happyz
/
llama.cpp
mirror of https://github.com/ggerganov/llama.cpp.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

memory : clarify comments for r_l and s_l tensors [no ci] (#19203) 

This commit updates the comments in state_write_data to clarify that it
is handling the R and S tensors and not Key and Value tensors.
This commit is contained in:
Daniel Bevenius 2026-01-30 15:18:41 +01:00 committed by GitHub
parent c3b87cebff
commit f3bc98890c
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 10 additions and 10 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

20
src/llama-memory-recurrent.cpp
View File

@ -785,21 +785,21 @@ void llama_memory_recurrent::state_write_data(llama_io_write_i & io, const std::
io.write(&s_trans, sizeof(s_trans));
io.write(&n_layer, sizeof(n_layer));
// Iterate and write all the keys first, each row is a cell
// Iterate and write all the R tensors first, each row is a cell
// Get whole range at a time
for (uint32_t il = 0; il < n_layer; ++il) {
// skip null layers (read_data will handle this by checking "r_l" and "s_l" for null)
if (r_l[il] == nullptr) continue;
// Write key type
// Write R tensor type
const int32_t r_type_i = (int32_t)r_l[il]->type;
io.write(&r_type_i, sizeof(r_type_i));
// Write row size of key
// Write row size of R tensor
const uint64_t r_size_row = ggml_row_size(r_l[il]->type, hparams.n_embd_r());
io.write(&r_size_row, sizeof(r_size_row));
// Read each range of cells of k_size length and write out
// Write each range of cells of r_size_row length
for (const auto & range : cell_ranges) {
const size_t range_size = range.second - range.first;
const size_t buf_size = range_size * r_size_row;
@ -812,15 +812,15 @@ void llama_memory_recurrent::state_write_data(llama_io_write_i & io, const std::
// skip null layers (read_data will handle this by checking "r_l" and "s_l" for null)
if (s_l[il] == nullptr) continue;
// Write value type
// Write S tensor type
const int32_t s_type_i = (int32_t)s_l[il]->type;
io.write(&s_type_i, sizeof(s_type_i));
// Write row size of value
// Write row size of S tensor
const uint64_t s_size_row = ggml_row_size(s_l[il]->type, hparams.n_embd_s());
io.write(&s_size_row, sizeof(s_size_row));
// Read each range of cells of s_size length and write out
// Write each range of S tensor rows
for (const auto & range : cell_ranges) {
const size_t range_size = range.second - range.first;
const size_t buf_size = range_size * s_size_row;
@ -828,7 +828,7 @@ void llama_memory_recurrent::state_write_data(llama_io_write_i & io, const std::
}
}
} else {
// When v is transposed, we also need the element size and get the element ranges from each row
// When S tensor is transposed, we also need the element size and get the element ranges from each row
const uint32_t mem_size = size;
for (uint32_t il = 0; il < n_layer; ++il) {
// skip null layers (read_data will handle this by checking "r_l" and "s_l" for null)
@ -836,7 +836,7 @@ void llama_memory_recurrent::state_write_data(llama_io_write_i & io, const std::
const uint32_t n_embd_s = hparams.n_embd_s();
// Write value type
// Write S tensor type
const int32_t s_type_i = (int32_t)s_l[il]->type;
io.write(&s_type_i, sizeof(s_type_i));
@ -849,7 +849,7 @@ void llama_memory_recurrent::state_write_data(llama_io_write_i & io, const std::
// For each row, we get the element values of each cell
for (uint32_t j = 0; j < n_embd_s; ++j) {
// Read each range of cells of v_size_el length and write out
// Write each range of cells of s_size_el length
for (const auto & range : cell_ranges) {
const size_t range_size = range.second - range.first;
const size_t src_offset = (range.first + j * mem_size) * s_size_el;