CS348Project/llama.cpp
5
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-10-31 08:51:55 +00:00
Code Issues Packages Projects Releases Wiki Activity

llama : use n_threads_batch only when n_tokens >= 32

Browse Source 
ggml-ci
This commit is contained in:
Georgi Gerganov
2023-11-27 20:48:44 +02:00
parent f815fe43d3
commit e9b7a5cbd0
1 changed files with 2 additions and 12 deletions
Download Patch File Download Diff File Expand all files Collapse all files

14
llama.cpp
View File

@@ -5433,7 +5433,7 @@ static int llama_decode_internal(
GGML_ASSERT(n_tokens <= n_batch); GGML_ASSERT(n_tokens <= n_batch);
int n_threads = n_tokens == 1 ? cparams.n_threads : cparams.n_threads_batch; int n_threads = n_tokens < 32 ? cparams.n_threads : cparams.n_threads_batch;
GGML_ASSERT((!batch.token && batch.embd) || (batch.token && !batch.embd)); // NOLINT GGML_ASSERT((!batch.token && batch.embd) || (batch.token && !batch.embd)); // NOLINT
const int64_t t_start_us = ggml_time_us(); const int64_t t_start_us = ggml_time_us();
@@ -5550,18 +5550,8 @@ static int llama_decode_internal(
n_threads = std::min(4, n_threads); n_threads = std::min(4, n_threads);
} }
// If all tensors can be run on the GPU then using more than 1 thread is detrimental.
const bool full_offload_supported =
model.arch == LLM_ARCH_LLAMA ||
model.arch == LLM_ARCH_BAICHUAN ||
model.arch == LLM_ARCH_FALCON ||
model.arch == LLM_ARCH_REFACT ||
model.arch == LLM_ARCH_MPT ||
model.arch == LLM_ARCH_STARCODER ||
model.arch == LLM_ARCH_STABLELM;
const bool fully_offloaded = model.n_gpu_layers >= (int) hparams.n_layer + 3; const bool fully_offloaded = model.n_gpu_layers >= (int) hparams.n_layer + 3;
if (ggml_cpu_has_cublas() && full_offload_supported && fully_offloaded) { if (ggml_cpu_has_cublas() && fully_offloaded) {
n_threads = 1; n_threads = 1;
} }