mirror of
https://github.com/ggml-org/llama.cpp.git
synced 2025-10-31 08:51:55 +00:00
f30ea47a87
* llama : add pipeline parallelism support for batch processing with multiple CUDA GPUs ggml-ci * server : add -ub, --ubatch-size parameter * fix server embedding test * llama : fix Mamba inference for pipeline parallelism Tested to work correctly with both `main` and `parallel` examples. * llama : limit max batch size to n_batch * add LLAMA_SCHED_MAX_COPIES to configure the number of input copies for pipeline parallelism default increase to 4 (from 2) changing this value may improve performance for some systems, but increases memory usage * fix hip build * fix sycl build (disable cpy_tensor_async) * fix hip build * llama : limit n_batch and n_ubatch to n_ctx during context creation * llama : fix norm backend * batched-bench : sync after decode * swiftui : sync after decode * ggml : allow ggml_get_rows to use multiple threads if they are available * check n_ubatch >= n_tokens with non-casual attention * llama : do not limit n_batch to n_ctx with non-casual attn * server : construct batch with size of llama_n_batch * ggml_backend_cpu_graph_compute : fix return value when alloc fails * llama : better n_batch and n_ubatch comment * fix merge * small fix * reduce default n_batch to 2048 --------- Co-authored-by: Francis Couture-Harpin <git@compilade.net> Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
Server tests
Python based server tests scenario using BDD and behave:
- issues.feature Pending issues scenario
- parallel.feature Scenario involving multi slots and concurrent requests
- security.feature Security, CORS and API Key
- server.feature Server base scenario: completion, embedding, tokenization, etc...
Tests target GitHub workflows job runners with 4 vCPU.
Requests are using aiohttp, asyncio based http client.
Note: If the host architecture inference speed is faster than GitHub runners one, parallel scenario may randomly fail.
To mitigate it, you can increase values in n_predict, kv_size.
Install dependencies
pip install -r requirements.txt
Run tests
- Build the server
cd ../../..
mkdir build
cd build
cmake ../
cmake --build . --target server
- Start the test:
./tests.sh
It's possible to override some scenario steps values with environment variables:
| variable | description |
|---|---|
PORT |
context.server_port to set the listening port of the server during scenario, default: 8080 |
LLAMA_SERVER_BIN_PATH |
to change the server binary path, default: ../../../build/bin/server |
DEBUG |
"ON" to enable steps and server verbose mode --verbose |
SERVER_LOG_FORMAT_JSON |
if set switch server logs to json format |
N_GPU_LAYERS |
number of model layers to offload to VRAM -ngl --n-gpu-layers |
Run @bug, @wip or @wrong_usage annotated scenario
Feature or Scenario must be annotated with @llama.cpp to be included in the default scope.
@bugannotation aims to link a scenario with a GitHub issue.@wrong_usageare meant to show user issue that are actually an expected behavior@wipto focus on a scenario working in progress@slowheavy test, disabled by default
To run a scenario annotated with @bug, start:
DEBUG=ON ./tests.sh --no-skipped --tags bug
After changing logic in steps.py, ensure that @bug and @wrong_usage scenario are updated.
./tests.sh --no-skipped --tags bug,wrong_usage || echo "should failed but compile"