*/
#include "cpu.h"
-#include "dyngen-exec.h"
#include "helper.h"
-#define DT0 (env->dt0)
-#define DT1 (env->dt1)
-#define QT0 (env->qt0)
-#define QT1 (env->qt1)
-
/* This function uses non-native bit order */
#define GET_FIELD(X, FROM, TO) \
((X) >> (63 - (TO)) & ((1ULL << ((TO) - (FROM) + 1)) - 1))
GET_FIELD_SP(pixel_addr, 11, 12);
}
-target_ulong helper_alignaddr(target_ulong addr, target_ulong offset)
-{
- uint64_t tmp;
-
- tmp = addr + offset;
- env->gsr &= ~7ULL;
- env->gsr |= tmp & 7ULL;
- return tmp & ~7ULL;
-}
-
-void helper_faligndata(void)
-{
- uint64_t tmp;
-
- tmp = (*((uint64_t *)&DT0)) << ((env->gsr & 7) * 8);
- /* on many architectures a shift of 64 does nothing */
- if ((env->gsr & 7) != 0) {
- tmp |= (*((uint64_t *)&DT1)) >> (64 - (env->gsr & 7) * 8);
- }
- *((uint64_t *)&DT0) = tmp;
-}
-
#ifdef HOST_WORDS_BIGENDIAN
#define VIS_B64(n) b[7 - (n)]
#define VIS_W64(n) w[3 - (n)]
float32 f;
} VIS32;
-void helper_fpmerge(void)
+uint64_t helper_fpmerge(uint64_t src1, uint64_t src2)
{
VIS64 s, d;
- s.d = DT0;
- d.d = DT1;
+ s.ll = src1;
+ d.ll = src2;
/* Reverse calculation order to handle overlap */
d.VIS_B64(7) = s.VIS_B64(3);
d.VIS_B64(1) = s.VIS_B64(0);
/* d.VIS_B64(0) = d.VIS_B64(0); */
- DT0 = d.d;
+ return d.ll;
}
-void helper_fmul8x16(void)
+uint64_t helper_fmul8x16(uint64_t src1, uint64_t src2)
{
VIS64 s, d;
uint32_t tmp;
- s.d = DT0;
- d.d = DT1;
+ s.ll = src1;
+ d.ll = src2;
#define PMUL(r) \
tmp = (int32_t)d.VIS_SW64(r) * (int32_t)s.VIS_B64(r); \
PMUL(3);
#undef PMUL
- DT0 = d.d;
+ return d.ll;
}
-void helper_fmul8x16al(void)
+uint64_t helper_fmul8x16al(uint64_t src1, uint64_t src2)
{
VIS64 s, d;
uint32_t tmp;
- s.d = DT0;
- d.d = DT1;
+ s.ll = src1;
+ d.ll = src2;
#define PMUL(r) \
tmp = (int32_t)d.VIS_SW64(1) * (int32_t)s.VIS_B64(r); \
PMUL(3);
#undef PMUL
- DT0 = d.d;
+ return d.ll;
}
-void helper_fmul8x16au(void)
+uint64_t helper_fmul8x16au(uint64_t src1, uint64_t src2)
{
VIS64 s, d;
uint32_t tmp;
- s.d = DT0;
- d.d = DT1;
+ s.ll = src1;
+ d.ll = src2;
#define PMUL(r) \
tmp = (int32_t)d.VIS_SW64(0) * (int32_t)s.VIS_B64(r); \
PMUL(3);
#undef PMUL
- DT0 = d.d;
+ return d.ll;
}
-void helper_fmul8sux16(void)
+uint64_t helper_fmul8sux16(uint64_t src1, uint64_t src2)
{
VIS64 s, d;
uint32_t tmp;
- s.d = DT0;
- d.d = DT1;
+ s.ll = src1;
+ d.ll = src2;
#define PMUL(r) \
tmp = (int32_t)d.VIS_SW64(r) * ((int32_t)s.VIS_SW64(r) >> 8); \
PMUL(3);
#undef PMUL
- DT0 = d.d;
+ return d.ll;
}
-void helper_fmul8ulx16(void)
+uint64_t helper_fmul8ulx16(uint64_t src1, uint64_t src2)
{
VIS64 s, d;
uint32_t tmp;
- s.d = DT0;
- d.d = DT1;
+ s.ll = src1;
+ d.ll = src2;
#define PMUL(r) \
tmp = (int32_t)d.VIS_SW64(r) * ((uint32_t)s.VIS_B64(r * 2)); \
PMUL(3);
#undef PMUL
- DT0 = d.d;
+ return d.ll;
}
-void helper_fmuld8sux16(void)
+uint64_t helper_fmuld8sux16(uint64_t src1, uint64_t src2)
{
VIS64 s, d;
uint32_t tmp;
- s.d = DT0;
- d.d = DT1;
+ s.ll = src1;
+ d.ll = src2;
#define PMUL(r) \
tmp = (int32_t)d.VIS_SW64(r) * ((int32_t)s.VIS_SW64(r) >> 8); \
PMUL(0);
#undef PMUL
- DT0 = d.d;
+ return d.ll;
}
-void helper_fmuld8ulx16(void)
+uint64_t helper_fmuld8ulx16(uint64_t src1, uint64_t src2)
{
VIS64 s, d;
uint32_t tmp;
- s.d = DT0;
- d.d = DT1;
+ s.ll = src1;
+ d.ll = src2;
#define PMUL(r) \
tmp = (int32_t)d.VIS_SW64(r) * ((uint32_t)s.VIS_B64(r * 2)); \
PMUL(0);
#undef PMUL
- DT0 = d.d;
+ return d.ll;
}
-void helper_fexpand(void)
+uint64_t helper_fexpand(uint64_t src1, uint64_t src2)
{
VIS32 s;
VIS64 d;
- s.l = (uint32_t)(*(uint64_t *)&DT0 & 0xffffffff);
- d.d = DT1;
+ s.l = (uint32_t)src1;
+ d.ll = src2;
d.VIS_W64(0) = s.VIS_B32(0) << 4;
d.VIS_W64(1) = s.VIS_B32(1) << 4;
d.VIS_W64(2) = s.VIS_B32(2) << 4;
d.VIS_W64(3) = s.VIS_B32(3) << 4;
- DT0 = d.d;
+ return d.ll;
}
#define VIS_HELPER(name, F) \
- void name##16(void) \
+ uint64_t name##16(uint64_t src1, uint64_t src2) \
{ \
VIS64 s, d; \
\
- s.d = DT0; \
- d.d = DT1; \
+ s.ll = src1; \
+ d.ll = src2; \
\
d.VIS_W64(0) = F(d.VIS_W64(0), s.VIS_W64(0)); \
d.VIS_W64(1) = F(d.VIS_W64(1), s.VIS_W64(1)); \
d.VIS_W64(2) = F(d.VIS_W64(2), s.VIS_W64(2)); \
d.VIS_W64(3) = F(d.VIS_W64(3), s.VIS_W64(3)); \
\
- DT0 = d.d; \
+ return d.ll; \
} \
\
uint32_t name##16s(uint32_t src1, uint32_t src2) \
return d.l; \
} \
\
- void name##32(void) \
+ uint64_t name##32(uint64_t src1, uint64_t src2) \
{ \
VIS64 s, d; \
\
- s.d = DT0; \
- d.d = DT1; \
+ s.ll = src1; \
+ d.ll = src2; \
\
d.VIS_L64(0) = F(d.VIS_L64(0), s.VIS_L64(0)); \
d.VIS_L64(1) = F(d.VIS_L64(1), s.VIS_L64(1)); \
\
- DT0 = d.d; \
+ return d.ll; \
} \
\
uint32_t name##32s(uint32_t src1, uint32_t src2) \
VIS_HELPER(helper_fpsub, FSUB)
#define VIS_CMPHELPER(name, F) \
- uint64_t name##16(void) \
+ uint64_t name##16(uint64_t src1, uint64_t src2) \
{ \
VIS64 s, d; \
\
- s.d = DT0; \
- d.d = DT1; \
+ s.ll = src1; \
+ d.ll = src2; \
\
d.VIS_W64(0) = F(s.VIS_W64(0), d.VIS_W64(0)) ? 1 : 0; \
d.VIS_W64(0) |= F(s.VIS_W64(1), d.VIS_W64(1)) ? 2 : 0; \
return d.ll; \
} \
\
- uint64_t name##32(void) \
+ uint64_t name##32(uint64_t src1, uint64_t src2) \
{ \
VIS64 s, d; \
\
- s.d = DT0; \
- d.d = DT1; \
+ s.ll = src1; \
+ d.ll = src2; \
\
d.VIS_L64(0) = F(s.VIS_L64(0), d.VIS_L64(0)) ? 1 : 0; \
d.VIS_L64(0) |= F(s.VIS_L64(1), d.VIS_L64(1)) ? 2 : 0; \
VIS_CMPHELPER(helper_fcmpeq, FCMPEQ)
VIS_CMPHELPER(helper_fcmple, FCMPLE)
VIS_CMPHELPER(helper_fcmpne, FCMPNE)
+
+uint64_t helper_pdist(uint64_t sum, uint64_t src1, uint64_t src2)
+{
+ int i;
+ for (i = 0; i < 8; i++) {
+ int s1, s2;
+
+ s1 = (src1 >> (56 - (i * 8))) & 0xff;
+ s2 = (src2 >> (56 - (i * 8))) & 0xff;
+
+ /* Absolute value of difference. */
+ s1 -= s2;
+ if (s1 < 0) {
+ s1 = -s1;
+ }
+
+ sum += s1;
+ }
+
+ return sum;
+}
+
+uint32_t helper_fpack16(uint64_t gsr, uint64_t rs2)
+{
+ int scale = (gsr >> 3) & 0xf;
+ uint32_t ret = 0;
+ int byte;
+
+ for (byte = 0; byte < 4; byte++) {
+ uint32_t val;
+ int16_t src = rs2 >> (byte * 16);
+ int32_t scaled = src << scale;
+ int32_t from_fixed = scaled >> 7;
+
+ val = (from_fixed < 0 ? 0 :
+ from_fixed > 255 ? 255 : from_fixed);
+
+ ret |= val << (8 * byte);
+ }
+
+ return ret;
+}
+
+uint64_t helper_fpack32(uint64_t gsr, uint64_t rs1, uint64_t rs2)
+{
+ int scale = (gsr >> 3) & 0x1f;
+ uint64_t ret = 0;
+ int word;
+
+ ret = (rs1 << 8) & ~(0x000000ff000000ffULL);
+ for (word = 0; word < 2; word++) {
+ uint64_t val;
+ int32_t src = rs2 >> (word * 32);
+ int64_t scaled = (int64_t)src << scale;
+ int64_t from_fixed = scaled >> 23;
+
+ val = (from_fixed < 0 ? 0 :
+ (from_fixed > 255) ? 255 : from_fixed);
+
+ ret |= val << (32 * word);
+ }
+
+ return ret;
+}
+
+uint32_t helper_fpackfix(uint64_t gsr, uint64_t rs2)
+{
+ int scale = (gsr >> 3) & 0x1f;
+ uint32_t ret = 0;
+ int word;
+
+ for (word = 0; word < 2; word++) {
+ uint32_t val;
+ int32_t src = rs2 >> (word * 32);
+ int64_t scaled = src << scale;
+ int64_t from_fixed = scaled >> 16;
+
+ val = (from_fixed < -32768 ? -32768 :
+ from_fixed > 32767 ? 32767 : from_fixed);
+
+ ret |= (val & 0xffff) << (word * 16);
+ }
+
+ return ret;
+}
+
+uint64_t helper_bshuffle(uint64_t gsr, uint64_t src1, uint64_t src2)
+{
+ union {
+ uint64_t ll[2];
+ uint8_t b[16];
+ } s;
+ VIS64 r;
+ uint32_t i, mask, host;
+
+ /* Set up S such that we can index across all of the bytes. */
+#ifdef HOST_WORDS_BIGENDIAN
+ s.ll[0] = src1;
+ s.ll[1] = src2;
+ host = 0;
+#else
+ s.ll[1] = src1;
+ s.ll[0] = src2;
+ host = 15;
+#endif
+ mask = gsr >> 32;
+
+ for (i = 0; i < 8; ++i) {
+ unsigned e = (mask >> (28 - i*4)) & 0xf;
+ r.VIS_B64(i) = s.b[e ^ host];
+ }
+
+ return r.ll;
+}
projects / qemu.git / blobdiff