Skip to content

Support using mmap when applying LoRA #2095

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 3 commits into from
Jul 11, 2023
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
3 changes: 1 addition & 2 deletions examples/common.cpp
Original file line number Diff line number Diff line change
Expand Up @@ -267,7 +267,6 @@ bool gpt_params_parse(int argc, char ** argv, gpt_params & params) {
break;
}
params.lora_adapter = argv[i];
params.use_mmap = false;
} else if (arg == "--lora-base") {
if (++i >= argc) {
invalid_param = true;
Expand Down Expand Up @@ -497,7 +496,7 @@ void gpt_print_usage(int /*argc*/, char ** argv, const gpt_params & params) {
fprintf(stderr, " --mtest compute maximum memory usage\n");
fprintf(stderr, " --export export the computation graph to 'llama.ggml'\n");
fprintf(stderr, " --verbose-prompt print prompt before generation\n");
fprintf(stderr, " --lora FNAME apply LoRA adapter (implies --no-mmap)\n");
fprintf(stderr, " --lora FNAME apply LoRA adapter\n");
fprintf(stderr, " --lora-base FNAME optional model to use as a base for the layers modified by the LoRA adapter\n");
fprintf(stderr, " -m FNAME, --model FNAME\n");
fprintf(stderr, " model path (default: %s)\n", params.model.c_str());
Expand Down
2 changes: 1 addition & 1 deletion examples/main/README.md
Original file line number Diff line number Diff line change
Expand Up @@ -293,5 +293,5 @@ These options provide extra functionality and customization when running the LLa
- `-mg i, --main-gpu i`: When using multiple GPUs this option controls which GPU is used for small tensors for which the overhead of splitting the computation across all GPUs is not worthwhile. The GPU in question will use slightly more VRAM to store a scratch buffer for temporary results. By default GPU 0 is used. Requires cuBLAS.
- `-ts SPLIT, --tensor-split SPLIT`: When using multiple GPUs this option controls how large tensors should be split across all GPUs. `SPLIT` is a comma-separated list of non-negative values that assigns the proportion of data that each GPU should get in order. For example, "3,2" will assign 60% of the data to GPU 0 and 40% to GPU 1. By default the data is split in proportion to VRAM but this may not be optimal for performance. Requires cuBLAS.
- `-lv, --low-vram`: Do not allocate a VRAM scratch buffer for holding temporary results. Reduces VRAM usage at the cost of performance, particularly prompt processing speed. Requires cuBLAS.
- `--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. 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.
2 changes: 1 addition & 1 deletion examples/server/README.md
Original file line number Diff line number Diff line change
Expand Up @@ -16,7 +16,7 @@ Command line options:
- `--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.
- `--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.
- `--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. 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.
- `-to N`, `--timeout N`: Server read/write timeout in seconds. Default `600`.
- `--host`: Set the hostname or ip address to listen. Default `127.0.0.1`.
Expand Down
3 changes: 1 addition & 2 deletions examples/server/server.cpp
Original file line number Diff line number Diff line change
Expand Up @@ -632,7 +632,7 @@ static void server_print_usage(const char *argv0, const gpt_params &params,
fprintf(stderr, " model path (default: %s)\n", params.model.c_str());
fprintf(stderr, " -a ALIAS, --alias ALIAS\n");
fprintf(stderr, " set an alias for the model, will be added as `model` field in completion response\n");
fprintf(stderr, " --lora FNAME apply LoRA adapter (implies --no-mmap)\n");
fprintf(stderr, " --lora FNAME apply LoRA adapter\n");
fprintf(stderr, " --lora-base FNAME optional model to use as a base for the layers modified by the LoRA adapter\n");
fprintf(stderr, " --host ip address to listen (default (default: %s)\n", sparams.hostname.c_str());
fprintf(stderr, " --port PORT port to listen (default (default: %d)\n", sparams.port);
Expand Down Expand Up @@ -820,7 +820,6 @@ static void server_params_parse(int argc, char **argv, server_params &sparams,
break;
}
params.lora_adapter = argv[i];
params.use_mmap = false;
}
else if (arg == "--lora-base")
{
Expand Down
6 changes: 3 additions & 3 deletions llama-util.h
Original file line number Diff line number Diff line change
Expand Up @@ -175,13 +175,13 @@ struct llama_mmap {
llama_mmap(struct llama_file * file, size_t prefetch = (size_t) -1 /* -1 = max value */, bool numa = false) {
size = file->size;
int fd = fileno(file->fp);
int flags = MAP_SHARED;
int flags = MAP_PRIVATE;
// prefetch/readahead impairs performance on NUMA systems
if (numa) { prefetch = 0; }
#ifdef __linux__
if (prefetch) { flags |= MAP_POPULATE; }
#endif
addr = mmap(NULL, file->size, PROT_READ, flags, fd, 0);
addr = mmap(NULL, file->size, PROT_READ | PROT_WRITE, flags, fd, 0);
if (addr == MAP_FAILED) {
throw std::runtime_error(format("mmap failed: %s", strerror(errno)));
}
Expand Down Expand Up @@ -223,7 +223,7 @@ struct llama_mmap {
throw std::runtime_error(format("CreateFileMappingA failed: %s", llama_format_win_err(error).c_str()));
}

addr = MapViewOfFile(hMapping, FILE_MAP_READ, 0, 0, 0);
addr = MapViewOfFile(hMapping, FILE_MAP_COPY, 0, 0, 0);
error = GetLastError();
CloseHandle(hMapping);

Expand Down