aaron/ispc
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

Add separate variants of memory built-ins for floats and doubles.

Browse Source 
Previously, we'd bitcast e.g. a vector of floats to a vector of i32s and then
use the i32 variant of masked_load/masked_store/gather/scatter.  Now, we have
separate float/double variants of each of those.
This commit is contained in:
Matt Pharr
2012-06-07 14:29:17 -07:00
parent 1ac3e03171
commit 89a2566e01
17 changed files with 593 additions and 41 deletions
Download Patch File Download Diff File Expand all files Collapse all files

64
ctx.cpp
View File

@@ -2518,12 +2518,16 @@ FunctionEmitContext::gather(llvm::Value *ptr, const PointerType *ptrType,
if (pt != NULL)
funcName = g->target.is32Bit ? "__pseudo_gather32_i32" :
"__pseudo_gather64_i64";
else if (llvmReturnType == LLVMTypes::DoubleVectorType ||
llvmReturnType == LLVMTypes::Int64VectorType)
else if (llvmReturnType == LLVMTypes::DoubleVectorType)
funcName = g->target.is32Bit ? "__pseudo_gather32_double" :
"__pseudo_gather64_double";
else if (llvmReturnType == LLVMTypes::Int64VectorType)
funcName = g->target.is32Bit ? "__pseudo_gather32_i64" :
"__pseudo_gather64_i64";
else if (llvmReturnType == LLVMTypes::FloatVectorType ||
llvmReturnType == LLVMTypes::Int32VectorType)
else if (llvmReturnType == LLVMTypes::FloatVectorType)
funcName = g->target.is32Bit ? "__pseudo_gather32_float" :
"__pseudo_gather64_float";
else if (llvmReturnType == LLVMTypes::Int32VectorType)
funcName = g->target.is32Bit ? "__pseudo_gather32_i32" :
"__pseudo_gather64_i32";
else if (llvmReturnType == LLVMTypes::Int16VectorType)
@@ -2538,15 +2542,15 @@ FunctionEmitContext::gather(llvm::Value *ptr, const PointerType *ptrType,
llvm::Function *gatherFunc = m->module->getFunction(funcName);
AssertPos(currentPos, gatherFunc != NULL);
llvm::Value *call = CallInst(gatherFunc, NULL, ptr, mask, name);
llvm::Value *gatherCall = CallInst(gatherFunc, NULL, ptr, mask, name);
// Add metadata about the source file location so that the
// optimization passes can print useful performance warnings if we
// can't optimize out this gather
if (disableGSWarningCount == 0)
addGSMetadata(call, currentPos);
addGSMetadata(gatherCall, currentPos);
return BitCastInst(call, llvmReturnType, LLVMGetName(call, "_gather_bitcast"));
return gatherCall;
}
@@ -2709,25 +2713,20 @@ FunctionEmitContext::maskedStore(llvm::Value *value, llvm::Value *ptr,
StoreInst(final, ptr);
return;
}
else if (Type::Equal(valueType, AtomicType::VaryingDouble) ||
Type::Equal(valueType, AtomicType::VaryingInt64) ||
else if (Type::Equal(valueType, AtomicType::VaryingDouble)) {
maskedStoreFunc = m->module->getFunction("__pseudo_masked_store_double");
}
else if (Type::Equal(valueType, AtomicType::VaryingInt64) ||
Type::Equal(valueType, AtomicType::VaryingUInt64)) {
ptr = BitCastInst(ptr, LLVMTypes::Int64VectorPointerType,
LLVMGetName(ptr, "_to_int64vecptr"));
value = BitCastInst(value, LLVMTypes::Int64VectorType,
LLVMGetName(value, "_to_int64"));
maskedStoreFunc = m->module->getFunction("__pseudo_masked_store_i64");
}
else if (Type::Equal(valueType, AtomicType::VaryingFloat) ||
Type::Equal(valueType, AtomicType::VaryingBool) ||
else if (Type::Equal(valueType, AtomicType::VaryingFloat)) {
maskedStoreFunc = m->module->getFunction("__pseudo_masked_store_float");
}
else if (Type::Equal(valueType, AtomicType::VaryingBool) ||
Type::Equal(valueType, AtomicType::VaryingInt32) ||
Type::Equal(valueType, AtomicType::VaryingUInt32) ||
CastType<EnumType>(valueType) != NULL) {
ptr = BitCastInst(ptr, LLVMTypes::Int32VectorPointerType,
LLVMGetName(ptr, "_to_int32vecptr"));
if (Type::Equal(valueType, AtomicType::VaryingFloat))
value = BitCastInst(value, LLVMTypes::Int32VectorType,
LLVMGetName(value, "_to_int32"));
maskedStoreFunc = m->module->getFunction("__pseudo_masked_store_i32");
}
else if (Type::Equal(valueType, AtomicType::VaryingInt16) ||
@@ -2827,27 +2826,34 @@ FunctionEmitContext::scatter(llvm::Value *value, llvm::Value *ptr,
llvm::Type *type = value->getType();
const char *funcName = NULL;
if (pt != NULL)
if (pt != NULL) {
funcName = g->target.is32Bit ? "__pseudo_scatter32_i32" :
"__pseudo_scatter64_i64";
else if (type == LLVMTypes::DoubleVectorType ||
type == LLVMTypes::Int64VectorType) {
}
else if (type == LLVMTypes::DoubleVectorType) {
funcName = g->target.is32Bit ? "__pseudo_scatter32_double" :
"__pseudo_scatter64_double";
}
else if (type == LLVMTypes::Int64VectorType) {
funcName = g->target.is32Bit ? "__pseudo_scatter32_i64" :
"__pseudo_scatter64_i64";
value = BitCastInst(value, LLVMTypes::Int64VectorType, "value2int");
}
else if (type == LLVMTypes::FloatVectorType ||
type == LLVMTypes::Int32VectorType) {
else if (type == LLVMTypes::FloatVectorType) {
funcName = g->target.is32Bit ? "__pseudo_scatter32_float" :
"__pseudo_scatter64_float";
}
else if (type == LLVMTypes::Int32VectorType) {
funcName = g->target.is32Bit ? "__pseudo_scatter32_i32" :
"__pseudo_scatter64_i32";
value = BitCastInst(value, LLVMTypes::Int32VectorType, "value2int");
}
else if (type == LLVMTypes::Int16VectorType)
else if (type == LLVMTypes::Int16VectorType) {
funcName = g->target.is32Bit ? "__pseudo_scatter32_i16" :
"__pseudo_scatter64_i16";
else if (type == LLVMTypes::Int8VectorType)
}
else if (type == LLVMTypes::Int8VectorType) {
funcName = g->target.is32Bit ? "__pseudo_scatter32_i8" :
"__pseudo_scatter64_i8";
}
llvm::Function *scatterFunc = m->module->getFunction(funcName);
AssertPos(currentPos, scatterFunc != NULL);