model : avoid ggml_cont_3d for fused QKV weights

[ggerganov]

ref #15602 (comment)

• Avoid Vcur = ggml_cont_3d(..) when the QKV weights are merged in a single tensor
• Make llama_kv_cache:: cpy_k and cpy_v more readable

[CISC]

Tested with CodeQwen1.5, Phi2, jina-embeddings-v3 and PLaMo2.

[CISC]

Hmmm, https://github.com/ggml-org/ci/blob/results/llama.cpp/60/d6e7c6fd8bacac0892b8722f5d5c585139cb43/ggml-4-x86-cuda-v100/stdall#L1957

[ggerganov]

Hmmm, https://github.com/ggml-org/ci/blob/results/llama.cpp/60/d6e7c6fd8bacac0892b8722f5d5c585139cb43/ggml-4-x86-cuda-v100/stdall#L1957

This is due to #15687

[CISC]

Hmmm, https://github.com/ggml-org/ci/blob/results/llama.cpp/60/d6e7c6fd8bacac0892b8722f5d5c585139cb43/ggml-4-x86-cuda-v100/stdall#L1957

This is due to #15687

Ah, I get a segfault locally though at the first REPEAT test after ARGMAX.

[ggerganov]

On my end, all tests except GET_ROWS and the new IM2COL_3D are passing.

[CISC]

On my end, all tests except GET_ROWS and the new IM2COL_3D are passing.

Nvm, must have been some other issue pre-rebase, I pulled latest changes and applied #15868 and everything is fine now.

Edit: Eh, almost, got GGML_ASSERT(ggml_is_contiguous(src0)) on PAD, but that's surely not related. It's pad_ext test with v == true. Fixed in #15869