Classes
struct	FPerPrioritizerInfo

Public Types
enum	EConstants : unsigned { MaxObjectCountPerBatch = 1024U }

enum	EBatchProcessStatus : unsigned { ProcessFullBatchesAndContinue , ProcessAllBatchesAndStop , NothingToProcess }

Public Member Functions
	FPrioritizerBatchHelper (uint32 PrioritizerCount)

void	InitForConnection ()

EBatchProcessStatus	PrepareBatch (FPerConnectionInfo &ConnInfo, const FNetBitArrayView Objects, const uint8 PrioritizerIndices, const float InDefaultPriorities)

Public Attributes
TArray< FPerPrioritizerInfo, TInlineAllocator< 16 > >	PerPrioritizerInfos

Member Enumeration Documentation

◆ EBatchProcessStatus

enum UE::Net::Private::FReplicationPrioritization::FPrioritizerBatchHelper::EBatchProcessStatus : unsigned

Enumerator
ProcessFullBatchesAndContinue
ProcessAllBatchesAndStop
NothingToProcess

◆ EConstants

enum UE::Net::Private::FReplicationPrioritization::FPrioritizerBatchHelper::EConstants : unsigned

Enumerator
MaxObjectCountPerBatch

Constructor & Destructor Documentation

◆ FPrioritizerBatchHelper()

UE::Net::Private::FReplicationPrioritization::FPrioritizerBatchHelper::FPrioritizerBatchHelper ( uint32 PrioritizerCount )

inlineexplicit

Member Function Documentation

◆ InitForConnection()

void UE::Net::Private::FReplicationPrioritization::FPrioritizerBatchHelper::InitForConnection ( )

inline

◆ PrepareBatch()

EBatchProcessStatus UE::Net::Private::FReplicationPrioritization::FPrioritizerBatchHelper::PrepareBatch	(	FPerConnectionInfo &	ConnInfo,
		const FNetBitArrayView	Objects,
		const uint8 *	PrioritizerIndices,
		const float *	InDefaultPriorities
	)

inline

Algorithm will get MaxBatchObjectCount dirty objects and sort them into the correct prioritizer. Return on the following conditions:

If a prioritizer has reached at least MaxBatchObjectCount
If all dirty objects have been sorted.

If we've processed all objects return ProcessAll. If any prioritizer has a full batch then return ProcessFull. This limits the memory footprint to a maximum of two chunks per prioritizer. Else continue.

Is there a full batch for any prioritizer? We expect checking it after the fact to be a lot faster in typical scenarios versus checking after each object addition to a prioritizer.

Member Data Documentation

◆ PerPrioritizerInfos

TArray<FPerPrioritizerInfo, TInlineAllocator<16> > UE::Net::Private::FReplicationPrioritization::FPrioritizerBatchHelper::PerPrioritizerInfos

The documentation for this class was generated from the following file:

Engine/Source/Runtime/Net/Iris/Private/Iris/ReplicationSystem/Prioritization/ReplicationPrioritization.cpp

Classes

Public Types

Public Member Functions

Public Attributes