t aimm)
-{
- unsigned int base = (ext ? 0x80 | opc : opc) << 24;
-
- if (aimm > 0xfff) {
- assert((aimm & 0xfff) == 0);
- aimm >>= 12;
- base |= 1 << 22; /* apply LSL 12 */
- assert(aimm <= 0xfff);
- }
- tcg_out32(s, base | (aimm << 10) | (rn << 5) | rd);
-}
-
-static void tcg_out_limm(TCGContext *s, enum aarch64_arith_opc opc, int ext,
- TCGReg rd, TCGReg rn, tcg_target_long val)
-{
- int index = (ext ? find_bitmask64(val) : find_bitmask32(val));
- unsigned base = (ext ? 0x80 | opc : opc) << 24;
-
- assert(index >= 0);
- tcg_out32(s, base | bitmask_enc[index] << 10 | rn << 5 | rd);
-}
-
-static inline void tcg_out_mul(TCGContext *s, int ext,
- TCGReg rd, TCGReg rn, TCGReg rm)
+static inline void tcg_out_data2(TCGContext *s, AArch64Insn insn,
+ AArch64Ext ext, TCGReg rd,
+ TCGReg rn, TCGReg rm)
{
- /* Using MADD 0x1b000000 with Ra = wzr alias MUL 0x1b007c00 */
- unsigned int base = ext ? 0x9b007c00 : 0x1b007c00;
- tcg_out32(s, base | rm << 16 | rn << 5 | rd);
+ tcg_out32(s, insn | ext | rm << 16 | rn << 5 | rd);
}
-static inline void tcg_out_shiftrot_reg(TCGContext *s,
- enum aarch64_srr_opc opc, int ext,
- TCGReg rd, TCGReg rn, TCGReg rm)
+static inline void tcg_out_ubfm(TCGContext *s, AArch64Ext ext, TCGReg rd,
+ TCGReg rn, unsigned int a, unsigned int b)
{
- /* using 2-source data processing instructions 0x1ac02000 */
- unsigned int base = ext ? 0x9ac02000 : 0x1ac02000;
- tcg_out32(s, base | rm << 16 | opc << 8 | rn << 5 | rd);
-}
-
-static inline void tcg_out_ubfm(TCGContext *s, int ext, TCGReg rd, TCGReg rn,
- unsigned int a, unsigned int b)
-{
- /* Using UBFM 0x53000000 Wd, Wn, a, b */
- unsigned int base = ext ? 0xd3400000 : 0x53000000;
+ unsigned int base = INSN_UBFM | (ext ? 0x80400000 : 0);
tcg_out32(s, base | a << 16 | b << 10 | rn << 5 | rd);
}
-static inline void tcg_out_sbfm(TCGContext *s, int ext, TCGReg rd, TCGReg rn,
- unsigned int a, unsigned int b)
+static inline void tcg_out_sbfm(TCGContext *s, AArch64Ext ext, TCGReg rd,
+ TCGReg rn, unsigned int a, unsigned int b)
{
- /* Using SBFM 0x13000000 Wd, Wn, a, b */
- unsigned int base = ext ? 0x93400000 : 0x13000000;
+ unsigned int base = INSN_SBFM | (ext ? 0x80400000 : 0);
tcg_out32(s, base | a << 16 | b << 10 | rn << 5 | rd);
}
-static inline void tcg_out_extr(TCGContext *s, int ext, TCGReg rd,
+static inline void tcg_out_extr(TCGContext *s, AArch64Ext ext, TCGReg rd,
TCGReg rn, TCGReg rm, unsigned int a)
{
- /* Using EXTR 0x13800000 Wd, Wn, Wm, a */
- unsigned int base = ext ? 0x93c00000 : 0x13800000;
- tcg_out32(s, base | rm << 16 | a << 10 | rn << 5 | rd);
+ tcg_out32(s, INSN_EXTR | ext | rm << 16 | a << 10 | rn << 5 | rd);
}
-static inline void tcg_out_shl(TCGContext *s, int ext,
- TCGReg rd, TCGReg rn, unsigned int m)
+static inline void tcg_out_shli(TCGContext *s, AArch64Ext ext, TCGReg rd,
+ TCGReg rn, unsigned int m)
{
int bits, max;
bits = ext ? 64 : 32;
@@ -596,29 +609,29 @@ static inline void tcg_out_shl(TCGContext *s, int ext,
tcg_out_ubfm(s, ext, rd, rn, bits - (m & max), max - (m & max));
}
-static inline void tcg_out_shr(TCGContext *s, int ext,
- TCGReg rd, TCGReg rn, unsigned int m)
+static inline void tcg_out_shri(TCGContext *s, AArch64Ext ext, TCGReg rd,
+ TCGReg rn, unsigned int m)
{
int max = ext ? 63 : 31;
tcg_out_ubfm(s, ext, rd, rn, m & max, max);
}
-static inline void tcg_out_sar(TCGContext *s, int ext,
- TCGReg rd, TCGReg rn, unsigned int m)
+static inline void tcg_out_sari(TCGContext *s, AArch64Ext ext, TCGReg rd,
+ TCGReg rn, unsigned int m)
{
int max = ext ? 63 : 31;
tcg_out_sbfm(s, ext, rd, rn, m & max, max);
}
-static inline void tcg_out_rotr(TCGContext *s, int ext,
- TCGReg rd, TCGReg rn, unsigned int m)
+static inline void tcg_out_rotri(TCGContext *s, AArch64Ext ext, TCGReg rd,
+ TCGReg rn, unsigned int m)
{
int max = ext ? 63 : 31;
tcg_out_extr(s, ext, rd, rn, rn, m & max);
}
-static inline void tcg_out_rotl(TCGContext *s, int ext,
- TCGReg rd, TCGReg rn, unsigned int m)
+static inline void tcg_out_rotli(TCGContext *s, AArch64Ext ext, TCGReg rd,
+ TCGReg rn, unsigned int m)
{
int bits, max;
bits = ext ? 64 : 32;
@@ -626,29 +639,31 @@ static inline void tcg_out_rotl(TCGContext *s, int ext,
tcg_out_extr(s, ext, rd, rn, rn, bits - (m & max));
}
-static void tcg_out_cmp(TCGContext *s, int ext, TCGReg a,
+static void tcg_out_cmp(TCGContext *s, AArch64Ext ext, TCGReg a,
tcg_target_long b, bool const_b)
{
if (const_b) {
/* Using CMP alias SUBS xzr, Xn, const */
- enum aarch64_arith_opc opc = ARITH_SUBSI;
+ AArch64Insn insn = INSN_SUBSI;
if (b < 0) {
b = ~b;
- opc = ARITH_ADDSI;
+ insn = INSN_ADDSI;
}
- tcg_out_aimm(s, opc, ext, TCG_REG_XZR, a, b);
+ tcg_out_aimm(s, insn, ext, TCG_REG_XZR, a, b);
} else {
/* Using CMP alias SUBS wzr, Wn, Wm */
- tcg_out_arith(s, ARITH_SUBS, ext, TCG_REG_XZR, a, b, 0);
+ tcg_out_arith(s, INSN_SUBS, ext, TCG_REG_XZR, a, b, 0);
}
}
-static inline void tcg_out_cset(TCGContext *s, int ext, TCGReg rd, TCGCond c)
+static inline void tcg_out_cset(TCGContext *s, AArch64Ext ext,
+ TCGReg rd, TCGCond c)
{
/* Using CSET alias of CSINC 0x1a800400 Xd, XZR, XZR, invert(cond) */
- unsigned int base = ext ? 0x9a9f07e0 : 0x1a9f07e0;
- tcg_out32(s, base | tcg_cond_to_aarch64[tcg_invert_cond(c)] << 12 | rd);
+ unsigned int base = INSN_CSINC | TCG_REG_XZR << 16 | TCG_REG_XZR << 5;
+ unsigned int cond = tcg_cond_to_aarch64[tcg_invert_cond(c)];
+ tcg_out32(s, base | ext | cond << 12 | rd);
}
static inline void tcg_out_goto(TCGContext *s, tcg_target_long target)
@@ -661,7 +676,7 @@ static inline void tcg_out_goto(TCGContext *s,
tcg_target_long target)
tcg_abort();
}
- tcg_out32(s, 0x14000000 | (offset & 0x03ffffff));
+ tcg_out32(s, INSN_B | (offset & 0x03ffffff));
}
static inline void tcg_out_goto_noaddr(TCGContext *s)
@@ -672,7 +687,7 @@ static inline void tcg_out_goto_noaddr(TCGContext *s)
Mask away possible garbage in the high bits for the first translation,
while keeping the offset bits for retranslation. */
uint32_t insn;
- insn = (tcg_in32(s) & 0x03ffffff) | 0x14000000;
+ insn = (tcg_in32(s) & 0x03ffffff) | INSN_B;
tcg_out32(s, insn);
}
@@ -681,7 +696,7 @@ static inline void tcg_out_goto_cond_noaddr(TCGContext *s,
TCGCond c)
/* see comments in tcg_out_goto_noaddr */
uint32_t insn;
insn = tcg_in32(s) & (0x07ffff << 5);
- insn |= 0x54000000 | tcg_cond_to_aarch64[c];
+ insn |= INSN_B_C | tcg_cond_to_aarch64[c];
tcg_out32(s, insn);
}
@@ -697,17 +712,17 @@ static inline void tcg_out_goto_cond(TCGContext *s,
TCGCond c,
}
offset &= 0x7ffff;
- tcg_out32(s, 0x54000000 | tcg_cond_to_aarch64[c] | offset << 5);
+ tcg_out32(s, INSN_B_C | tcg_cond_to_aarch64[c] | offset << 5);
}
static inline void tcg_out_callr(TCGContext *s, TCGReg reg)
{
- tcg_out32(s, 0xd63f0000 | reg << 5);
+ tcg_out32(s, INSN_BLR | reg << 5);
}
static inline void tcg_out_gotor(TCGContext *s, TCGReg reg)
{
- tcg_out32(s, 0xd61f0000 | reg << 5);
+ tcg_out32(s, INSN_BR | reg << 5);
}
static inline void tcg_out_call(TCGContext *s, tcg_target_long target)
@@ -720,7 +735,7 @@ static inline void tcg_out_call(TCGContext *s,
tcg_target_long target)
tcg_out_movi(s, TCG_TYPE_I64, TCG_REG_TMP, target);
tcg_out_callr(s, TCG_REG_TMP);
} else {
- tcg_out32(s, 0x94000000 | (offset & 0x03ffffff));
+ tcg_out32(s, INSN_BL | (offset & 0x03ffffff));
}
}
@@ -741,7 +756,7 @@ aarch64_limm(unsigned int m, unsigned int r)
to test a 32bit reg against 0xff000000, pass M = 8, R = 8.
to test a 32bit reg against 0xff0000ff, pass M = 16, R = 8.
*/
-static inline void tcg_out_tst(TCGContext *s, int ext, TCGReg rn,
+static inline void tcg_out_tst(TCGContext *s, AArch64Ext ext, TCGReg rn,
unsigned int m, unsigned int r)
{
/* using TST alias of ANDS XZR, Xn,#bimm64 0x7200001f */
@@ -750,8 +765,8 @@ static inline void tcg_out_tst(TCGContext *s, int ext,
TCGReg rn,
}
/* and a register with a bit pattern, similarly to TST, no flags change */
-static inline void tcg_out_andi(TCGContext *s, int ext, TCGReg rd, TCGReg rn,
- unsigned int m, unsigned int r)
+static inline void tcg_out_andi(TCGContext *s, AArch64Ext ext, TCGReg rd,
+ TCGReg rn, unsigned int m, unsigned int r)
{
/* using AND 0x12000000 */
unsigned int base = ext ? 0x92400000 : 0x12000000;
@@ -760,8 +775,7 @@ static inline void tcg_out_andi(TCGContext *s, int ext,
TCGReg rd, TCGReg rn,
static inline void tcg_out_ret(TCGContext *s)
{
- /* emit RET { LR } */
- tcg_out32(s, 0xd65f03c0);
+ tcg_out32(s, INSN_RET | TCG_REG_LR << 5);
}
void aarch64_tb_set_jmp_target(uintptr_t jmp_addr, uintptr_t addr)
@@ -804,25 +818,23 @@ static inline void tcg_out_goto_label_cond(TCGContext *s,
}
}
-static inline void tcg_out_rev(TCGContext *s, int ext, TCGReg rd, TCGReg rm)
+static inline void tcg_out_rev(TCGContext *s, AArch64Ext ext,
+ TCGReg rd, TCGReg rm)
{
- /* using REV 0x5ac00800 */
- unsigned int base = ext ? 0xdac00c00 : 0x5ac00800;
- tcg_out32(s, base | rm << 5 | rd);
+ AArch64Insn insn = ext ? INSN_REVx : INSN_REVw;
+ tcg_out32(s, insn | rm << 5 | rd);
}
-static inline void tcg_out_rev16(TCGContext *s, int ext, TCGReg rd, TCGReg rm)
+static inline void tcg_out_rev16(TCGContext *s, AArch64Ext ext,
+ TCGReg rd, TCGReg rm)
{
- /* using REV16 0x5ac00400 */
- unsigned int base = ext ? 0xdac00400 : 0x5ac00400;
- tcg_out32(s, base | rm << 5 | rd);
+ tcg_out32(s, INSN_REV16 | ext | rm << 5 | rd);
}
-static inline void tcg_out_sxt(TCGContext *s, int ext, int s_bits,
+static inline void tcg_out_sxt(TCGContext *s, AArch64Ext ext, int s_bits,
TCGReg rd, TCGReg rn)
{
- /* using ALIASes SXTB 0x13001c00, SXTH 0x13003c00, SXTW 0x93407c00
- of SBFM Xd, Xn, #0, #7|15|31 */
+ /* using ALIASes SXTB, SXTH, SXTW of SBFM Xd, Xn, #0, #7|15|31 */
int bits = 8 * (1 << s_bits) - 1;
tcg_out_sbfm(s, ext, rd, rn, 0, bits);
}
@@ -830,38 +842,37 @@ static inline void tcg_out_sxt(TCGContext *s, int ext,
int s_bits,
static inline void tcg_out_uxt(TCGContext *s, int s_bits,
TCGReg rd, TCGReg rn)
{
- /* using ALIASes UXTB 0x53001c00, UXTH 0x53003c00
- of UBFM Wd, Wn, #0, #7|15 */
+ /* using ALIASes UXTB, UXTH, of UBFM Wd, Wn, #0, #7|15 */
int bits = 8 * (1 << s_bits) - 1;
tcg_out_ubfm(s, 0, rd, rn, 0, bits);
}
-static void tcg_out_addi(TCGContext *s, int ext, TCGReg rd, TCGReg rn,
+static void tcg_out_addi(TCGContext *s, AArch64Ext ext, TCGReg rd, TCGReg rn,
tcg_target_long aimm)
{
- enum aarch64_arith_opc opc = ARITH_ADDI;
+ AArch64Insn insn = INSN_ADDI;
tcg_target_long lo, hi;
if (aimm < 0) {
aimm = -aimm;
- opc = ARITH_SUBI;
+ insn = INSN_SUBI;
}
hi = aimm & 0xfff000;
lo = aimm & 0xfff;
assert(aimm == hi + lo);
if (hi != 0) {
- tcg_out_aimm(s, opc, ext, rd, rn, hi);
+ tcg_out_aimm(s, insn, ext, rd, rn, hi);
rn = rd;
}
if (lo != 0 || rd != rn) {
- tcg_out_aimm(s, opc, ext, rd, rn, lo);
+ tcg_out_aimm(s, insn, ext, rd, rn, lo);
}
}
static inline void tcg_out_nop(TCGContext *s)
{
- tcg_out32(s, 0xd503201f);
+ tcg_out32(s, INSN_NOP);
}
#ifdef CONFIG_SOFTMMU
@@ -887,17 +898,23 @@ static const void * const qemu_st_helpers[4] = {
static void tcg_out_qemu_ld_slow_path(TCGContext *s, TCGLabelQemuLdst *lb)
{
+ int opc = lb->opc;
+ int s_bits = opc & 3;
+
reloc_pc19(lb->label_ptr[0], (tcg_target_long)s->code_ptr);
- tcg_out_movr(s, 1, TCG_REG_X0, TCG_AREG0);
- tcg_out_movr(s, (TARGET_LONG_BITS == 64), TCG_REG_X1, lb->addrlo_reg);
+
+ tcg_out_mov(s, TCG_TYPE_PTR, TCG_REG_X0, TCG_AREG0);
+ tcg_out_mov(s, TCG_TYPE_TL, TCG_REG_X1, lb->addrlo_reg);
tcg_out_movi(s, TCG_TYPE_I32, TCG_REG_X2, lb->mem_index);
tcg_out_movi(s, TCG_TYPE_I64, TCG_REG_TMP,
- (tcg_target_long)qemu_ld_helpers[lb->opc & 3]);
+ (tcg_target_long)qemu_ld_helpers[s_bits]);
tcg_out_callr(s, TCG_REG_TMP);
- if (lb->opc & 0x04) {
- tcg_out_sxt(s, 1, lb->opc & 3, lb->datalo_reg, TCG_REG_X0);
+
+ if (opc & 0x04) {
+ tcg_out_sxt(s, E64, s_bits, lb->datalo_reg, TCG_REG_X0);
} else {
- tcg_out_movr(s, 1, lb->datalo_reg, TCG_REG_X0);
+ tcg_out_mov(s, s_bits == 3 ? TCG_TYPE_I64 : TCG_TYPE_I32,
+ lb->datalo_reg, TCG_REG_X0);
}
tcg_out_goto(s, (tcg_target_long)lb->raddr);
@@ -905,14 +922,17 @@ static void tcg_out_qemu_ld_slow_path(TCGContext *s,
TCGLabelQemuLdst *lb)
static void tcg_out_qemu_st_slow_path(TCGContext *s, TCGLabelQemuLdst *lb)
{
+ int s_bits = lb->opc;
+
reloc_pc19(lb->label_ptr[0], (tcg_target_long)s->code_ptr);
- tcg_out_movr(s, 1, TCG_REG_X0, TCG_AREG0);
- tcg_out_movr(s, (TARGET_LONG_BITS == 64), TCG_REG_X1, lb->addrlo_reg);
- tcg_out_movr(s, 1, TCG_REG_X2, lb->datalo_reg);
+ tcg_out_mov(s, TCG_TYPE_PTR, TCG_REG_X0, TCG_AREG0);
+ tcg_out_mov(s, TCG_TYPE_TL, TCG_REG_X1, lb->addrlo_reg);
+ tcg_out_mov(s, s_bits == 3 ? TCG_TYPE_I64 : TCG_TYPE_I32,
+ TCG_REG_X2, lb->datalo_reg);
tcg_out_movi(s, TCG_TYPE_I32, TCG_REG_X3, lb->mem_index);
tcg_out_movi(s, TCG_TYPE_I64, TCG_REG_TMP,
- (tcg_target_long)qemu_st_helpers[lb->opc & 3]);
+ (tcg_target_long)qemu_st_helpers[s_bits]);
tcg_out_callr(s, TCG_REG_TMP);
tcg_out_nop(s);
@@ -980,10 +1000,10 @@ static void tcg_out_tlb_read(TCGContext *s, TCGReg
addr_reg,
/* Add any "high bits" from the tlb offset to the env address into X2,
to take advantage of the LSL12 form of the addi instruction.
X2 = env + (tlb_offset & 0xfff000) */
- tcg_out_addi(s, 1, TCG_REG_X2, base, tlb_offset & 0xfff000);
+ tcg_out_addi(s, E64, TCG_REG_X2, base, tlb_offset & 0xfff000);
/* Merge the tlb index contribution into X2.
X2 = X2 + (X0 << CPU_TLB_ENTRY_BITS) */
- tcg_out_arith(s, ARITH_ADD, 1, TCG_REG_X2, TCG_REG_X2,
+ tcg_out_arith(s, INSN_ADD, E64, TCG_REG_X2, TCG_REG_X2,
TCG_REG_X0, -CPU_TLB_ENTRY_BITS);
/* Merge "low bits" from tlb offset, load the tlb comparator into X0.
X0 = load [X2 + (tlb_offset & 0x000fff)] */
@@ -997,7 +1017,7 @@ static void tcg_out_tlb_read(TCGContext *s, TCGReg
addr_reg,
(is_read ? offsetof(CPUTLBEntry, addr_read)
: offsetof(CPUTLBEntry, addr_write)));
/* Perform the address comparison. */
- tcg_out_cmp(s, (TARGET_LONG_BITS == 64), TCG_REG_X0, TCG_REG_X3, 0);
+ tcg_out_cmp(s, EXT(TARGET_LONG_BITS == 64), TCG_REG_X0, TCG_REG_X3, 0);
*label_ptr = s->code_ptr;
/* If not equal, we jump to the slow path. */
tcg_out_goto_cond_noaddr(s, TCG_COND_NE);
@@ -1024,8 +1044,8 @@ static void tcg_out_qemu_ld_direct(TCGContext *s, int
opc, TCGReg data_r,
case 1 | 4:
if (TCG_LDST_BSWAP) {
tcg_out_ldst_r(s, LDST_16, LDST_LD, data_r, addr_r, off_r);
- tcg_out_rev16(s, 0, data_r, data_r);
- tcg_out_sxt(s, 1, 1, data_r, data_r);
+ tcg_out_rev16(s, E32, data_r, data_r);
+ tcg_out_sxt(s, E64, 1, data_r, data_r);
} else {
tcg_out_ldst_r(s, LDST_16, LDST_LD_S_X, data_r, addr_r, off_r);
}
@@ -1033,14 +1053,14 @@ static void tcg_out_qemu_ld_direct(TCGContext *s, int
opc, TCGReg data_r,
case 2:
tcg_out_ldst_r(s, LDST_32, LDST_LD, data_r, addr_r, off_r);
if (TCG_LDST_BSWAP) {
- tcg_out_rev(s, 0, data_r, data_r);
+ tcg_out_rev(s, E32, data_r, data_r);
}
break;
case 2 | 4:
if (TCG_LDST_BSWAP) {
tcg_out_ldst_r(s, LDST_32, LDST_LD, data_r, addr_r, off_r);
- tcg_out_rev(s, 0, data_r, data_r);
- tcg_out_sxt(s, 1, 2, data_r, data_r);
+ tcg_out_rev(s, E32, data_r, data_r);
+ tcg_out_sxt(s, E64, 2, data_r, data_r);
} else {
tcg_out_ldst_r(s, LDST_32, LDST_LD_S_X, data_r, addr_r, off_r);
}
@@ -1048,7 +1068,7 @@ static void tcg_out_qemu_ld_direct(TCGContext *s, int
opc, TCGReg data_r,
case 3:
tcg_out_ldst_r(s, LDST_64, LDST_LD, data_r, addr_r, off_r);
if (TCG_LDST_BSWAP) {
- tcg_out_rev(s, 1, data_r, data_r);
+ tcg_out_rev(s, E64, data_r, data_r);
}
break;
default:
@@ -1065,7 +1085,7 @@ static void tcg_out_qemu_st_direct(TCGContext *s, int
opc, TCGReg data_r,
break;
case 1:
if (TCG_LDST_BSWAP) {
- tcg_out_rev16(s, 0, TCG_REG_TMP, data_r);
+ tcg_out_rev16(s, E32, TCG_REG_TMP, data_r);
tcg_out_ldst_r(s, LDST_16, LDST_ST, TCG_REG_TMP, addr_r, off_r);
} else {
tcg_out_ldst_r(s, LDST_16, LDST_ST, data_r, addr_r, off_r);
@@ -1073,7 +1093,7 @@ static void tcg_out_qemu_st_direct(TCGContext *s, int
opc, TCGReg data_r,
break;
case 2:
if (TCG_LDST_BSWAP) {
- tcg_out_rev(s, 0, TCG_REG_TMP, data_r);
+ tcg_out_rev(s, E32, TCG_REG_TMP, data_r);
tcg_out_ldst_r(s, LDST_32, LDST_ST, TCG_REG_TMP, addr_r, off_r);
} else {
tcg_out_ldst_r(s, LDST_32, LDST_ST, data_r, addr_r, off_r);
@@ -1081,7 +1101,7 @@ static void tcg_out_qemu_st_direct(TCGContext *s, int
opc, TCGReg data_r,
break;
case 3:
if (TCG_LDST_BSWAP) {
- tcg_out_rev(s, 1, TCG_REG_TMP, data_r);
+ tcg_out_rev(s, E64, TCG_REG_TMP, data_r);
tcg_out_ldst_r(s, LDST_64, LDST_ST, TCG_REG_TMP, addr_r, off_r);
} else {
tcg_out_ldst_r(s, LDST_64, LDST_ST, data_r, addr_r, off_r);
@@ -1196,7 +1216,7 @@ static void tcg_out_op(TCGContext *s, TCGOpcode opc,
{
/* ext will be set in the switch below, which will fall through to the
common code. It triggers the use of extended regs where appropriate. */
- int ext = 0;
+ AArch64Ext ext = E32;
switch (opc) {
case INDEX_op_exit_tb:
@@ -1252,9 +1272,10 @@ static void tcg_out_op(TCGContext *s, TCGOpcode opc,
break;
case INDEX_op_mov_i64:
- ext = 1; /* fall through */
+ tcg_out_mov(s, TCG_TYPE_I64, args[0], args[1]);
+ break;
case INDEX_op_mov_i32:
- tcg_out_movr(s, ext, args[0], args[1]);
+ tcg_out_mov(s, TCG_TYPE_I32, args[0], args[1]);
break;
case INDEX_op_movi_i64:
@@ -1265,123 +1286,122 @@ static void tcg_out_op(TCGContext *s, TCGOpcode opc,
break;
case INDEX_op_add_i64:
- ext = 1; /* fall through */
+ ext = E64; /* fall through */
case INDEX_op_add_i32:
if (const_args[2]) {
tcg_out_addi(s, ext, args[0], args[1], args[2]);
} else {
- tcg_out_arith(s, ARITH_ADD, ext, args[0], args[1], args[2], 0);
+ tcg_out_arith(s, INSN_ADD, ext, args[0], args[1], args[2], 0);
}
break;
case INDEX_op_sub_i64:
- ext = 1; /* fall through */
+ ext = E64; /* fall through */
case INDEX_op_sub_i32:
if (const_args[2]) {
tcg_out_addi(s, ext, args[0], args[1], -args[2]);
} else {
- tcg_out_arith(s, ARITH_SUB, ext, args[0], args[1], args[2], 0);
+ tcg_out_arith(s, INSN_SUB, ext, args[0], args[1], args[2], 0);
}
break;
case INDEX_op_and_i64:
- ext = 1; /* fall through */
+ ext = E64; /* fall through */
case INDEX_op_and_i32:
if (const_args[2]) {
- tcg_out_limm(s, ARITH_ANDI, ext, args[0], args[1], args[2]);
+ tcg_out_limm(s, INSN_ANDI, ext, args[0], args[1], args[2]);
} else {
- tcg_out_arith(s, ARITH_AND, ext, args[0], args[1], args[2], 0);
+ tcg_out_arith(s, INSN_AND, ext, args[0], args[1], args[2], 0);
}
break;
case INDEX_op_or_i64:
- ext = 1; /* fall through */
+ ext = E64; /* fall through */
case INDEX_op_or_i32:
if (const_args[2]) {
- tcg_out_limm(s, ARITH_ORI, ext, args[0], args[1], args[2]);
+ tcg_out_limm(s, INSN_ORRI, ext, args[0], args[1], args[2]);
} else {
- tcg_out_arith(s, ARITH_OR, ext, args[0], args[1], args[2], 0);
+ tcg_out_arith(s, INSN_ORR, ext, args[0], args[1], args[2], 0);
}
break;
case INDEX_op_xor_i64:
- ext = 1; /* fall through */
+ ext = E64; /* fall through */
case INDEX_op_xor_i32:
if (const_args[2]) {
- tcg_out_limm(s, ARITH_XORI, ext, args[0], args[1], args[2]);
+ tcg_out_limm(s, INSN_EORI, ext, args[0], args[1], args[2]);
} else {
- tcg_out_arith(s, ARITH_XOR, ext, args[0], args[1], args[2], 0);
+ tcg_out_arith(s, INSN_EOR, ext, args[0], args[1], args[2], 0);
}
break;
case INDEX_op_mul_i64:
- ext = 1; /* fall through */
+ ext = E64; /* fall through */
case INDEX_op_mul_i32:
- tcg_out_mul(s, ext, args[0], args[1], args[2]);
+ tcg_out_data2(s, INSN_MUL, ext, args[0], args[1], args[2]);
break;
case INDEX_op_shl_i64:
- ext = 1; /* fall through */
+ ext = E64; /* fall through */
case INDEX_op_shl_i32:
if (const_args[2]) { /* LSL / UBFM Wd, Wn, (32 - m) */
- tcg_out_shl(s, ext, args[0], args[1], args[2]);
+ tcg_out_shli(s, ext, args[0], args[1], args[2]);
} else { /* LSL / LSLV */
- tcg_out_shiftrot_reg(s, SRR_SHL, ext, args[0], args[1], args[2]);
+ tcg_out_data2(s, INSN_LSLV, ext, args[0], args[1], args[2]);
}
break;
case INDEX_op_shr_i64:
- ext = 1; /* fall through */
+ ext = E64; /* fall through */
case INDEX_op_shr_i32:
if (const_args[2]) { /* LSR / UBFM Wd, Wn, m, 31 */
- tcg_out_shr(s, ext, args[0], args[1], args[2]);
+ tcg_out_shri(s, ext, args[0], args[1], args[2]);
} else { /* LSR / LSRV */
- tcg_out_shiftrot_reg(s, SRR_SHR, ext, args[0], args[1], args[2]);
+ tcg_out_data2(s, INSN_LSRV, ext, args[0], args[1], args[2]);
}
break;
case INDEX_op_sar_i64:
- ext = 1; /* fall through */
+ ext = E64; /* fall through */
case INDEX_op_sar_i32:
if (const_args[2]) { /* ASR / SBFM Wd, Wn, m, 31 */
- tcg_out_sar(s, ext, args[0], args[1], args[2]);
+ tcg_out_sari(s, ext, args[0], args[1], args[2]);
} else { /* ASR / ASRV */
- tcg_out_shiftrot_reg(s, SRR_SAR, ext, args[0], args[1], args[2]);
+ tcg_out_data2(s, INSN_ASRV, ext, args[0], args[1], args[2]);
}
break;
case INDEX_op_rotr_i64:
- ext = 1; /* fall through */
+ ext = E64; /* fall through */
case INDEX_op_rotr_i32:
if (const_args[2]) { /* ROR / EXTR Wd, Wm, Wm, m */
- tcg_out_rotr(s, ext, args[0], args[1], args[2]);
+ tcg_out_rotri(s, ext, args[0], args[1], args[2]);
} else { /* ROR / RORV */
- tcg_out_shiftrot_reg(s, SRR_ROR, ext, args[0], args[1], args[2]);
+ tcg_out_data2(s, INSN_RORV, ext, args[0], args[1], args[2]);
}
break;
case INDEX_op_rotl_i64:
- ext = 1; /* fall through */
+ ext = E64; /* fall through */
case INDEX_op_rotl_i32: /* same as rotate right by (32 - m) */
if (const_args[2]) { /* ROR / EXTR Wd, Wm, Wm, 32 - m */
- tcg_out_rotl(s, ext, args[0], args[1], args[2]);
+ tcg_out_rotli(s, ext, args[0], args[1], args[2]);
} else {
- tcg_out_arith(s, ARITH_SUB, 0,
+ tcg_out_arith(s, INSN_SUB, E32,
TCG_REG_TMP, TCG_REG_XZR, args[2], 0);
- tcg_out_shiftrot_reg(s, SRR_ROR, ext,
- args[0], args[1], TCG_REG_TMP);
+ tcg_out_data2(s, INSN_RORV, ext, args[0], args[1], TCG_REG_TMP);
}
break;
case INDEX_op_brcond_i64:
- ext = 1; /* fall through */
- case INDEX_op_brcond_i32: /* CMP 0, 1, cond(2), label 3 */
+ ext = E64; /* fall through */
+ case INDEX_op_brcond_i32:
tcg_out_cmp(s, ext, args[0], args[1], const_args[1]);
tcg_out_goto_label_cond(s, args[2], args[3]);
break;
case INDEX_op_setcond_i64:
- ext = 1; /* fall through */
+ ext = E64; /* fall through */
case INDEX_op_setcond_i32:
tcg_out_cmp(s, ext, args[1], args[2], const_args[2]);
tcg_out_cset(s, 0, args[0], args[3]);
@@ -1425,7 +1445,7 @@ static void tcg_out_op(TCGContext *s, TCGOpcode opc,
break;
case INDEX_op_bswap64_i64:
- ext = 1; /* fall through */
+ ext = E64; /* fall through */
case INDEX_op_bswap32_i64:
case INDEX_op_bswap32_i32:
tcg_out_rev(s, ext, args[0], args[1]);
@@ -1436,17 +1456,17 @@ static void tcg_out_op(TCGContext *s, TCGOpcode opc,
break;
case INDEX_op_ext8s_i64:
- ext = 1; /* fall through */
+ ext = E64; /* fall through */
case INDEX_op_ext8s_i32:
tcg_out_sxt(s, ext, 0, args[0], args[1]);
break;
case INDEX_op_ext16s_i64:
- ext = 1; /* fall through */
+ ext = E64; /* fall through */
case INDEX_op_ext16s_i32:
tcg_out_sxt(s, ext, 1, args[0], args[1]);
break;
case INDEX_op_ext32s_i64:
- tcg_out_sxt(s, 1, 2, args[0], args[1]);
+ tcg_out_sxt(s, E64, 2, args[0], args[1]);
break;
case INDEX_op_ext8u_i64:
case INDEX_op_ext8u_i32:
@@ -1457,7 +1477,7 @@ static void tcg_out_op(TCGContext *s, TCGOpcode opc,
tcg_out_uxt(s, 1, args[0], args[1]);
break;
case INDEX_op_ext32u_i64:
- tcg_out_movr(s, 0, args[0], args[1]);
+ tcg_out_movr(s, E32, args[0], args[1]);
break;
default:
@@ -1616,7 +1636,7 @@ static void tcg_target_qemu_prologue(TCGContext *s)
TCG_REG_FP, TCG_REG_LR, frame_size_callee_saved);
/* FP -> callee_saved */
- tcg_out_movr_sp(s, 1, TCG_REG_FP, TCG_REG_SP);
+ tcg_out_mov(s, TCG_TYPE_PTR, TCG_REG_FP, TCG_REG_SP);
/* store callee-preserved regs x19..x28 using FP -> callee_saved */
for (r = TCG_REG_X19; r <= TCG_REG_X27; r += 2) {
@@ -1625,7 +1645,7 @@ static void tcg_target_qemu_prologue(TCGContext *s)
}
/* make stack space for TCG locals */
- tcg_out_addi(s, 1, TCG_REG_SP, TCG_REG_SP,
+ tcg_out_addi(s, E64, TCG_REG_SP, TCG_REG_SP,
-frame_size_tcg_locals * TCG_TARGET_STACK_ALIGN);
/* inform TCG about how to find TCG locals with register, offset, size */
tcg_set_frame(s, TCG_REG_SP, TCG_STATIC_CALL_ARGS_SIZE,
@@ -1644,7 +1664,7 @@ static void tcg_target_qemu_prologue(TCGContext *s)
tb_ret_addr = s->code_ptr;
/* remove TCG locals stack space */
- tcg_out_addi(s, 1, TCG_REG_SP, TCG_REG_SP,
+ tcg_out_addi(s, E64, TCG_REG_SP, TCG_REG_SP,
frame_size_tcg_locals * TCG_TARGET_STACK_ALIGN);
/* restore registers x19..x28.
--
1.8.3.1
