CS348Project/llama.cpp
5
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-11-10 10:27:03 +00:00
Code Issues Packages Projects Releases Wiki Activity

llama: fix ASAN error with M-RoPE (#16848)

Browse Source
This commit is contained in:
Xuan-Son Nguyen
2025-10-29 20:11:39 +01:00
committed by GitHub
parent e3af5563bd
commit 3464bdac37
1 changed files with 8 additions and 5 deletions
Download Patch File Download Diff File Expand all files Collapse all files

13
src/llama-batch.cpp
View File

@@ -669,10 +669,8 @@ llama_ubatch llama_batch_allocr::ubatch_add(const std::vector<int32_t> & idxs, u
auto udata = std::make_shared<llama_ubatch::data_t>();
const int32_t n_pos_cur = batch.embd ? n_pos_per_embd : 1;
const int64_t n_embd_all = batch.embd ? (int64_t) n_tokens*n_embd : 0;
const int64_t n_pos_all = (int64_t) n_tokens*n_pos_cur;
const int64_t n_pos_all = (int64_t) n_tokens*n_pos_per_embd;
udata->token .resize(n_tokens);
udata->embd .resize(n_embd_all);
@@ -694,8 +692,13 @@ llama_ubatch llama_batch_allocr::ubatch_add(const std::vector<int32_t> & idxs, u
memcpy(udata->embd.data() + i*n_embd, batch.embd + (int64_t) idxs[i]*n_embd, n_embd*sizeof(float));
}
for (int j = 0; j < n_pos_cur; ++j) {
udata->pos[j*n_tokens + i] = batch.pos[j*batch.n_tokens + idxs[i]];
for (size_t j = 0; j < (size_t)n_pos_per_embd; ++j) {
// if we are using M-RoPE
// if the current batch is text, we need to broadcast the same position across all RoPE sections
// otherwise, the input batch is image embeddings, we copy the positions as-is
// if we are not using M-RoPE, there is only one position per token (this loop runs only once)
size_t src_off = batch.token ? 0 : j*batch.n_tokens;
udata->pos[j*n_tokens + i] = batch.pos[src_off + idxs[i]];
}
udata->n_seq_id[i] = batch.n_seq_id[idxs[i]];