diff options
| author | slaren <slarengh@gmail.com> | 2024-03-18 11:03:04 +0100 |
|---|---|---|
| committer | GitHub <noreply@github.com> | 2024-03-18 11:03:04 +0100 |
| commit | 2bf8d0f7c4cc1235755ad06961ca761e458c5e55 (patch) | |
| tree | d2a462deb3c0e34cfb26eab6881a65bfb9fc3b28 /ggml-backend-impl.h | |
| parent | 496bc79bc2b79bfd6124b8687a8dbd6a646e9b06 (diff) | |
backend : offload large batches to GPU (#6083)
* backend : offload large batches to GPU
* fix hip
* code cleanup
* fix CUDA split buffers
* Update ggml-backend-impl.h
Co-authored-by: Johannes Gäßler <johannesg@5d6.de>
* cuda : fix memset without set_device
* imatrix : remove sched affix from weight names
* sched : add a new split if the current one has too many inputs
reduce max inputs per split
more cleanup
* update backends
ggml-ci
---------
Co-authored-by: Johannes Gäßler <johannesg@5d6.de>
Diffstat (limited to 'ggml-backend-impl.h')
| -rw-r--r-- | ggml-backend-impl.h | 5 |
1 files changed, 5 insertions, 0 deletions
diff --git a/ggml-backend-impl.h b/ggml-backend-impl.h index e475e20e..f121e1de 100644 --- a/ggml-backend-impl.h +++ b/ggml-backend-impl.h @@ -103,6 +103,11 @@ extern "C" { // check if the backend supports an operation bool (*GGML_CALL supports_op)(ggml_backend_t backend, const struct ggml_tensor * op); + // check if the backend wants to run an operation, even if the weights are allocated in a CPU buffer + // these should be expensive operations with large batch sizes that may benefit from running on this backend + // even if the weight has to be copied from the CPU temporarily + bool (*GGML_CALL offload_op)(ggml_backend_t backend, const struct ggml_tensor * op); + // (optional) event synchronization ggml_backend_event_t (*GGML_CALL event_new) (ggml_backend_t backend); void (*GGML_CALL event_free) (ggml_backend_event_t event); |