mirror of
https://github.com/hrydgard/ppsspp.git
synced 2026-11-12 00:59:41 +01:00
use vmx for QuickTexHash
This commit is contained in:
1 parent
f3612b2afb
commit
1c7cae2139
1 file changed
+6
-8
@@ -879,20 +879,18 @@ static inline u32 QuickTexHash(u32 addr, int bufw, int w, int h, GETextureFormat
|
||||
{
|
||||
#endif
|
||||
#ifdef _XBOX
|
||||
if (0) {
|
||||
if ((((u32)(intptr_t)checkp | sizeInRAM) & 0x1f) == 0) {
|
||||
__vector4 add, xor;
|
||||
__vector4 cur = __vzero();
|
||||
const __vector4 * ptr = (const __vector4*)checkp;
|
||||
for (u32 i = 0; i < sizeInRAM / 16; i+=2) {
|
||||
add = __loadunalignedvector(&ptr[i]);
|
||||
xor = __loadunalignedvector(&ptr[i+1]);
|
||||
//check += *checkp++;
|
||||
//check ^= *checkp++;
|
||||
add = __lvx(&ptr[i],0);
|
||||
xor = __lvx(&ptr[i+1],0);
|
||||
cur = __vadduwm(cur, add);
|
||||
cur = __vand(cur, add);
|
||||
cur = __vxor(cur, xor);
|
||||
}
|
||||
// Add the four parts into the low i32. // todo
|
||||
check = cur.u[0];
|
||||
// Add the four parts into the low i32. // is it possible with vmx ?
|
||||
check = cur.u[0]+cur.u[1]+cur.u[2]+cur.u[3];
|
||||
} else
|
||||
#endif
|
||||
for (u32 i = 0; i < sizeInRAM / 8; ++i) {
|
||||
|
||||
Reference in new issue
Block a user