mirror of
https://git.suyu.dev/suyu/suyu.git
synced 2024-11-24 20:36:24 -05:00
dyncom: const correctness changes
This commit is contained in:
parent
87df493b5b
commit
d03e7f08ff
3 changed files with 7 additions and 7 deletions
|
@ -51,7 +51,7 @@ enum {
|
|||
|
||||
typedef unsigned int (*shtop_fp_t)(ARMul_State* cpu, unsigned int sht_oper);
|
||||
|
||||
static bool CondPassed(ARMul_State* cpu, unsigned int cond) {
|
||||
static bool CondPassed(const ARMul_State* cpu, unsigned int cond) {
|
||||
const bool n_flag = cpu->NFlag != 0;
|
||||
const bool z_flag = cpu->ZFlag != 0;
|
||||
const bool c_flag = cpu->CFlag != 0;
|
||||
|
|
|
@ -30,7 +30,7 @@
|
|||
* @return If the PC is being read, then the word-aligned PC value is returned.
|
||||
* If the PC is not being read, then the value stored in the register is returned.
|
||||
*/
|
||||
static inline u32 CHECK_READ_REG15_WA(ARMul_State* cpu, int Rn) {
|
||||
static inline u32 CHECK_READ_REG15_WA(const ARMul_State* cpu, int Rn) {
|
||||
return (Rn == 15) ? ((cpu->Reg[15] & ~0x3) + cpu->GetInstructionSize() * 2) : cpu->Reg[Rn];
|
||||
}
|
||||
|
||||
|
@ -43,6 +43,6 @@ static inline u32 CHECK_READ_REG15_WA(ARMul_State* cpu, int Rn) {
|
|||
* @return If the PC is being read, then the incremented PC value is returned.
|
||||
* If the PC is not being read, then the values stored in the register is returned.
|
||||
*/
|
||||
static inline u32 CHECK_READ_REG15(ARMul_State* cpu, int Rn) {
|
||||
static inline u32 CHECK_READ_REG15(const ARMul_State* cpu, int Rn) {
|
||||
return (Rn == 15) ? ((cpu->Reg[15] & ~0x1) + cpu->GetInstructionSize() * 2) : cpu->Reg[Rn];
|
||||
}
|
||||
|
|
|
@ -249,7 +249,7 @@ enum : u32 {
|
|||
VFP_SNAN = (VFP_NAN|VFP_NAN_SIGNAL)
|
||||
};
|
||||
|
||||
static inline int vfp_single_type(vfp_single* s)
|
||||
static inline int vfp_single_type(const vfp_single* s)
|
||||
{
|
||||
int type = VFP_NUMBER;
|
||||
if (s->exponent == 255) {
|
||||
|
@ -293,7 +293,7 @@ static inline void vfp_single_unpack(vfp_single* s, s32 val, u32* fpscr)
|
|||
|
||||
// Re-pack a single-precision float. This assumes that the float is
|
||||
// already normalised such that the MSB is bit 30, _not_ bit 31.
|
||||
static inline s32 vfp_single_pack(vfp_single* s)
|
||||
static inline s32 vfp_single_pack(const vfp_single* s)
|
||||
{
|
||||
u32 val = (s->sign << 16) +
|
||||
(s->exponent << VFP_SINGLE_MANTISSA_BITS) +
|
||||
|
@ -335,7 +335,7 @@ struct vfp_double {
|
|||
#define vfp_double_packed_exponent(v) (((v) >> VFP_DOUBLE_MANTISSA_BITS) & ((1 << VFP_DOUBLE_EXPONENT_BITS) - 1))
|
||||
#define vfp_double_packed_mantissa(v) ((v) & ((1ULL << VFP_DOUBLE_MANTISSA_BITS) - 1))
|
||||
|
||||
static inline int vfp_double_type(vfp_double* s)
|
||||
static inline int vfp_double_type(const vfp_double* s)
|
||||
{
|
||||
int type = VFP_NUMBER;
|
||||
if (s->exponent == 2047) {
|
||||
|
@ -379,7 +379,7 @@ static inline void vfp_double_unpack(vfp_double* s, s64 val, u32* fpscr)
|
|||
|
||||
// Re-pack a double-precision float. This assumes that the float is
|
||||
// already normalised such that the MSB is bit 30, _not_ bit 31.
|
||||
static inline s64 vfp_double_pack(vfp_double* s)
|
||||
static inline s64 vfp_double_pack(const vfp_double* s)
|
||||
{
|
||||
u64 val = ((u64)s->sign << 48) +
|
||||
((u64)s->exponent << VFP_DOUBLE_MANTISSA_BITS) +
|
||||
|
|
Loading…
Reference in a new issue