Skip to content

Fix KV cache for split mode graph with layers left on CPU #1506

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
ikawrakow merged 1 commit into from
Mar 25, 2026
+49 −52
Merged

Fix KV cache for split mode graph with layers left on CPU #1506

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
101 changes: 49 additions & 52 deletions src/llama.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -916,62 +916,59 @@ static bool llama_kv_cache_init(
if (split_cache && (!K || !V || !K->extra || !V->extra)) {
ctx = offload ? ctx_map.at(model.buft_layer[i].buft) : cache.ctxs.front();
split_cache_i = false;
} else {
int n_embd_head_v = hparams.n_embd_head_v;
k = ggml_new_tensor_2d(ctx, type_k, n_embd_head_k, n_head_kv*kv_size);

int64_t v_ne = int64_t(n_embd_v_row)*kv_size;
v = ggml_new_tensor_1d(ctx, type_v, v_ne);

auto k_name = std::string{"cache_k_l"} + std::to_string(i);
auto v_name = std::string{"cache_v_l"} + std::to_string(i);
ggml_set_name(k, k_name.c_str());
ggml_set_name(v, v_name.c_str());
//ggml_format_name(k, "cache_k_l%d", i);
//ggml_format_name(v, "cache_v_l%d", i);

if (split_cache_i) {
bool use_V_for_K = model.layers[i].attn_k_norm && model.layers[i].attn_k_norm->ne[0] == K->ne[1] ? true : false;
auto extra_K = (const ggml_split_tensor_t *)K->extra;
auto extra_V = (const ggml_split_tensor_t *)V->extra;
auto & split_k_l = cache.split_k_l.emplace_back();
auto & split_v_l = cache.split_v_l.emplace_back();
split_k_l.tensor_splits.resize(extra_K->n_device, nullptr);
split_v_l.tensor_splits.resize(extra_V->n_device, nullptr);
for (int is = 0; is < extra_K->n_device; ++is) {
auto split = use_V_for_K ? extra_V->splits[is] : extra_K->splits[is];
if (!split) continue;
int nhead_kv = use_V_for_K ? split->ne[1] / n_embd_head_v : split->ne[1]/n_embd_head_k;
if (use_V_for_K) {
LLAMA_LOG_DEBUG("K_cache(%d, %d): using %d instead of %ld heads\n",
i, is, nhead_kv, extra_K->splits[is]->ne[1]/n_embd_head_k);
}
split_k_l.tensor_splits[is] = ggml_new_tensor_2d(ctx, type_k, n_embd_head_k, nhead_kv * kv_size);
auto split_name = k_name + '.' + std::to_string(is);
ggml_set_name(split_k_l.tensor_splits[is], split_name.c_str());
mem_split[is] += ggml_nbytes(split_k_l.tensor_splits[is]);
}
split_k_l.ggml.n_device = extra_K->n_device;
split_k_l.ggml.split_dim = 0;
split_k_l.ggml.splits = split_k_l.tensor_splits.data();
for (int is = 0; is < extra_V->n_device; ++is) {
auto split = extra_V->splits[is];
if (!split) continue;
split_v_l.tensor_splits[is] = ggml_new_tensor_1d(ctx, type_v, split->ne[1] * kv_size);
auto split_name = v_name + '.' + std::to_string(is);
ggml_set_name(split_v_l.tensor_splits[is], split_name.c_str());
mem_split[is] += ggml_nbytes(split_v_l.tensor_splits[is]);
}
int n_embd_head_v = hparams.n_embd_head_v;
k = ggml_new_tensor_2d(ctx, type_k, n_embd_head_k, n_head_kv*kv_size);

int64_t v_ne = int64_t(n_embd_v_row)*kv_size;
v = ggml_new_tensor_1d(ctx, type_v, v_ne);

auto k_name = std::string{"cache_k_l"} + std::to_string(i);
auto v_name = std::string{"cache_v_l"} + std::to_string(i);
ggml_set_name(k, k_name.c_str());
ggml_set_name(v, v_name.c_str());

if (split_cache_i) {
bool use_V_for_K = model.layers[i].attn_k_norm && model.layers[i].attn_k_norm->ne[0] == K->ne[1] ? true : false;
auto extra_K = (const ggml_split_tensor_t *)K->extra;
auto extra_V = (const ggml_split_tensor_t *)V->extra;
auto & split_k_l = cache.split_k_l.emplace_back();
auto & split_v_l = cache.split_v_l.emplace_back();
split_k_l.tensor_splits.resize(extra_K->n_device, nullptr);
split_v_l.tensor_splits.resize(extra_V->n_device, nullptr);
for (int is = 0; is < extra_K->n_device; ++is) {
auto split = use_V_for_K ? extra_V->splits[is] : extra_K->splits[is];
if (!split) continue;
int nhead_kv = use_V_for_K ? split->ne[1] / n_embd_head_v : split->ne[1]/n_embd_head_k;
if (use_V_for_K) {
LLAMA_LOG_DEBUG("K_cache(%d, %d): using %d instead of %ld heads\n",
i, is, nhead_kv, extra_K->splits[is]->ne[1]/n_embd_head_k);
}
split_v_l.ggml.n_device = extra_V->n_device;
split_v_l.ggml.split_dim = 0;
split_v_l.ggml.splits = split_v_l.tensor_splits.data();
k->extra = (void *)&split_k_l.ggml;
v->extra = (void *)&split_v_l.ggml;
split_k_l.tensor_splits[is] = ggml_new_tensor_2d(ctx, type_k, n_embd_head_k, nhead_kv * kv_size);
auto split_name = k_name + '.' + std::to_string(is);
ggml_set_name(split_k_l.tensor_splits[is], split_name.c_str());
mem_split[is] += ggml_nbytes(split_k_l.tensor_splits[is]);
}
split_k_l.ggml.n_device = extra_K->n_device;
split_k_l.ggml.split_dim = 0;
split_k_l.ggml.splits = split_k_l.tensor_splits.data();
for (int is = 0; is < extra_V->n_device; ++is) {
auto split = extra_V->splits[is];
if (!split) continue;
split_v_l.tensor_splits[is] = ggml_new_tensor_1d(ctx, type_v, split->ne[1] * kv_size);
auto split_name = v_name + '.' + std::to_string(is);
ggml_set_name(split_v_l.tensor_splits[is], split_name.c_str());
mem_split[is] += ggml_nbytes(split_v_l.tensor_splits[is]);
}
split_v_l.ggml.n_device = extra_V->n_device;
split_v_l.ggml.split_dim = 0;
split_v_l.ggml.splits = split_v_l.tensor_splits.data();
k->extra = (void *)&split_k_l.ggml;
v->extra = (void *)&split_v_l.ggml;
}
cache.k_l.push_back(k);
cache.v_l.push_back(v);
}
cache.k_l.push_back(k);
cache.v_l.push_back(v);
}
if (is_mla_attn && cparams.mla_attn && n_mla < n_layer && n_mla > 0) {
LLAMA_LOG_ERROR("%s: unexpected situation with %d out of %d layers having MLA enabled\n", __func__, n_mla, int(n_layer));
Expand Down