jobData | The job to schedule. |
transforms | The TransformAccessArray to run the job on. |
batchSize | Granularity in which workstealing is performed. For example, a value of 32 means the job queue will steal 32 iterations and then perform them in an efficient inner loop. |
dependsOn | A JobHandle containing any jobs that must finish executing before this job begins. You can combine multiple jobs with JobHandle.CombineDependencies. Use dependencies to ensure that two jobs that read or write to the same data don't run in parallel. |
JobHandle The handle identifying the scheduled job., which you can use as a dependency for a later job or to ensure completion on the main thread.
Schedules an IJobParallelForTransform job with read-only access to the transform data by reference.
This method provides better parallelization because it can read all transforms in parallel instead of just parallelizing over different hierarchies. Transforms aren't guaranteed to be valid when this scheduling mode is used. Invalid transform references in the input array are still processed. Use TransformAccess.isValid to check whether a specific instance is valid.