highperfocused/ollama
1
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2025-08-27 20:32:35 +02:00
Code Issues Packages Projects Releases Wiki Activity

disable output_all (#11959)

Browse Source
This commit is contained in:
Michael Yang
2025-08-18 17:45:40 -07:00
committed by GitHub
parent 9cfbffafc5
commit f804e8a460
3 changed files with 25 additions and 3 deletions
Download Patch File Download Diff File Expand all files Collapse all files

3
llama/llama.cpp/src/llama-context.cpp vendored
View File

@@ -962,8 +962,7 @@ int llama_context::decode(const llama_batch & batch_inp) {
const int64_t n_vocab = vocab.n_tokens();
const int64_t n_embd = hparams.n_embd;
// when computing embeddings, all tokens are output
const bool output_all = cparams.embeddings;
const bool output_all = false;
if (!balloc->init(batch_inp, vocab, memory.get(), n_embd, cparams.kv_unified ? LLAMA_MAX_SEQ : cparams.n_seq_max, output_all)) {
LLAMA_LOG_ERROR("%s: failed to initialize batch\n", __func__);

2
llama/patches/0019-Enable-CUDA-Graphs-for-gemma3n.patch
View File

@@ -13,7 +13,7 @@ checks.
1 file changed, 18 insertions(+)
diff --git a/ggml/src/ggml-cuda/ggml-cuda.cu b/ggml/src/ggml-cuda/ggml-cuda.cu
index 57eae461..9db0c8b5 100644
index 57eae461..c7f9dc3a 100644
--- a/ggml/src/ggml-cuda/ggml-cuda.cu
+++ b/ggml/src/ggml-cuda/ggml-cuda.cu
@@ -2671,12 +2671,24 @@ static bool check_node_graph_compatibility_and_refresh_copy_ops(ggml_backend_cud

23
llama/patches/0023-decode-disable-output_all.patch Normal file
View File

@@ -0,0 +1,23 @@
From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
From: Michael Yang <git@mxy.ng>
Date: Mon, 18 Aug 2025 16:58:39 -0700
Subject: [PATCH] decode: disable output_all
---
src/llama-context.cpp | 3 +--
1 file changed, 1 insertion(+), 2 deletions(-)
diff --git a/src/llama-context.cpp b/src/llama-context.cpp
index 26a5cf9c..6ece5263 100644
--- a/src/llama-context.cpp
+++ b/src/llama-context.cpp
@@ -962,8 +962,7 @@ int llama_context::decode(const llama_batch & batch_inp) {
const int64_t n_vocab = vocab.n_tokens();
const int64_t n_embd = hparams.n_embd;
- // when computing embeddings, all tokens are output
- const bool output_all = cparams.embeddings;
+ const bool output_all = false;
if (!balloc->init(batch_inp, vocab, memory.get(), n_embd, cparams.kv_unified ? LLAMA_MAX_SEQ : cparams.n_seq_max, output_all)) {
LLAMA_LOG_ERROR("%s: failed to initialize batch\n", __func__);