Takahito Tejima
fbc139d39c
fix compile errors on gcd kernel
2013-03-07 22:43:47 -08:00
Takahito Tejima
8efecb0fca
Batching stuffs: generalized kernel batches, table/dispatcher refactoring, multiMeshFactory, drawContext, etc.
...
2 client APIs are changed.
- VertexBuffer::UpdateData() takes start vertex offset
- ComputeController::Refine() takes FarKernelBatchVector
Also, ComputeContext no longer holds farmesh.
Client can free farmesh after OsdComputeContext is created.
(but still need FarKernelBatchVector to apply subdivision kernels)
2013-03-07 17:50:15 -08:00
Aras Pranckevicius
e8ceaba176
gcd: convert face loop to work in 32 item size batches.
...
MBP6,2 (2 core 4 thread), catmark_car Lv4: 31.2ms CPU, 15.8ms GCD (was 22ms)
2013-02-03 10:05:58 +02:00
Aras Pranckevicius
c4b463668a
gcd: convert most of loops to submit work in 32 item size batches.
...
MBP6,2 (2 core 4 thread), catmark_car Lv3: 6.9ms CPU, 5.2ms GCD; Lv4: 31.2ms CPU, 22.ms GCD
2013-02-03 09:38:33 +02:00
Aras Pranckevicius
b556e9d6cd
gcd: naive implementation using dispatch_apply for all loops. Not faster, dispatch overhead too high
2013-02-02 23:22:04 +02:00
Aras Pranckevicius
54c7329ba4
gcd: files for GCD dispatcher (right now just copy of OMP one), add items to glViewer
2013-02-02 22:53:33 +02:00
