root/llama.cpp
1
0
Fork 0
mirror of https://github.com/ggerganov/llama.cpp.git synced 2024-12-24 10:24:35 +00:00
Code Issues Actions 5 Packages Projects Releases Wiki Activity

server : add --numa support (#2524)

Browse Source
This commit is contained in:
Cheng Shao 2023-08-14 15:36:42 +02:00 committed by GitHub
parent 348acf188c
commit d75561df20
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
2 changed files with 6 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

1
examples/server/README.md
View File

@ -16,6 +16,7 @@ Command line options:
- `--memory-f32`: Use 32-bit floats instead of 16-bit floats for memory key+value. Not recommended. - `--memory-f32`: Use 32-bit floats instead of 16-bit floats for memory key+value. Not recommended.
- `--mlock`: Lock the model in memory, preventing it from being swapped out when memory-mapped. - `--mlock`: Lock the model in memory, preventing it from being swapped out when memory-mapped.
- `--no-mmap`: Do not memory-map the model. By default, models are mapped into memory, which allows the system to load only the necessary parts of the model as needed. - `--no-mmap`: Do not memory-map the model. By default, models are mapped into memory, which allows the system to load only the necessary parts of the model as needed.
- `--numa`: Attempt optimizations that help on some NUMA systems.
- `--lora FNAME`: Apply a LoRA (Low-Rank Adaptation) adapter to the model (implies --no-mmap). This allows you to adapt the pretrained model to specific tasks or domains. - `--lora FNAME`: Apply a LoRA (Low-Rank Adaptation) adapter to the model (implies --no-mmap). This allows you to adapt the pretrained model to specific tasks or domains.
- `--lora-base FNAME`: Optional model to use as a base for the layers modified by the LoRA adapter. This flag is used in conjunction with the `--lora` flag, and specifies the base model for the adaptation. - `--lora-base FNAME`: Optional model to use as a base for the layers modified by the LoRA adapter. This flag is used in conjunction with the `--lora` flag, and specifies the base model for the adaptation.
- `-to N`, `--timeout N`: Server read/write timeout in seconds. Default `600`. - `-to N`, `--timeout N`: Server read/write timeout in seconds. Default `600`.

5
examples/server/server.cpp
View File

@ -666,6 +666,7 @@ static void server_print_usage(const char *argv0, const gpt_params &params,
{ {
fprintf(stdout, " --no-mmap do not memory-map model (slower load but may reduce pageouts if not using mlock)\n"); fprintf(stdout, " --no-mmap do not memory-map model (slower load but may reduce pageouts if not using mlock)\n");
} }
fprintf(stdout, " --numa attempt optimizations that help on some NUMA systems\n");
#ifdef LLAMA_SUPPORTS_GPU_OFFLOAD #ifdef LLAMA_SUPPORTS_GPU_OFFLOAD
fprintf(stdout, " -ngl N, --n-gpu-layers N\n"); fprintf(stdout, " -ngl N, --n-gpu-layers N\n");
fprintf(stdout, " number of layers to store in VRAM\n"); fprintf(stdout, " number of layers to store in VRAM\n");
@ -940,6 +941,10 @@ static void server_params_parse(int argc, char **argv, server_params &sparams,
{ {
params.use_mmap = false; params.use_mmap = false;
} }
else if (arg == "--numa")
{
params.numa = true;
}
else if (arg == "--embedding") else if (arg == "--embedding")
{ {
params.embedding = true; params.embedding = true;