Diamond75/Panda3DS
1
0
Fork 0
mirror of https://github.com/wheremyfoodat/Panda3DS.git synced 2025-04-18 03:31:31 +12:00
Code Issues Projects Releases Packages Wiki Activity Actions

[Shader JIT] Fix swizzle when loading from indexed reg

Browse source
This commit is contained in:
wheremyfoodat 2023-07-01 00:49:34 +03:00
parent 5fecb6023a
commit 726897231c
1 changed files with 31 additions and 6 deletions
Download patch file Download diff file Expand all files Collapse all files

37
src/core/PICA/dynapica/shader_rec_emitter_x64.cpp
View file

@ -253,6 +253,14 @@ void ShaderEmitter::loadRegister(Xmm dest, const PICAShader& shader, u32 src, u3
Helpers::panic("[ShaderJIT]: Unimplemented source index type %d", index); Helpers::panic("[ShaderJIT]: Unimplemented source index type %d", index);
} }
// Swizzle and load register into dest, from [state pointer + rcx + offset] and apply the relevant swizzle
auto swizzleAndLoadReg = [this, &dest, &compSwizzle, &convertedSwizzle](size_t offset) {
if (compSwizzle == noSwizzle) // Avoid emitting swizzle if not necessary
movaps(dest, xword[statePointer + rcx + offset]);
else // Swizzle is not trivial so we need to emit a shuffle instruction
pshufd(dest, xword[statePointer + rcx + offset], convertedSwizzle);
};
// Here we handle what happens when using indexed addressing & we can't predict what register will be read at compile time // Here we handle what happens when using indexed addressing & we can't predict what register will be read at compile time
// The index of the access is assumed to be in rax // The index of the access is assumed to be in rax
// Add source register (src) and index (rax) to form the final register // Add source register (src) and index (rax) to form the final register
@ -268,7 +276,7 @@ void ShaderEmitter::loadRegister(Xmm dest, const PICAShader& shader, u32 src, u3
jae(maybeTemp); jae(maybeTemp);
mov(rcx, rax); mov(rcx, rax);
shl(rcx, 4); // rcx = rax * sizeof(vec4 of floats) = rax * 16 shl(rcx, 4); // rcx = rax * sizeof(vec4 of floats) = rax * 16
movaps(dest, xword[statePointer + rcx + inputOffset]); swizzleAndLoadReg(inputOffset);
jmp(end); jmp(end);
// If (reg < 0x1F) return tempRegisters[reg - 0x10] // If (reg < 0x1F) return tempRegisters[reg - 0x10]
@ -277,7 +285,7 @@ void ShaderEmitter::loadRegister(Xmm dest, const PICAShader& shader, u32 src, u3
jae(maybeUniform); jae(maybeUniform);
lea(rcx, qword[rax - 0x10]); lea(rcx, qword[rax - 0x10]);
shl(rcx, 4); shl(rcx, 4);
movaps(dest, xword[statePointer + rcx + tempOffset]); swizzleAndLoadReg(tempOffset);
jmp(end); jmp(end);
// If (reg < 0x80) return floatUniforms[reg - 0x20] // If (reg < 0x80) return floatUniforms[reg - 0x20]
@ -286,7 +294,7 @@ void ShaderEmitter::loadRegister(Xmm dest, const PICAShader& shader, u32 src, u3
jae(unknownReg); jae(unknownReg);
lea(rcx, qword[rax - 0x20]); lea(rcx, qword[rax - 0x20]);
shl(rcx, 4); shl(rcx, 4);
movaps(dest, xword[statePointer + rcx + uniformOffset]); swizzleAndLoadReg(uniformOffset);
jmp(end); jmp(end);
L(unknownReg); L(unknownReg);
@ -844,7 +852,24 @@ void ShaderEmitter::recLOOP(const PICAShader& shader, u32 instruction) {
add(eax, 1); // The iteration count is actually uniform.x + 1 add(eax, 1); // The iteration count is actually uniform.x + 1
mov(dword[statePointer + loopRegOffset], ecx); // Set loop counter mov(dword[statePointer + loopRegOffset], ecx); // Set loop counter
Helpers::panic("Unimplemented LOOP instruction"); // TODO: This might break if an instruction in a loop decides to yield...
push(rax); // Push loop iteration counter
push(rdx); // Push loop increment
Label loopStart;
L(loopStart);
compileUntil(shader, dest + 1);
const size_t stackOffsetOfLoopIncrement = 0;
const size_t stackOffsetOfIterationCounter = stackOffsetOfLoopIncrement + 8;
mov(ecx, dword[rsp + stackOffsetOfLoopIncrement]); // ecx = Loop increment
add(dword[statePointer + loopRegOffset], ecx); // Increment loop counter
sub(dword[rsp + stackOffsetOfIterationCounter], 1); // Subtract 1 from loop iteration counter
jnz(loopStart); // Back to loop start if not over
add(rsp, 16);
loopLevel--;
} }
void ShaderEmitter::printLog(const PICAShader& shaderUnit) { void ShaderEmitter::printLog(const PICAShader& shaderUnit) {
@ -852,12 +877,12 @@ void ShaderEmitter::printLog(const PICAShader& shaderUnit) {
for (int i = 0; i < shaderUnit.tempRegisters.size(); i++) { for (int i = 0; i < shaderUnit.tempRegisters.size(); i++) {
const auto& r = shaderUnit.tempRegisters[i]; const auto& r = shaderUnit.tempRegisters[i];
printf("t%d: (%f, %f, %f, %f)\n", i, r[0].toFloat64(), r[1].toFloat64(), r[2].toFloat64(), r[3].toFloat64()); printf("t%d: (%.2f, %.2f, %.2f, %.2f)\n", i, r[0].toFloat64(), r[1].toFloat64(), r[2].toFloat64(), r[3].toFloat64());
} }
for (int i = 0; i < shaderUnit.outputs.size(); i++) { for (int i = 0; i < shaderUnit.outputs.size(); i++) {
const auto& r = shaderUnit.outputs[i]; const auto& r = shaderUnit.outputs[i];
printf("o%d: (%f, %f, %f, %f)\n", i, r[0].toFloat64(), r[1].toFloat64(), r[2].toFloat64(), r[3].toFloat64()); printf("o%d: (%.2f, %.2f, %.2f, %.2f)\n", i, r[0].toFloat64(), r[1].toFloat64(), r[2].toFloat64(), r[3].toFloat64());
} }
printf("addr: (%d, %d)\n", shaderUnit.addrRegister[0], shaderUnit.addrRegister[1]); printf("addr: (%d, %d)\n", shaderUnit.addrRegister[0], shaderUnit.addrRegister[1]);