Skip to content
This repository was archived by the owner on Jul 4, 2025. It is now read-only.

Commit a2ce055

Browse files
hiro-vhiro-v
authored
fix: windows build with -DBUILD_SHARED_LIBS=OFF (#399)
1 parent 937ec23 commit a2ce055

File tree

5 files changed

+44
-53
lines changed

5 files changed

+44
-53
lines changed

‎.github/scripts/e2e-test-llama-windows.bat‎

Lines changed: 5 additions & 9 deletions
Original file line numberDiff line numberDiff line change
@@ -25,10 +25,6 @@ set /a max=11000
2525
set /a range=max-min+1
2626
set /a PORT=%min% + %RANDOM% %% %range%
2727

28-
rem Kill any existing Nitro processes
29-
echo Killing any existing Nitro processes...
30-
taskkill /f /im nitro.exe 2>nul
31-
3228
rem Start the binary file
3329
start /B "" "%BINARY_PATH%" 1 "127.0.0.1" %PORT% > %TEMP%\nitro.log 2>&1
3430

@@ -64,9 +60,9 @@ echo curl_data1=%curl_data1%
6460
echo curl_data2=%curl_data2%
6561

6662
rem Run the curl commands and capture the status code
67-
curl.exe -o "%TEMP%\response1.log" -s -w "%%{http_code}" --location "http://127.0.0.1:%PORT%/inferences/llamacpp/loadModel" --header "Content-Type: application/json" --data "%curl_data1%" > %TEMP%\response1.log 2>&1
63+
curl.exe --connect-timeout 60 -o "%TEMP%\response1.log" -s -w "%%{http_code}" --location "http://127.0.0.1:%PORT%/inferences/llamacpp/loadModel" --header "Content-Type: application/json" --data "%curl_data1%" > %TEMP%\response1.log 2>&1
6864

69-
curl.exe -o "%TEMP%\response2.log" -s -w "%%{http_code}" --location "http://127.0.0.1:%PORT%/inferences/llamacpp/chat_completion" ^
65+
curl.exe --connect-timeout 60 -o "%TEMP%\response2.log" -s -w "%%{http_code}" --location "http://127.0.0.1:%PORT%/inferences/llamacpp/chat_completion" ^
7066
--header "Content-Type: application/json" ^
7167
--header "Accept: text/event-stream" ^
7268
--header "Access-Control-Allow-Origin: *" ^
@@ -78,13 +74,13 @@ rem Read the status codes from the log files
7874
for /f %%a in (%TEMP%\response1.log) do set "response1=%%a"
7975
for /f %%a in (%TEMP%\response2.log) do set "response2=%%a"
8076

81-
if "%response1%" neq "000" (
77+
if "%response1%" neq "200" (
8278
echo The first curl command failed with status code: %response1%
8379
type %TEMP%\response1.log
8480
set "error_occurred=1"
8581
)
8682

87-
if "%response2%" neq "000" (
83+
if "%response2%" neq "200" (
8884
echo The second curl command failed with status code: %response2%
8985
type %TEMP%\response2.log
9086
set "error_occurred=1"
@@ -111,4 +107,4 @@ echo Nitro test run successfully!
111107

112108
rem Kill the server process
113109
@REM taskkill /f /pid %pid%
114-
taskkill /f /im nitro.exe 2>nul || exit /B 0
110+
taskkill /f /im nitro.exe 2>nul || exit /B 0

‎.github/scripts/e2e-test-whisper-windows.bat‎

Lines changed: 11 additions & 23 deletions
Original file line numberDiff line numberDiff line change
@@ -25,10 +25,6 @@ set /a max=11000
2525
set /a range=max-min+1
2626
set /a PORT=%min% + %RANDOM% %% %range%
2727

28-
rem Kill any existing Nitro processes
29-
echo Killing any existing Nitro processes...
30-
taskkill /f /im nitro.exe 2>nul
31-
3228
rem Start the binary file
3329
start /B "" "%BINARY_PATH%" 1 "127.0.0.1" %PORT% > %TEMP%\nitro.log 2>&1
3430

@@ -49,28 +45,21 @@ if not defined pid (
4945

5046
rem Wait for a few seconds to let the server start
5147

52-
rem Check if %TEMP%\testmodel exists, if not, download it
48+
rem Check if %TEMP%\testwhisper exists, if not, download it
5349
if not exist "%MODEL_PATH%" (
5450
bitsadmin.exe /transfer "DownloadTestModel" %DOWNLOAD_URL% "%MODEL_PATH%"
5551
)
5652

5753
rem Define JSON strings for curl data
5854
call set "MODEL_PATH_STRING=%%MODEL_PATH:\=\\%%"
59-
set "curl_data1={\"model_path\":\"%MODEL_PATH_STRING%\",\"model_id\":\"whisper.cpp\"}"
60-
61-
rem Print the values of curl_data1 for debugging
62-
echo curl_data1=%curl_data1%
55+
set "curl_data1={\"model_path\":\"%MODEL_PATH_STRING%\",\"model_id\":\"whisper\"}"
6356

6457
rem Run the curl commands and capture the status code
65-
curl.exe -o %TEMP%\response1_code.log -s -w "%%{http_code}" --location "http://127.0.0.1:%PORT%/v1/audio/load_model" --header "Content-Type: application/json" --data "%curl_data1%" > %TEMP%\response1_code.log 2>&1
58+
curl.exe -o %TEMP%\response1.log -s -w "%%{http_code}" --location "http://127.0.0.1:%PORT%/v1/audio/load_model" --header "Content-Type: application/json" --data "%curl_data1%" > %TEMP%\response1_code.log 2>&1
6659

67-
curl.exe -o %TEMP%\response2_code.log -s -w "%%{http_code}" --location "http://127.0.0.1:%PORT%/v1/audio/transcriptions" ^
68-
--header "Access-Control-Allow-Origin: *" ^
69-
--form 'model_id="whisper.cpp"' ^
70-
--form 'file=@"..\whisper.cpp\samples\jfk.wav"' ^
71-
--form 'temperature="0.0"' ^
72-
--form 'prompt="The transcript is about OpenAI which makes technology like DALL·E, GPT-3, and ChatGPT with the hope of one day building an AGI system that benefits all of humanity. The president is trying to raly people to support the cause."' ^
73-
> %TEMP%\response2_code.log 2>&1
60+
curl -o %TEMP%\response2.log -s -w "%%{http_code}" --location "http://localhost:%PORT%/v1/audio/transcriptions" ^
61+
--form "file=@../..//whisper.cpp/samples/jfk.wav" ^
62+
--form "model_id=whisper" > %TEMP%\response2_code.log 2>&1
7463

7564
set "error_occurred=0"
7665

@@ -80,13 +69,13 @@ for /f %%a in (%TEMP%\response2_code.log) do set "response2=%%a"
8069

8170
if "%response1%" neq "200" (
8271
echo The first curl command failed with status code: %response1%
83-
type %TEMP%\response1_code.log
72+
type %TEMP%\response1.log
8473
set "error_occurred=1"
8574
)
8675

87-
if "%response2%" neq "000" (
76+
if "%response2%" neq "200" (
8877
echo The second curl command failed with status code: %response2%
89-
type %TEMP%\response2_code.log
78+
type %TEMP%\response2.log
9079
set "error_occurred=1"
9180
)
9281

@@ -101,14 +90,13 @@ if "%error_occurred%"=="1" (
10190

10291
echo ----------------------
10392
echo Log load model:
104-
type %TEMP%\response1_code.log
93+
type %TEMP%\response1.log
10594

10695
echo ----------------------
10796
echo "Log run test:"
108-
type %TEMP%\response2_code.log
97+
type %TEMP%\response2.log
10998

11099
echo Nitro test run successfully!
111100

112101
rem Kill the server process
113-
@REM taskkill /f /pid %pid%
114102
taskkill /f /im nitro.exe 2>nul || exit /B 0

‎.github/workflows/build.yml‎

Lines changed: 20 additions & 16 deletions
Original file line numberDiff line numberDiff line change
@@ -50,7 +50,7 @@ on:
5050
env:
5151
BRANCH_NAME: ${{ github.head_ref || github.ref_name }}
5252
LLM_MODEL_URL: https://huggingface.co/TheBloke/TinyLlama-1.1B-Chat-v0.3-GGUF/resolve/main/tinyllama-1.1b-chat-v0.3.Q2_K.gguf
53-
WHISPER_MODEL_URL: https://huggingface.co/ggerganov/whisper.cpp/resolve/main/ggml-tiny.bin
53+
WHISPER_MODEL_URL: https://huggingface.co/ggerganov/whisper.cpp/resolve/main/ggml-tiny-q5_1.bin
5454

5555
jobs:
5656
create-draft-release:
@@ -377,11 +377,11 @@ jobs:
377377
run: |
378378
# To test with CoreML
379379
if [[ ! -f "/tmp/testwhisper-encoder.mlmodelc" ]]; then
380-
wget https://huggingface.co/ggerganov/whisper.cpp/resolve/main/ggml-tiny.en-encoder.mlmodelc.zip
381-
unzip ggml-tiny.en-encoder.mlmodelc.zip
382-
rm ggml-tiny.en-encoder.mlmodelc.zip
380+
wget https://huggingface.co/ggerganov/whisper.cpp/resolve/main/ggml-tiny-encoder.mlmodelc.zip
381+
unzip ggml-tiny-encoder.mlmodelc.zip
382+
rm ggml-tiny-encoder.mlmodelc.zip
383383
rm -rf /tmp/testwhisper-encoder.mlmodelc
384-
mv ggml-tiny.en-encoder.mlmodelc /tmp/testwhisper-encoder.mlmodelc
384+
mv ggml-tiny-encoder.mlmodelc /tmp/testwhisper-encoder.mlmodelc
385385
fi
386386
# run e2e testing
387387
cd nitro
@@ -586,7 +586,7 @@ jobs:
586586
cmake --build ./build_deps/nitro_deps --config Release
587587
mkdir -p build
588588
cd build
589-
cmake .. -DLLAMA_NATIVE=OFF -DLLAMA_BUILD_SERVER=ON -DLLAMA_BLAS=ON -DBUILD_SHARED_LIBS=ON -DCMAKE_BUILD_TYPE=RELEASE -DWHISPER_SDL2=ON -DNITRO_VERSION=${{ needs.set-nitro-version.outputs.version }}
589+
cmake .. -DLLAMA_NATIVE=OFF -DLLAMA_BUILD_SERVER=ON -DLLAMA_BLAS=ON -DBUILD_SHARED_LIBS=OFF -DCMAKE_BUILD_TYPE=RELEASE -DWHISPER_SDL2=ON -DNITRO_VERSION=${{ needs.set-nitro-version.outputs.version }}
590590
cmake --build . --config Release -j "%NUMBER_OF_PROCESSORS%"
591591
592592
- name: Pack artifacts
@@ -602,15 +602,19 @@ jobs:
602602
7z a -ttar temp.tar .\build\Release\*
603603
7z a -tgzip nitro.tar.gz temp.tar
604604
605-
# - name: Run e2e testing - Llama.cpp
606-
# shell: cmd
607-
# run: |
608-
# .\.github\scripts\e2e-test-llama-windows.bat .\build\Release\nitro.exe ${{ env.LLM_MODEL_URL }}
605+
- name: Run e2e testing - Llama.cpp
606+
shell: cmd
607+
run: |
608+
cd build\Release
609+
..\..\.github\scripts\e2e-test-llama-windows.bat nitro.exe ${{ env.LLM_MODEL_URL }}
610+
rmdir /S /Q .\build\Release\uploads
609611
610-
# - name: Run e2e testing - Whisper.cpp
611-
# shell: cmd
612-
# run: |
613-
# .\.github\scripts\e2e-test-whisper-windows.bat .\build\Release\nitro.exe ${{ env.WHISPER_MODEL_URL }}
612+
- name: Run e2e testing - Whisper.cpp
613+
shell: cmd
614+
run: |
615+
cd build\Release
616+
..\..\.github\scripts\e2e-test-whisper-windows.bat nitro.exe ${{ env.WHISPER_MODEL_URL }}
617+
rmdir /S /Q .\build\Release\uploads
614618
615619
- name: Upload Artifact
616620
uses: actions/upload-artifact@v2
@@ -679,7 +683,7 @@ jobs:
679683
cmake --build ./build_deps/nitro_deps --config Release
680684
mkdir -p build
681685
cd build
682-
cmake .. -DLLAMA_NATIVE=OFF -DLLAMA_BUILD_SERVER=ON -DLLAMA_VULKAN=ON -DBUILD_SHARED_LIBS=ON -DCMAKE_BUILD_TYPE=RELEASE -DWHISPER_SDL2=ON -DNITRO_VERSION=${{ needs.set-nitro-version.outputs.version }}
686+
cmake .. -DLLAMA_NATIVE=OFF -DLLAMA_BUILD_SERVER=ON -DLLAMA_VULKAN=ON -DBUILD_SHARED_LIBS=OFF -DCMAKE_BUILD_TYPE=RELEASE -DWHISPER_SDL2=ON -DNITRO_VERSION=${{ needs.set-nitro-version.outputs.version }}
683687
cmake --build . --config Release -j "%NUMBER_OF_PROCESSORS%"
684688
685689
- name: Pack artifacts
@@ -770,7 +774,7 @@ jobs:
770774
cmake --build ./build_deps/nitro_deps --config Release
771775
mkdir -p build
772776
cd build
773-
cmake .. -DLLAMA_NATIVE=OFF -DLLAMA_BUILD_SERVER=ON -DLLAMA_CUBLAS=ON -DBUILD_SHARED_LIBS=ON -DCMAKE_BUILD_TYPE=RELEASE -DWHISPER_SDL2=ON -DWHISPER_CUBLAS=ON -DNITRO_VERSION=${{ needs.set-nitro-version.outputs.version }}
777+
cmake .. -DLLAMA_NATIVE=OFF -DLLAMA_BUILD_SERVER=ON -DLLAMA_CUBLAS=ON -DBUILD_SHARED_LIBS=OFF -DCMAKE_BUILD_TYPE=RELEASE -DWHISPER_SDL2=ON -DWHISPER_CUBLAS=ON -DNITRO_VERSION=${{ needs.set-nitro-version.outputs.version }}
774778
cmake --build . --config Release -j "%NUMBER_OF_PROCESSORS%"
775779
776780
- name: Pack artifacts

‎CMakeLists.txt‎

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -59,7 +59,7 @@ add_executable(${PROJECT_NAME} main.cc)
5959
#
6060
# and comment out the following lines
6161
find_package(Drogon CONFIG REQUIRED)
62-
target_link_libraries(${PROJECT_NAME} PRIVATE Drogon::Drogon common llama llava whisper
62+
target_link_libraries(${PROJECT_NAME} PRIVATE Drogon::Drogon common llama whisper llava
6363
${CMAKE_THREAD_LIBS_INIT})
6464

6565
# ##############################################################################

‎controllers/whisperCPP.cc‎

Lines changed: 7 additions & 4 deletions
Original file line numberDiff line numberDiff line change
@@ -872,7 +872,9 @@ void whisperCPP::load_model(
872872
(*jsonBody)["warm_up_audio_path"].asString();
873873
// Return 400 error if warm up audio path is not found
874874
if (!is_file_exist(warm_up_audio_path.c_str())) {
875-
std::string error_msg = "Warm up audio " + warm_up_audio_path + " not found, please provide a valid path or don't specify it at all";
875+
std::string error_msg =
876+
"Warm up audio " + warm_up_audio_path +
877+
" not found, please provide a valid path or don't specify it at all";
876878
LOG_INFO << error_msg;
877879
Json::Value jsonResp;
878880
jsonResp["message"] = error_msg;
@@ -881,9 +883,10 @@ void whisperCPP::load_model(
881883
callback(resp);
882884
return;
883885
} else {
884-
LOG_INFO << "Warming up model " << model_id << " with audio " << warm_up_audio_path << " ...";
885-
std::string warm_up_result = whisper.inference(
886-
warm_up_audio_path, "en", "", text_format, 0, false);
886+
LOG_INFO << "Warming up model " << model_id << " with audio "
887+
<< warm_up_audio_path << " ...";
888+
std::string warm_up_result = whisper.inference(warm_up_audio_path, "en",
889+
"", text_format, 0, false);
887890
LOG_INFO << "Warm up model " << model_id << " completed";
888891
}
889892
} else {

0 commit comments

Comments
 (0)