Differences From Artifact [b494f24572]:
- File
src/OFMatrix4x4.m
— part of check-in
[d53c87e7bb]
at
2023-10-31 20:27:32
on branch trunk
— OFMatrix4x4: Partially unroll multiplication loop
This way, `result` doesn't need to be initialized. (user: js, size: 6634) [annotate] [blame] [check-ins using]
To Artifact [240d8b02dd]:
- File src/OFMatrix4x4.m — part of check-in [fa3c89f802] at 2023-10-31 22:14:33 on branch trunk — OFMatrix4x4: Add #pragma GCC target("3dnow") (user: js, size: 6771) [annotate] [blame] [check-ins using]
| ︙ | ︙ | |||
26 27 28 29 30 31 32 33 34 35 36 37 38 39 |
{ 0, 1, 0, 0 },
{ 0, 0, 1, 0 },
{ 0, 0, 0, 1 }
};
@implementation OFMatrix4x4
#if defined(OF_AMD64) || defined(OF_X86)
static void
multiplyWithMatrix_3DNow(OFMatrix4x4 *self, SEL _cmd, OFMatrix4x4 *matrix)
{
float result[4][4];
for (uint_fast8_t i = 0; i < 4; i++) {
for (uint_fast8_t j = 0; j < 4; j++) {
| > > > > | 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 |
{ 0, 1, 0, 0 },
{ 0, 0, 1, 0 },
{ 0, 0, 0, 1 }
};
@implementation OFMatrix4x4
#if defined(OF_AMD64) || defined(OF_X86)
# ifndef __clang__
# pragma GCC push_options
# pragma GCC target("3dnow")
# endif
static void
multiplyWithMatrix_3DNow(OFMatrix4x4 *self, SEL _cmd, OFMatrix4x4 *matrix)
{
float result[4][4];
for (uint_fast8_t i = 0; i < 4; i++) {
for (uint_fast8_t j = 0; j < 4; j++) {
|
| ︙ | ︙ | |||
113 114 115 116 117 118 119 120 121 122 123 124 125 126 |
"femms"
:: "r"(&result), "r"(&self->_values), "r"(&vector)
: "mm0", "mm1", "mm2", "mm3", "mm4", "memory"
);
return result;
}
+ (void)initialize
{
if (self != [OFMatrix4x4 class])
return;
if ([OFSystemInfo supports3DNow]) {
| > > > | 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 |
"femms"
:: "r"(&result), "r"(&self->_values), "r"(&vector)
: "mm0", "mm1", "mm2", "mm3", "mm4", "memory"
);
return result;
}
# ifndef __clang__
# pragma GCC pop_options
# endif
+ (void)initialize
{
if (self != [OFMatrix4x4 class])
return;
if ([OFSystemInfo supports3DNow]) {
|
| ︙ | ︙ |