llama : remove redundant reshape in build_kv_store (#6369)

* llama: remove redundant reshape in build_kv_store

This commit removes the reshape of the V matrix in the build_kv_store.

The motivation for this is that V matrix has the shape:
```console
(gdb) p *v_cur
$46 = {type = GGML_TYPE_F32, backend = GGML_BACKEND_TYPE_CPU,
       buffer = 0x0, ne = {4096, 512, 1, 1}, nb = {4, 16384, 8388608,
       8388608}, op = GGML_OP_MUL_MAT, op_params = {
       0 <repeats 16 times>}, flags = 0, grad = 0x0,
       src = {0xb496b0, 0x7ffef1c40950, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0,
       0x0, 0x0}, perf_runs = 0, perf_cycles = 0, perf_time_us = 0,
       view_src = 0x0, view_offs = 0, data = 0x0,
       name = "Vcur-0", '\000' <repeats 57 times>, extra = 0x0,
       padding = "\000\000\000\000\000\000\000"}
```
And after reshaping this tensor we get:
```console
gdb) p *ggml_reshape_2d(ctx, v_cur, n_embd_v_gqa, n_tokens)
$44 = {type = GGML_TYPE_F32, backend = GGML_BACKEND_TYPE_CPU,
       buffer = 0x0, ne = {4096, 512, 1, 1}, nb = {4, 16384, 8388608,
       8388608}, op = GGML_OP_RESHAPE, op_params = {
       0 <repeats 16 times>}, flags = 0, grad = 0x0,
       src = {0x7ffef1c40e00, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0,
       0x0}, perf_runs = 0, perf_cycles = 0, perf_time_us = 0,
       view_src = 0x7ffef1c40e00, view_offs = 0, data = 0x0,
       name = "Vcur-0 (reshaped)", '\000' <repeats 46 times>, extra = 0x0,
       padding = "\000\000\000\000\000\000\000"}
```
I noticed that the `src` and `view_src` fields are different but that the
dimensions are the same. From the code comment it seems like the reshape
call is not needed and perhaps the above can motivate the removal of the
reshape call.

Signed-off-by: Daniel Bevenius <redacted>
* llama : add assert

---------

Signed-off-by: Daniel Bevenius <redacted>
Co-authored-by: Georgi Gerganov <redacted>

diff --git a/llama.cpp b/llama.cpp
index 77ec9b7a1935d6ba3b6d422132b13ddc0fd946af..1875e247168411249d407d63af2ec4b78746fb77 100644 (file)
--- a/llama.cpp
+++ b/llama.cpp
@@ -5523,8 +5523,8 @@ static void llm_build_kv_store(
     GGML_ASSERT(kv.size == n_ctx);
 
     // compute the transposed [n_tokens, n_embd] V matrix
-    struct ggml_tensor * v_cur_t = ggml_transpose(ctx, ggml_reshape_2d(ctx, v_cur, n_embd_v_gqa, n_tokens));
-    //struct ggml_tensor * v_cur_t = ggml_transpose(ctx, v_cur); // TODO: reshape above is likely not needed
+    assert(v_cur->ne[0] == n_embd_v_gqa && v_cur->ne[1] == n_tokens);
+    struct ggml_tensor * v_cur_t = ggml_transpose(ctx, v_cur);
     cb(v_cur_t, "v_cur_t", il);
 
     struct ggml_tensor * k_cache_view = ggml_view_1d(ctx, kv.k_l[il], n_tokens*n_embd_k_gqa,