Revision 02acedf9 target-arm/neon_helper.c
| b/target-arm/neon_helper.c | ||
|---|---|---|
| 1693 | 1693 |
float32 f1 = float32_abs(vfp_itos(b)); |
| 1694 | 1694 |
return (float32_compare_quiet(f0, f1, NFS) > 0) ? ~0 : 0; |
| 1695 | 1695 |
} |
| 1696 |
|
|
| 1697 |
#define ELEM(V, N, SIZE) (((V) >> ((N) * (SIZE))) & ((1ull << (SIZE)) - 1)) |
|
| 1698 |
|
|
| 1699 |
void HELPER(neon_qunzip8)(CPUState *env, uint32_t rd, uint32_t rm) |
|
| 1700 |
{
|
|
| 1701 |
uint64_t zm0 = float64_val(env->vfp.regs[rm]); |
|
| 1702 |
uint64_t zm1 = float64_val(env->vfp.regs[rm + 1]); |
|
| 1703 |
uint64_t zd0 = float64_val(env->vfp.regs[rd]); |
|
| 1704 |
uint64_t zd1 = float64_val(env->vfp.regs[rd + 1]); |
|
| 1705 |
uint64_t d0 = ELEM(zd0, 0, 8) | (ELEM(zd0, 2, 8) << 8) |
|
| 1706 |
| (ELEM(zd0, 4, 8) << 16) | (ELEM(zd0, 6, 8) << 24) |
|
| 1707 |
| (ELEM(zd1, 0, 8) << 32) | (ELEM(zd1, 2, 8) << 40) |
|
| 1708 |
| (ELEM(zd1, 4, 8) << 48) | (ELEM(zd1, 6, 8) << 56); |
|
| 1709 |
uint64_t d1 = ELEM(zm0, 0, 8) | (ELEM(zm0, 2, 8) << 8) |
|
| 1710 |
| (ELEM(zm0, 4, 8) << 16) | (ELEM(zm0, 6, 8) << 24) |
|
| 1711 |
| (ELEM(zm1, 0, 8) << 32) | (ELEM(zm1, 2, 8) << 40) |
|
| 1712 |
| (ELEM(zm1, 4, 8) << 48) | (ELEM(zm1, 6, 8) << 56); |
|
| 1713 |
uint64_t m0 = ELEM(zd0, 1, 8) | (ELEM(zd0, 3, 8) << 8) |
|
| 1714 |
| (ELEM(zd0, 5, 8) << 16) | (ELEM(zd0, 7, 8) << 24) |
|
| 1715 |
| (ELEM(zd1, 1, 8) << 32) | (ELEM(zd1, 3, 8) << 40) |
|
| 1716 |
| (ELEM(zd1, 5, 8) << 48) | (ELEM(zd1, 7, 8) << 56); |
|
| 1717 |
uint64_t m1 = ELEM(zm0, 1, 8) | (ELEM(zm0, 3, 8) << 8) |
|
| 1718 |
| (ELEM(zm0, 5, 8) << 16) | (ELEM(zm0, 7, 8) << 24) |
|
| 1719 |
| (ELEM(zm1, 1, 8) << 32) | (ELEM(zm1, 3, 8) << 40) |
|
| 1720 |
| (ELEM(zm1, 5, 8) << 48) | (ELEM(zm1, 7, 8) << 56); |
|
| 1721 |
env->vfp.regs[rm] = make_float64(m0); |
|
| 1722 |
env->vfp.regs[rm + 1] = make_float64(m1); |
|
| 1723 |
env->vfp.regs[rd] = make_float64(d0); |
|
| 1724 |
env->vfp.regs[rd + 1] = make_float64(d1); |
|
| 1725 |
} |
|
| 1726 |
|
|
| 1727 |
void HELPER(neon_qunzip16)(CPUState *env, uint32_t rd, uint32_t rm) |
|
| 1728 |
{
|
|
| 1729 |
uint64_t zm0 = float64_val(env->vfp.regs[rm]); |
|
| 1730 |
uint64_t zm1 = float64_val(env->vfp.regs[rm + 1]); |
|
| 1731 |
uint64_t zd0 = float64_val(env->vfp.regs[rd]); |
|
| 1732 |
uint64_t zd1 = float64_val(env->vfp.regs[rd + 1]); |
|
| 1733 |
uint64_t d0 = ELEM(zd0, 0, 16) | (ELEM(zd0, 2, 16) << 16) |
|
| 1734 |
| (ELEM(zd1, 0, 16) << 32) | (ELEM(zd1, 2, 16) << 48); |
|
| 1735 |
uint64_t d1 = ELEM(zm0, 0, 16) | (ELEM(zm0, 2, 16) << 16) |
|
| 1736 |
| (ELEM(zm1, 0, 16) << 32) | (ELEM(zm1, 2, 16) << 48); |
|
| 1737 |
uint64_t m0 = ELEM(zd0, 1, 16) | (ELEM(zd0, 3, 16) << 16) |
|
| 1738 |
| (ELEM(zd1, 1, 16) << 32) | (ELEM(zd1, 3, 16) << 48); |
|
| 1739 |
uint64_t m1 = ELEM(zm0, 1, 16) | (ELEM(zm0, 3, 16) << 16) |
|
| 1740 |
| (ELEM(zm1, 1, 16) << 32) | (ELEM(zm1, 3, 16) << 48); |
|
| 1741 |
env->vfp.regs[rm] = make_float64(m0); |
|
| 1742 |
env->vfp.regs[rm + 1] = make_float64(m1); |
|
| 1743 |
env->vfp.regs[rd] = make_float64(d0); |
|
| 1744 |
env->vfp.regs[rd + 1] = make_float64(d1); |
|
| 1745 |
} |
|
| 1746 |
|
|
| 1747 |
void HELPER(neon_qunzip32)(CPUState *env, uint32_t rd, uint32_t rm) |
|
| 1748 |
{
|
|
| 1749 |
uint64_t zm0 = float64_val(env->vfp.regs[rm]); |
|
| 1750 |
uint64_t zm1 = float64_val(env->vfp.regs[rm + 1]); |
|
| 1751 |
uint64_t zd0 = float64_val(env->vfp.regs[rd]); |
|
| 1752 |
uint64_t zd1 = float64_val(env->vfp.regs[rd + 1]); |
|
| 1753 |
uint64_t d0 = ELEM(zd0, 0, 32) | (ELEM(zd1, 0, 32) << 32); |
|
| 1754 |
uint64_t d1 = ELEM(zm0, 0, 32) | (ELEM(zm1, 0, 32) << 32); |
|
| 1755 |
uint64_t m0 = ELEM(zd0, 1, 32) | (ELEM(zd1, 1, 32) << 32); |
|
| 1756 |
uint64_t m1 = ELEM(zm0, 1, 32) | (ELEM(zm1, 1, 32) << 32); |
|
| 1757 |
env->vfp.regs[rm] = make_float64(m0); |
|
| 1758 |
env->vfp.regs[rm + 1] = make_float64(m1); |
|
| 1759 |
env->vfp.regs[rd] = make_float64(d0); |
|
| 1760 |
env->vfp.regs[rd + 1] = make_float64(d1); |
|
| 1761 |
} |
|
| 1762 |
|
|
| 1763 |
void HELPER(neon_unzip8)(CPUState *env, uint32_t rd, uint32_t rm) |
|
| 1764 |
{
|
|
| 1765 |
uint64_t zm = float64_val(env->vfp.regs[rm]); |
|
| 1766 |
uint64_t zd = float64_val(env->vfp.regs[rd]); |
|
| 1767 |
uint64_t d0 = ELEM(zd, 0, 8) | (ELEM(zd, 2, 8) << 8) |
|
| 1768 |
| (ELEM(zd, 4, 8) << 16) | (ELEM(zd, 6, 8) << 24) |
|
| 1769 |
| (ELEM(zm, 0, 8) << 32) | (ELEM(zm, 2, 8) << 40) |
|
| 1770 |
| (ELEM(zm, 4, 8) << 48) | (ELEM(zm, 6, 8) << 56); |
|
| 1771 |
uint64_t m0 = ELEM(zd, 1, 8) | (ELEM(zd, 3, 8) << 8) |
|
| 1772 |
| (ELEM(zd, 5, 8) << 16) | (ELEM(zd, 7, 8) << 24) |
|
| 1773 |
| (ELEM(zm, 1, 8) << 32) | (ELEM(zm, 3, 8) << 40) |
|
| 1774 |
| (ELEM(zm, 5, 8) << 48) | (ELEM(zm, 7, 8) << 56); |
|
| 1775 |
env->vfp.regs[rm] = make_float64(m0); |
|
| 1776 |
env->vfp.regs[rd] = make_float64(d0); |
|
| 1777 |
} |
|
| 1778 |
|
|
| 1779 |
void HELPER(neon_unzip16)(CPUState *env, uint32_t rd, uint32_t rm) |
|
| 1780 |
{
|
|
| 1781 |
uint64_t zm = float64_val(env->vfp.regs[rm]); |
|
| 1782 |
uint64_t zd = float64_val(env->vfp.regs[rd]); |
|
| 1783 |
uint64_t d0 = ELEM(zd, 0, 16) | (ELEM(zd, 2, 16) << 16) |
|
| 1784 |
| (ELEM(zm, 0, 16) << 32) | (ELEM(zm, 2, 16) << 48); |
|
| 1785 |
uint64_t m0 = ELEM(zd, 1, 16) | (ELEM(zd, 3, 16) << 16) |
|
| 1786 |
| (ELEM(zm, 1, 16) << 32) | (ELEM(zm, 3, 16) << 48); |
|
| 1787 |
env->vfp.regs[rm] = make_float64(m0); |
|
| 1788 |
env->vfp.regs[rd] = make_float64(d0); |
|
| 1789 |
} |
|
Also available in: Unified diff