mirror of
https://github.com/paboyle/Grid.git
synced 2026-06-04 19:24:36 +01:00
Peter Boyle
49 lines
1.5 KiB
Markdown
49 lines
1.5 KiB
Markdown
---
|
|
name: ref_lattice_vs_vector
|
|
description: When to use Lattice<T> vs std::vector<T> for GPU-portable field storage in Grid
|
|
metadata:
|
|
node_type: memory
|
|
type: reference
|
|
originSessionId: 956e80aa-401d-481a-80bb-17f8abe1c131
|
|
---
|
|
|
|
## Rule
|
|
|
|
Use `Lattice<vobj>` (or `std::vector<Lattice<vobj>>`) for any field that will be read or written inside `accelerator_for`. `std::vector<vobj>` is host memory and is NOT device-accessible.
|
|
|
|
## Before vs after GPU offload
|
|
|
|
```cpp
|
|
// CPU-only (host memory, not GPU accessible)
|
|
std::vector<SpinColourVector_v> tloopv(oSites, Zero());
|
|
// accessed directly: tloopv[ss]
|
|
|
|
// GPU-portable
|
|
Lattice<SpinColourVector_v> tloop(grid);
|
|
// accessed via view: autoView(tloop_v, tloop, AcceleratorWrite);
|
|
// coalescedWrite(tloop_v[ss], val);
|
|
```
|
|
|
|
## Corollary: function signatures
|
|
|
|
CPU-only version:
|
|
```cpp
|
|
void PackLeft(const std::vector<std::vector<vobj>> &leftv);
|
|
```
|
|
|
|
GPU-portable version:
|
|
```cpp
|
|
void PackLeft(const std::vector<Lattice<vobj>> &leftv);
|
|
```
|
|
|
|
## deviceVector for raw device buffers
|
|
|
|
`deviceVector<T>` (defined in Grid) is like `std::vector<T>` but in device-accessible memory. Use for raw scalar scratch/pack buffers (e.g. GEMM input/output staging). Not for structured lattice data.
|
|
|
|
## Pointer arrays for batched BLAS
|
|
|
|
`deviceVector<scalar *>` holds batch pointer arrays. Populate with `acceleratorPut(ptrs[t], base + offset)` — sets device-side pointer from host. See `A2ASpatialSum::Allocate`.
|
|
|
|
## Related
|
|
[[ref_coalesced_views]] [[ref_accelerator_for]]
|