vrr/koboldcpp
2
0
Fork 0
mirror of https://github.com/LostRuins/koboldcpp.git synced 2025-09-10 17:14:36 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

Merge branch 'upstream' into concedo_experimental

Browse source 
# Conflicts:
#	.github/ISSUE_TEMPLATE/010-bug-compilation.yml
#	.github/ISSUE_TEMPLATE/011-bug-results.yml
#	.github/labeler.yml
#	.github/workflows/build.yml
#	.github/workflows/release.yml
#	.gitmodules
#	CMakeLists.txt
#	ggml/CMakeLists.txt
#	ggml/src/CMakeLists.txt
#	ggml/src/ggml-cann/aclnn_ops.cpp
#	ggml/src/ggml-cann/ggml-cann.cpp
#	ggml/src/ggml-opencl/ggml-opencl.cpp
#	ggml/src/ggml-opencl/kernels/softmax_4_f16.cl
#	ggml/src/ggml-opencl/kernels/softmax_4_f32.cl
#	ggml/src/ggml-opencl/kernels/softmax_f16.cl
#	ggml/src/ggml-opencl/kernels/softmax_f32.cl
#	ggml/src/ggml-sycl/element_wise.cpp
#	ggml/src/ggml-sycl/element_wise.hpp
#	ggml/src/ggml-sycl/ggml-sycl.cpp
#	scripts/sync-ggml-am.sh
#	scripts/sync-ggml.last
#	scripts/sync-ggml.sh
#	tests/test-backend-ops.cpp
#	tests/test-c.c
This commit is contained in:
Concedo 2025-07-05 12:16:28 +08:00
commit 57ce374240
64 changed files with 2944 additions and 979 deletions
Download patch file Download diff file Expand all files Collapse all files

28
src/llama-batch.cpp
View file

@ -166,6 +166,8 @@ bool llama_batch_allocr::init(
// note: tracking the other way around is not necessary for now
//seq_cpl[s0][s1] = true;
has_cpl = true;
}
}
}
@ -404,6 +406,10 @@ uint32_t llama_batch_allocr::get_n_outputs() const {
return n_outputs;
}
uint32_t llama_batch_allocr::get_n_used() const {
return n_used;
}
std::vector<int32_t> & llama_batch_allocr::get_out_ids() {
return out_ids;
}
@ -419,6 +425,8 @@ llama_pos llama_batch_allocr::seq_pos_max(llama_seq_id seq_id) const {
void llama_batch_allocr::split_reset() {
out_ids.clear();
n_used = 0;
used.clear();
used.resize(get_n_tokens(), false);
@ -443,6 +451,7 @@ llama_ubatch llama_batch_allocr::split_simple(uint32_t n_ubatch) {
idxs.push_back(cur_idx);
used[cur_idx] = true;
++n_used;
++cur_idx;
@ -458,9 +467,17 @@ llama_ubatch llama_batch_allocr::split_simple(uint32_t n_ubatch) {
return ubatch_add(idxs, idxs.size(), false);
}
llama_ubatch llama_batch_allocr::split_equal(uint32_t n_ubatch) {
llama_ubatch llama_batch_allocr::split_equal(uint32_t n_ubatch, bool sequential) {
if (sequential && has_cpl) {
LLAMA_LOG_ERROR("%s: sequential split is not supported when there are coupled sequences in the input batch\n", __func__);
return {};
}
std::vector<seq_set_t> cur_seq_set;
llama_seq_id last_seq_id = -1;
// determine the non-overlapping sequence sets participating in this ubatch
for (int32_t i = 0; i < batch.n_tokens; ++i) {
if (used[i]) {
@ -477,9 +494,16 @@ llama_ubatch llama_batch_allocr::split_equal(uint32_t n_ubatch) {
}
}
// accept only increasing sequence ids
if (sequential) {
add = add && (cur_seq_set.empty() || batch.seq_id[i][0] == last_seq_id + 1);
}
if (add) {
cur_seq_set.push_back(seq_set[i]);
last_seq_id = batch.seq_id[i][0];
if (cur_seq_set.size() > n_ubatch) {
break;
}
@ -528,6 +552,7 @@ llama_ubatch llama_batch_allocr::split_equal(uint32_t n_ubatch) {
idxs_per_seq[s].push_back(idx);
used[idx] = true;
++n_used;
++cur_idx[s];
}
@ -569,6 +594,7 @@ llama_ubatch llama_batch_allocr::split_seq(uint32_t n_ubatch) {
idxs.push_back(cur_idx);
used[cur_idx] = true;
++n_used;
if (idxs.size() >= n_ubatch) {
break;