mirror of
https://github.com/ggml-org/llama.cpp.git
synced 2026-09-11 04:56:56 +02:00
Salvatore MesoracaandGeorgi Gerganov
efe6a83e30
* ggml_cont: fix issue with transposed tensors when one dimension is 1 when using multiple threads, it is not enough to check for the tensors to be contiguous for ggml_compute_forward_dup_same_cont to work correctly. The tensors strides also need to match. Signed-off-by: Salvatore Mesoraca <[email protected]> * Add ggml_cont tests Signed-off-by: Salvatore Mesoraca <[email protected]> * Remove dead code it isn't possible to reach this code because all these functions are invoked by ggml_compute_forward_dup if and only if src0->type != dst->type Signed-off-by: Salvatore Mesoraca <[email protected]> * Make ggml_compute_forward_dup_same_cont work with contiguous tensors Co-authored-by: Georgi Gerganov <[email protected]> Signed-off-by: Salvatore Mesoraca <[email protected]> --------- Signed-off-by: Salvatore Mesoraca <[email protected]> Co-authored-by: Georgi Gerganov <[email protected]>