* Copyright (c) 2007, 2008 CodeSourcery.
* Written by Paul Brook
*
- * This code is licenced under the GNU GPL v2.
+ * This code is licensed under the GNU GPL v2.
*/
#include <stdlib.h>
#include <stdio.h>
#include "cpu.h"
-#include "exec-all.h"
-#include "helpers.h"
+#include "exec/exec-all.h"
+#include "exec/helper-proto.h"
#define SIGNBIT (uint32_t)0x80000000
#define SIGNBIT64 ((uint64_t)1 << 63)
-#define SET_QC() env->vfp.xregs[ARM_VFP_FPSCR] = CPSR_Q
-
-static float_status neon_float_status;
-#define NFS &neon_float_status
+#define SET_QC() env->vfp.xregs[ARM_VFP_FPSCR] |= CPSR_Q
#define NEON_TYPE1(name, type) \
typedef struct \
NEON_VOP_BODY(vtype, n)
#define NEON_VOP_ENV(name, vtype, n) \
-uint32_t HELPER(glue(neon_,name))(CPUState *env, uint32_t arg1, uint32_t arg2) \
+uint32_t HELPER(glue(neon_,name))(CPUARMState *env, uint32_t arg1, uint32_t arg2) \
NEON_VOP_BODY(vtype, n)
/* Pairwise operations. */
#undef NEON_FN
#undef NEON_USAT
-uint32_t HELPER(neon_qadd_u32)(CPUState *env, uint32_t a, uint32_t b)
+uint32_t HELPER(neon_qadd_u32)(CPUARMState *env, uint32_t a, uint32_t b)
{
uint32_t res = a + b;
if (res < a) {
return res;
}
-uint64_t HELPER(neon_qadd_u64)(CPUState *env, uint64_t src1, uint64_t src2)
+uint64_t HELPER(neon_qadd_u64)(CPUARMState *env, uint64_t src1, uint64_t src2)
{
uint64_t res;
#undef NEON_FN
#undef NEON_SSAT
-uint32_t HELPER(neon_qadd_s32)(CPUState *env, uint32_t a, uint32_t b)
+uint32_t HELPER(neon_qadd_s32)(CPUARMState *env, uint32_t a, uint32_t b)
{
uint32_t res = a + b;
if (((res ^ a) & SIGNBIT) && !((a ^ b) & SIGNBIT)) {
return res;
}
-uint64_t HELPER(neon_qadd_s64)(CPUState *env, uint64_t src1, uint64_t src2)
+uint64_t HELPER(neon_qadd_s64)(CPUARMState *env, uint64_t src1, uint64_t src2)
{
uint64_t res;
return res;
}
+/* Unsigned saturating accumulate of signed value
+ *
+ * Op1/Rn is treated as signed
+ * Op2/Rd is treated as unsigned
+ *
+ * Explicit casting is used to ensure the correct sign extension of
+ * inputs. The result is treated as a unsigned value and saturated as such.
+ *
+ * We use a macro for the 8/16 bit cases which expects signed integers of va,
+ * vb, and vr for interim calculation and an unsigned 32 bit result value r.
+ */
+
+#define USATACC(bits, shift) \
+ do { \
+ va = sextract32(a, shift, bits); \
+ vb = extract32(b, shift, bits); \
+ vr = va + vb; \
+ if (vr > UINT##bits##_MAX) { \
+ SET_QC(); \
+ vr = UINT##bits##_MAX; \
+ } else if (vr < 0) { \
+ SET_QC(); \
+ vr = 0; \
+ } \
+ r = deposit32(r, shift, bits, vr); \
+ } while (0)
+
+uint32_t HELPER(neon_uqadd_s8)(CPUARMState *env, uint32_t a, uint32_t b)
+{
+ int16_t va, vb, vr;
+ uint32_t r = 0;
+
+ USATACC(8, 0);
+ USATACC(8, 8);
+ USATACC(8, 16);
+ USATACC(8, 24);
+ return r;
+}
+
+uint32_t HELPER(neon_uqadd_s16)(CPUARMState *env, uint32_t a, uint32_t b)
+{
+ int32_t va, vb, vr;
+ uint64_t r = 0;
+
+ USATACC(16, 0);
+ USATACC(16, 16);
+ return r;
+}
+
+#undef USATACC
+
+uint32_t HELPER(neon_uqadd_s32)(CPUARMState *env, uint32_t a, uint32_t b)
+{
+ int64_t va = (int32_t)a;
+ int64_t vb = (uint32_t)b;
+ int64_t vr = va + vb;
+ if (vr > UINT32_MAX) {
+ SET_QC();
+ vr = UINT32_MAX;
+ } else if (vr < 0) {
+ SET_QC();
+ vr = 0;
+ }
+ return vr;
+}
+
+uint64_t HELPER(neon_uqadd_s64)(CPUARMState *env, uint64_t a, uint64_t b)
+{
+ uint64_t res;
+ res = a + b;
+ /* We only need to look at the pattern of SIGN bits to detect
+ * +ve/-ve saturation
+ */
+ if (~a & b & ~res & SIGNBIT64) {
+ SET_QC();
+ res = UINT64_MAX;
+ } else if (a & ~b & res & SIGNBIT64) {
+ SET_QC();
+ res = 0;
+ }
+ return res;
+}
+
+/* Signed saturating accumulate of unsigned value
+ *
+ * Op1/Rn is treated as unsigned
+ * Op2/Rd is treated as signed
+ *
+ * The result is treated as a signed value and saturated as such
+ *
+ * We use a macro for the 8/16 bit cases which expects signed integers of va,
+ * vb, and vr for interim calculation and an unsigned 32 bit result value r.
+ */
+
+#define SSATACC(bits, shift) \
+ do { \
+ va = extract32(a, shift, bits); \
+ vb = sextract32(b, shift, bits); \
+ vr = va + vb; \
+ if (vr > INT##bits##_MAX) { \
+ SET_QC(); \
+ vr = INT##bits##_MAX; \
+ } else if (vr < INT##bits##_MIN) { \
+ SET_QC(); \
+ vr = INT##bits##_MIN; \
+ } \
+ r = deposit32(r, shift, bits, vr); \
+ } while (0)
+
+uint32_t HELPER(neon_sqadd_u8)(CPUARMState *env, uint32_t a, uint32_t b)
+{
+ int16_t va, vb, vr;
+ uint32_t r = 0;
+
+ SSATACC(8, 0);
+ SSATACC(8, 8);
+ SSATACC(8, 16);
+ SSATACC(8, 24);
+ return r;
+}
+
+uint32_t HELPER(neon_sqadd_u16)(CPUARMState *env, uint32_t a, uint32_t b)
+{
+ int32_t va, vb, vr;
+ uint32_t r = 0;
+
+ SSATACC(16, 0);
+ SSATACC(16, 16);
+
+ return r;
+}
+
+#undef SSATACC
+
+uint32_t HELPER(neon_sqadd_u32)(CPUARMState *env, uint32_t a, uint32_t b)
+{
+ int64_t res;
+ int64_t op1 = (uint32_t)a;
+ int64_t op2 = (int32_t)b;
+ res = op1 + op2;
+ if (res > INT32_MAX) {
+ SET_QC();
+ res = INT32_MAX;
+ } else if (res < INT32_MIN) {
+ SET_QC();
+ res = INT32_MIN;
+ }
+ return res;
+}
+
+uint64_t HELPER(neon_sqadd_u64)(CPUARMState *env, uint64_t a, uint64_t b)
+{
+ uint64_t res;
+ res = a + b;
+ /* We only need to look at the pattern of SIGN bits to detect an overflow */
+ if (((a & res)
+ | (~b & res)
+ | (a & ~b)) & SIGNBIT64) {
+ SET_QC();
+ res = INT64_MAX;
+ }
+ return res;
+}
+
+
#define NEON_USAT(dest, src1, src2, type) do { \
uint32_t tmp = (uint32_t)src1 - (uint32_t)src2; \
if (tmp != (type)tmp) { \
#undef NEON_FN
#undef NEON_USAT
-uint32_t HELPER(neon_qsub_u32)(CPUState *env, uint32_t a, uint32_t b)
+uint32_t HELPER(neon_qsub_u32)(CPUARMState *env, uint32_t a, uint32_t b)
{
uint32_t res = a - b;
if (res > a) {
return res;
}
-uint64_t HELPER(neon_qsub_u64)(CPUState *env, uint64_t src1, uint64_t src2)
+uint64_t HELPER(neon_qsub_u64)(CPUARMState *env, uint64_t src1, uint64_t src2)
{
uint64_t res;
#undef NEON_FN
#undef NEON_SSAT
-uint32_t HELPER(neon_qsub_s32)(CPUState *env, uint32_t a, uint32_t b)
+uint32_t HELPER(neon_qsub_s32)(CPUARMState *env, uint32_t a, uint32_t b)
{
uint32_t res = a - b;
if (((res ^ a) & SIGNBIT) && ((a ^ b) & SIGNBIT)) {
return res;
}
-uint64_t HELPER(neon_qsub_s64)(CPUState *env, uint64_t src1, uint64_t src2)
+uint64_t HELPER(neon_qsub_s64)(CPUARMState *env, uint64_t src1, uint64_t src2)
{
uint64_t res;
#undef NEON_FN
/* The addition of the rounding constant may overflow, so we use an
- * intermediate 64 bits accumulator. */
+ * intermediate 64 bit accumulator. */
uint32_t HELPER(neon_rshl_s32)(uint32_t valop, uint32_t shiftop)
{
int32_t dest;
return dest;
}
-/* Handling addition overflow with 64 bits inputs values is more
- * tricky than with 32 bits values. */
+/* Handling addition overflow with 64 bit input values is more
+ * tricky than with 32 bit values. */
uint64_t HELPER(neon_rshl_s64)(uint64_t valop, uint64_t shiftop)
{
int8_t shift = (int8_t)shiftop;
#undef NEON_FN
/* The addition of the rounding constant may overflow, so we use an
- * intermediate 64 bits accumulator. */
+ * intermediate 64 bit accumulator. */
uint32_t HELPER(neon_rshl_u32)(uint32_t val, uint32_t shiftop)
{
uint32_t dest;
return dest;
}
-/* Handling addition overflow with 64 bits inputs values is more
- * tricky than with 32 bits values. */
+/* Handling addition overflow with 64 bit input values is more
+ * tricky than with 32 bit values. */
uint64_t HELPER(neon_rshl_u64)(uint64_t val, uint64_t shiftop)
{
int8_t shift = (uint8_t)shiftop;
NEON_VOP_ENV(qshl_u32, neon_u32, 1)
#undef NEON_FN
-uint64_t HELPER(neon_qshl_u64)(CPUState *env, uint64_t val, uint64_t shiftop)
+uint64_t HELPER(neon_qshl_u64)(CPUARMState *env, uint64_t val, uint64_t shiftop)
{
int8_t shift = (int8_t)shiftop;
if (shift >= 64) {
NEON_VOP_ENV(qshl_s32, neon_s32, 1)
#undef NEON_FN
-uint64_t HELPER(neon_qshl_s64)(CPUState *env, uint64_t valop, uint64_t shiftop)
+uint64_t HELPER(neon_qshl_s64)(CPUARMState *env, uint64_t valop, uint64_t shiftop)
{
int8_t shift = (uint8_t)shiftop;
int64_t val = valop;
NEON_VOP_ENV(qshlu_s16, neon_u16, 2)
#undef NEON_FN
-uint32_t HELPER(neon_qshlu_s32)(CPUState *env, uint32_t valop, uint32_t shiftop)
+uint32_t HELPER(neon_qshlu_s32)(CPUARMState *env, uint32_t valop, uint32_t shiftop)
{
if ((int32_t)valop < 0) {
SET_QC();
return helper_neon_qshl_u32(env, valop, shiftop);
}
-uint64_t HELPER(neon_qshlu_s64)(CPUState *env, uint64_t valop, uint64_t shiftop)
+uint64_t HELPER(neon_qshlu_s64)(CPUARMState *env, uint64_t valop, uint64_t shiftop)
{
if ((int64_t)valop < 0) {
SET_QC();
return helper_neon_qshl_u64(env, valop, shiftop);
}
-/* FIXME: This is wrong. */
#define NEON_FN(dest, src1, src2) do { \
int8_t tmp; \
tmp = (int8_t)src2; \
#undef NEON_FN
/* The addition of the rounding constant may overflow, so we use an
- * intermediate 64 bits accumulator. */
-uint32_t HELPER(neon_qrshl_u32)(CPUState *env, uint32_t val, uint32_t shiftop)
+ * intermediate 64 bit accumulator. */
+uint32_t HELPER(neon_qrshl_u32)(CPUARMState *env, uint32_t val, uint32_t shiftop)
{
uint32_t dest;
int8_t shift = (int8_t)shiftop;
return dest;
}
-/* Handling addition overflow with 64 bits inputs values is more
- * tricky than with 32 bits values. */
-uint64_t HELPER(neon_qrshl_u64)(CPUState *env, uint64_t val, uint64_t shiftop)
+/* Handling addition overflow with 64 bit input values is more
+ * tricky than with 32 bit values. */
+uint64_t HELPER(neon_qrshl_u64)(CPUARMState *env, uint64_t val, uint64_t shiftop)
{
int8_t shift = (int8_t)shiftop;
if (shift >= 64) {
#undef NEON_FN
/* The addition of the rounding constant may overflow, so we use an
- * intermediate 64 bits accumulator. */
-uint32_t HELPER(neon_qrshl_s32)(CPUState *env, uint32_t valop, uint32_t shiftop)
+ * intermediate 64 bit accumulator. */
+uint32_t HELPER(neon_qrshl_s32)(CPUARMState *env, uint32_t valop, uint32_t shiftop)
{
int32_t dest;
int32_t val = (int32_t)valop;
return dest;
}
-/* Handling addition overflow with 64 bits inputs values is more
- * tricky than with 32 bits values. */
-uint64_t HELPER(neon_qrshl_s64)(CPUState *env, uint64_t valop, uint64_t shiftop)
+/* Handling addition overflow with 64 bit input values is more
+ * tricky than with 32 bit values. */
+uint64_t HELPER(neon_qrshl_s64)(CPUARMState *env, uint64_t valop, uint64_t shiftop)
{
int8_t shift = (uint8_t)shiftop;
int64_t val = valop;
return x;
}
+/* Reverse bits in each 8 bit word */
+uint32_t HELPER(neon_rbit_u8)(uint32_t x)
+{
+ x = ((x & 0xf0f0f0f0) >> 4)
+ | ((x & 0x0f0f0f0f) << 4);
+ x = ((x & 0x88888888) >> 3)
+ | ((x & 0x44444444) >> 1)
+ | ((x & 0x22222222) << 1)
+ | ((x & 0x11111111) << 3);
+ return x;
+}
+
#define NEON_QDMULH16(dest, src1, src2, round) do { \
uint32_t tmp = (int32_t)(int16_t) src1 * (int16_t) src2; \
if ((tmp ^ (tmp << 1)) & SIGNBIT) { \
return ((x >> 16) & 0xffff) | ((x >> 32) & 0xffff0000);
}
-uint32_t HELPER(neon_unarrow_sat8)(CPUState *env, uint64_t x)
+uint32_t HELPER(neon_unarrow_sat8)(CPUARMState *env, uint64_t x)
{
uint16_t s;
uint8_t d;
return res;
}
-uint32_t HELPER(neon_narrow_sat_u8)(CPUState *env, uint64_t x)
+uint32_t HELPER(neon_narrow_sat_u8)(CPUARMState *env, uint64_t x)
{
uint16_t s;
uint8_t d;
return res;
}
-uint32_t HELPER(neon_narrow_sat_s8)(CPUState *env, uint64_t x)
+uint32_t HELPER(neon_narrow_sat_s8)(CPUARMState *env, uint64_t x)
{
int16_t s;
uint8_t d;
return res;
}
-uint32_t HELPER(neon_unarrow_sat16)(CPUState *env, uint64_t x)
+uint32_t HELPER(neon_unarrow_sat16)(CPUARMState *env, uint64_t x)
{
uint32_t high;
uint32_t low;
return low | (high << 16);
}
-uint32_t HELPER(neon_narrow_sat_u16)(CPUState *env, uint64_t x)
+uint32_t HELPER(neon_narrow_sat_u16)(CPUARMState *env, uint64_t x)
{
uint32_t high;
uint32_t low;
return low | (high << 16);
}
-uint32_t HELPER(neon_narrow_sat_s16)(CPUState *env, uint64_t x)
+uint32_t HELPER(neon_narrow_sat_s16)(CPUARMState *env, uint64_t x)
{
int32_t low;
int32_t high;
return (uint16_t)low | (high << 16);
}
-uint32_t HELPER(neon_unarrow_sat32)(CPUState *env, uint64_t x)
+uint32_t HELPER(neon_unarrow_sat32)(CPUARMState *env, uint64_t x)
{
if (x & 0x8000000000000000ull) {
SET_QC();
return x;
}
-uint32_t HELPER(neon_narrow_sat_u32)(CPUState *env, uint64_t x)
+uint32_t HELPER(neon_narrow_sat_u32)(CPUARMState *env, uint64_t x)
{
if (x > 0xffffffffu) {
SET_QC();
return x;
}
-uint32_t HELPER(neon_narrow_sat_s32)(CPUState *env, uint64_t x)
+uint32_t HELPER(neon_narrow_sat_s32)(CPUARMState *env, uint64_t x)
{
if ((int64_t)x != (int32_t)x) {
SET_QC();
return (a - b) ^ mask;
}
-uint64_t HELPER(neon_addl_saturate_s32)(CPUState *env, uint64_t a, uint64_t b)
+uint64_t HELPER(neon_addl_saturate_s32)(CPUARMState *env, uint64_t a, uint64_t b)
{
uint32_t x, y;
uint32_t low, high;
return low | ((uint64_t)high << 32);
}
-uint64_t HELPER(neon_addl_saturate_s64)(CPUState *env, uint64_t a, uint64_t b)
+uint64_t HELPER(neon_addl_saturate_s64)(CPUARMState *env, uint64_t a, uint64_t b)
{
uint64_t result;
return result;
}
-#define DO_ABD(dest, x, y, type) do { \
- type tmp_x = x; \
- type tmp_y = y; \
+/* We have to do the arithmetic in a larger type than
+ * the input type, because for example with a signed 32 bit
+ * op the absolute difference can overflow a signed 32 bit value.
+ */
+#define DO_ABD(dest, x, y, intype, arithtype) do { \
+ arithtype tmp_x = (intype)(x); \
+ arithtype tmp_y = (intype)(y); \
dest = ((tmp_x > tmp_y) ? tmp_x - tmp_y : tmp_y - tmp_x); \
} while(0)
{
uint64_t tmp;
uint64_t result;
- DO_ABD(result, a, b, uint8_t);
- DO_ABD(tmp, a >> 8, b >> 8, uint8_t);
+ DO_ABD(result, a, b, uint8_t, uint32_t);
+ DO_ABD(tmp, a >> 8, b >> 8, uint8_t, uint32_t);
result |= tmp << 16;
- DO_ABD(tmp, a >> 16, b >> 16, uint8_t);
+ DO_ABD(tmp, a >> 16, b >> 16, uint8_t, uint32_t);
result |= tmp << 32;
- DO_ABD(tmp, a >> 24, b >> 24, uint8_t);
+ DO_ABD(tmp, a >> 24, b >> 24, uint8_t, uint32_t);
result |= tmp << 48;
return result;
}
{
uint64_t tmp;
uint64_t result;
- DO_ABD(result, a, b, int8_t);
- DO_ABD(tmp, a >> 8, b >> 8, int8_t);
+ DO_ABD(result, a, b, int8_t, int32_t);
+ DO_ABD(tmp, a >> 8, b >> 8, int8_t, int32_t);
result |= tmp << 16;
- DO_ABD(tmp, a >> 16, b >> 16, int8_t);
+ DO_ABD(tmp, a >> 16, b >> 16, int8_t, int32_t);
result |= tmp << 32;
- DO_ABD(tmp, a >> 24, b >> 24, int8_t);
+ DO_ABD(tmp, a >> 24, b >> 24, int8_t, int32_t);
result |= tmp << 48;
return result;
}
{
uint64_t tmp;
uint64_t result;
- DO_ABD(result, a, b, uint16_t);
- DO_ABD(tmp, a >> 16, b >> 16, uint16_t);
+ DO_ABD(result, a, b, uint16_t, uint32_t);
+ DO_ABD(tmp, a >> 16, b >> 16, uint16_t, uint32_t);
return result | (tmp << 32);
}
{
uint64_t tmp;
uint64_t result;
- DO_ABD(result, a, b, int16_t);
- DO_ABD(tmp, a >> 16, b >> 16, int16_t);
+ DO_ABD(result, a, b, int16_t, int32_t);
+ DO_ABD(tmp, a >> 16, b >> 16, int16_t, int32_t);
return result | (tmp << 32);
}
uint64_t HELPER(neon_abdl_u64)(uint32_t a, uint32_t b)
{
uint64_t result;
- DO_ABD(result, a, b, uint32_t);
+ DO_ABD(result, a, b, uint32_t, uint64_t);
return result;
}
uint64_t HELPER(neon_abdl_s64)(uint32_t a, uint32_t b)
{
uint64_t result;
- DO_ABD(result, a, b, int32_t);
+ DO_ABD(result, a, b, int32_t, int64_t);
return result;
}
#undef DO_ABD
return low | ((uint64_t)high << 32);
}
-/* FIXME: There should be a native op for this. */
-uint64_t HELPER(neon_negl_u64)(uint64_t x)
-{
- return -x;
-}
-
-/* Saturnating sign manuipulation. */
+/* Saturating sign manipulation. */
/* ??? Make these use NEON_VOP1 */
#define DO_QABS8(x) do { \
if (x == (int8_t)0x80) { \
} else if (x < 0) { \
x = -x; \
}} while (0)
-uint32_t HELPER(neon_qabs_s8)(CPUState *env, uint32_t x)
+uint32_t HELPER(neon_qabs_s8)(CPUARMState *env, uint32_t x)
{
neon_s8 vec;
NEON_UNPACK(neon_s8, vec, x);
} else { \
x = -x; \
}} while (0)
-uint32_t HELPER(neon_qneg_s8)(CPUState *env, uint32_t x)
+uint32_t HELPER(neon_qneg_s8)(CPUARMState *env, uint32_t x)
{
neon_s8 vec;
NEON_UNPACK(neon_s8, vec, x);
} else if (x < 0) { \
x = -x; \
}} while (0)
-uint32_t HELPER(neon_qabs_s16)(CPUState *env, uint32_t x)
+uint32_t HELPER(neon_qabs_s16)(CPUARMState *env, uint32_t x)
{
neon_s16 vec;
NEON_UNPACK(neon_s16, vec, x);
} else { \
x = -x; \
}} while (0)
-uint32_t HELPER(neon_qneg_s16)(CPUState *env, uint32_t x)
+uint32_t HELPER(neon_qneg_s16)(CPUARMState *env, uint32_t x)
{
neon_s16 vec;
NEON_UNPACK(neon_s16, vec, x);
}
#undef DO_QNEG16
-uint32_t HELPER(neon_qabs_s32)(CPUState *env, uint32_t x)
+uint32_t HELPER(neon_qabs_s32)(CPUARMState *env, uint32_t x)
{
if (x == SIGNBIT) {
SET_QC();
return x;
}
-uint32_t HELPER(neon_qneg_s32)(CPUState *env, uint32_t x)
+uint32_t HELPER(neon_qneg_s32)(CPUARMState *env, uint32_t x)
{
if (x == SIGNBIT) {
SET_QC();
return x;
}
-/* NEON Float helpers. */
-uint32_t HELPER(neon_min_f32)(uint32_t a, uint32_t b)
+uint64_t HELPER(neon_qabs_s64)(CPUARMState *env, uint64_t x)
{
- float32 f0 = make_float32(a);
- float32 f1 = make_float32(b);
- return (float32_compare_quiet(f0, f1, NFS) == -1) ? a : b;
+ if (x == SIGNBIT64) {
+ SET_QC();
+ x = ~SIGNBIT64;
+ } else if ((int64_t)x < 0) {
+ x = -x;
+ }
+ return x;
}
-uint32_t HELPER(neon_max_f32)(uint32_t a, uint32_t b)
+uint64_t HELPER(neon_qneg_s64)(CPUARMState *env, uint64_t x)
{
- float32 f0 = make_float32(a);
- float32 f1 = make_float32(b);
- return (float32_compare_quiet(f0, f1, NFS) == 1) ? a : b;
+ if (x == SIGNBIT64) {
+ SET_QC();
+ x = ~SIGNBIT64;
+ } else {
+ x = -x;
+ }
+ return x;
}
-uint32_t HELPER(neon_abd_f32)(uint32_t a, uint32_t b)
+/* NEON Float helpers. */
+uint32_t HELPER(neon_abd_f32)(uint32_t a, uint32_t b, void *fpstp)
{
+ float_status *fpst = fpstp;
float32 f0 = make_float32(a);
float32 f1 = make_float32(b);
- return float32_val((float32_compare_quiet(f0, f1, NFS) == 1)
- ? float32_sub(f0, f1, NFS)
- : float32_sub(f1, f0, NFS));
+ return float32_val(float32_abs(float32_sub(f0, f1, fpst)));
}
-uint32_t HELPER(neon_add_f32)(uint32_t a, uint32_t b)
+/* Floating point comparisons produce an integer result.
+ * Note that EQ doesn't signal InvalidOp for QNaNs but GE and GT do.
+ * Softfloat routines return 0/1, which we convert to the 0/-1 Neon requires.
+ */
+uint32_t HELPER(neon_ceq_f32)(uint32_t a, uint32_t b, void *fpstp)
{
- return float32_val(float32_add(make_float32(a), make_float32(b), NFS));
+ float_status *fpst = fpstp;
+ return -float32_eq_quiet(make_float32(a), make_float32(b), fpst);
}
-uint32_t HELPER(neon_sub_f32)(uint32_t a, uint32_t b)
+uint32_t HELPER(neon_cge_f32)(uint32_t a, uint32_t b, void *fpstp)
{
- return float32_val(float32_sub(make_float32(a), make_float32(b), NFS));
+ float_status *fpst = fpstp;
+ return -float32_le(make_float32(b), make_float32(a), fpst);
}
-uint32_t HELPER(neon_mul_f32)(uint32_t a, uint32_t b)
+uint32_t HELPER(neon_cgt_f32)(uint32_t a, uint32_t b, void *fpstp)
{
- return float32_val(float32_mul(make_float32(a), make_float32(b), NFS));
-}
-
-/* Floating point comparisons produce an integer result. */
-#define NEON_VOP_FCMP(name, cmp) \
-uint32_t HELPER(neon_##name)(uint32_t a, uint32_t b) \
-{ \
- if (float32_compare_quiet(make_float32(a), make_float32(b), NFS) cmp 0) { \
- return ~0; \
- } else { \
- return 0; \
- } \
+ float_status *fpst = fpstp;
+ return -float32_lt(make_float32(b), make_float32(a), fpst);
}
-NEON_VOP_FCMP(ceq_f32, ==)
-NEON_VOP_FCMP(cge_f32, >=)
-NEON_VOP_FCMP(cgt_f32, >)
-
-uint32_t HELPER(neon_acge_f32)(uint32_t a, uint32_t b)
+uint32_t HELPER(neon_acge_f32)(uint32_t a, uint32_t b, void *fpstp)
{
+ float_status *fpst = fpstp;
float32 f0 = float32_abs(make_float32(a));
float32 f1 = float32_abs(make_float32(b));
- return (float32_compare_quiet(f0, f1,NFS) >= 0) ? ~0 : 0;
+ return -float32_le(f1, f0, fpst);
}
-uint32_t HELPER(neon_acgt_f32)(uint32_t a, uint32_t b)
+uint32_t HELPER(neon_acgt_f32)(uint32_t a, uint32_t b, void *fpstp)
{
+ float_status *fpst = fpstp;
float32 f0 = float32_abs(make_float32(a));
float32 f1 = float32_abs(make_float32(b));
- return (float32_compare_quiet(f0, f1, NFS) > 0) ? ~0 : 0;
+ return -float32_lt(f1, f0, fpst);
+}
+
+uint64_t HELPER(neon_acge_f64)(uint64_t a, uint64_t b, void *fpstp)
+{
+ float_status *fpst = fpstp;
+ float64 f0 = float64_abs(make_float64(a));
+ float64 f1 = float64_abs(make_float64(b));
+ return -float64_le(f1, f0, fpst);
+}
+
+uint64_t HELPER(neon_acgt_f64)(uint64_t a, uint64_t b, void *fpstp)
+{
+ float_status *fpst = fpstp;
+ float64 f0 = float64_abs(make_float64(a));
+ float64 f1 = float64_abs(make_float64(b));
+ return -float64_lt(f1, f0, fpst);
}
#define ELEM(V, N, SIZE) (((V) >> ((N) * (SIZE))) & ((1ull << (SIZE)) - 1))
-void HELPER(neon_qunzip8)(CPUState *env, uint32_t rd, uint32_t rm)
+void HELPER(neon_qunzip8)(CPUARMState *env, uint32_t rd, uint32_t rm)
{
uint64_t zm0 = float64_val(env->vfp.regs[rm]);
uint64_t zm1 = float64_val(env->vfp.regs[rm + 1]);
env->vfp.regs[rd + 1] = make_float64(d1);
}
-void HELPER(neon_qunzip16)(CPUState *env, uint32_t rd, uint32_t rm)
+void HELPER(neon_qunzip16)(CPUARMState *env, uint32_t rd, uint32_t rm)
{
uint64_t zm0 = float64_val(env->vfp.regs[rm]);
uint64_t zm1 = float64_val(env->vfp.regs[rm + 1]);
env->vfp.regs[rd + 1] = make_float64(d1);
}
-void HELPER(neon_qunzip32)(CPUState *env, uint32_t rd, uint32_t rm)
+void HELPER(neon_qunzip32)(CPUARMState *env, uint32_t rd, uint32_t rm)
{
uint64_t zm0 = float64_val(env->vfp.regs[rm]);
uint64_t zm1 = float64_val(env->vfp.regs[rm + 1]);
env->vfp.regs[rd + 1] = make_float64(d1);
}
-void HELPER(neon_unzip8)(CPUState *env, uint32_t rd, uint32_t rm)
+void HELPER(neon_unzip8)(CPUARMState *env, uint32_t rd, uint32_t rm)
{
uint64_t zm = float64_val(env->vfp.regs[rm]);
uint64_t zd = float64_val(env->vfp.regs[rd]);
env->vfp.regs[rd] = make_float64(d0);
}
-void HELPER(neon_unzip16)(CPUState *env, uint32_t rd, uint32_t rm)
+void HELPER(neon_unzip16)(CPUARMState *env, uint32_t rd, uint32_t rm)
{
uint64_t zm = float64_val(env->vfp.regs[rm]);
uint64_t zd = float64_val(env->vfp.regs[rd]);
env->vfp.regs[rd] = make_float64(d0);
}
-void HELPER(neon_qzip8)(CPUState *env, uint32_t rd, uint32_t rm)
+void HELPER(neon_qzip8)(CPUARMState *env, uint32_t rd, uint32_t rm)
{
uint64_t zm0 = float64_val(env->vfp.regs[rm]);
uint64_t zm1 = float64_val(env->vfp.regs[rm + 1]);
env->vfp.regs[rd + 1] = make_float64(d1);
}
-void HELPER(neon_qzip16)(CPUState *env, uint32_t rd, uint32_t rm)
+void HELPER(neon_qzip16)(CPUARMState *env, uint32_t rd, uint32_t rm)
{
uint64_t zm0 = float64_val(env->vfp.regs[rm]);
uint64_t zm1 = float64_val(env->vfp.regs[rm + 1]);
env->vfp.regs[rd + 1] = make_float64(d1);
}
-void HELPER(neon_qzip32)(CPUState *env, uint32_t rd, uint32_t rm)
+void HELPER(neon_qzip32)(CPUARMState *env, uint32_t rd, uint32_t rm)
{
uint64_t zm0 = float64_val(env->vfp.regs[rm]);
uint64_t zm1 = float64_val(env->vfp.regs[rm + 1]);
env->vfp.regs[rd + 1] = make_float64(d1);
}
-void HELPER(neon_zip8)(CPUState *env, uint32_t rd, uint32_t rm)
+void HELPER(neon_zip8)(CPUARMState *env, uint32_t rd, uint32_t rm)
{
uint64_t zm = float64_val(env->vfp.regs[rm]);
uint64_t zd = float64_val(env->vfp.regs[rd]);
env->vfp.regs[rd] = make_float64(d0);
}
-void HELPER(neon_zip16)(CPUState *env, uint32_t rd, uint32_t rm)
+void HELPER(neon_zip16)(CPUARMState *env, uint32_t rd, uint32_t rm)
{
uint64_t zm = float64_val(env->vfp.regs[rm]);
uint64_t zd = float64_val(env->vfp.regs[rd]);
env->vfp.regs[rm] = make_float64(m0);
env->vfp.regs[rd] = make_float64(d0);
}
+
+/* Helper function for 64 bit polynomial multiply case:
+ * perform PolynomialMult(op1, op2) and return either the top or
+ * bottom half of the 128 bit result.
+ */
+uint64_t HELPER(neon_pmull_64_lo)(uint64_t op1, uint64_t op2)
+{
+ int bitnum;
+ uint64_t res = 0;
+
+ for (bitnum = 0; bitnum < 64; bitnum++) {
+ if (op1 & (1ULL << bitnum)) {
+ res ^= op2 << bitnum;
+ }
+ }
+ return res;
+}
+uint64_t HELPER(neon_pmull_64_hi)(uint64_t op1, uint64_t op2)
+{
+ int bitnum;
+ uint64_t res = 0;
+
+ /* bit 0 of op1 can't influence the high 64 bits at all */
+ for (bitnum = 1; bitnum < 64; bitnum++) {
+ if (op1 & (1ULL << bitnum)) {
+ res ^= op2 >> (64 - bitnum);
+ }
+ }
+ return res;
+}
projects / qemu.git / blobdiff