jdelony/llama.cpp
1
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2026-06-27 23:50:20 -05:00
Code Issues Packages Projects Releases Wiki Activity

binaries : Improve rpc-server and export-graph-ops names. (#25045)

Browse Source 
Tests are generally prefixed with -test, so rename export-graph-ops
accordingly.

rpc-server is probably too generic a name for /usr/bin. Because it
should work with any ggml application, it is renamed to ggml-rpc-server.
This commit is contained in:
Christian Kastner 2026-06-27 09:31:29 +02:00 committed by GitHub
parent 0275c0f800
commit c299a92c38
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
6 changed files with 25 additions and 25 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
SECURITY.md
View File

@ -80,7 +80,7 @@ To protect sensitive data from potential leaks or unauthorized access, it is cru
### Untrusted environments or networks ### Untrusted environments or networks
If you can't run your models in a secure and isolated environment or if it must be exposed to an untrusted network, make sure to take the following security precautions: If you can't run your models in a secure and isolated environment or if it must be exposed to an untrusted network, make sure to take the following security precautions:
* Do not use the RPC backend, [rpc-server](https://github.com/ggml-org/llama.cpp/tree/master/tools/rpc) and [llama-server](https://github.com/ggml-org/llama.cpp/tree/master/tools/server) functionality (see https://github.com/ggml-org/llama.cpp/pull/13061). * Do not use the RPC backend, [ggml-rpc-server](https://github.com/ggml-org/llama.cpp/tree/master/tools/rpc) and [llama-server](https://github.com/ggml-org/llama.cpp/tree/master/tools/server) functionality (see https://github.com/ggml-org/llama.cpp/pull/13061).
* Confirm the hash of any downloaded artifact (e.g. pre-trained model weights) matches a known-good value. * Confirm the hash of any downloaded artifact (e.g. pre-trained model weights) matches a known-good value.
* Encrypt your data if sending it over the network. * Encrypt your data if sending it over the network.

8
tests/CMakeLists.txt
View File

@ -302,9 +302,9 @@ target_link_libraries(${TEST_TARGET} PRIVATE llama)
llama_build_and_test(test-alloc.cpp) llama_build_and_test(test-alloc.cpp)
target_include_directories(test-alloc PRIVATE ${PROJECT_SOURCE_DIR}/ggml/src) target_include_directories(test-alloc PRIVATE ${PROJECT_SOURCE_DIR}/ggml/src)
llama_build(export-graph-ops.cpp) llama_build(test-export-graph-ops.cpp)
target_include_directories(export-graph-ops PRIVATE ${PROJECT_SOURCE_DIR}/ggml/src) target_include_directories(test-export-graph-ops PRIVATE ${PROJECT_SOURCE_DIR}/ggml/src)
if (TARGET gguf-model-data) if (TARGET gguf-model-data)
target_link_libraries(export-graph-ops PRIVATE gguf-model-data) target_link_libraries(test-export-graph-ops PRIVATE gguf-model-data)
target_compile_definitions(export-graph-ops PRIVATE LLAMA_HF_FETCH) target_compile_definitions(test-export-graph-ops PRIVATE LLAMA_HF_FETCH)
endif() endif()

2
tests/test-backend-ops.cpp
View File

@ -9943,7 +9943,7 @@ static void usage(char ** argv) {
printf(" --output specifies output format (default: console, options: console, sql, csv)\n"); printf(" --output specifies output format (default: console, options: console, sql, csv)\n");
printf(" --list-ops lists all available GGML operations\n"); printf(" --list-ops lists all available GGML operations\n");
printf(" --show-coverage shows test coverage\n"); printf(" --show-coverage shows test coverage\n");
printf(" --test-file reads test operators from a test file generated by llama-export-graph-ops\n"); printf(" --test-file reads test operators from a test file generated by test-export-graph-ops\n");
printf(" -j <n> runs tests using <n> parallel worker threads (default: 1, test mode only)\n"); printf(" -j <n> runs tests using <n> parallel worker threads (default: 1, test mode only)\n");
} }

2
tests/export-graph-ops.cpp → tests/test-export-graph-ops.cpp
View File

@ -185,7 +185,7 @@ int main(int argc, char ** argv) {
return 1; return 1;
} }
#else #else
LOG_ERR("export-graph-ops compiled without HF fetch support\n"); LOG_ERR("test-export-graph-ops compiled without HF fetch support\n");
return 1; return 1;
#endif #endif
} }

2
tools/rpc/CMakeLists.txt
View File

@ -1,4 +1,4 @@
set(TARGET rpc-server) set(TARGET ggml-rpc-server)
add_executable(${TARGET} rpc-server.cpp) add_executable(${TARGET} rpc-server.cpp)
target_link_libraries(${TARGET} PRIVATE ggml) target_link_libraries(${TARGET} PRIVATE ggml)
target_compile_features(${TARGET} PRIVATE cxx_std_17) target_compile_features(${TARGET} PRIVATE cxx_std_17)

34
tools/rpc/README.md
View File

@ -4,8 +4,8 @@
> This example and the RPC backend are currently in a proof-of-concept development stage. As such, the functionality is fragile and > This example and the RPC backend are currently in a proof-of-concept development stage. As such, the functionality is fragile and
> insecure. **Never run the RPC server on an open network or in a sensitive environment!** > insecure. **Never run the RPC server on an open network or in a sensitive environment!**
The `rpc-server` allows exposing `ggml` devices on a remote host. The `ggml-rpc-server` allows exposing `ggml` devices on a remote host.
The RPC backend communicates with one or several instances of `rpc-server` and offloads computations to them. The RPC backend communicates with one or several instances of `ggml-rpc-server` and offloads computations to them.
This can be used for distributed LLM inference with `llama.cpp` in the following way: This can be used for distributed LLM inference with `llama.cpp` in the following way:
```mermaid ```mermaid
@ -14,15 +14,15 @@ flowchart TD
rpcb<-->|TCP|srvb rpcb<-->|TCP|srvb
rpcb<-.->|TCP|srvn rpcb<-.->|TCP|srvn
subgraph hostn[Host N] subgraph hostn[Host N]
srvn[rpc-server]<-.->dev4["CUDA0"] srvn[ggml-rpc-server]<-.->dev4["CUDA0"]
srvn[rpc-server]<-.->dev5["CPU"] srvn[ggml-rpc-server]<-.->dev5["CPU"]
end end
subgraph hostb[Host B] subgraph hostb[Host B]
srvb[rpc-server]<-->dev3["Metal"] srvb[ggml-rpc-server]<-->dev3["Metal"]
end end
subgraph hosta[Host A] subgraph hosta[Host A]
srva[rpc-server]<-->dev["CUDA0"] srva[ggml-rpc-server]<-->dev["CUDA0"]
srva[rpc-server]<-->dev2["CUDA1"] srva[ggml-rpc-server]<-->dev2["CUDA1"]
end end
subgraph host[Main Host] subgraph host[Main Host]
local["Local devices"]<-->ggml[llama-cli] local["Local devices"]<-->ggml[llama-cli]
@ -33,7 +33,7 @@ flowchart TD
class local,dev,dev2,dev3,dev4,dev5 devcls class local,dev,dev2,dev3,dev4,dev5 devcls
``` ```
By default, `rpc-server` exposes all available accelerator devices on the host. By default, `ggml-rpc-server` exposes all available accelerator devices on the host.
If there are no accelerators, it exposes a single `CPU` device. If there are no accelerators, it exposes a single `CPU` device.
## Usage ## Usage
@ -41,7 +41,7 @@ If there are no accelerators, it exposes a single `CPU` device.
### Remote hosts ### Remote hosts
On each remote host, build the backends for each accelerator by adding `-DGGML_RPC=ON` to the build options. On each remote host, build the backends for each accelerator by adding `-DGGML_RPC=ON` to the build options.
For example, to build the `rpc-server` with support for CUDA accelerators: For example, to build the `ggml-rpc-server` with support for CUDA accelerators:
```bash ```bash
mkdir build-rpc-cuda mkdir build-rpc-cuda
@ -50,10 +50,10 @@ cmake .. -DGGML_CUDA=ON -DGGML_RPC=ON
cmake --build . --config Release cmake --build . --config Release
``` ```
When started, the `rpc-server` will detect and expose all available `CUDA` devices: When started, the `ggml-rpc-server` will detect and expose all available `CUDA` devices:
```bash ```bash
$ bin/rpc-server $ bin/ggml-rpc-server
ggml_cuda_init: GGML_CUDA_FORCE_MMQ: no ggml_cuda_init: GGML_CUDA_FORCE_MMQ: no
ggml_cuda_init: GGML_CUDA_FORCE_CUBLAS: no ggml_cuda_init: GGML_CUDA_FORCE_CUBLAS: no
ggml_cuda_init: found 1 CUDA devices: ggml_cuda_init: found 1 CUDA devices:
@ -67,14 +67,14 @@ Devices:
You can control the set of exposed CUDA devices with the `CUDA_VISIBLE_DEVICES` environment variable or the `--device` command line option. The following two commands have the same effect: You can control the set of exposed CUDA devices with the `CUDA_VISIBLE_DEVICES` environment variable or the `--device` command line option. The following two commands have the same effect:
```bash ```bash
$ CUDA_VISIBLE_DEVICES=0 bin/rpc-server -p 50052 $ CUDA_VISIBLE_DEVICES=0 bin/ggml-rpc-server -p 50052
$ bin/rpc-server --device CUDA0 -p 50052 $ bin/ggml-rpc-server --device CUDA0 -p 50052
``` ```
### Main host ### Main host
On the main host build `llama.cpp` with the backends for the local devices and add `-DGGML_RPC=ON` to the build options. On the main host build `llama.cpp` with the backends for the local devices and add `-DGGML_RPC=ON` to the build options.
Finally, when running `llama-cli` or `llama-server`, use the `--rpc` option to specify the host and port of each `rpc-server`: Finally, when running `llama-cli` or `llama-server`, use the `--rpc` option to specify the host and port of each `ggml-rpc-server`:
```bash ```bash
$ llama-cli -hf ggml-org/gemma-3-1b-it-GGUF -ngl 99 --rpc 192.168.88.10:50052,192.168.88.11:50052 $ llama-cli -hf ggml-org/gemma-3-1b-it-GGUF -ngl 99 --rpc 192.168.88.10:50052,192.168.88.11:50052
@ -90,7 +90,7 @@ This can speed up model loading significantly, especially when using large model
To enable the cache, use the `-c` option: To enable the cache, use the `-c` option:
```bash ```bash
$ bin/rpc-server -c $ bin/ggml-rpc-server -c
``` ```
By default, the cache is stored in the `$HOME/.cache/llama.cpp/rpc` directory and can be controlled via the `LLAMA_CACHE` environment variable. By default, the cache is stored in the `$HOME/.cache/llama.cpp/rpc` directory and can be controlled via the `LLAMA_CACHE` environment variable.
@ -103,8 +103,8 @@ RDMA is enabled by default when `libibverbs` is found at build time.
### Troubleshooting ### Troubleshooting
Use the `GGML_RPC_DEBUG` environment variable to enable debug messages from `rpc-server`: Use the `GGML_RPC_DEBUG` environment variable to enable debug messages from `ggml-rpc-server`:
```bash ```bash
$ GGML_RPC_DEBUG=1 bin/rpc-server $ GGML_RPC_DEBUG=1 bin/ggml-rpc-server
``` ```